Full text of Accidents and Accident Prevention in Machine Building. Revision of Bulletin 216 : Bulletin of the United States Bureau of Labor Statistics, No. 256

View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

U. S. DEPARTMENT OF LABOR

BUREAU OF LABOR STATISTICS
ROYAL MEEKER, Commissioner

BULLETIN OF THE UNITED STATES \
BU REAU OF LA B O R S T A T IS T IC S / • • • ’
IN D U STRIA L

A C C ID EN TS

AN D

H YG IEN E

j
| IN U .

L tJ D

SERIES

ACCIDENTS AND ACCIDENT PRE
VENTION IN MACHINE BUILDING

REVISION OF BULLETIN 216

LUCIAN W. CHA NEY

NO VEM BER, 1919

WASHINGTON
GOVERNMENT PRINTING OFFICE

1920

CONTENTS.
Page.

Summary............................................................................................................................. 7-14
Purpose of the investigation....................................................................................
7
Scope of the report....................................................................................................
7, 8
Accident severity rates: A method of measuring the seriousness of acci
dents .........................................................................................................................
8
Accident rates for the industry as a w hole..........................................................
9
Accident rates, b y character of product...............................................................
9
Accident rates, b y departments.............................................................................
9
Course of accident rates over a series of years..................................................... 9,10
Effect of safety systems upon accident occurrence............................................
10
Accident rates for a large machine-building establishment, by occupations. 10,11
Important causes of accident..................................................................................
11
Nature of in ju ry ......................................................................................................... 11,12
Inability to speak English as related to accidents.............................................
12
Accident rates higher at night................................................................................
12
Distribution of accidents by months.....................................................................
12
United States Government arsenals and navy yards........................................ 12,13
Methods of accident prevention............................................................................. 13,14
Chapter I.— Industrial accident rates........................................................................... 15-25
Definition of “ acciden t’ ’ ....................................................................................... 15,16
The meaning of accident rates................................................................................
16
Accident frequency rates......................................................................................... 16-18
Accident severity rates............................................................................................ 18,19
Fatalities.............................................................................................................. 19, 20
Permanent disabilities......................................................................................20, 21
Illustrations of the use of severity rates....................................................... 21-25
Chapter II .—Accident experience................................................................................ 27-78
Accident rates for 1912, by character of product................................................27-29
Accident rates for 1912, by departments............................................................. 29-33
Accident rates over a series of years..................................................................... 34-36
Fatal accidents in engine building................................................................
36
Course of accident rates, by departments............................................................ 37-40
Electrical assembly shops................................................................................
37
Forge shops..........................................................................................................
38
Foundries............................................................................................................. 38, 39
Machine shops.................................................................................................... 39, 40
Woodworking shops...........................................................................................
40
Effect of safety systems upon accident occurrence............................................40-42
Occupational accident rates.................................................................................... 42-46
Accident causes..........................................................................................................46-52
Accident causes, by plant groups and by departments............................ 46-48
Accident causes over a series of years...........................................................48-51
Necessity of rates for the measurement of accident causes................t . . 51, 52
Nature of in ju ry.........................................................................................................
53
Permanent results of injury.....................................................................................54, 55

CO N TE N TS.

Chapter I I .— A ccident experience— Concluded.
PageInability to speak English as related to accidents.............................................55-57
Day and night accident rates..................................................................................57-60
Distribution of accidents by months..................................................................... 61, 62
Government arsenals and navy yards................................................................... 62-68
Incomplete reporting b y Government shops...............................................65-68
Influence of the war on accident rates in machine building...........................68-78
Rates in munition plants................................................................................. 71, 72
Government plants............................................................................................ 72-74
Comparisons with the iron and steel industry............................................ 74-76
Cause of fluctuation in accident rates........................................................... 76-78
Chapter I I I .— Safety organization................................................................................. 79-84
The inspector.............................................................................................................. 79, 80
The safety com m ittee............................................................................................... 80, 81
Maintenance of interest............................................................................................ 81-83
Surgical care............................................................................................................... 83, 84
Chapter IY .— Direct safeguarding methods in machine building..........................85-97
Shop conditions in foundries.................................................................................. 85-87
Safeguarding in foundries........................................................................................ 87-91
Conditions in machine shops...................................................................................91-94
Safeguarding in machine shops.............................................................................. 94, 95
Safeguarding in electrical manufacture................................................................ 95, 96
Safeguarding in woodworking shops......................................................................
97
Chapter V .—Machine design as a factor of safety.................................................... 99-115
Machinery for the steel industry.......................................................................... 99-101
Cranes and hoists.................................................................................................... 101-104
Electrical apparatus.............................................................................................. 104-106
Locomotives............................................................................................................ 106,107
Other prime movers............................................................................................... 107,108
Machine tools.......................................................................................................... 108-113
Metal planers................................................................................................... 109,110
Boring m ills..................................................................................................... 110,111
Lathes............................................................................................................... 111,112
Drills.....................................................................................................................
112
Milling machines............................................................................................ 112,113
Machine tool accessories...........................................................................................
113
Transmission gearing............................................................................................. 113,114
Other products........................................................................................................ 114,115
Appendix A .— Results of accidents in 194 machine-building plants in 1912,
b y departments and by products........................................................................... 116,117
Appendix B .— Results of accidents in a machine-building plant, 1907 to
1913, b y occupations................................................................................................ 118,119
CHARTS.
Chart A .—Accident rates in iron and steel industry, 1905 to 1917....................
Chart B .— Frequency and severity of accidents in the machine-building in
dustry, 1912, classified b y products...........................................................................
Chart C.— Frequency and severity of accidents in the machine-building in
dustry, 1912, classified b y departments....................................................................
Chart D .— Inability to speak English as related to accidents................................
Chart E.— Night and day accident rates.....................................................................
Chart F.— Comparison of the trend of several related items by reduction to
index numbers, smoothing the curves and plotting on a percentage scale___

24
28
31
56
59
78

C O N TE N TS.

ILLUSTRATIONS.
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate
Plate

1.— Exterior of foundry.........................................................................................
2.— Interior of foundry..........................................................................................
3.—Machine for testing goggles...........................................................................
4.— Tumbler and sand-blasting machines, with exhausts..........................
5.— Belt-driven screw machines.........................................................................
6.— Lathes driven b y individual motors...........................................................
7.— Old type of testing switchboard.................................................................
8.— Standard testing switchboard.......................................................................
9.— Ripsaw guard—position when saw is used.......................... ....................
10.— Old style wire-drawing bench....................................................................
31.— Modern wire-drawing ben ch .......................................................................
12.— Crane trolley, unguarded............................................................................
13.— Crane trolley, guarded.................................................................................
14.— Power house, with reciprocating engine belted to generator..........
15.— Power house, with turbo-generators.........................................................
16.—Boring mill, with unguarded gears...........................................................
17.—Boring mill, with guarded gears................................................................
18.—-Large gun lathe, with unguarded gears................ : .................................
19.— Large gun lathe, with guarded gears........................................................
20.— Old type of drill, with unguarded gears..................................................
21.— New type of drill, with guarded gears....... ] ............................................
22.— Milling machine, with unguarded cutter................................................
23.—Milling machine, with guarded cutter.....................................................
24.—Belt shifting by hand...................................................................................
25.—Belt shifting with shifter.............................................................................

5
Facing
page —

86
88
90
91
94
94
95
95
97
100
100
102
102
104
104
110
110
110
110
112
112
112
112
112
112

BULLETIN OF THE

U. S. BUREAU OF LABOR STATISTICS.
WASHINGTON.

n o . 256.

N o v e m b e r , 1919.

ACCIDENTS AND ACCIDENT PREVENTION IN MACHINE
BUILDING.
SUMMARY.
PU RPOSE OF TH E IN V E S T IG A T IO N .

The purpose of this investigation was not only to ascertain the
frequency and severity of accidents in the machine-building in
dustry but also to study and analyze these accidents in such a way
as to supply the information necessary for effective safety work.
To this end the report seeks, as far as possible, to locate the accident
hazards in particular departments and occupations, to discover the
reason and causes for the occurrence of accidents, and to point out
some of the more successful methods for their prevention.
SCOPE OF THE REPORT.

The machine-building industry covers the production of a very
large variety of machinery and machine appliances. The number of
plants engaged in the various branches of the industry is extremely
large. It was, therefore, necessary to limit the investigation to cer
tain selected plants, representing, as fairly as could be determined,
the more important classes of product. In all, the investigation
covered 194 plants. These plants worked 347,109,000 man hours
in 1912, which is equivalent to 115,703 full-time or 300-day workers.
The 300-day worker, or full-time worker, as defined by the joint
committee of the International Congress on Social Insurance and the
International Institute of Statistics, is one who works 300 days a
year, 10 hours per day— 3,000 hours per annum.1
For each of the 194 plants full data relating to accidents were ob
tained for the year 1912. In addition, similar data for a series of
years were obtained from such of these plants as had the necessary
records in available form.
All accidents causing loss of time beyond the day on which the
accident occurred are included in the statistics in this report. The
frequency of accident occurrences is expressed in rates. These rates
give the number of accidents per thousand workers employed 300

For fuller discussion of this, see p. 17.

ACCID ENTS IN

M A C H IN E B U IL D IN G .

days of 10 hours each. This is the same method of presentation
used in the report on accidents in the iron and steel industry.1
A CCIDENT S E V E R IT Y R A T E S : A METHOD OF M E A S U R IN G THE
SE R IO U SN ESS OF A CC ID E N T S.2

In this report for the first time an attempt is made to show the
seriousness of accidents by what has been called “ severity rates.”
The meaning of this term may be best expressed by means of an
example: Assume that a plant employing 1,000 300-day workers
during the course of a year had 200 accidents, and that the total
time lost by the men injured was 5,000 working-days; the accident
frequency rate for the year would be 200 per 1,000 workers; the
“ severity” rate would be 5,000 days lost per 1,000 workers, or, more
conveniently expressed, an average of 5 days per individual worker.
To make such computations it is necessary, of course, to express
fatal and permanent injuries, as well as temporary disabilities, in
terms of workdays lost. This is done by valuing a fatal injury
(assuming the employee killed of an average age of 30) as equiv
alent to the loss of 20 years’ work time— 6,000 days.3 Other in
juries— such as loss of hand or foot— are credited with lower time
losses, in proportion to their probable effect upon earning capacity—
3.000 days for a hand, 2,400 days for a foot, etc. This method of
evaluating permanent injury in terms of time loss, although based
upon somewhat rough estimates, is by no means arbitrary.
Severity rates, thus computed, constitute a very much more
accurate measure of accident hazard than do the older frequency
rates. A striking example may be cited: The machine-building
industry, in one year, had an accident frequency rate of 118 per
1.000 300-day workers. This was, as it happened, actually higher
than the accident frequency in a large steel plant in one year, the
rate there being 114.5 cases per 1,000 workers. But even a casual
acquaintance with the two industries would indicate that the steel
plant represented the more hazardous employment, inasmuch as its
accidents are, on the whole, of a more serious character than those
occurring in machine building. This was evident when severity
rates were computed according to the method described, the steel
plant having a severity rate of 15.7 days lost per full-time or 300-day
worker, as against only 5.1 days lost per worker in machine building.
In this case the severity rate is clearly more valuable than, the fre
quency rate in indicating the relative hazards of the two industries.
1 Conditions of Em ploym ent in the Iron and Steel Industry in the United States (S. Doc. No. 110, 62d
Cong., 1st sess.), Vol. IV.
2 See Ch. I for a full discussion of meaning and importance of “ severity ” rates.
3 For full explanation of the method followed, see p. 18.

SUM M ARY.

ACCIDENT R A T E S FOR THE IN D U S T R Y A S A W H O LE.1

The number of accidents occurring during the year 1912 was 13,647,
resulting in 37 deaths, 411 permanent injuries, and 13,199 temporary
disabilities. This is equivalent to an accident frequency rate of 118
per 1,000 full-time workers and a severity rate of 5.1 days lost per
worker. These rates may be contrasted with the experience of a rep
resentative steel plant during the same year, for which the frequency
rate was 153.5 and the severity rate 13.3 days lost. Accidents in the
steel plant were thus only about one-third more frequent than in
machine building, but their severity was more than twice as great.
ACCIDENT R A TE S, B Y C H A R A C TE R OF PRODUCT.1

The accident hazards of the machine-building plants vary greatly
with the character of their products. Those engaged in the making
of locomotives have the highest severity rate— 8.5 days lost per
worker— and the builders of ships have the next highest— 6.7 days
lost per worker. The severity rates for the other classified groups are
as follows: Machines for the steel industry, 6.6 days; cranes and
hoists, 3.4 days; electrical apparatus, 2.8 days; power-transmission
machinery, 3.2 days; mining machinery, 3 days; and machine tools,
2 days.
ACCIDENT R A TE S, B Y D E P A R T M E N T S.2

Classifying the combined plants by departmental divisions, boiler
shops and yard labor show by far the greatest hazards. Boiler shops
have a frequency rate of 224.1 per 1,000 workers and a severity rate of
20.7 days lost per worker, while yard labor has a frequency rate of
221.1 and a severity rate of 21.6 days lost. These rates are, roughly,
as high as those in the iron and steel industry, which is recognized as
inherently a much more hazardous industry than machine building.
The high rates of the boiler shops are, primarily, the result of insecure
trestles and scaffolding. For the excessive rates in the yard depart
ment responsibility rests upon the general neglect of safety precau
tions in the transportation work of many plants.
Other important departments show the following severity rates:
Power, 15.3 days lost; forge shops, 10.9 days; erecting shops, 7.8
days; iron foundries, 5.4 days; woodworking, 5 days; machine shops,
3.4 days; electric shops, 3 days.
COURSE OF ACCID ENT R A T E S O V E R A SE R IE S OF Y E A R S .3

One of the fundamental inquiries in a study of this character is
whether or not accidents are decreasing. A precise answer is diffi
cult, because of the fact that very few plants had reliable accident
records over a period of years. Such information was obtainable for
i See p. 27 et seq.

See p. 29 et seq.

s See p. 34 et seq.

ACCIDENTS IN

M A C H IN E B U IL D IN G .

two groups of plants. The first group covers the years 1907 to 1912.
No significant change in accident occurrence is observable. The
second group covers the years 1910 to 1913. The frequency rate
shows no decrease, but the severity rate, after running as high as 5.2
days in 1910, 6.2 days in 1911, and 5.5 days in 1912, drops to 2.7 days
in 1913. This decrease undoubtedly reflects the more thorough
safety organization effected in some of these plants in 1912. The fact
that the frequency rate shows no decline is certainly due to the more
complete reporting of minor accidents in the later years.
A study of the course of accident rates in individual departments is
offered, but it is not very conclusive, as in some cases the records were
available for only a very limited period and in others the number of
employees was too small for the drawing of conclusions.
EFFECT OF S A F E T Y SY ST E M S UPON ACCID ENT OCCURRENCE.1

A striking method of showing the effect of a good safety system in
accident prevention is to compare the accident rates in plants having,
with those in plants not having, well-organized systems. This is
done for three important groups of plants. In every case the plants
not having a good safety organization show accident frequency rates
three or four times as high as those having a well-developed system.
ACCIDENT R A T E S FO R A L A R G E M A C H IN E -B U ILD IN G E S T A B L IS H
M E N T, B Y OCCUPATIONS.2

The study of accident rates by occupations is of particular impor
tance from the standpoint of safety work. It is by locating accident
hazards in particular occupational groups that precise knowledge is
gained as to the proper place for applying preventive measures. The
great handicap to such a study is the difficulty of determining the
number of 300-day workers in the individual occupation, and this
information is necessary for the proper computation of rates. This
difficulty was overcome for one large machine-building concern and a
very interesting group of occupational rates was obtained.
The highest severity rate among the occupations listed in this estab
lishment is that of cranemen— 19.6 days— due to an exceedingly large
number of fatal injuries. This fatality hazard is attributable to the
faulty construction of cranes. Since this investigation cranes have
been greatly modified by the safety features incorporated in their
original design.
Common labor has a high severity rate— 13.1 days. This group
is so large that a high rate among them is very significant. There is
no one particular preventive measure by which the rate can be
reduced. There is demanded a general application of all possible
remedial measures.

’See p. 40 et seq.

2 See p. 42 et seq.

SU M M A R Y .

Machine hands show the high rate of 11.6 days lost per worker.
As the group is small this may not be a normal rate, but, in some
measure, it certainly reflects the fact that a good many automatic
machines have been constructed with their moving parts needlessly
exposed.
IM POR TAN T C A U SE S OF A C C ID E N T .1

The analysis of accident causes, together with the determination
of occupational rates, is at present the most important practical sub
ject to be considered in accident studies. A careful study of accident
causes was made in selected groups of machine-building plants and is
presented in Tables 16 and 17.
For the industry as a whole “ falling objects” stands out as the
most frequent cause of accidents, the frequency rate for 5 machinebuilding plants from 1907 to 1912 being 14.44 cases per 1,000 300day workers, and for 4 machine-building plants, 1910 to 1913, 14.35
cases per 1,000 300-day workers. As measured by severity “ cranes
and hoists” assumes first place, the severity rate being 1.75 days lost
per 300-day worker in the group of 5 plants, 1907 to 1912, and 0.98
day lost per 300-day worker in the group of 4 plants, 1910 to 1913.2
In foundries “ hot metal” appears as the accident cause with most
serious effects, the severity rate being 2.32 days lost per worker out
of a total of 6.10 days lost for all foundry causes. In the more pro
gressive foundries provision of proper shoes, leggings, and eye
protectors has nearly eliminated many of the dangers of handling
hot metal.
In machine shops 1 1 operating machines ” is responsible for a severity
rate of 1.18 days out of 2.82 days for all causes; in electrical assembly
shops “ operating machines” is responsible for a severity rate of 0.81
day out of a total of 1.98 days for all causes.
Comparing accident causes in machine building with those in
steel making, it develops that, for the most part, the accident causes
are of similar importance in both industries. Thus “ falling and fly
ing objects” is by far the most frequent cause of injury in both
steel making and machine building. In both industries, also, cranes
and hoists are fertile causes of accidents, with severe resulting injuries.
“ Using tools,” on the other hand, is the cause of a considerable per
centage of accidents, but the resulting severity is inconsiderable.
N A T U R E OF IN J U R Y .3

Nature of injury is of much less importance from the standpoint
of accident prevention than is cause of injury. Perhaps the most
interesting point in this connection is the comparison of the char
acter of the injuries occurring in machine building with those occur
i See p. 46 et seq.

See p. 48 et seq.

a See p. 63 et seq.

ACCIDENTS IN

M A C H IN E B U IL D IN G .

ring in steel making. Injuries by bruising and lacerating the hands
and fingers are much more important both as to frequency and
severity in machine building than in steel making, as would be
expected from the character of the processes carried on. Also, eye
injuries are much more important both as to frequency and severity
in machine building. On the other hand, bums stand out promi
nently in the steel industry. In other respects there is a remarkable
similarity between the two industries.
IN A B IL IT Y TO SP E A K E N G LISH A S R E L A T E D TO A CC ID E NTS.1

It was not possible in the plants covered to separate the employees
into English and non-English speakers. For one large machinebuilding plant, however, separation was possible between the Ameri
can born and foreign born. The foreign born showed an accident
rate approximately double that for the native born. This excess
rate among foreign born is clearly attributable to the same causes
which lead to a constant excess among non-English speaking steel
workers— partly to their failure to understand clearly the orders given
them, and partly to the fact that the recent immigrant suffers from
lack of experience, and thus falls largely into the group of unskilled
occupations involving exposure to inherently high accident hazards.
ACCIDENT R A T E S H IG H E R AT N IG H T.2

The compilation of the accident experience of a large machinebuilding plant which worked both day and night developed that the
accident frequency rate was 50 per cent higher among night workers
than among day workers. This is in keeping with the experience
of the iron and steel industry.
D ISTR IB U T IO N OF A CC ID ENTS B Y M ONTHS.3

An analysis of the monthly distribution of accidents in three
important plants shows the highest frequency rates in the months
of August and September. This would indicate that the depressing
influence of summer heat may be a factor in accident hazard and
suggests the value of proper ventilating apparatus as an accident
preventive measure as well as a needed contribution to the comfort
of the workers.
U N IT E D STA TES G O V E R N M E N T A R S E N A L S A N D N A V Y Y A R D S .4

The navy yards and arsenals of the United States Government are
machine-building plants. But the discussion of the accident rates
therein is separated from that for private plants because of the fact
that accident reports from the Government shops for the period cov
ered by this study were extremely incomplete for short-time disa
i See p. 55 et seq.

See p. 57 et seq.

3 See p. 61 et seq.

See p. 62 et seq.

SU M M A R Y .

bilities. To make fair comparisons between the rates for the two
groups of plants it is therefore necessary to exclude all disabilities
of less than two weeks’ duration. This requires recomputation of
the rates previously considered, which are based on the inclusion of
all disabilities of over a day’s duration.
The most obvious comparison is that between Government navy
yards and private shipbuilding plants. The Government yards show
the lower frequency rate, 74.8 cases per 1,000 workers, as against 99.4
cases in private plants, but, on the other hand, show' the higher
severity rate, 9.7 days lost per worker as against 6 days in private
plants.
The Government arsenals, because of their varied activities, may
perhaps be compared with fairness with the entire machine-building
industry. When this is done, the arsenals show a higher frequency
rate— 51.7 cases against 32.2 cases per 1,000 workers— and also a
slightly higher severity rate— 5.4 against 4.7 days lost per worker.
This somewhat unsatisfactory showing of the Government shops
calls for careful consideration. Certain factors in their conduct im
press the outside observer as favorable to low accident rates—
extreme orderliness and cleanliness, freedom from violent fluctuations
in employment, the quality and stability of the working force, and the
eight-hour day. Certain other factors, existing at least in 1912,
would seem to have a contrary effect— imperfect mechanical safe
guarding, lack of safety organization, and an honest conviction on
the part of the supervising authorities that such accidents as occurred
were without remedy.
Comparing the accident rates of the Government arsenals with
those of the Government navy yards, it is found that the navy yards
have much the higher rates. This is in accordance with the known
character of the relative hazards of the work done.
METHODS OF ACCIDENT P R E V E N T IO N .

Experience has everywhere shown that the most effective work for
the prevention of accidents must come from a proper safety organiza
tion within the plant itself.1 Such an organization involves some
form of a safety committee system, with representatives of both em
ployer and employees working together to develop the best safety
methods, not only in the field of mechanical safeguards, but also in the
education of the employees in the observance of proper precautions
and the maintenance of the safety spirit. It is important to note, in
this connection, that the existence of compensation laws in most of
the States now furnishes an economic incentive for accident reduc
tion, which was so often absent under the old liability system.
* For a fuller discussion of the subject of safety organization see Ch. III.

ACCIDENTS IN

M A C H IN E B U IL D IN G .

The plant safety organization, however, does not itself do away
with the need of mechanical safeguards. It is rather an assurance
that the proper safeguards will be adopted and will be properly used.
For practically all of the dangers attending the use of the machinery
and processes in machine building, excellent safeguards have been
devised and are in use in certain plants.1
In discussing the question of safety in the machine-building indus
try it is important to remember that that industry not only uses ma
chinery which needs to be safeguarded, but that its work consists of
the production of machines for use in other industries. The extent
to which the machines thus manufactured will later be a source of
danger to the workers in those other industries depends in consider
able measure upon the character of their original construction. The
subject of machine design— of building a machine in such a way as
to offer the minimum of hazard to its future operators— thus becomes
of very great significance. This subject is discussed in Chapter Y.

i For an illustrated description of these safeguards see Ch. IV .

CHAPTER I.—INDUSTRIAL ACCIDENT RATES.
The purpose of accident studies is the very practical one of finding
out where and why accidents occur and how they may be prevented.
The first stage in every such study is necessarily the counting and
analysis of the accidents reported. In attempting this two serious
difficulties present themselves: First, the lack of a uniform definition
of what is to be regarded as an “ accident” ; and, second, a confusion
as to the proper determination and use of accident rates. Failure to
grasp the importance of these two points has been responsible for
much loose thinking and many false conclusions, and has also been
responsible for the present unsatisfactory character of accident
statistics in this country.
DE FIN IT IO N OF “ A C C ID E N T .”

First, then, what is to be regarded as an industrial accident for
the purposes of statistical study? No definition has as yet been
universally accepted. Some establishments and States attempt to
take account of all injuries, however trivial. Others exclude those
of a minor character and take account only of such as cause a loss of
a definite amount of time. It is evident that the accident showing
of a plant may be completely altered by a change in definition of
accident, and that in the absence of a uniform definition all compari
sons of the accident data of different plants, industries, or other
groups become almost worthless. The precise definition is not so
important. The important thing is that the same definition should
be everywhere observed.
The most significant step so far taken toward such uniformity in
this country is the action of the International Association of Indus
trial Accident Boards and Commissions in adopting a definition of
“ tabulatable accidents” — i. e., a definition not necessarily to be
followed in the original reporting of accidents, but to be used in all
statistical tabulations. The definition is substantially the same as
the one long used by the Bureau of Labor Statistics in its accident
investigations and employed in the present report:
“ Tabulatable accidents, diseases, and injuries.—All accidents,
diseases, and injuries arising out of employment and resulting in
death, permanent disability, or any loss of time other than the
remainder of the day, shift, or turn in which the injury was incurred,
shall be classified as ‘ tabulatable accidents, diseases, and injuries/
and a report of all such cases to some State or National authority
shall be required.”
124566°— 20— Bull. 256------ 2

ACCIDENTS IN

M A C H IN E B U IL D IN G .

The States which belong to the International Association of Indus
trial Accident Boards and Commissions are thus committed to a
uniform standard definition of the accidents which are to be tabu
lated. Some States may at first find it impossible to tabulate all
accidents as required by the definition, but the desirability of doing
so is apparent and many have already made a beginning.
THE M E A N IN G OF ACCIDENT R A TE S.

The second of the two above-mentioned difficulties— the deter
mination and use of accurate accident rates— presents a more serious
problem than that involved in the definition of accident. Here it is
necessary not only to have uniformity, but to decide upon a correct
method. In the early attempts at accident statistics, attention was
limited to the number of accidents occurring in a given plant or
group. But mere numbers, of course, mean nothing unless related
to the number of persons exposed to accident. This led to the cus
tom of expressing accidents in terms of so many per 1,000 workers,
and constituted an approach to a correct method. To say that a
given industry had an accident rate of 100 per 1,000 workers does
convey a definite idea, and can be compared with a rate of, say,
300 per 1,000 workers in another industry. But the method was
extremely crude, because the basic figure “ 1,000 workers” was
indefinite and variable. Usually it was derived by rough estimate
as to the number of persons employed, such as averaging the number
employed at different times of the year or averaging the pay rolls of
the year. But no such average could be at all an accurate measure
of what was wanted. The number of days worked varies in different
plants, as do also the daily hours of labor. Two plants may have the
same yearly accident rate— say, 200 per “ 1,000 workers,” estimated
on the above basis— but if one worked only 8 hours a day for 250
days and the other worked 12 hours a day for 365 days, it is clear
that the real accident hazard is much higher in the former plant,
inasmuch as the same number of accidents per 1,000 workers occurred
during a much more limited period of time.
ACCIDENT FR E Q U E N C Y R A T E S.

From this weakness, it became evident that in order to get a rate
that would measure real hazard, it is necessary to know not only the
number of men employed, but also the time of their employment.
The only way to obtain this is to ascertain the actual number of
hours worked by all employees for the year. This gives the number
of man-hours, i. e., the theoretical number of men required to pro
duce the output of the plant in one hour, or, what is the same thing,
the theoretical number of hours required by one man to turn out
the same product. Man-hours so derived constitute the correct basis

C H A PT E R I .---- IN D U ST R IA L AC C ID E N T RATES.

upon which to calculate accident rates. But the term is unfamiliar
and for practical purposes it is convenient to convert man-hours
into full-time workers. The full-time worker, as defined by the
joint committee of the International Congress on Social Insurance
and the International Institute of Statistics, is one who works 10
hours per day for 300 days per annum, making a total of 3,000 hours
per annum.
Thus, if a plant having 1,000 machines, each requiring one man
to operate, worked 10 hours a day for 300 days, the total number of
man-hours worked would be 3,000,000 per year (i. e., 1,000 X 10 X 300),
and the number of full-time workers would be 1,000, although indi
vidual employees may have changed many times during the year.
Another plant having exactly the same equipment, but working
only 9 hours a day for only 200 days would have a total of only
1,800,000 man-hours (1,000X9X200), the equivalent of only 600
full-time workers (1,800,000^3,000). By thus reducing the em
ployment in the two plants to a common unit, accident hazard may
be accurately compared. Thus, if each of these plants had 200 acci
dents during the course of the year, the rate for the first plant would
be 200 per 1,000 full-time workers, while for the latter it would be
333 per 1,000 full-time workers (i. e., 200 X ^ f ) and the relation
200 to 333 would correctly express the relative accident frequency
in the two plants.
The full-time worker or 300-day worker, so defined, may seem at
first thought to be a mere statistical abstraction. It is true that
the full-time worker, like the average man, is a unit of measure,
not a living, breathing man, but for the purpose of accident statis
tics a standardized workman to serve as a unit of measure is abso
lutely essential. Furthermore, the statistical full-time workman
who is assumed to work 10 hours a day for 300 days in the year
conforms very closely in most industries to the actual workman who
enjoys good health and works every day the establishment is running.
Accident statistics, to be comparable, must be stated in terms of a
common unit of measure. The 300-day worker is merely a unit of
measure of the quantity of labor, just as the yard is the unit of
measure for length. The number of 300-day or full-time workers is
obtained by dividing the number of man-hours actually worked in
an establishment by 3,000, the number of hours per annum assumed
to be worked by the 300-day worker.
In those establishments which keep accurate records of the hours
worked by each employee every day, the man-hours worked by the
establishment can easily be obtained from the records and hence the
number of full-time or 300-day workers can easily be computed.
Few small establishments, however, keep any such accurate records of

ACCIDENTS IN

M A C H IN E B U IL D IN G .

time worked. For the majority of small plants it is necessary to
compute the number of man-hours worked and the full-time (300day) workers. The method suggested by the conference called by
Commissioner Meeker, which met in Chicago October 12 and 13, 1914,
was as follows: “ If this exact information is not available in this form
in the records, then an approximation should be computed by taking
the number of men at work (or enrolled) on a certain day of each
month in the year and the average of these numbers multiplied by
the number of hours worked by the establishment for the year would
be the number of man-hours measuring the exposure to risk for the
year.”
This resolution has not been adopted by the committee on statistics
and compensation insurance cost of the International Association of
Industrial Accident Boards and Commissions, but the necessity of
reaching an approximation to the man-hours worked in establish
ments which keep no accurate records has been tacitly accepted.
The establishments covered by this study kept accurate time
records, so tihe man-hours were taken from these records.
By the method outlined, true rates are obtained as regards the
risk of accident occurrence or frequency. These rates may be called
accident frequency rates. Thus if the accident frequency rate, so
derived, for the steel industry is 114 per 1,000 full-time workers, and
is 118 for the machine-building industry, it is correct to conclude that
accidents are less frequent in the steel industry than in machine
building, in the proportion of 114 to 118. All differences in the
hours of labor, number of days worked, etc., in the two industries
have been duly taken into account. Again, if a given plant shows
an accident frequency rate of 100 one year and 90 the next, it is a
correct conclusion that accidents have decreased 10 per cent in
frequency.
A CCIDENT S E V E R IT Y R A TE S.

Frequency rates of this character were computed and used in the
Report on Accidents in the Iron and Steel Industry, issued by the
Bureau of Labor Statistics in 1913. In all the establishments covered
the number of man-hours worked per year was obtained and the work
ing force then reduced to so many full-time or 300-day workers.
The method was found practicable and, within limits, highly useful.
But it had one serious weakness, namely, that frequency rates, as
the name indicates, measure the frequency of accidents, but take no
account of the severity of the resulting injuries, and experience has
shown that the two things do not necessarily move in the same direc
tion. The frequency rates may be the same in two plants in the same
industry, and the hazards may be entirely different because one plant
has very few severe accidents, while the other has a large proportion
of serious accidents. To put all industries and all plants on a com

C H A PT E R I .---- IN D U ST R IA L A C CID EN T RATES.

mon basis a system of computing accident rates must be devised
which will take into account the difference in economic significance
between the accident which bruises the workman’s thumb and the
accident which breaks his back.
In other words, what is needed is some method of weighting inju
ries according to their severity. Several methods suggest themselves
as possible— compensation paid, wage loss, or time loss. A compensa
tion system necessarily weights the importance of accidents in fixing
a scale of benefits which aims to apportion the payment to the hurt.
But compensation payments do not offer the universal measure
desired because the benefits differ from State to State and are also
subject to change within the same State. Wage loss due to injury
offers perhaps a better measure of severity, but this, too, suffers under
the handicap that wages differ from place to place and from time to
time. Time loss as a measure does not suffer from these objections.
An accident that causes 6 days' disability is precisely twice as serious
as one causing only 3 days' disability, and this relation is always and
everywhere the same.
The days lost because of injury may thus be taken as the most
satisfactory measure of the true hazards of industry— of the burden
imposed upon the worker and the community because of industrial
accidents. The only difficulty in its practical application is that in
case of death and permanent injuries the time lost must be estimated.
For temporary disabilities, from which recovery is complete, the time
losses are matters of record— 2 days, 10 days, 6 weeks, as the case
may be. But, if the accident results in death, the time loss is not so
clearly measurable. It exists, however, and may be estimated as the
number of working days by which the worker's life was curtailed.
Similar estimates are possible in case of permanent injuries, such as
loss of hand or foot.
F ATA L IT IE S.

In case of an injury causing death the time loss to the family and
society is the expectancy of productive working life of the deceased
workman. It is not possible to learn the age of all workmen killed in
industrial accidents; but from estimates made by the Wisconsin
Industrial Commission, from, statistics obtained by several compensa
tion commissions, and from the investigations of the Bureau of Labor
Statistics, it seems reasonable to estimate that the average age of vic
tims of fatal accidents is approximately 30 years. According to the
American life tables, the life expectancy at age 30 is 35 years. This
is for the population as a whole. Workingmen exposed to all the
hazards of illness and accident in industry have a shorter expectancy
of life than the average for the whole population. The expected pro
ductive life of workers is even shorter than their life expectancy.
Exact data are lacking, but in the light of all obtainable information

ACCIDENTS IN

M A C H IN E B U IL D IN G .

it seems fair to estimate the working time lost on the average by
relatives and the community for each workman killed by accident as
20 years, or 6,000 working days, counting 300 working days to the
year. This is admittedly an estimate. A mathematically accurate
measure is obviously impossible. It is also unimportant. The main
thing is to get the best possible approximation and to apply it to
existing accident statistics for the purpose of comparing accident
records plant by plant, industry by industry, and year by year.
PER M AN EN T DISAB ILITIES.

The system of time allowances used in the present report is the
result of a process of evolution which may properly be summarily
stated here. In the volume on Accidents and Accident Prevention
of the series Conditions of Employment in the Iron and Steel Indus
try there is, beginning at page 124, an extended study of severity
rates due to temporary disability. In 1914 the author of this report
prepared charts for the Panama-Pacific Exposition showing severity
rates based on the time allowances of the Wisconsin Industrial Com
mission. Later there was worked out, at the suggestion of Commis
sioner Meeker, the system based on life expectancy used in Bulletin
216 on Machine Building and Bulletin 234 on the Safety Movement
in the Iron and Steel Industry. Still later, Commissioner Meeker
brought about the formation of the Committee on Statistics of the
International Association of Industrial Accident Boards and Com
missions. After prolonged study and consultation this committee
proposed the scale of allowances published in Bulletin 234, page
278, and, with the committee's comment and explanation, in Bulletin
248, page 115. This scale, to which the Bureau is conforming its
publications as rapidly as possible, is as follows:
SCALE

OF TIME

LOSSES FOR W EIG H TIN G IN D U ST R IA L ACCIDENTS
SHOW SE V E R IT Y OF ACCIDENTS.

R esult of injury.
D ea th ....................................
Permanent total disability.
Loss of arm...........................
Loss of han d.........................
Loss of leg.............................
Loss of foot............................
Loss of e y e............................
Loss of thum b.......................
Loss of 1 finger......................
Loss of 2 fingers....................
Loss of 3 fingers....................
Loss of 4 fingers....................
Loss of thumb and 1 finger.
Loss of thum b and 2 fingers
Loss of thum b and 3 fingers
Loss of thum b and 4 fingers
Loss of great to e ...................
Loss of any tw o toes............

Days
allowed.

6,000

4.000
3.000
4.000
2.400
1,800
600
300
750

1,200

1,800

1,200

1,500

2.000

2.400
300
300

SO A S TO

C H A PT E R I .— IN D U ST R IA L A C CID EN T RATES.

This schedule supplies a series of constants by which death and
permanent injuries may be weighted in terms of a common un,it—
time lost in days—which is also the same unit as that used for meas
uring temporary disabilities. Multiplying the number of deaths
and permanent disabilities by the time loss determined for each and
adding the products to the days lost through temporary disabilities,
a figure is obtained which represents the total days lost from injuries.
Dividing this number, representing total days lost, by the number
of full-time workers gives as a quotient the average number of days
lost per full-time worker. This last figure may be called the acci
dent severity rate, since it shows the burdensomeness or seriousness of
the accidents analyzed.
The whole process of working out the accident severity rate may
be illustrated as follows: Plant A operated 4,200,000 man-hours in
1915, requiring 1,400 full-time (300-day, 10-hour-per-day) workers.
During the year 324 accidents occurred, resulting in 1 death and
the loss of the following members: 2 arms, 1 foot, 5 thumbs, 25 first
fingers, while the 290 temporary disabilities showed a time loss of
2,790 days. Applying the time losses in the above scale to these data
the following results are obtained:
T able

1 — TIME ALLOW ANCES IN ONE PLA N T.
Time allowances (in
days).
Item.
Per case.
6,000
4,000
2,400
600
300

1 death......................................
2 arms.......................................
1 foot.........................................
5 thum bs..................................
25 fingers...................................
290 temporary disabilities
Total............... ..................

Total.
6,000
8,000
2,400
3,000
7,500
2,790
29,690

The total number of days lost, 29,690, divided by the number of
full-time workers, 1,400, gives an average of 21 days per full-time
worker. This is what is here called the accident severity rate,
expressed in terms of days. The accident frequency rate for the same
group per 1,000 full-time 300-day workers would be 3 2 4 = 231.
ILLU STR ATIO N S OF TH E USE OF S E V E R IT Y RATES.

The preceding paragraphs explain the meaning of accident severity
rates and the method by which they are obtained. The significance
of such rates in their practical application is indicated in the two
following illustrations.
In the table following comparison is made of the accident experi
ence for a year of the iron and steel industry as represented by a
large plant, and of the machine-building industry as represented

ACCIDEN TS IN

by a group of plants.
in parallel columns.
T able

M A C H IN E B U IL D IN G .

Frequency rates and severity rates are shown

2 .—ACCIDENT RA TES IN ST E E L M ANUFACTURE A N D MACHINE B U IL D IN G .
Accident severity
rates (days lost per
300-day worker).

Accident frequency
rates (per 1,000
300-day workers).

Number of cases.

N um
ber of
300Per Tem
Per Tem
dav
Per Tem
po
po
ma rary
ma rary
ma po To
work
To
To Death. nent
Death.
nent
ers. Death. nent rary
dis
tal.
tal.
dis
dis
in abil tal.
in abil
in
jury.
jury. abil
jury- ity.
ity.
ity.

Industry.

Iron and steel (1913)... 7,562
M a c h in e b u i l d
ing (1912).................. 115,703

14
37

866

1.9

4.6 108.0 114.5

11.1

2.2

2.4 15.7

411 13,199 13,647

3.6 114.1 118.0

1.9

2.1

1.1 5.1

817

Examination of the columns giving total frequency rates and total
severity rates shows that, on the basis of frequency, the machinebuilding plants were more hazardous than the steel plant— the re
spective rates being 118 as against 114.5 per 1,000 full-time workers.
On the basis of severity, however, the steel industry was about three
times as hazardous as machine building— the days lost per full-time
worker being 15.7 and 5.1, respectively. It is clear that as between
these diametrically opposite showings of the relative hazards of the
two industries, the severity rates offer a decidedly more accurate
measure of true hazard. In machine building there is opportunity
for many minor injuries, but the danger of serious injury is much
less than in the steel industry. The severity rate brings out this
fact.
The second illustration shows how, over a period of years, within
the same establishment, accident severity rates may run counter to
accident frequency rates. The next table gives data of this char
acter. It shows the accident experience of a large steel plant over a
period of four years. The plant is one in which most serious atten
tion has been devoted to the prevention of accidents. Chart A (p. 24)
presents the same material in graphic form.
T able

3 .—ACCIDENT E X P E R IE N C E OF A LARGE ST E E L P L A N T , 1910 TO 1913.

Number of cases.

Year.

1910....................
1911....................
1912....................
1913....................

Accident frequency
rates (per 1,000
300-day workers).

N um
ber of
300day
Per Tem
Per
work
ma po
ma
ers. Death. nent rary To Death. nent
dis
tal.
in abil
in
jury. ity.
jury.
7,642
5,774
7,396
7,562

13
9
5
14

33
974 1,020
21
616
646
48 1,082 1,135
35
817
866

1.7
1.6
.7
1.9

4.3
3.6
6.5
4.6

Accident severity
rates (days lost per
300-day worker).

Tem
Per
po
rary To Death. ma
nent
dis tal.
in
abil
jury.
ity.
127.5
106.6
146.3
108.0

133.5
111.8
153.5
114.5

10.2
9.4
5.0
11.1

2.4
2.1
5.5
2.2

Tem
po
rary To
dis tal.
abil
ity.
2.2
2.4
2.8
2.4

14.8
13.9
13.3
15.7

C H A P T E R I .---- IN D U ST R IA L A CCID EN T RATES.

Limiting attention to the columns showing total rates, it will be
noted that in 1910 the frequency rate was 133.5 per 1,000 300-day
workers and the severity rate was 14.8 days lost per 300-day worker.
The next year, 1911, shows a decrease in both frequency and severity.
In 1912, however, there was a marked increase in frequency—from
111.8 to 153.5—but the severity rate dropped from 13.9 to 13.3. In
other words accidents had considerably increased in frequency, but
they were less serious in their total results. In 1913 this experience
was reversed. A marked reduction occurred in accident frequency—
from 153.5 to 114.5— while the severity rate jumped from 13.3 to
15.7. In other words, the year 1913, instead of being a “ good ” year,
as it might be assumed to be under the system of frequency rates,
was the worst of the four years covered by the table.
These illustrations bring up certain points which it seems desirable
to emphasize. The first concerns the use of terms. Severity rates
derived in the manner explained are expressed for convenience in
terms of workdays lost. For instance, the steel plant referred to
above is represented as having a severity rate, in 1913, of 15.7 days
lost per 300-day worker. The term “ days lost” as thus used is to
some extent a statistical abstraction, but it is close enough to con
crete fact to permit of its use in. its ordinary sense without any con
siderable degree of error, provided that the weighting scale employed
is a reasonable one. In any case, however, the real significance
of severity rates is in their use, not as positive amounts, but as
relative amounts, as indicating the relation between groups. Thus,
to recur to the example of the steel plant mentioned, the important
fact is that the severity rate for 1913 shows an increase over that for
1912 in the relation of 15.7 to 13.3.
This leads to a second point which can not be too much emphasized:
The fact that, inasmuch as the real significance of severity rates is in
the measurement of relative hazards, the character of the weighting
scale used becomes comparatively unimportant. Thus by changing
the weights in the scale offered above the resulting severity rates may
be considerably altered in their positive amounts, but unless the
changes are of a very radical character the relation between the rates
for different groups will remain substantially the same. In other
words, it is desirable to have the scale used as accurate as possible,
but the fact that a completely accurate scale can not be devised does
not impair the value of accident severity rating.

RATES

IRON AND S TE E L

IN OUSTRY,

1905 T O

I9 IZ

rf^
A C C ID E N T
PER IOOO

10 0

FREQUEN CY

RATES

A C C ID E N T

150

200

250

S E V E R IT Y

R A TES

num ber

W O R K E R .)

300«DAY

3 0 0 -0 A Y WORKERS

CO AY5 L O S T

300

PER 300*D A Y

WORKERS

6,406
7.4 94.
7.565

ACCIDENTS

4.5 75
IN
6,215
7, £42
5»774

7,562
4.741
5,599
9, 634
1 0 ,8 6 3 ;

IOO

150

200

250

300

CHART A.

BUILDING.

7,3 96

MACHINE

A C C ID E N T

C H A P T E R I .— IN D U ST R IA L ACC ID E N T RATES.

Another fact deserving emphasis is that severity rates have a very
important advantage over frequency rates, in that the errors in acci
dent reporting are minimized. Accident reports are probably never
absolutely complete, and, as a rule, the completeness of reporting is
in direct proportion to the seriousness of injury. The more serious
the injury the greater the likelihood of its being reported. Frequently
the reporting of minor injuries is extremely incomplete. Inasmuch
as the accuracy of frequency rates depends upon the completeness of
accident reports, and as all accidents have the same weight, a failure
to report any considerable number of minor accidents renders the
rates obtained of very little value. Such is not the case with severity
rates. Here the disabilities are weighted according to their impor
tance, and a large group of minor disabilities has comparatively little
effect upon the derived severity rate. Thus, from the material avail
able concerning the iron and steel industry, it is estimated that the
total exclusion of all disabilities of less than two weeks will rarely
diminish the total severity rate for that industry as much as 1 per
cent, whereas such an exclusion would diminish frequency rates as
much as 60 per cent. In the machine-building industry, according to
data collected by the Bureau, the corresponding percentages are 7
and 70.

CHAPTER IL—ACCIDENT EXPERIENCE.
As a basis for this report, accident data for a single year, 1912,
were obtained from 194 machine-building plants, employing for that
year a total of 115,703 300-day workers, which is equivalent to
347,109,000 man hours.1 It is believed that the plants thus covered
are sufficient in number and are well enough distributed to be fairly
representative of the industry in its important branches. Also, in
addition to the data for the year 1912, information regarding acci
dents over a series of years was secured from such of these plants as
had records of sufficient completeness for this purpose.
An analysis of the information obtained is presented in this chapter.
All accident rates are given in two forms— accident frequency rates
(i. e., the number of accidents per 1,000 300-day workers) and acci
dent severity rates (i. e., the average number of days lost per 300-day
worker). The significance of severity rates and their method of com
putation were explained in the preceding chapter. Also, the reasons
were there pointed out for believing that severity rates are a much
more accurate measure of true hazard than are frequency rates,
especially for purposes of comparison between industries.
A CC ID E NT R A T E S FOR 1 9 1 2 , B Y C H A R A C TE R OF PRODUCT.

The machine-building industry covers the manufacture of a large
variety of machines and machine tools as finished products, and the
accident hazard of a plant varies greatly with the character of the
product. This is shown in the following table, which classifies the
machine-building plants according to their products and shows for
each class, for the year 1912, the number of 300-day workers, the
number of cases of accidents, and the resulting accident rates:
T able

4 . —FR E Q U E N C Y A N D SE V E R IT Y OF ACCIDENTS IN 194 M ACHINE-BUILDING
PL A N T S IN 1912, C LASSIFIED B Y PRODUCTS.
1

Product.

Accident s e v e r i t y
A c c i d e n t frequency
Number of cases.
rates (days lost per
rates (per 1,000 300N um
day workers).
300-day worker).
ber
of
300Per Tem
Per Tem
Per Tem
day
ma po To
ma po To
ma po To
work Death. nent
rary
nent
Death.
Death.
nent
rary
rary
ers.
in disa tal.
in disa tal.
in disa tal.
jury. bility.
jury. bility.
jury. bility.

Ships.......................... 6,615
Mining m achinery... 3.994
cranes, e t c ................ 4,362
Iron and steel ma
chinery................... 2,692
L o c o m o t iv e s , e n
gines. e tc ................ 31,229
Electrical apparatus. 35,674
Transmission .......... 2,226
Machine tools............ 24,359
Unclassified
4,552
T otal................ 115,703

3
1
1
22
5
3
2
37

15 1,422 1.440
12
755
767
15
829
813
16

438

455

2.3
3.0
3.4

215.0 217.8
2.7
189.0 192.0
186.4 190.0 " ‘i.Y

1.7
1.1
.5

2.3
1.9
1.6

6.7
3.0
3.4

0.5

5.9

162.7 169.0

2.2

2.9

1.5

6.6

160 4,348 4,530
100 3,455 3 , 5C0
9
186
195
68 1,486 1,557
16
296
314

.7 5.1
•1 i 2.8
4.0
.1 ! 2.8
.4 i 3.5

139.2 145.0
96.8 99.7
85.6 89.6
61.0 63.9
65.0 68.9

4.2
.7
.7
2.7

2.9
1.1
2.1
.8
1.1

1.4
1.0
1.1
.5
.7

8.5
2.8
3 2
2.0
4.5

411 13,199 13,647 !

114.1 118.0

1.9

2.1

1.1

5.1

3.6

1 This does not include the navy yards and arsenals of the Federal Government.
was available regarding these Government shops is presented on p. 62 et seq.

Such information as

FREQUENCY AND SEVERITY OF ACCIDENTS
IN THE MACHINE BUILDING INDUSTRY.

<912.

COMBINED DATA FOR 194 P L A N TS .

CLASSIFIED BY PRODUCTS .
ACCIDENT FREQUENCY RATES
PRODUCTS

DAYS LOST

A C C ID E N T SEVERITY RATES
PER 300* DAY WORKER

PER 1000 300*DAY WORKERS

FREQUENCY

PER

RATES

300-DAY
WORKER

6.7

0,61 S

MINING MACHINES

3.0

3.m

CRANES, ETC .

3.4-

oss

CHARGING CARS.i

6.6

2,692

LOCOMOTIVES. ETC.

6.5

S1.22S

MACHINE

ELECTRICAL

2.8

. 35.674

TRANSMISSION

3.2

2.226

MACHINE TOOLS

2.0

24,3 S t

NOT SPECIFIED.

4.5

ALL PRODUCTS

4*S1
IIS,701

CHART B.

BUILDING,

smps

ACCIDENTS

C H A PT E R I I .---- ACCID EN T EXPE R IE N C E .

Considering, first, the last line of the table, it will be seen that
among the 115,703 300-day workers there occurred during the course
of the year a total of 13,647 injuries, consisting of 37 deaths, 411
permanent disabilities, and 13,199 temporary disabilities. This is
equivalent, as shown, to an accident frequency rate of 118 per 1,000
workers and a severity rate (i. e., time lost) of 5.1 days per worker.
These rates may be contrasted with the experience of a large steel
plant during the same year, for which the frequency rate was 153.5
cases and the severity rate 13.3 days lost.1 Accidents in the steel
plant were thus only about one-third more frequent than in the
machine-building plants, but their severity was more than twice as
great.
It is also to be noted that the accident rates for the different groups
are very dissimilar. The highest frequency rate (217.8 cases) and
the next to the highest severity rate (6.7 days) appear in shipbuilding.
This is the result largely of the building construction involved in the
putting together of ship hulls, this work having the high hazard inci
dent to the operation of reaming, riveting, and other construction
machines. Also, the scaffolding used is more temporary in character
than is customary in general building construction.
Of the three groups having the largest number of 300-day workers—
i. e., locomotives and engines, electrical apparatus, and machine
tools— “ locomotives and engines” shows the highest frequency rate
(145 cases) and also the highest severity rate (8.5 days), whereas for
“ electrical apparatus” the corresponding rates are at a much lower
level (99.7 cases and 2.8 days), and for “ machine tools” are at a still
lower level (63.9 cases and 2 days). In the case of these three
groups, therefore, frequency rates and severity rates vary uniformly.
But this is not the case with all of the groups listed. Thus,
“ cranes and hoists” has a higher frequency rate than “ locomotives
and engines” (190 against 145 cases), but a very much lower severity
rate (3.4 against 8.5 days lost). Other striking contrasts of this
character are shown in Chart B, which is a graphic presentation of the
data in the table. The existence of these contrasts reinforces the
earlier statement that no fair comparison of hazard in diverse in
dustries or industrial groups can be made on the basis of accident
frequency alone. The frequency may be great when the severity is
relatively low.
ACCIDENT R A T E S FOR 1912, B Y D E P A R TM E N TS.

The preceding section has shown how the accident hazards of
machine-building plants vary with the character of the product.
From the standpoint of accident prevention, however, it is of much
i See B ui. 234, Safety Movement in the Iron and Steel Industry, p. 61.

ACCIDENTS IN

M A C H IN E B U IL D IN G .

more importance to study the accident experience of the various
departments. A knowledge of the accident rates in particular de
partments, and better still in particular occupations, of a plant or
industry, indicates the fields in which safety work is most called for,
or, at least, in which investigation is most needed fop the proper
devising of safeguarding methods.
The following table shows the combined departmental accident
experience for the year 1912, of the 194 plants covered. Chart C
presents the same information in graphic form. In using these data
it is to be borne in mind that while, as a rule, the number of 300day workers in each department is sufficiently large to make de
ductions fairly conclusive, this is probably not true in the case of
the brass foundries and the power departments. In each of these
instances there are less than 1,000 300-day workers concerned, and
it is the general rule of this report to require at least 1,000 workers
in order to justify the computation of rates. Nevertheless, rates
based on a lesser number of workers may be sometimes, as at present;
of sufficient interest to justify presentation, provided the results are
interpreted with caution.
T able

5 .—FR E Q U E N C Y

A N D SE V E R IT Y OF ACCIDENTS IN 194 M ACH INE-BUILDING
PL A N T S, IN 1912, C LASSIFIED B Y D E PA R TM E N TS.

Number of cases.
Department.

N um
ber of
300day
work Death.
ers.

2,994
1,221
11,373
2,776
12,307
37,595
877
1,468
3,571
20,144
717
20,660

6
3
6
3
4
7
2

T otal.................. 115,703

Boiler shops................
Y ards..........................
Erecting shops...........
Forge shops................
Foundries (iron).......
Machine shops............
Power houses.............
Maintenance...............
Woodworking.............
Electric shops.............
Foundries (brass).......
Unclassified................

1
4
1

Accident f r e q u e n c y
rates (per 1,000 300day workers).

Per Tem
ma po
nent rary Total. D eath.
in disa
jury. bility.
29
7
56
13
49
132
2
1
27
65

636
260
1,992
439
1,670
3,923
87
137
264
1,579
52
30 2,160

671
270
2,054
455
1,723
4,062
91
138
292
1,648
52
2,191

411 13,199 13,647

2.0
2.5
.5
1.1
.3
.2
2.3

Per Tem
ma po
nent rary Total. D eath.
in disa
jury. bility.
9.7
5.7
4.9
4.7
4.0
3.5
2.3
.7
7.6
3.2

Per Tem
ma po
nent rary To
in disa tal.
jury. bility.

224.1
12.0 6.3
221.1
14.7 4.3
3.1 2 .8
180.6
163.9
6.4 2.8
140.0
1.9 2.1
108.1
1.1 1.4
103.8
13.7
.3
94.0
.8
81.8 ” 'i .y 2.4
81.8
1.2 1.0
72.5
106.2
” .*5*

212.4
212.9
175.2
158.1
135.7
104.4
99.2
93.3
73.9
78.4
72.5
1.5 104.6

3.6 114.1 118.0

Accident s e v e r i t y
rates (days lost per
300-day worker).

1.9

2.1

2.4
2.6
1.9
1.7
1.4
.9
1.3
1.1
.9
.8
.8
1.1

20.7
21.6
7.8
10.9
5.4
3.4
15.3
1.9
5.0
3.0
.8
1.9

1.1

5.1

The departments are arranged in the table, as also in the chart, in
the descending order of frequency rates.
Boiler shops have by far the highest accident rates of any of the
strictly machine-building departments. Its frequency rate (224.1
cases per 1,000 workers) is the highest of any department. Its
severity rate of 20.7 days lost per worker is exceeded only by the
yard department which has a rate of 21.6 days; and its death rate of
12.0 days is exceeded only by yards with 14.7 days and by the power
department with 13.7 days. The frequency rate of boiler shops,

FREQUENCY AND SEVERITY OF ACCIDENTS
124566°— 20— Bull. 256-------- 3

1912.

COMBINED DATA FOR 194. PLANTS.

SHOWING THE VARIATION IN ACCIDENT RATES IN THE IMPORTANT DEPARTMENTS. AND CONTRASTING ACCIDENT FREQUENCY
AND ACCIDENT SEVERITY

( s e v e r i t y b e in g m e a s u r e d in t e r m s o r d a y s l o s t ) .

A C C ID E N T S E V E R ITY RATES

ACCIDENT FREQUENCY RATES
PER 1000 300-DAY WORKERS
50

100

ISO

200

PER
250

»0

3 0 0 -DAY
...

W ORKER

2 Q -------------------- & £ -----------------

ACCIDENT DAYS LOST NUMBER
PER
FREQUENCY
3 0 0 -DAY
3 0 0 -OAY WORKERS
RATES
WORKER

%
2 0 .7

221

2 1 .6

1.221

161

f .t

1 1.373

1 0 .9

p H H H H H H H H H
p H B H IH H H H H I

164

(RON FOUNDRY

jH H H B H H H H

140

p H H H H H H

108

■ H

104

i s.S

877

1 .3

1.468

S .O

3.571
2 0 .1 4 4

MACHINE SHOPS

i
a

WOODWORKING

■ H H H I I

E LE C TR IC A L

■ H H H B

BRASS FOUNDRY

K K ttK K M

c la s s i f i e d
DEPARTMENTS

H
s

jH B IH H B B i
1

io o

s s i
i^o

20b

2io

2 ,776

S . A.

12.307

3 .4 -

37.S9S

3 .0

0 .8

717

106

1 .3

20 ;6 6 0

tie

S .t

115.703

EXPERIENCE

e r e c t in g

II.— ACCIDENT

2.994

224

CHAPTER

til TH E MACHINE BUILDING INDUSTRY.

00
CHART C.

ACCID EN TS IN

M A C H IN E B U IL D IN G .

224.1 cases, is indeed almost as high as the rate of 245.2 cases per
1,000 workers in the iron and steel industry in 1910.1
As the boiler shops show such high accident rates, it is appropriate
to inquire more particularly regarding accident causes in that depart
ment. In general, it may be stated that the work of constructing
boilers is similar in character to the erection of structural iron,
which is generally agreed to be extrahazardous. This extra hazard
arises, in part at least, from the necessary use of temporary staging
and trestles which afford an insecure footing, and also from the fact
that many of the air riveters, reamers, and other pneumatic tools
can not be given fixed location, but must be moved and handled
with constant danger of loss of control and of consequent loss of foot
ing. The substitution of fixed apparatus in many fabricating shops
has tended somewhat to reduce the frequency of accident, but prob
ably the most effective point of attack would be the temporary
structures used in the building processes. A very great improve
ment in scaffolding used in building construction has been made in
recent years, and there is no reason why inventive genius could not
improve the conditions of the boiler shop. In another direction
also improvement is possible. Boiler making is rough and not par
ticularly precise work, and is often relegated to very dark and illventilated buildings, and improvement in the character of the lightin^would, in many cases, tend to lessen the number of accidents.
Next to the boiler shops in frequency of accident, and exceeding
them in severity, is the relatively small yard department. The
exact limits of the employment in this department and the assign
ment to it of its proper quota of accidents are matters of consider
able difficulty. If the high rates shown in the table were not con1 Conditions of Em ploym ent in the Iron and Steel Industry (S. Doc. No. 110, 62d Cong., 1st sess.),
V ol. IV, p. 43.
Summary of accidents in iron and steel industry, by departments, year ending June SO, 1910.
Accident frequency rates (per
1,000 300-day workers).

Number of injuries.

Department.

300-day
work
ers.

Perma
nent
Fatal.
disa
bility.

Tem
porary
disa
bility,
1 day
and
over.

Total.

5,047
1,552
3,067
62
4,199
3,911
4,147
2,483
2,198
9,372

3.06
4.36
3.33

2.55
3.00
4. 21

2.06
1.03
1.32
2.86
.49
1.32

36,038

1.86

19,604
3,668
9,017
1,239
13,566
10,675
17,421
16,441
16,480
38,868

60
16
30

50
11
38

28
11
23
47
8
51

40
28
31
23
37
142

4,937
1,525
2,999
62
4,131
3,872
4,093
2,413
2,153
9,179

Total.............................. 146,979

274

400

35,364

Blast furnaces.........................
Bessemer steel works.............
Open-hearth steel works.......
Puddling.................................
Rolling mills (m echanical)...
Rolling m ills (hand)..............
Mechanical..............................
Y ards.......................................
Steel foundries........................
Departments not specified...

Tem
porary
Perma disa
Fatal.
nent
bility,
injury. 1 day
and
over.

Total.

2.95
2.62
1.78
1.40
2.25
3.67

251.8
415.8
332.6
50.0
304.5
362.7
234.9
146.8
130.6
237.1

257.4
423.1
340.1
50.0
309.5
366.4
238.0
151.0
133.4
242.0

2. 72

240.6 |

245.2

CH A PT E R I I .---- ACCIDEN T E X PERIEN C E.

firmed by the previous more extensile study of similar occupations
in iron and steel, some doubt might be entertained regarding their
validity. The death frequency for yard employees in iron and steel
is 2.86 cases per 1,000 300-day workers in an exposure of 16,441
persons. In machine building the death frequency for yard em
ployees is 2.5 cases per 1,000 300-day workers in an exposure of
1,221 persons. The liability to fatal injury in strictly yard opera
tions is indicated by the fact that switchmen in the iron and steel
industry in 1910 had a death frequency rate of no less than 27.3
cases per 1,000 300-day workers.1
It is of interest to note some of the reasons which have tended,
and evidently still tend, to keep yard conditions from responding to
safety efforts as promptly and as fully as have other departments.
In the first place, there is the fact of divided authority. In large
plants the tracks of adjacent railways are directly connected with
those of the plant, and it may easily happen that the railway does
not observe the same safety precautions as does the plant. The
railway may even relax the rules of its own yards when it passes
into territory under another jurisdiction.
In the second place, it may be that in matters of its transportation
equipment the plant is deficient in those elements of safety required
by law for public carriers. And in the third place, transportation,
though so vital an element in plant operation, is detachea.in a
measure from the other operations and has not been as carefully
considered from the safety standpoint as have other plant activities.
That the frequency of death should be nearly as great in the yard
operations of these machine-building plants as in the iron and steel
industry indicates the need for rigorous revisions of yard methods.
The third department in order of accident frequency is erecting.
The work in this department partakes of the nature of fabricating,
such as goes on in boiler shops, and of assembling, which is a con
spicuous element in the work of electrical shops. Its frequency rate
is 180.6 cases per 1,000 300-dav workers, while the severity rate is
7.8 days lost per worker. This severity rate may be compared with
the one of 20.7 days for boiler shops and 21.6 days for yards.
From the standpoint of size, machine shops constitute much the
most important department. Both in frequency of accident (108.1
cases per 1,000 300-day workers) and in severity (3.4 days lost per
worker) this department stands somewhat below the average for
all departments. The reasons for the existing rates are brought out
more clearly in the section on causes of injury in machine building.2
1 Conditions of Employm ent in the Iron and Steel Industry in the U nited States (S. Doc. No. 110, 62d
Cong., 1st sess.), Vol. IV. p. 99.
2 See p. 47.

ACCID ENTS IN M A C H IN E B U IL D IN G .

A CCID ENT R A T E S O V E R A SE R IE S OF Y E A R S .

The data of the preceding sections exhibit the accident hazards
of the machine-building industry as a whole and by departments.
The next step is to study the course of accidents by plants and by
departments over a series of years.
Such a study involves several difficulties. In the first place, there
is the very practical difficulty that accident data for a series of years
could be obtained from only a limited number of the plants from
which information was obtained for the single year 1912. Most of
the plants either had no records for earlier years or, if they had, the
records were too incomplete for use.
In the second place, many of the most desired analyses dea* with
groups of such small size that the influence of unusual causes may
become unduly prominent. All small groups, therefore, must be
interpreted with great caution. Finally, it is to be emphasized that
with improvement in accident prevention work the reporting of
accident cases tends to be more complete. In the analyses presented
below, every possible precaution was exercised in selecting plants
whose records have been kept with care and uniformity, but in spite
of this effort improvements which have undoubtedly been made are
masked in a measure by more exact reporting.
The following table shows the accident rates in a group of five im
portant machine-building plants over a period of six years, 1907 to
1912. This is the largest group of plants for which accident data
could be obtained for a period of any considerable length. These
5 plants, while included in the total of 194 plants for which detailed
information was presented for the year 1912 (see Table 5), are engaged
in work causing injuries of less frequency but of greater severity
than the average for the larger group. This fact, however, in no
way interferes with the value of the following comparisons of the
same group of plants over a series of years.
T able

6 .—FR E Q U E N C Y A N D SE V E R IT Y OF ACCIDENTS IN F IV E M ACHINE-BUILDING
PL A N T S, B Y Y E A R S , 1907 TO 1912.

Number of cases.

Year.

Accident frequency rates Accident severity rates
(per 1,000 300-day
(days lost per 300-day
workers).
worker).

N um
ber of
300day
Per Tem
work
ma po
ers. Death. nent rary
Total. Death.
in disa
bil
jury. ity
.

Per
ma
nent
in
jury.

Tem
po
rary
disa Total. Death.
bil
ity.

Per
ma
nent
in
jury.

Tem
po
rary Total.
disa
bil
ity .

22,023
8,261
11,303
18,729
16,481
17,233

124
27
41
80
53
81

1,809
324
713
1,642
1,329
1,744

0.8

4
9
20
12

.4
.5
1.2
.7

5.6
3. 2
3 .6
4.3
3 .2
4.7

75.7 82.1
36.0 39.2
59.1 63.1
82.9 87.7
76.2 80.6
95.8 101.2

2.1
2.9
7.3
4.2

3.9
1.6
3.3
3.3
2.1
3.0

0.8
.3
.5
.7
.7
.9

9.3
1.9
5.9
6.9
10.1
8.1

Total.............. 94,030

406 7,093 7,561

4.3

75.4

80.4

3.9

3.0

7.6

1907...........................
1908...........................
1909...........................
1910...........................
1911...........................
1912...........................

1,668
297
668
1,553
1,256
1,651

4.6

C H A PT E R I I .---- ACCID EN T EX PE R IE N C E .

In the tabulation of accident data the year 1908 appears as abnor
mal, due to the depressed industrial conditions of the time. The
data for that year should, therefore, be omitted from any deductions
drawn from this table. With this exclusion it appears that the
accident rates, for both frequency and severity of injury, show con
siderable variation from 1907 to 1912, but do not show any tendency
to decline.
It is possible, of course, that a better reporting of accidents in the
later years masks an actual reduction in accident hazards, and per
sonal knowledge of conditions in the plants leads to the belief that
improvement in equipment and other safety activities had actually
resulted in a reduction of general hazard. But it is evident that
the improvement could not have been very great. Better accident
reporting may account for the stationary, or indeed rising, frequency
rate. But, as was noted in Chapter I, the severity rate is not
markedly influenced by a fuller reporting of accidents.
In any case the showing made by these machine-building plants is
in marked contrast to the showing during the same years of the iron
and steel industry, in which there occurred some very remarkable
instances of progressive accident reduction.1 Undoubtedly, one im
portant reason for this difference in the experience of the two indus
tries is that there never were such seriously hazardous conditions in
machine building as those prevailing in the iron and steel industry.
#In iron and steel plants as late as 1910 general frequency rates of
500 cases and over per 1,000 300-day workers were common and
fatality frequency rates of nearly 5 cases per 1,000 300-day workers
were not unknown. When it is noted that the fatality rate in coal
mining,2 admittedly the most hazardous industry so far as fatalities
are concerned, has varied from 4.23 to 6.33 cases per 1,000 300-day
workers during the period 1896 to 1913, it is evident that the con
ditions in iron and steel manufacture, even in 1910, were such as to
call for the* most serious attention. On the other hand, in machine
building, with a fatality rate of 0.7 case and a general frequency rate
of less than 100, the need of immediate preventive measures would not
be forced so strongly upon the attention of employers.
As a result of this lesser degree of accident hazard, machine builders
did not begin thorough organization for accident prevention as early
as such efforts were undertaken by steel works, mines, and railways.
Up to the year 1912 their efforts had been largely directed to mechani
cal safeguards without the thorough organization elsewhere found
essential. This was not true in all machine-building plants, as is evi
dent from the comparisons of plants having and plants not having
i Conditions of Employm ent in the Iron and Steel Industry in the United States (S. Doc. No. 11062d,
Cong., 1st sess.), Vol. IV.
a U. S. Bulletin No. 157, Bureau of Labor Statistics, p. 104.

ACCIDENTS IN

M A C H IN E B U IL D IN G .

good safety systems (see p. 41), but it was sufficiently general to
influence the accident rates for machine building as a whole.
The next table is of the same character and of similar effect as the
preceding one, although it is for a more limited period of time. It
shows the accident experience of another group of four machinebuilding plants, over a period of four years, 1910 to 1913. This group
contains plants of low intrinsic hazard on the whole and so presents
rather lower accident rates than the group considered above.
Ta b l e

7 .—FR E Q U E N C Y

Year.

1910...........................
1911...........................
1912...........................
1913...........................

A N D SE V E R IT Y OF ACCIDENTS
B U IL D IN G PL A N T S, 1910 TO 1913.

FO UR

MACHINE-

Accident frequency rates Accident severity rates
(per 1,000 300-day work (days lost per 300-day
Number of cases.
ers.)
worker).
Num
ber of
300Per Tem
Per Tem
day
Per Tem
ma po To
ma po To
work
ma po
Death. nent rary tal.
Death. nent rary To
ers. Death. nent rary tal.
in disa
in disa
in disa tal.
jury. bility.
jury. bility.
jury- bility.
t
28,584
25,997
28,042
32,101

T o ta l............. 114,724

11
17
13
5

122
91
110
97

2,072
1,777
2,415
2,671

2,205
1,885
2,538
2,773

0.4
.7
.5
.2

4.3
3.5
3.9
3.0

72.5
68.4
86.1
83.2

77.2
72.6
90.5
86.4

2.3
3.9
2.8
.9

2.2
1.7
1.9
1.0

0.7
.6
.8
.8

5.2
6.2
5.5
2.7

420 8,935 9,401

3.7

77.9

81.9

2.3

1.8

4.8

Increased exactness in the reporting of accidents undoubtedly
affects, to some extent, the course of the frequency rates. The se
verity rates, it will be noted, show no considerable change until 1913,
when there was a marked decrease from a rate of 5.5 days lost to 2.7*
days. The decrease took place among the deaths and permanent
injuries, and is almost certainly accounted for by the more thorough
safety organization effected in some of these plants in 1912.
F A T A L A CC ID EN T S IN ENGINE B U ILD IN G .

For the engine-building division of the machine-building trades
records of fatal accidents, together with the amount of exposure, are
available for a long series of years, from 1902 to 1913. This permits
the computation of fatal accident rates. The course of these rates
is too irregular to permit of any important deductions, but are per
haps of sufficient interest to warrant presentation in tabular form.
T

able

8 .—FA T A L ACCIDENT R A T E S IN E N G IN E B U IL D IN G , B Y Y E A R S , 1902 TO 1913.

Year.

1902............................
1903............................
1904............................
1905............................
1906............................
1907
1908............................

Number
of
300-day
workers.

Fatali
ties.

Fatal
accident
rates
per 1,000
300-day
workers.

10,156
12,220
8,595
12,269
14,215
15,433
4,238

13
10
5
12
8
16

1.28
.82
.58
.98
.56
1.04

Year.

1909..........................
1910..........................
1911..........................
1912..........................
1913..........................
T otal...........

Number
of
300-day
workers.

Fatal
accident
Fatali
rates
ties.
per 1,000
300-day
workers.

5,967
11,936
10,832
11,146
12,984

4
7
14
12
2

0.67
.59
1.29
1.08
.15

129,991

103

.79

C H A PT E R I I .— A CCID EN T EXPE R IE N C E .

COURSE OF ACCIDENT R A TE S, B Y D E P A R TM E N TS.

Whenever a department has a sufficient number of 300-day work
ers, it is highly desirable to ascertain its accident experience over a
period of time. It is entirely possible that one department may
show a considerable reduction of accident hazard when no such
change is observable in the accident rates for the plant as a whole or
for a group of plants. And, on the other hand, conditions in one
department may be growing worse but the fact be concealed in the
general accident rate for the plant as a whole by the improvement
in other departments. The experience of individual departments
through excessive years is thus essential for the purpose of effective
safety work.
In the following series of tables there is presented information of
this character for such departments and for such periods of time as
the existing records make possible. Some of the data are necessarily
somewhat fragmentary.
ELECTR ICAL A SSEM BLY SHOPS.

Accident rates for two groups of electrical assembly shops, covering
different periods of time, are shown in the table below. Group A
presents data for two plants for the three-year period, 1910 to 1912;
group B for the three plants for the two-year period, 1912 and 1913.
T able

9 . —FR E Q U E N C Y A N D SE V E R IT Y OF ACCIDENTS IN ELECTRICAL A SSEM BLY
SHO PS, B Y Y E A R S.
A ccident f r e q u e n c y
rates (per 1,000 300day workers).

Number of cases.
Year.

Accident severity rates
(days lost per 300-day
worker).

N um
ber of
300Per Tem
Per Tem
Per Tem
day
ma po
work
ma po
ma po
ers. Death. nent rary Total. Death. nent rary Total. Death. nent rary Total.
in disa
in disa
in disa
jury. bility.
jury. bility.
jury. bility.

Group A (2 shops):
1910.................... 7,109
1911.................... 6,636
1912.................... 7,688

28
23
28

387
322
491

0.1
.1

3.9
3.5
3.6

50.4
45.1
60.1

54.4
48.6
63.8

0.9

1.6
1.9
1.1

0.6
.4
.5

3.1
2.3
2.4

358
299
462

T otal.............. 21,433

79 1,119 1,200

3.7

52.2

56.0

1.5

2.5

Group B (3 shops):
1912.................... 18,219
1913.................... 19,033

3
2

56 1,360 1,419
69 1,529 1,600

.2
.1

3.1
3.6

74.7
80.3

78.0
84.0

1.0
.6

.7
.9

.7
.8

2.4
2.3

Total.............. 37,252

125 2,889 3,019

3.4

77.6

81.1

2.4

Both frequency rates and severity rates are low in all of these
shops, and no material change is observable from year to year in
either of the groups.

ACCID ENTS IN M A C H IN E B U IL D IN G .
FORGE SHOPS.

Accident rates for six forge shops during the years 1912 and 1913
are shown in the following table:
T

able

1 0 .—FR E Q U E N C Y A N D S E V E R IT Y OF ACCIDENTS IN SIX FO RG E SH O PS, 1912
A N D 1913.

Year.

A ccid en t f r e q u e n c y Accident severity rates
(days lost per 300-day
rates (per 1,000 300Number of cases.
worker).
day workers).
Num
ber of
300Per Tem
Per Tem
Per Tem
day
ma po
work
ma po
ma po
ers. Death. nent rary Total. Death. nent rary Total. Death. nent rary Total.
in disa
in disa
in disa
jury. bility.
jury. bility.
jury. bility.
1,255
1,359

2
1

6
7

145
128

153
136

1.6
.7

4.8 115.5 121.9
5.2 94.2 100.1

9.5
4.4

2.5
1.2

1.5
.9

13.5
6.5

Total.............. 2,614

273

289

1.1

5.0 104.4 110.5

6.9

1.9

1.2

10.0

191 2
191 3

Inasmuch as the data cover a period of only two years, and as the
amount of exposure is small, the rather marked reduction in rates
can not be accepted as very conclusive. Especially is this so as no
information is available regarding the working conditions, which
would be likely to affect the rates.
FOUNDRIES.

For a group of four foundries it was possible to secure data for a
period of six years— 1907 to 1912— and for a second group of five
foundries information was obtainable for a two-year period— 1912
and 1913.
T able

1 1 .—FR E Q U E N C Y A N D SE V E R IT Y OF ACCIDENTS IN F O U N D R IE S , B Y Y E A R S .

Year.

A ccid en t f r e q u e n c y Accident severity rates
rates (per 1,000 300Number of cases.
(days lost per 300-day
worker).
day workers).
Num
ber of
300Per Tem
day
Per Tem
Per Tem
ma po
work
ma po
ma po
ers. Death. nent rary Total. Death. nent rary Total. Death. nent rary Total.
in disa
in disa
in disa
jury. bility.
jury. bility.
jury. bility.

Group A ( 4 p lan ts):
190 7
190 8
190 9
191 0
191 1
191 2

2 ,222
993
1,363
2,010
1,709
1,826

5
5
2
4

134

132
31
57
123
104
138

550

3 ,54;2
3,427

1
2

5
12

Total.............. 6 ,969

T otal.............. 10,123
Group B (5 plants):
191 2
191 3

118
31
52
ITS

2.7

4.3

17.5

5.’6’
3 .2
.2
.2

0.5
.2
.3
.4
.5
.3

7.5
.2
5.3
3.6
18.2
.5

57.8

3 .5

2.3

6.3

.3
.6

1.4 63.5 65.2
3 .5 110.3 114.4

1 .7
3 .5

1.4
.6

1.0
1.3

4.1
5.4

2 .5

87.0

2.6

1 .0

1.2

4.8

0.5

5.9

2.9

3 .7
2.5
1 .2
2 .2

53.1
31.2
38.2
58.7
56.8
73.4

59.5
31.2
4 1.9
61.2
6 0.9
7 5.6

585

2 .9

54.3

225
378

231
392

603

623

89.9

C H A PT E R I I .---- ACCID EN T EXPE R IE N C E .

The normal accident hazard of group A in the above table is best
understood by the omission of fatalities, as it is known chat the high
rate of 1911 was due to a single mass accident of very unusual char
acter. With this omission, the severity rates are distinctly lower in
the later years. This may be attributed to a decrease in the number
of foot burns, resulting from the adoption of impro ved shoes and leg
gings for the workers. An increase in the frequency rates of this
group during the same years, 1910 to 1912, in which there were marked
decreases in severity rates (except fatalities) is partly due to the bet
ter reporting of minor injuries.
Group B shows an increase in both frequency and severity rates for
1913 over 1912, but the period is too short to make these results very
significant.
M ACHINE SHOPS.

Accident rates for a group of five machine shops over a six-vear
period and for a group of eight shops over a two-year period are given
below.
T able

1 2 . —F R E Q U E N C Y A N D S E V E R IT Y OF ACCIDENTS IN M ACHINE SHOPS, BY
Y E A R S.

Year.

Group A (5 shops):
19.07....................
190 8
190 9
191 0
191 1
191 2

A ccid en t f r e q u e n c y Accident severity rates
Number of cases.
rates (per 1,000 300(days lost per 300-day
day workers).
worker).
Num
ber of
I
300day
Per Tem
Per Tem
Per Tem
work
ma po
ma po
ma po
ers. Death. nent rary Total. Death. nent rary Total. Death. nent rary Total
in disa
in disa
in disa
jury. bility.
jury. bility.
jury. bility.

7,817
3,520
4,747
6,688
6,303
6,647

1
1
1

48
9
15
28
21
20

660
96
256
624
449
543

709
105
271
653
471
564

0.1

84.4
27.3
53.9
93.3
71.2
81.7

90.6
2 9.9
57.1
97.7
74.7
84.9

0.8

.2
.2
.2

6.1
2.6
3 .2
4.2
3.3
3.0

73.6

.9
.9
.9

2.7
.8
1.6
2.0
1.7
1.6

0.5
.2
.5
.7
.6
.7

4.0
1.0
2.1
3.6
3.2
3.2

T o ta l............. 35, 722

141 2,628 2,773

4.0

77.7

1.9

3.2

Group B (8 shops):
9,676
191 2
10,472
191 3

2
3

38 1,048 1,088
27 1,230 1,260

.2
.3

3.9 108.3 112.4
2.6 117.5 120.4

1.3
1.7

1.9
1.0

1.0
1.1

4.2
3.8

T otal.............. 20,148

65 2,278 2,348

3.2 113.1 116.6

1.5

1.4

1.0

3.9

These machine-shop groups contain the largest number of 300day workers of any of the departmental groups which it has been
possible to assemble. Their accident experience, therefore, should
carry considerable weight, especially so as mechanical safeguarding
has been undertaken by all machine shops and here, if anywhere,
the effect of mechanical safeguarding should show itself.
As a
matter of fact the frequency rates for group A show little change,
their smallness in 1908 and 1909 corresponding with the usual ex
perience of accident rates for those years. The severity rates are

ACCIDENTS IN

M A C H IN E B U IL D IN G .

scarcely more conclusive, although some argument could be offered
for a very slight improvement. On the whole, it would seem fairly
justifiable to say that this machine-shop experience indicates that
mechanical safeguarding in the absence of other preventive methods
does not accomplish much.
Group B, covering only two years, also shows a condition of sub
stantially uniform accident rates.
Both of these groups contain individual machine shops in which
very possibly the accident hazard had been definitely reduced, but
the exposures of such shops were too small to justify separate com
putation.
W O O D W O R K IN G SHOPS.

The next table shows the accident rates for eight woodworking
shops for the years 1912 and 1913. The period is too short to permit
of any valuable deductions, but the table affords some additional
rates for comparison with those shown for the woodworking depart
ment in Table 5.
T able

1 3 . —FR E Q U E N C Y A N D S E V E R IT Y OF A C CIDENTS IN E IG H T W OODW O R K IN G
SH O PS, 1912 A N D 1913.
Accident frequency rates Accident severity rates
(per 1,000 300-day work (days lost per 300-day
worker).
ers).

Number of cases.

Year.

N um
ber of
300-day
work
ers.

1912....................
1913....................

1.442
1,561

14
9

90
111

104
120

9.7
5.8

62.4
71.1

72.1
76.9

2.9
1.1

0.8
1.0

3.7
2.1

T otal.......

3,003

201

224

7.7

66.9

74.6

2.0

2.9

Per
ma
Death. nent
in
jury.

Tem
Per
po
ma
rary Total. Death. nent
dis
in
abil
jury.
ity.

Tem
Per
po
rary Total. Death. ma
nent
dis
in
abil
jury.
ity .

TemPOrary Total.
dis
abil
ity.

EFFEC T OF SA F E T Y SY ST E M S UPON ACCID ENT OCCURRENCE.

A primary purpose of this report is to emphasize the possibilities of
constructive effort for the reduction of accidents. A pointed method
of doing this is to contrast the accident experience of plants having
well-developed safety systems with that of those in which safety
systems are in process of development. A contrast of this kind is
here attempted for the machine-building industry. The comparison
is limited to three groups of plants employing the largest number of
workers— electrical apparatus, locomotives and engines, and machine
tools— inasmuch as the relatively small size of the other groups
would render the results of questionable value.

C H A PT E R I I .---- ACCID EN T EXPE R IE N C E .

As a basis for this comparison, careful study was made of the
various plants, their methods of mechanical safeguarding, their
committee organizations, and their safety-education work. With
the knowledge thus obtained the plants were divided into two classes.
In class A were placed those plants having in considerable measure
the requisites of a good safety organization, namely:
1. Safeguarding by signs, warnings, and mechanical contrivances.
2. Adequate safety inspection.
3. Safety committees of superintendents, foremen, and workmen.
4. Emergency and hospital care of the injured.
5. A compensation or relief system.
In class B were placed the plants in which some important element
of this combination was lacking.
After thus classifying the plants the accident data for each class
were compiled, with the results shown in the following table:
1 4 . —FR E Q U E N C Y A N D SE V E R IT Y OF ACCIDENTS IN T H R E E G R OUPS OF
M A CH INE-BUILDING PL A N T S C L ASSIFIED ACCORDING TO C H A RACTER OF S A F E T Y
SYSTEM .

table

Number of cases.

Safety organization.

Accident frequency rates Accident severity rates
(per 1,000 300-day
(days lost per 300
workers).
day worker).

N um
ber of
300Per Tem
Per Tem
Per Tem
day
po
po
work
ma po To
ma rary
ma rary
ers. Death. nent dis Total. Death. nent dis Total. Death. nent rary
in abil
in dis tal.
in abil
jury.
jury. abil
jury. ity.
ity .
ity.

Electrical apparatus:
Class A ................. 23,012
Class B .................. 9,538
Unclassified......... 3,124

2
2
1

55 1,441 1,498
32 1,735 1,769
293
13
279

0.1
.2
.3

2.4 62.6 65.1
3.4 181.9 185.5
4.2 89.3 93.8

0 .5
1.3
1.9

0.4
2.6
1.1

0.7
1.8
.9

1.6
5.7
3.9

96.9

99.8

1.1

1.0

2.8

T otal.................. 35,674

100 3,455 3,560

Locomotives, etc.:
Class A .................. 4,971
Class B ................. 19,355
Unclassified......... 6,903

19
3

594
17
577
112 2,610 2,741
31 1,161 1,195

1.0
.4

3 .4 116.1 119.5
5.8 134.9 141.7
4 .5 168.2 173.1

5.9
2.6

1.8
2.5
5.2

1.0
1.4
1.8

2.8
9.8
9.6

160 4,348 4,530

5.1 139.2 145.0

4.2

2.9

1.4

8.5

2 .8

T otal.................. 31,229

Machine tools:
Class A .................. 6,769
Class B .................. 1,955
Unclassified......... 15,635

8
6
54

285
277
241
235
974 1,031

1.2 40.9 42.1
3.1 120.3 123.4
3.5 62.3 66.0

1.1

.2
1.2
1.0

.3
.9
.5

.5
2.1
2.6

Total.................. 24,359

68 1,486 1,557

2.8

61.0

2.0

63.9

Inspection of this table shows that in each of the three groups
class A has a very much lower accident rate than has class B, as
measured both by frequency and severity. Thus in the case of the
electrical apparatus group the plants of class A show a frequency rate
of only 65.1 as against 185.5 for the plants of class B, and a severity
rate of only 1.6 as against 5.7 days lost per worker. It is evident,
therefore, that the development of safety methods and safety organ

ACCID EN TS IN

M A C H IN E B U IL D IN G .

ization does have a marked effect in reducing the accident rate.1
Details as to the character and possibilities of accident preventive
work in machine-building plants are discussed in Chapter IV.
OCCU PATIONAL AC C ID E N T R A T E S .

It was one of the chief efforts in this study to accumulate sufficient
data for the presentation of satisfactory occupational accident rates.
The importance of accident rates b y individual occupations as an aid
to effective preventive methods can not well be overemphasized. But
the difficulties in the way of obtaining the material for such a pres
entation are very serious. One great difficulty is that occupational
names are used with a variety of meanings. Thus some concerns
distinguish between a machine hand who tends a screw machine
and a machinist who operates a lathe, but others make no such dis
tinction, using machinist as an extremely inclusive term. This un
certainty as to the meaning of occupational names interferes seriously
with the attempt to consolidate the records of different plants, and
such consolidation is necessary in order to obtain a sufficient volume
of data for rate computation. A second and still greater difficulty in
the way of getting occupational rates is that of determining the
exposure of each occupation; that is to say, the number or proportion
of 300-day workers engaged in each occupation. This is especially
difficult when the men in different occupations have workdays of

different lengths.
1 The method of comparison employed above was also used in the first iron and steel report (1913), the
plants being divided into three classes according as the safety system was (A ) well developed, (B ) in
process of development, or (C) not developed at all. In the classification of machine-building plants in
the above tex t classes A and B correspond to those in the iron and steel report. There is no class C, inas
much as all of the machine-building plan ts had made more than a beginning in safety work and safety
organization.
For an account of th e use of th is method of comparison in the iron and steel industry, see Conditions of
E m ploym ent in th e Iron and Steel Industry in th e U nited States (S. Doc. No. 110, 62d Cong., 1st sess.)
Vol. IV , pp. 43-49. The results of th e comparison there made are shown in th e following table, repro
duced from the report referred to:
Comparison of accident rates in iron and steel plants classified according to degree of development of safety
systems, year ending June 80, 1910.

! Accident frequency rates (per
1
1,000 300-d'ay workers).

Number of accidents.
Plants having safety system s 300-Jday
work
of specified class.
ers.

Class A. System well devel
oped...................................... 24,411
Class B. System in process of
developm ent....................... 28,830
Class C. System not devel
oped...................................... 14,916

Fatal.

Perma
nent
disabil
ity.

Tem
porary
disa
bility,
1 day
and
over.

Total.

Fatal.

Tem
Perma porary
disa
nent
disabil bility,
1
day
ity.
and
over.

Total.

3,993

4,079

1.73

1. 79

163.6

167.1

105

7,674

7,852

2.53

3. 64

266.2

272.4

7,505

7,576

2.48

2.28

503. 2

507.9

C H A PT E R I I .---- ACCID EN T EX PE R IE N C E .

Because of these handicaps the study of occupational rates in
machine building, as presented below, is necessarily somewhat
limited. Only those cases have been included in which all the facts
were ascertainable with accuracy.
In considering the rates presented it is to be clearly understood
that they are strictly valid only for the particular group for the
particular period covered. But, in addition, they may be accepted
as fairly representative of relative occupational hazards throughout
the industry.1
T able

1 5 .—FR E Q U E N C Y A N D SE V E R IT Y OF ACCIDENTS IN A M ACH INE-BUILDING
P L A N T D U R IN G 7 Y E A R S, 1907 TO 1913, B Y OCCUPATIONS.

Occupation.

Bench and vise hands__
Blacksmiths and helpers.
B oiler makers and helpers
Calkers and chippers___
Carpenters......................
Core makers.....................
Cranemen.........................
Drillers and helpers........
Erectors and helpers...
Laborers...........................
Machinists and helpers. .
Machine hands................
Pattern makers................
Reamers and riveters___
Sheet-iron workers..........
Other occupations...........

Accident
frequency Accident severity rates
Number of cases.
(days lost per 300rates (per 1,000 300N um
day worker).
day workers).
ber
of
300Per Tem
Per Tem
Per Tem
day
ma po To
ma po To
po To
work Death. ma rary
D
eath.
Death.
rary
rary
nent
nent
nent
ers.
in disa tal.
in disa tal.
in disa tal.
jury. bility.
jury. bility.
jury. bility.
2,937
4,350
2,413
1,769
1,074
920
1,011
3,269
2,360
10,035
18,534
1,290
1,246
2,734
1,946
16,648

T otal....................... 72,536

2
1
1

19
12
9
17

3
1
2
14
5
1
1
3

3
22
26
73
65
8
20
18
2
44

22
56

174 178
318 339
607 620
311 321
97 114
12
12
55
61
429 452
408 436
1,278 1,365
1,301 1,371
225 234
28
49
490 511
48
46
917 983

342 6,696 7,094

0.5
.4
.6
3 .0
.3
.9
1.4
.3
.8
.8
1.1
1.3
.8

1.4 59. 2 60.6
4.4 73.1 78.0
5.0 251.6 257.0
5.1 176.4 182.1
15. 8 90.3 106.1
13.0 13.0
3 .0 54.4 60.4
6.4 131.2 137.9
11.0 172.9 184.8
7.3 127.4 136.1
3.5 70.2 74.0
6.2 174.4 181.4
15.3 22.5 38.6
6.6 179.2 186.9
1. 0 23.6 24.6
2.6 55.1 59.0
4.7

92.3 97.8

2.7
2.5
3.4

0.9
2.5
2.5
6.0
5. 7

17."8 " l . 2
1.9 2.7
5.1 7.2
8.3 3.6
1.7 1.8
4.7 5.5
4.8 8 .4
6.6 3.9
1.3
7.9 1.6
4.6

2.5

0.5 1.4
.7 5.9
2.3 7.3
1.3 10.7
1.0 6. 7
. 1 .1
.6 19.6
1.1 5.7
1.6 13.9
1.2 13.1
.6 4.1
1.4 11.6
.2 13.4
1.5 12.0
.2 1.5
.7 10.2
.8 7.9

The close correspondence of the rates to the known hazards of the
occupations will appear as comment is made upon the individual
occupations. Since the severity rates’ are the more exact measure
of relative hazard they will be mainly used in this discussion.
The highest severity rate (19.6 days) is among cranemen. It is
entirely due to the number of deaths. In permanent and temporary
disability cranemen rank low. This fatality hazard is attributable
to the faulty construction of cranes, now fortunately greatly modified
by the safety features incorporated by all makers. Formerly it was
frequently necessary for a craneman to climb out on the girders of his
crane. There being no protective railings and no secure footway
a fall to the floor many feet below could easily occur, with fatal
results. With the improved patterns of cranes in use this high rate
should be much reduced and the occupation become of relatively
small hazard.
i Appendix B, p. 118, gives details regarding the results of the individual injuries upon which th istablo
is based.

ACCIDENTS IN

M A C H IN E B U IL D IN G .

Common labor ranks high in severity (13.1 days). This group is
so large (10,035 300-day workers) that a high rate among them is of
the greatest significance. There is no one particular preventive
measure by which the rate can be reduced. The only recourse is
the vigorous and persistent application of all the preventive methods
which experience has shown to be successful.
Erectors have a severity rate of 13.9 days. Permanent injuries of
a severe character are a large factor, two cases of loss of foot, one
case of loss of hand, and five cases of loss of eye being included.
The erector is constantly engaged in operations involving the moving
and adjusting of heavy parts and the use of more or less temporary
and insecure trestles and scaffolds. Improvement in this rate will
depend largely on improvement in crane methods, in better design
and construction of scaffolds, and in other means of reaching the
structures which are being erected.
The high rate (10.7 days) of calkers and chippers is due, in this
particular group, to two cases of the loss of an arm, an injury which in
a group of small size has considerable influence. This injury might
not be as serious in other groupings of this occupation, but the loss
of eyes from flying chips is sufficiently frequent to give these workers,
where due precautions are not used, a uniformly high severity rate.
Improved crane methods affecting the moving of the parts on which
they work and the use of proper protective goggles will bring down
this rate.
In the case of machine hands the small size of the group introduces
the possibility that the high rate of 11.6 days lost per worker may be
due to some unusual occurrence. In some measure, however, it
certainly reflects the fact that a good many automatic machines to
which the operator need only feed the material have been constructed
with their moving parts unduly and needlessly exposed. The con
tinuance of such a rate with modern and well-protected machines
must be regarded as impossible. Manufacturers who have machines
of this type which are not well protected should be warned by this
rate that under compensation they are likely to suffer heavily if the
defects are not remedied.
Reamers and riveters with a severity rate of 12.0 days and boiler
makers with 7.3 days may be regarded as belonging essentially in
the same class as erectors. They suffer from similar dangers, with
the additional dangers incident to certain machines which they
operate. These are frequently actuated by compressed air, and since
they must often be held in place by the workman there is a chance
that the pressure will swing the machine and throw the man from
the trestle or scaffold. A very frequent cause of injury appearing
in the records is “ lost control of air machine.” Machines electrically
actuated are now being used for reaming and riveting and it is clear

CH A PT E R I I .---- ACCID EN T E X PERIEN C E.

that they considerably lessen the danger. In general the precautions
suggested for erectors should be here applied.
Drillers were separated from the main body of machinists under
the impression that the drill presented greater hazards than the
lathe, miller, and other machine-shop apparatus. This impression
seems to have been justified by the resulting severity rate of 5.7
days for drillers and 4.1 days for the larger group of machinists. The
special hazard of the drill is that of the clothing being caught and
twisted upon it. Numberless illustrations could be given. One
will suffice. An apprentice boy reached around the drill to get a
tool. His sleeve was caught and before the machine could be stopped
all his clothing was torn off except his shoes. The possible remedies
for this hazard are two, both of which should be applied if possible:
(1) Clothing for drill operators which has no loose ends or slack
places. Since this is a matter of personal adjustment it has some
difficulties, but these have been overcome wherever effort has been
made. (2) Inventive genius has been trying to devise a chuck for
drills which will allow the tool to revolve freely when the drill is not
pressing upon the work. Such safety chucks have not been wholly
successful but have gone far enough to indicate usefulness in some
circumstances.
The modern construction of machine-shop apparatus should prac
tically eliminate accidents, formerly very common, arising in the
adjustment of gears, belts, and other parts of the machine. This
leaves the danger of being caught between the moving work and the
tool or parts of the machine. Practically, the reduction of this danger
must rest entirely upon the devising of safer methods and upon
greater care and skill on the part of the worker.
In the discussion of causes of injury, page 46, it will appear that
objects flying from the machine are a considerable source of danger.
It will probably be hard to convince machinists that this menace
of loss of sight is frequent enough to justify the use of protective
goggles, but if they could realize that loss of eyes is the most seri
ous single result of accidents causing permanent injury, protection
might seem worth while. In a group of plants having an exposure
of 94,030 during a series of years eye losses had a severity rate of 0.58
day while all permanent injuries to hands and fingers, including loss,
had a rate of 1.38 days.1
Among the other occupations the low rates of bench and vise hands
and sheet-iron workers, both in frequency and severity, are note
worthy.
Both carpenters and pattern makers have high severity rates in
permanent disability. This is due to the operation of three machines—•
the ripsaw, the band saw, and the wood planer or jointer. Fatalities

i See Table 21, p. 54.

ACCID EN TS IN

M A C H IN E B U IL D IN G .

among these workmen are nearly always due to what is known as
the “ kick back” from the ripsaw. The board is caught by the saw
teeth and thrown violently backward, striking the man in the abdo
men. One concern employing between 3,000 and 4,000 men has
had but two fatalities in the course of its history. These were due
to accidents of this kind. The efficient guarding of saws is a matter
of no small difficulty, but there are a number of reliable devices for
this purpose, among which any manufacturer should be able to find
something suited to his needs and which his workpeople will use.
The inclosure of the band saw to the point of almost perfect safety
is now so nearly universal that the finding of one not so inclosed
causes a distinct shock of surprise. There are still in use on a good
many planers the square cutter heads, which allow the hand to drop
as far as the knuckles between the blades, although none was ob
served in 1912 in the plants included in this study. With the sub
stitution of the cylinder head and the application of convenient
covers for the portion of the knives not in use, the high permanent
injury rate should drop materially. An absolutely perfect safety
device for the jointer, an automatic feed, was observed in use in one
of the Government arsenals. The initial expense of this device was
considerable, but the man in charge was of the opinion that the
increased efficiency made it a good investment.
ACCIDENT C A U S E S .1

The analysis of the causes of accident and the determination of
occupational rates seem at present to be the most important practical
subjects to be considered in accident studies. Occupational rates
in machine building have been considered in the preceding section.
The subject of accident causes will now be considered in as much
detail as the available material makes possible.
A CCIDEN T CAUSES, B Y P L A N T GROUPS A N D B Y D EPAR TM EN TS.

The relative importance of the various accident causes in machine
building is indicated in Table 16. This table distributes frequency
and severity rates by causes for three important machine-building
departments— electrical assembly shops, foundries, and machine
shops— and also for two groups of entire plants. Only three depart
ments were chosen for presentation, because for none of the others
was there a sufficient volume of exposure to permit of fair comparison.
1 For a classification of causes adopted by the International Association of Industrial A ccident Boards
and Commissions in May, 1916, see B ulletin No. 201, U. S. Bureau of Labor Statistics. This classification
w ill be used in future by the Bureau in common w ith other organizations concerned in its preparation.
The tabulation of the present report had progressed too far to permit the application to it of this later
classification.

CH A PT E R I I .— ACCIDEN T E X PERIEN C E.

Table 16.—CAUSES OF IN JU R Y IN SP EC IFIE D M ACH INE-BUILDING D E PA R TM E N TS
AND

PLANTS.
Frequency rates (cases per 1,000 300-day workers).
Electrical Foun Machine Machine building.
assembly dries,
shops,
shops,
plants, 6 plants, 5 plants, 4 plants,
6 plants, 5 1907
to
1907
to
1910 to
1907 to
1907 to
1913.
1913.
1912.
1913.
1913.

Cause of injury.

Number of 300-day workers...........................................
Cranes and hoists............................................................
Falling objects.................................................................
Falls of worker................................................................
H ot m etal.........................................................................
Handling material...........................................................
Operating machines........................................................
Objects flying from machines........................................
Objects falling from machines.......................................
U sing tools.......................................................................
Objects flying from tools................................................
Other flying objects.........................................................
Reaming, riveting, and chipping..................................
Objects flying during reaming, riveting, etc................
Other causes.....................................................................
Causes not reported.........................................................
Total.......................................................................

30,906

15,1*89

47,412

94,030

114,724

8.16
2.10
8.35
14.88
4.01
4.50
13.56
1.55
7.83
8.09
16.53
1.84
.72
2.36
.42
4.27 ....... i.*38*
32
.26
2.17
.54
.06
.59
4.15
.32
7.97
11.10
.26
.23

5.97
12.17
4.39
.34
9.24
22.29
7.87
2.53
5.86
.93
1.56
.30
.57
4.91
.38

7.15
14.44
8.11
1.99
8.10
13.04
4.02
1.07
5.38
.99
3.19
1.29
2.34
8.89
.40

5.74
14.35
7.80
2.43
8.88
14.12
4.65
1.09
5.20
1.03
2.60
1.11
1.79
10.70
.45

67.78

79.31

80. 40

81.94

60. 74

Severity rates (days lost per 300-day worker).
Cranes and hoists............................................................
Falling objects.................................................................
Falls of worker.................................................................
H ot m etal.........................................................................
Handling material...........................................................
Operating machines........................................................ 1
Objects flving from m achines........................................
Objects falling from m achines.......................................
Using tools.......................................................................
Objects flying from tools................................................
Other flying objects
..................................................
Reaming, riveting, and chipp ing.................................
Objects flying dunng reaming, riveting, etc..............
Other causes....................................................................
Causes not reported........................................................
Total.......................................................................

0.03
.08
.06
.01
.09
.81
.14
0)
.03
.05
.05
0)
0)
.60
0)
1.98

.01
C1)
.10
0)
.16
.63
0)

0.22
.64
.06
0)
.10
1.18
.25
.03
.06
.13
.03
0)
0)
.09
.01

1.75
.86
.76
.35
.11
.88
.36
.02
.05
.09
.26
.04
.28
.95
.16

0.98
.71
.37
.02
.11
.83
.27
.01
.04
.09
.21
.03
.15
.98
.07

6.10

2.82

6.92

4.87

0.87
1.09
.53
2.32
.15
.22
C1)

1 Less than 0.005.

From the upper half of this table it appears that, for the industry
as a whole, as represented by the two groups of plants, the most
important single cause of accident as regards frequency is “ falling
objects.” This is also the most frequent cause in foundry work.
But in the electrical assembly and machine shops primacy shifts
to the “ operation of machines’ ’ as the most fertile cause of accident.
As regards the severity of resulting injuries, as shown in the lower
half of the table, “ cranes and hoists” stands out for the industry as
a whole as the most serious single cause. In foundries “ hot metal”
appears as the accident cause of most serious after effects, the severity
rate for hot metal being 2.32 days lost per worker out of 6.10 days
lost per worker for all causes. This preeminence of hot metal as a
hazard of foundry work emphasizes the point at which preventive
124566°— 20— Bull. 256------ 4

ACCIDENTS IN

M A C H IN E B U IL D IN G .

measures can be applied effectively. In the more progressive foun
dries provision of proper shoes, leggings, and eye protectors has nearly
eliminated certain kinds of burns.
ACCIDENT CAUSES OVER A SERIES OF Y E A R S.

In the study of accident causes it is of particular interest to trace
the importance of the several causes over a period of years. It has
been possible to obtain the necessary material for this purpose for
two groups of machine-building plants of sufficient size to make
yearly comparison reasonably safe. The first group consists of five
plants with accident experience over a period of six years, 1907 to
1912. This information is presented in Table 17. Table 18 shows,
for the purpose of comparison, the experience of a large iron and
steel plant.
T able

1 7 .—ACCIDENT CAUSES IN F IV E M ACHINE-BUILDING PL A N T S, B Y Y E A R S,
1907 TO 1912.
[Covering 7,561 cases of accident.]
Frequency rates (cases per 1,000 300-day workers).
Accident causes.

Number of 300-dav workers.
Cranes and hoists.............................................
F alling o b je cts...............................' ...............................

Falls of worker.................................................
Hot m etal.........................................................
Handling material...................................« ---Operating m achines........................................
Objects flying from m achines.........................
Objects falling from m achines........................
U sm g to o ls........................................................
Objects flying from tools..................................
Other flying objects..........................................
Reaming, riveting, e tc.................................... .
Objects flying during reaming, riveting, etc.
Other causes.....................................................
Causes not reported.........................................
Total.

1907

1908

1909

1910

1911

1912

22,023

8,261

11,303

18,729

16,481

17,233

94,030

6.67
16.07
7.58
1.91
8.72
14.76
2.69
1.14
6.04
.95
3.54
.54
2.63
8.63
.36

3.63
5.33
5.08
.73
4.12
5.81
1.69
.36
4.24
.61
.73
1.33
1.21
4f24
.12

6.72
10.71
6.72
1.33
6.37
11.77
2.39
.62
3.89
.88
1.68
1.33
1.24
7.08
.35

9.56
14.42
7.74
2.08
11.91
13.67
4.54
1.76
5.71
1.17
2. 99
1.23
1.71
8.97
.21

7.83
15.53
8.43
2.00
6.49
12.01
4.13
1.09
5.58
.91
3.16
1.94
2.61
8.56
.36

6.44
18.16
11.26
3.02
7.78
15.44
7.37
.87
5.51
1.16
5.16
1.62
3.66
12.88
.87

7.15
14.44
8.11
1.99
8.10
13.04
4. 02
1.07
5.38
.99
3.19
1.29
2.34
8.89
.40

82.23

39.23

63.08

87.67

80.63

101.20

80.40

Total.

Severity rates (days lost per 300-day worker).
Cranes and hoists......................
Falling objects...........................
Falls of worker...........................
H ot m eta l...................................
Handling m aterial.....................
Operating m achines..................
Objects flying from m achines..
Objects falling from m achines.
Usm g tools.................................
Objects flying from to o ls..........
Other flying objects.
Reaming, riveting, etc.
Objects flying during reaming, riveting, etc.
Other causes.......................................................
Causes not reported...........................................
T otal.........................................................

2.61
1.01

1.07
.17
.14

1.00
.34
.02
.05
.06
.14
.29

1.10
.40

0.19
.05
.15

1.11

.29
1.44

0.69
.91
.45

.05
.48
.03

.08
1.15
.34

.19
.77
.63

.01

.02
.05

.21
.03

.20
1.47

.01

.02
.01
.23
.01
.34
.07
5.10

.01
.01
.07
.15
.58

.02
.01

2.91
.67
.87
1.61
.05
.71
.56

.01
.02
.23
.15

.01

1.82
1.61
.51
.03

.12
1.12
.09
.02
.03
.23

.21
.01

.76
.35

.11

.88

.36

.02
.05
.09
.26

.04*

1.61
. .01

1.78

.95
.23
.41

.28
.95
.16

6.06

9.58

7.39

6.92

C H A PT E R I I .---- ACCID EN T EX PE R IE N C E .
T able 1 8 . —ACCIDENT CAUSES IN A ST E E L PL A N T , B Y Y E A R S, 1905 TO 1913.

[Covering 10,390 cases of accident.]
Frequency rates (cases per 1,000 300-day workers).
Accident causes.
1905

1906

1907

1908

1909

1910

1911

1912

1913 Total.

92.7
31.2
26.6
19.0
26.2
17.2
15.9
12.5
4.2
1.6
52.9

55.4
22.7
22.6
14.9
17.2
9.3
12.0
11.1
3.6
1.1
44.4

55.1
19.6
24.5
14.1
10.6
11.2
10.9
7.3
3.3
1.6
30.9

46.3
20.3
11.1
14.2
10.9
4.6
10.1
4.6
1.3
1.1
25.2

52.1
16.9
15.1
20.0
17.7
10.0
7.4
8.0
2.1
.2
24.7

38.3
12.3
9.8
12.2
16.2
7.6
7.6
5.0
2.6
.7
21.2

20.4
12.0
7.3
16.1
10.9
9.5
5.4
3.3
1.0
.5
25.4

39.7
14.7
12.3
18.9
14.8
10.5
4.3
6.9
2.4
.1
28.9

25.3
9.0
11.9
12.2
10.2
10.8
3.6
5.0
.7
.8
25.0

46.9
17.4
16.0
15.6
15.0
10.2
8.5
7.2
2.4
.8
31.3

T otal................................................. 300.0 214.3 189.1 149.7 174.2 133.5 111.8 153.5 114.5

171.3

Falling and flying objects.........................
Falls of worker............................................
H ot m etal...................................................
Cranes and hoists.......................................
Handling material and w ork...................
Using tools..................................................
Operating machines...................................
Locomotives, cars, etc...............................
E lectricity..................................................
Belts, shafts, and gears..............................
Unclassified.................................................

Severity rates (days lost per 300-day worker).
Falling and flying objects.........................
Falls of worker...........................................
H ot m etal...................................................
Cranes and hoists.......................................
Handling material and w ork...................
Using tools..................................................
Operating machines..................................
Locomotives, cars, etc...............................
E lectricity...................................................
Belts, shafts, and gears..............................
Unclassified.................................................

6.9
1.4
1.3
2.8
.2
.2
.5
2.8
.1
1.4
7.9

6.6
3.0
6.6
4.0
.2
.1
1.2
5.4
1.7
.9
9.0

5.8
3.0
.8
5.0
.2
.2
1.3
1.3
1.8
.9
7.9

1.5
3.4
.2
1.8
.3
C1)
2.1
1.7
0)
3.1
7.9

4.6
4.6
2.6
.4
.4
.2
.1
1.2
0)
0)
3.4

4.5
1.9
1.9
1.4
.2
.1
1.0
.3
0)
.4
3.0

1.0
2.8
1.3
4.9
.1
.1
1.3
1.2

3.7
1.6
2.7
1.1
.7
.3
.4
.8

w
1.0

T otal.................................................

25.5

38.7

28.2

22.0

17.5

14.8

13.9

(l)
2.0

3.5
2.7
3.9
3.3
.1
.1
.1
.2
0)
0)
1.6

4.4
2.7
2.4
2.7
.3
.1
.8
1.7
.4
.7
4.8

13.3

15.7

21.0

1 Less than 0.005.

Again it may be repeated that the year 1908, and to some extent
the year 1909, were abnormal years in machine building, and the acci
dent rates for those years are of little present significance. With this
in mind, an examination of Table 17 shows that in machine buflding
accident rates, for both frequency and severity, and both in total and
by separate causes, show considerable variations from year to year,
but do not show any downward tendency. To some extent, an actual
reduction in hazard may be concealed, particularly in the case of
frequency rates, by a better system of reporting. But, in any case,
it is clear that during the years covered no such marked reduction in
accident rate took place in machine building as is shown to have
taken place in the steel plant whose experience is presented in Table
18. This particular steel plant is one in which safety work was
vigorously pushed during the years covered, but which is in no way
exceptional and is fairly illustrative of the steel industry as a whole.
In making the above comparisons it is to be repeated that the
opportunity for accident reduction in machine building was much
less than that in the steel industry. The sources of danger in machine
building are both less serious and less easily brought under control
than many of those in iron and steel plants. Accident hazards in
machine building were never at the high point reached in some of the

ACCIDENTS IN

M A C H IN E B U IL D IN G .

steel plants in the years before active safety work was undertaken.
Thus, in Table 17 the frequency rate for the group of machine-building
plants, during the six years covered, only once rose above 100 cases
per 1,000 300-day workers, this being in 1912, when the rate was
101.2. In contrast to this, the steel plant shown in Table 18 (one
of the earliest to untertake safety work) had a frequency rate of
300 cases per 1,000 300-day workers in 1905 and of 189.1 cases in 1907.
But, even when all due allowance is made for these fundamental
differences, it remains true that up to the year 1912 the machinebuilding industry had not fully awakened to the possibilities of
accident prevention. By that time, however, many individual plants
had inaugurated important safety activities and these were being
rapidly taken up by the entire industry. That these efforts were
productive of success is indicated by the experience of a group of
plants for which data were obtainable for the year 1913 in compari
son with preceding years. This group consists of four plants with
experience extending over a period of four years, 1910 to 1913.
T a b l e 1 9 . — ACCIDENT

CAUSES IN F O U R M ACH INE-BUILDING PL A N T S, B Y Y E A R S , 1910
TO 1913.
[Covering 9,401 cases of accident.]
Frequency rates (cases per 1,000 300-day workers).

A c c id e n t causes.

1910

1911

1912

1913

Total.

28,584

25,997

28,042

32,101

.4,724

Cranes and hoists.............................................
Falling objects..................................................
F alls of worker.................................................
H ot m etal.........................................................
Handling material............................................
Operating machines........................................
Objects flying from m achines........................
Objects falling from mchines..........................
U sing to o ls........................................................
Objects flying from to o ls................................
Other flying objects.........................................
Reaming, riveting, and chipping...................
Objects flying dunng reaming, riveting,etc.
Other causes.....................................................
Causes not reported.........................................

7.56
12.77
6.79
1.99
10.04
13.58
3.36
1.22
4.97
.98
2.20
.84
1.22
9.45
.17

5.85
13.23
6.89
1.62
6.46
13.12
3.77
.92
4.77
.96
2.08
1.19
1.69
9.81
.15

5.03
15.55
9.84
2.71
8.42
16.33
5.49
.75
5.42
1.00
3.50
1.03
2.46
12.45
.53

4.67
15.61
7.66
3.24
10.22
13.49
5.79
1.40
5.54
1.15
2.59
1.34
1.78
11.03
.87

5.74
14.35
7.80
2.43
8.88
14.12
4.65
1.09
5.20
1.03
2.60
1.11
1.79
10.70
.45

Total.........................................................

77.14

72.51

90.51

86.38

81.94

Number of 300-day workers. . .

...................

Severity rates (days lost per 300-day worker).
Cranes and hoists.............................................
Falling objects..................................................
Falls of worker.................................................
H ot m eta l.........................................................
Handling material...........................................
Operating machines.........................................
Objects flying from m achines........................
Objects falling from m achines........................
U sm g to o ls..................... *.................................
Objects flying from to o ls................................
Other flying objects.........................................
Reaming, rivetm g, and chipping..................
Objects flying during reaming, riveting, etc
Other causes.....................................................
Causes not reported.........................................

0.46
.90
.33
.02
.16
.89
.19
.01
.06
.05
.45
.01
.01
1.64
.01

1.95
.48
.58
.01
.05
.70
.58
.01
.03
.10
.10
.10
.01
1.51
0)

1.20
1.03
.34
.02
.12
.92
.17
.01
.03
.14
.26
.01
.60
.45
.26

0.37
.46
.10
.03
.10
.83
.16
.01
.04
.05
.10
.03
.01
.39
.02

0.97
.71
.33
.02
.11
.83
.27
.01
.04
.09
.21
.03
.15
.98
.07

T otal........................................................

5.19

6.21

5.56

2.70

4.82

1 Less than 0.005.

C H APTER I I .---- ACCID EN T EXPE R IE N C E .

The frequency rates of this group of plants show an irregular but
rather upward tendency from 1910 to 1913, but the increase is no
greater than could be accounted for by the better reporting which
has certainly occurred in some of these concerns and probably in all
of them. The striking change is in the severity rates, which drop
from over 5.19 days lost per worker in 1910 to 1912 to 2.70 days lost
in 1913.
This brings up again the important fact that an increase in fre
quency rates may be entirely compatible with a decrease in severity.
The increase in frequency may be due simply to better reporting, and,
as better reporting is nearly always correlated with special efforts at
accident reduction, there may well be a corresponding reduction in
accident severity as measured by severity rates. It is possible
indeed that a very great improvement in reporting might give rising
figures for both frequency and severity rates when the real accident
hazards were actually diminishing. This fact suggests the danger
of concluding from slightly rising accident rates that no improve
ment in accident reduction has occurred. Hasty conclusions have
led in several plants to undue discouragement in their efforts toward
accident prevention.
On the other hand such a fall in severity rates as occurs between
1912 and 1913 in the plant group shown in the table (from 5.56 to
2.70 days lost) may be accepted as fairly conclusive, inasmuch as the
reduction was distributed among practically all causes. A drop at
only one point might very well be due to a fortuitous circumstance,
but, when nearly all causes show the same tendency, it is safe to infer
that some general and pervasive influence is at work. Personal
knowledge of conditions in these plants substantiates the indication
of the rates. Machine-building concerns generally had begun active
preventive work by 1912. Those whose activities were well under
way before that year were not sufficiently numerous to influence
materially the accident rates for the industry, but these rates began to
respond on a fairly general scale in 1913.
It would be instructive to know the causes of accident, by depart
ments, over a series of years in supplement to the data for groups of
plants as above given. But until a much larger amount of material
is available such a study would not be sufficiently conclusive to be
worth the labor of preparation.
NECESSITY OF R ATES FOR TH E M EASUREM ENT OF A CC ID E N T CAUSES.

Before leaving this discussion of accident causes it is desirable to
point out by means of a simple illustration a mathematical defect
which renders inconclusive a method which has frequently been used
as a means of determining the importance of accident causes from

ACCIDENTS IN

M A C H IN E B U IL D IN G .

year to year. The groups of accidents in the years which it is desired
to compare have been thrown into the form of percentages and the
conclusion drawn, for example, that the handling of cranes and hoists
had improved if they caused 10 per cent of all recorded accidents
in one year and only 8 per cent in the year following. An illustration
will serve to bring out the dangers which lurk in this procedure.
Let it be supposed that a concern employs 10,000 300-day workers
in 1910 and 9,000 in 1911. Ten causes of accident are recognized,
designated by letters A, B, C, D, and so on. In 1910 assign 100
accidents to each cause and in 1911 the same number except the last
cause, J, to which 150 are assigned. The following table presents
the resulting distribution by percentages and by frequency rates:
Per cent due to
each cause.

Cases.

Frequency rates.

Cause.
1910

1911

1910

1911

1910

1911

A ................
B ................
C................
D ................
E ................
F ..........
G ..........
H ................
I ...........
J..................

100
100
100
100
100
100
100
100
100
100

100
100
100
100
100
100
100
100
100
150

10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0

9.5
9.5
9.5
9.5
9.5
9.5
9.5
9.5
9.5
14.3

10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0

11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
16.7

T o ta l....

1,000

1,050

100.0

116.7

It might be concluded from the percentage column of this table
that there has been a decrease in the importance of each cause except
cause J. When, however, the frequency rates are considered, it is
clear that each of the causes has become more serious.
The inherent weakness of all such percentages as those of the
table is that they can not vary independently. If cause J changes,
it modifies the percentage for each of the other causes as well, and
thus completely conceals what has actually taken place. This same
difficulty has frequently been emphasized in the study of mortality,
as, for example, when an epidemic of typhoid fever produces a low
ered percentage of tuberculosis at the same time that the incidence
of tuberculosis, as shown by the rates, was increasing.
Thus, in the measurement of the changing importance of accident
causes, percentage distribution may be of misleading significance.
The only reliable guide is that offered by accident rates, either fre
quency or severity.

CH A PT E R I I .— ACCID EN T EX PE R IE N C E .

N A T U R E OP IN JU R Y.

From the standpoint of accident prevention, the subject of nature
of injury is of much less importance than is that of cause of accident.
Nevertheless, an analysis of the injuries according to their nature is
not without value in accident prevention work and is of much sig
nificance in the study of accident compensation.
The proper classification of nature of injury has been a matter of
considerable discussion. In the present study it has seemed that
the most useful classification is one which makes the pathological
condition (burns, crushing injuries, fractures, etc.) the basis and then
subdivides according to the anatomical region affected— head, hands,
etc.1 If this is supplemented by information showing the resulting
permanent injury, if any (such as loss of hand, loss of sight), a very
complete picture of the injury is offered. Thus, a particular injury
would be listed as follows: A crushing injury (pathological) to the
hand (anatomical region) causes the ultimate loss of the hand
(result).
A simple classification of the type outlined is used in the following
table, which gives an analysis of the nature of the injuries which
occurred in seven machine-building plants during a period of several
years. The table represents a total of 14,204 injuries, occurring
among a total of 179,956 300-day workers.
T able

3 0 . —N A T U R E OF IN JU R Y IN SE V E N M ACHINE-BUILDING PL A N TS, 1907 TO
1913.

Injury.

Bruises, cuts, and lacerations to—
Hands and fingers............................................................
Feet and to e s.................................................................
Other parts....................................................................
B u m s.....................................................................................
Crushing injuries to—
Hands and fingers.........................................................
Feet and to es.................................................................
Other parts........................................................................
Dislocations and sprains.....................................................
E ye injuries..........................................................................
Fractures..............................................................................
Punctured wounds..............................................................
Unclassified injuries.............................................................
Total................................................................................

Number of Days lost.
injuries.

Accident
frequency.

Accident
severity.

4,007
1,744
3,062
755

77,245
17,638
80,393
90,505

22.3
9.7
17.0
4.2

0.43
.10
.45
.50

425
37
29
869
1,498
1,112
439
227

105,222
14,466
167,592
9,929
78,449
192,947
3,094
142,786

2.4
.2
.2
4.8
8.3
6.2
2.4
1.3

.58
.08
.93
.06
.44
1.07
.02
.79

14,204

980,266

79.0

5.45

This table classifies the injuries according to their nature in a few
large groups, and shows the accident rates, according to both fre
quency and severity, for each group. The accident rates, standing
by themselves, are not so significant as they will become when similar
compilations are made for other industries or for other periods of
time.
1 For a classification of injuries by their nature and anatomical location proposed by the International
Association of Industrial Accident Boards and Commissions, see Bulletin No. 201, U. S. Bureau of Labor
Statistics, p. 81.

ACCIDENTS IN

M A C H IN E B U IL D IN G .

P E R M A N E N T R E SU L T S OF IN JU R Y .

The table following shows the frequency and severity of permanent
injuries in three important departments of machine building and
also in two groups of entire plants:
T able

2 1 — FR E Q U E N C Y A N D SE V E R IT Y OF PE R M A N E N T IN JU R IE S IN SP E C IFIE D
G R OUPS OF PL A N T S.
Frequency rates (cases per 1,000 300-day workers).

Injury.

Number of 300-day workers..
Loss of great to e ..........................
Loss of 1 joint of great to e .............
Loss of other toe or to es.................
Loss of 1 joint of other toe or to e s.
Loss of foot......................................
Loss of leg ........................................
Loss of thum b.................................
Loss of 1 Joint of th um b.................
Loss of 1 joint of finger or fingers.
Loss of first finger...........................
Loss of second finger.......................
Loss of third finger.........................
Loss of fourth finger.......................
Loss of h a n d ....................................
Loss of arm......................................
L oss of e y e ...................................................

Other permanent injuries..............
Total.

Electrical Found Machine
assembly
ries,
shops,
shops,
plants, 6 plants,
6 plants, 6 1907
to
1907
to
190* to
1913.
1913.
1913.

Machine building.
5 plants,
1907 to
1912
(Group
A).

194
plants,
1912
(Group
B).

30,906

15,925

47,412

94,030

0.06

0.13

0.04

0.13

0.05

” ” .‘03

.02

.06
.03

....... *07

.02

.06
.19
.39
1.37
.48
.29
.13

.16
1.61
.54

.10

.06
.15

.02

.06
.16
.32

.30

.13
.06
.38
.29

3.73

4.31

3.45

.02
.21

2.64

.01
.03
.01
.10

.10

.13
.38
1.41
.55
.17
.19
.04
.04

.42
1.62
.49
.19

115,703

.10

.04
.05
.04

.20

.02
.31
.41

3.55

Severity rates (days lost per 300-day worker).
Loss of great to e ..............................
Loss of 1 joint of great to e .............
Loss of other toe or to es.................
Loss of 1 joint of other toe or toes.
Loss of foot......................................
Loss of leg........................................
Loss of thum b.................................
Loss of 1 joint of th um b................
Loss of 1 joint of finger or fingers..
Loss of first finger...........................
Loss of second finger.......................
Loss of third finger.........................
Loss of fourth finger.......................
Loss of hand....................................
Loss of arm......................................
Loss of ey e.......................................
Other permanent injuries.............
Total.

0.05
.(V) " ‘ .

0.05

.01

0.01

0)
0).35

.07
.09
.13
.29
.30
.07
.05

0.06

0).01
(*).24
.22
.13
.14
.29
.26

.11

” .'25
.38

.54
.38
.45

.08
.32
.35

.04
.04
.37
.24
.58
.34

1.74

2.22

2.26

3.07

.18

.01
.12

0.03

.07
.03
.09
.05
.33
.29
.04

.01
.01
.12
.07
.47
.49

2.11

1 Less than 0.005.

The first part of this table makes it possible to compare the fre
quency of the specified forms of permanent injury in three important
departments and in two groups of plants which may be taken as
fairly representing the general industry. Injuries to the hand are
by far the most numerous. Adding together the injuries to the
various parts of the hand, the total numbers of cases of hand injury

C H A PT E R I I .---- ACCID EN T E X PERIEN C E.

per 1,000 workers are: In electrical assembly shops, 2.88; in found
ries, 1.51; in machine shops, 2.91; in group A of machine-building
plants, 3.18; and in group B, 2.64.
Also it may be noted that loss of eye occurs sufficiently often to
attract attention; but its real importance appears more clearly when
comparison is made of the frequency of several injuries, as shown in
the upper part of the table, with the severity, as shown in the lower
part. Such a comparison may be made between hand injuries and
eye injuries. Thus, in machine shops, hand injuries have a fre
quency of 2.91 cases per 1,000 workers, while eye injuries have a
frequency of only 0.21 case. But, as regards severity, hand injuries
show a time loss of ] .06 days per worker, while the time loss from eye
injuries is as much as 0.32 day per worker. That is to say, injuries to
the hand while 14 times as numerous as injuries to the eye are less
than four times as important from the standpoint of economic severity.
This comparison does more than illustrate the value of the severity
rates as a means of more exact analysis of accident hazard. It
points very directly to one of the most serious of preventable acci
dents. With proper precautions eye losses can be reduced almost
to the vanishing point. This has been accomplished by a number
of plants which, at the outset, had much more serious conditions to
face than those confronting any of the plants here included. Indeed,
as has been elsewhere noted, the mere fact that conditions are not
very severe may have a direct tendency to obscure the real impor
tance of preventive effort.
IN A B IL IT Y TO SP E A K E N G LISH A S R E L A T E D TO A CC ID E NTS.

In the first report of the Bureau on accidents in the iron and steel
industry 1 a careful study was presented of the comparative accident
rates of English speaking and non-English speaking workers, the
basis of the comparison being the experience of a large steel plant
over a period of years. The results of that study are shown in Chart
D. From the chart it appears that while the accident rates were re
duced for non-English speaking steel workers as well as those speak
ing English, the improvement in the case of the non-English speaking
workers was much less definite and much less steady.
It is not to be concluded from this fact that the evident handicap
upon the non-English speaking employees is entirely due to their
inability to understand directions and orders. This is unquestion
ably a factor in their less favorable accident rate. But another
factor also enters, namely, that the non-English speaking workers, as
a rule, suffer from lack of experience and thus are found largely in
the group of unskilled occupations involving inherently high accident
hazards.
1 Conditions of Em ploym ent in the Iron and Steel Industry in the U nited States (S. Doc. No. 110,
62d Cong., 1st sess.), Vol. IV.

INABILITY TO SPEAK ENGLISH AS RELATED TO ACCIDENTS
EXPERIENCE OF A LARGE STEEL PLANT, 1906 TO 1913.
SHOWING TH AT ACCIDENTS WERE MORE FREQUENT AND ALSO MORE SEVERE AMONG NON-ENGLISH SPEAKING
WORKERS. THAN AMONG THOSE SPEAKING ENGLISH.
I NON* ENGLISH SPEAKING

A ENGLISH SPEAKING

AC C ID EN T SEVERITY RATES'-

PER 1000 300-DAY WORKERS

PER 3 0 0 -DAY WOHKCR

yy////////////zm

1907 ( f

CHART D.

<40

4 0 .1

7.262

230

3-1.. 5

7.817

101

2 0 .6

6.357

193

2 1 .6

’

4.433

12. A

8.941

198

2 i. a

4.475

i o .3

8.773

20S

I4..I

6.185

19.9

3 1 .3 3 3

2 4 .0

2 2 .9 1 0

BUILDING.

TO-------- Z5------ r s s -------2T5--------5 5 ”

NUMBER
300-DAY
WORKERS

MACHINE

ACCIDENT
FREQUENCY
300-DAY
RATES
WORKER

isos-

ACCIDENT FREQUENCY RATES

ACCIDENTS

V t

C H A PT E R I I .— ACCID EN T EXPE R IE N C E .

For the machine-building industry it was not possible to obtain
such full information as to the effects of inability to Speak English
upon accident rates as was obtainable for the steel industry. Of
much interest, however, as bearing upon the same subject is the fol
lowing table, which compares the accident rates of the Americanborn worker and the foreign-born worker in a large machine-building
plant:
2 2 .—FR EQ U E N C Y A N D SE V E R IT Y OF ACCIDENTS AMONG AM ERICAN
FOREIGN-BORN W ORKM EN IN A M A CH INE-BUILDING
PL A N T D U R IN G
PE R IO D 1910 TO 1913.

T able

1
Number of cases.

Race group.

A ND
THE

Accident frequency rates Accident severity rates
(per 1,000 300-day
(days lost per 300-day
workers).
worker).

N um
ber of
300Per Tem
Per
day
po
ma rary
work
ma
ers. D eath. nent disa Total. Death. nent
in bil
in
jury.
jury. ity.

Tem
Per
po
ma
rary Total. Death. nent
disa
in
bil
jury.
ity.

Tem
po
rary
disa Total.
bil
ity.

American born........ 22,556
Foreign born........... 18,039

11
16

35 1,320 1,366
82 1,737 1,835

0.5
.9

1.6
4.6

58.5 60.6
96.3 101.8

2.9
5.3

0.9
3.4

0 .5
.9

4.3
9.6

T otal.............. 40,595

117 3,057 3,201

2.9

75.3

4.0

2.1

6.8

78.9

The foreign born are not entirely non-English speaking, but the
constant excess of the accident rates of the foreign born, as shown
in the table, may clearly be attributed to causes similar to those
affecting the accident rates of the non-English speaking workers in
the steel industry, referred to above. This conclusion is strengthened
by the accident experience of a group of Polish workers which it
was possible to isolate from the other foreign born. In this Polish
group, consisting of 4,798 300-day workers, is found the greatest
proportion of non-English speakers and also the greatest proportion
of those engaged in common labor. The accident frequency rate of
this group was 115 cases per 1,000 workers and the severity rate 13.5
days lost per worker. These are distinctly higher than the rates for
the foreign born as a whole (101.8 and 9.6 days).
D A Y A N D NIGH T ACCIDENT B A T E S.

The question of accident distribution through the hours of the day
has been illustrated elsewhere by so many and such extensive com
pilations that no special study of it need be made here. Attention
will be chiefly confined to the allied question of day and night accident
distribution, as illustrated by such limited data as could be obtained
from the machine-building plants covered.1
1 A considerable body of additional data regarding day and night accidents has recently been accumulated
by the United States Bureau of Labor Statistics and will be embodied in a later report.

ACCIDENTS IN

M A C H IN E B U IL D IN G .

The following table shows, by hours of the day and night, the
distribution of 6,075 accidents in a large machine-building plant in
1913. This number is chiefly composed of nondisabling accidents,
for which class of accidents full reports were available in this plant.
For the purpose of studying distribution of accidents those of a non
disabling character are just as useful as those causing disability.
2 3 . —D IST R IB U T IO N T H R O U G H TH E D A Y A N D NIG HT OF D ISA BLIN G A N D
N O N D ISA BLING IN JU R IE S IN A M ACHINE-BUILDING PL A N T , 1913.

T able

Nondisabling
injuries.

Disabling
injuries.

Hour ending at—
Day.

Day.

Night.

7 .................................
8 .................................
9 .................................
10...............................
11...............................
12...............................
1 .......................
2 .................................
3 .................................
4 ........................
5 .................................
6 .................................

31
362
499
628
574
396
263
463
510
429
290
80

43
53
44
52
42
25
14
46
36
32
22
27

9
87
102
159
119
98
41
84
107
93
72
29

19
20
19
16
10
8
1
4
6
3
5
3

T otal.................

4,525

436

1,000

114

As regards the hourly distribution of accidents shown in the table,
it is sufficient to note that it conforms entirely to the general type of
the compilations hitherto made.
There are two peaks of accident occurrence, one in each half of the
working period, with the peak tending to come earlier in the second
half.
For the purpose of accurate comparison of day and night accidents,
the data given in the preceding table are presented in the next table
in the form of day and night frequency rates.
T able

2 4 .—COMPARISON OF D A Y A N D NIG HT ACCIDENT R A TE S IN A MACHINEBU IL D IN G PLA N T, 1913.

Number of 300-day
workers.

Frequency rates
(cases per 1,000 300day workers).

Cases of injury.

Class of accident.
Day.

Night.

Nondisabling..................
Disabling.........................
T otal......................

13,359

881

Day.

Night.

Day.

Night.

4,525
1,000

436
114

338.73
74.86

494.89
129.40

5,525

550

413.59

624.29

The excess in night frequency rates is very marked for both non
disabling and disabling accidents. Combining both classes of acci
dents, the frequency rate appears as 413.59 cases for dayworkers as
against 624.29 cases for night workers. The night rate is thus almost
exactly 50 per cent higher than the day rate.

NIGHT AND DAY ACCIDENT RATES
EXPERIENCE OF A LARGE STEEL PLANT.
PART 2

BY DEPA R TM EN TS,

(co m b ine p d a ta f o r i? o s t o 1913 )

SHOWING MARKED EXCESS OF NIGHT ACCIDENTS OVER DAY ACCIDENTS. AS REGARDS BOTH FREQUENCY AND SEVERITY

NIGHT (6 P.M.TO 6 A .M .)

IY/A
A C C ID E N T

DAY (6 A.M. TO 6 P. M.)

SEVER ITY

R A TE S

PER 300-DAY WORKER

PER 1000 300-DAY WORKERS

“ TO--------20------ 3 0 ------- 3 0 --------50------ 50 --------70------- SO-

ACCIDENT oays l o s t NUMBER
FREQUENCY
per
3 0 0 -DAY
RATES
3 0 0 -DAY WORKERS
WORKER

229

5 5 .6

2,879

204

3 1 .0

6.413

S TE E L
WORKS

224

35.0

3.538

218

25.1

5.045

160

»7.2

6.S16

ISO

I4 -.7

10.789

•89

36.8

556

15.9

7.767

R O LLIN G
M IL L S

MECHANICAL

YARDS

122
299
>39

y //////////////M

OTHER
DEPARTMENTS
E N TIR E
PLAN T

1 0 ----- ZO— 50----- V 5-----50— 5 0 -----70-----50“"

CHART E.

7A..7

87 4

9.2

5.083

3 2 .9

1.209

12. t

9.880

214

3 5 .2

1 5.672

142

18.1

44 ,9 7 7

II.------------------------------------------------------------------------------------------------------

BLAST
1
FURNACES

CHAPTER

ACCIDENT FREQUENCY RATES

Cn
CO

ACCIDENTS IN

M A C H IN E B U IL D IN G .

That the excess of night accident rates over day rates is true of the
individual departments as it is of the plants as a whole is indicated in
the following table, which presents such information as could be
secured on this point. The table gives data and accident frequency
rates for three important departments of a large plant for the years
1907 and 1910 combined.
2 5 . —COMPARISON OF D A Y A N D N IG H T A C CIDENT R A T E S IN A M A C H IN E -B U ILD
ING P L A N T FO R TH E Y E A R S 1907 A N D 1910 COM BINED, B Y D E P A R T M E N T S .

T able

Number of 300-day
workers.

Frequency rates
(cases per 1,000 300-day
workers).

Cases of injury.

Department.
Day.

:Tight.

Day.

Night.

Day.

Night.

Boiler..........................................
E recting.....................................
Machine shops...........................
Other..........................................

2,090
6,596
4,947
6,073

937
2,955
2,120
1,651

347
266
460
422

173
213
512
228

166.03
40.32
92.98
69.49

184.63
72.08
243.40
138.10

T otal................................

19,706

7,663

1,495

1,126

75.87

146.94

It is of interest to compare the experience of the machine industry,
in this matter of night and day accident rates, with the experience
of the iron and steel industry as shown in Chart E. In the latter
industry the excess in night rates over day rates is shown to be
constant.
It is also of interest to compare the rates for machine shops in
machine building, as shown in the preceding table, with the mechan
ical department of the iron and steel industry as shown in the chart,
the two departments resembling each other in character of work.
The mechanical department of the steel industry shows a night
accident rate of 389 cases per 1,000 300-day workers, as against a
day rate of 122 cases. The corresponding figures for the machine
shops of the machine-building industry are: Night, 243.4; day, 92.98.
In both departments, therefore, the night accident rate is very much
higher than the day accident rate.
The constant excess of night accident rates over day accident
rates in all of the examples available indicates that such result is
due to the operation of definite causes. Two such causes suggest
themselves as of importance: (1) That the artificial light of the
night is not equal to natural daylight, and (2) that the physical
condition of the night worker is not so good as that of the man on
daywork. This comes, in part at least, from the fact that it is quite
impossible that conditions of restful sleep can be furnished in the day
comparable with those of the night.
In any case, whatever may be the causes of higher night accident
rates, the subject demands serious attention. A good deal of excel
lent work has been done on the lighting problem. Much remains to
be done in making known and usable the information available.

CH A PT E R I I .---- ACCID EN T E X PERIEN C E.

D ISTR IB U T IO N OF A CC ID ENTS B Y MONTHS.

There exists some disposition to regard an inquiry into the monthly
distribution of accidents rather as one of curious interest than of
practical utility. Nothing could be further from the fact. If it be
found that there is a constant monthly or seasonal peak of accidents,
it will be an indication of definite climatic or operating conditions
whose effects can be determined and against which provision can in
a measure be made. Data upon this subject were obtainable from
three machine-building plants engaged in different classes of manu
facture. Plant A produces engines, plant B produces electrical
apparatus, and plant C produces machine tools. The monthly
accident experience of these three plants is shown in the following
table. The experience of a large steel plant is added for purposes of
comparison.
T able

2 6 . —D IST R IB U T IO N OF ACCIDENTS IN THE M ONTHS OF THE Y E A R FO R 3
M ACHINE-BUILDING PL A N T S A N D 1 S T E E L PL A N T .

Number of 300-day workers.

Frequency rates (cases per
1,000 300-day workers).

Number of injuries.

M onth.
Plant Plant Plant Steel Plant Plant Plant Steel Plant Plant Plant Steel
A.
B.
C.
A.
C.
c. plant.
B.
plant.
B.
A.
plant.
January............
February..........
March...............
A pril.................
M ay..................
Jun e..................
J u ly ..................
A ugust.............
Septem ber.......
October............
N ovem ber........
December.........

3,199
3,304
3,366
3,310
3,420
3,409
3,313
3,336
3,362
3,339
3,349
3,210

492
358
447
485
580
616
705
745
792
675
633
567

212
199
196
213
237
233
206
265
247
292
180
197

88
84
99
104
105
116
95
54
90
97
66
67

663
668
641
620
618
634
644
743
640
695
577
607

78.21
63.41
80.92
85.39
99.21
105.37
117.70
117.12
121.85
104.88
99.54
93. 61

67.88
64.99
63.45
68.84
76. 75
74.63
65.38
83.20
75. 58
88.48
53.13
58.15

40. 31
38.06
44.51
46. 78
47.17
51.86
42.62
54.77
41.32
44.03
29.97
30.50

207.25
202.18
190.43
187.31
180.70
185.98
194.39
222.72
190.36
208.15
172.29
189.10

T otal___ 72,536 38,258 25,295 39,917

7,095

2,677

1,065

7,750

97.81

69.97

42.10

194.15

6,291
5,646
5,524
5,680
5,846
5,846
5,990
6,361
6,500
6,436
6,359
6,057

3,123
3,062
3,089
3,094
3,088
3,122
3,151
3,185
3,268
3,300
3,388
3,388

2,183
2,207
2,224
2,223
2,226
2,237
2,229
986
2,178
2,203
2,202
2,197

Inspection of the table brings out the fact that the high points in
these four groups occur as follows: Two in August, one in September,
and one in October. This concentration of the high points of acci
dent frequency becomes more definite if the data are combined in
groups of three months each—January to March, February to April,
etc. If this is done, it is found that in the three machine-building
plants the group of three months having the least frequency is always
a group including January, while the highest frequency is always in
a three-month group, including the month of August. The highest
point for the steel plant is also in the three-month group, including
August.
These facts, taken with the occurrence of the actual peak in two
plants in August and in one in September, seem to indicate that the
depressing influence of summer heat and humidity may be a factor
in the accident hazard. The use of ventilating fans may, therefore,

ACCIDENTS IN M A C H IN E B U IL D IN G .

be considered not only a needed contribution to the comfort of hardpressed workmen, but also as a distinct safety device. At the same
time it should be kept in mind that there may be factors, other than
heat, concealed in these high rates which may appear, upon further
study, to be of decisive importance.
One further point in connection with this subject may be com
mented upon. In the preceding table the steel plant shows a peak
for January as well as for August. An important compilation of
fatal cases shows, for three years, a peak in December and January,
with a smaller rise in the warm months.1 An explanation of this
December-January peak has been the deficiency of natural light
due to the shortness of the day and to cloudiness. The possible
importance of this factor is obvious. There is another which must
be influential. This is low temperature. The steel industry has
many occupations exposed to the outdoor cold. This has a twofold
effect upon accident hazard. First and directly, the cold benumbs
the muscles, and thus renders the worker less expert; second, the
use of gloves and mittens, made necessary by the cold, must at
times increase the liability to accident. A better design of glove
might have an importance similar to that of the foundry man’s shoe,
which has so much reduced foundry burns.
GOVERNMENT A R SE N A LS AN D N A V Y Y A R D S.

The navy yards and arsenals operated by the United States
Government are machine-building plants, and thus within the scope
of this study. But important differences in the character of the
accident records obtainable render difficult a comparison of the
accident experience of the two classes of plants.
Under the Federal compensation act of 1908 accident reports from
Government shops were made to the Bureau of Labor Statistics, which
was charged with the administration of that law. But until 1912
there was no available record of exposure— i. e., of the number of
persons employed— and, in consequence, no accident rates could be
computed for the earlier years. For the years 1912, 1913, and 1914
exact information regarding employment is available. But even with
this information at hand it has not been possible to compute accident
rates exactly comparable with those presented above for private plants,
owing to the fact that the accident reports from the Government
shops appear, on analysis, to be extremely incomplete for disabilities
of less than two weeks’ duration. The evidence for this statement
is presented in the next section. But there seems no doubt as to the
fact itself, and, because of it, it is necessary, in contrasting the acci
dent rates in Government shops with those in private plants, to
exclude all accidents causing disabilities of less than two weeks.
1 John, Calder, in Journal of American Society of Mechanical Engineers.

C H A PT E R II .— ACCIDEN T E X PE R IE N C E .

Such computations have been made for a few important groupings,
and are presented in the next table. This table shows accident
frequency and accident severity rates in Government arsenals and
navy yards for the three years 1912 to 1914, and brings into compari
son therewith the corresponding rates for the private machinebuilding plants as a whole and also for the shipbuilding departments
of those plants for the year 1912. These rates, it must be remem
bered, are based on the exclusion of disabilities of under two weeks’
duration, and thus differ from the rates earlier given, which are
based on all disabilities of a duration of one day and over.
T able

2 7 . —FR E Q U E N C Y A N D SE V E R IT Y OF ACCIDENTS IN A R SE N A L S A N D N A V Y
Y A R D S, 1912 TO 1914, A N D IN MACHINE A N D SH IP B U IL D IN G P L A N T S, 1912.
Accident f r e q u e n c y
rates (per 1,000 300day workers).

Number rf cases.

Year.

N um
ber
of
300day
work Death.
ers.

Accident s e v e r i t y
rates (days lost per
300-day worker).

Tem
Tem
Tem
Per po
Per po
Per po
ma rary To
ma rary To
ma rary To
nent disa tal. Death. nent disa tal. Death. nent disa tal.
in bility
in bility
in bility
jury: over 2
jury. over 2
jury. over 2
weeks
weeks
weeks

Arsenals:
1912.........................
1913....................
1914.........................

3,992
3,950
4,612

1
3
1

10
13
14

192
190
226

203
206
241

0 3
.8
.2

2.5
3.3
3.2

48.1 50.9
48.1 52.2
49.0 52.4

1.5
4.5
1.3

1.1
2.7
1.6

2. 0
1.2
1.6

4.6
8.4
4.5

Total...................

608

650

2.9

48.4 51.7

2.4

1.8

1.2

5.4

25 1,063 1,107
32 1,181 1,227
32 1,058 1,102

1.2
.9
.8

1.6
2.1
2.1

68.1 70.9
77.6 80.6
70.1 73.0

6.5
5.5
4.8

.9
1.8
1.3

2. 5
2.6
2.5

9.9
9.9
8.6

12,554

N a v y yards:
1912......................... 15,608
1913......................... 15,226
1914....................
15,094

19
14
12

Total.................... 45,928

89 3,302 3,436

1.0

1.9

71.9 74.8

5.9

1.3

2.5

9.7

Machine building......... 115,703
Shipbuilding................. 6,615

37
3

411 3,279 3,727
15
635 653

.3
.5

3.6
2.3

28.3 32.2
96.6 99.4

1.9
2.7

2.1
1.6

.7
1.7

4.7
6.0

The first comparison suggested is between the two classes of
Government shops. The rates for the arsenals show a much smaller
degree of hazard than those for the navy yards— the frequency rates
for the three-year period being for the arsenals 51.7 cases per 1,000
300-day workers as against 74.8 cases for the navy yards; and the
severity rates being for the arsenals 5.4 days lost per worker as
against 9.7 days for the navy yards. This is in accordance with the
known character of the relative hazards of the work done.
In attempting to contrast the accident rates of Government shops
with those of private plants, the most obvious comparison is that be
tween Government navy yards and private shipbuilding plants. The
Government yards have the lower frequency rate— 74.8 cases per
1,000 workers for the three-year period, as against 99.4 cases in private
yards in 1912. But this condition is reversed when comparison is
made by severity, the severity rate for the Government shops for
124566°— 20— Bull. 256------ 5

ACCIDENTS IN M A C H IN E B U IL D IN G .

the three-year period being 9.7 days as against only 6.0 days for the
private yards. The high severity rate in Government yards is due,
in large part, to a very high fatality rate. On the other hand, the
group of employees for private yards is small and the fatality rate,
based on a small number of deaths, may be abnormal, although a
careful study of the plants is convincing that they fairly represent
the actual conditions in private yards.
There is no particular group of the private plants studied with
which the Government arsenals may fairly be compared. The
variety of their activities places them more on a par with the entire
machine-building industry. Comparing these two groups, it will be
noted that the arsenals have a higher frequency rate— 51.7 cases as
against 32.2 cases per 1,000 workers, and also a higher severity rate—
5.4 days lost against 4.7 days lost per worker in machine-building
plants.
This somewhat unsatisfactory showing of the Government plants
calls for careful consideration. The following comments upon this
point, it should be clearly understood, apply to the years 1912 and
1913, during which the working conditions of the Government
plants were reviewed. Changes are known to have occurred since
that time which may well account for the steady reduction in severity
shown by later navy-yard reports. The reduction of the originally
lower accident rates of the arsenals would necessarily be a more
difficult matter.
The following points impress the outside observer as highly favor
able to a low accident rate in Government plants:
First, extreme orderliness and cleanliness. Only the very best
private plans approach the Government shops in this particular.
Second, freedom from violent fluctuations in the amount of employ
ment. This is illustrated by the very constant number of 300-day
workers shown in the three years under consideration. Under normal
conditions the demands for output are fairly constant, but if war were
threatened and the work were speeded up accordingly the rising
accident curve which always accompanies the beginning of increased
activity would no doubt occur.
Third, the quality and stability of the working force. The certainty
of regular employment attracts men of skill and serious purpose.
Perhaps no private employer can hope to maintain a force of such
high quality.
Fourth, the reasonable working hours. The eight-hour day pre
vails. It may be that individual workmen abuse the leisure that the
shorter day affords and are more liable to accident on that account,
but anyone who contends that this condition is general has given the
matter very superficial consideration and is basing conclusions upon
striking cases instead of upon a solid body of facts.

C H A PT E R I I .— ACCID EN T EX PE R IE N C E .

The conditions tending to counteract the favorable effect of the
items named above are the following:
First, imperfect mechanical safeguarding. Improvements in this
respect began before 1912, but there were still at that time in many of
the Government shops conditions that the expert safety man in
private employ would view with the greatest surprise.
Second, lack of safety organization. In 1912 it was not possible to
find in the plants visited anything remotely resembling the method
of inspection, committee work, and safety education which had then
become nearly universal in the iron and steel industry, railways, and
mines and which was taking root in other branches of industry.
Third, and strongly influencing the other two, an honest conviction
on the part of the supervising authorities that such accidents as
occurred were without remedy. The old view of such accidents led
to constant search for some reckless behavior on the part of the man
employed. Of course, such instances could be found and steady
attention to them developed a conviction that the fundamental cause
of accidents was a hopelessly ingrained carelessness. Nothing has
become clearer during the progress of the safety movement than the
effective response of the men who do the work. The results attained
are largely due to this response and the supervising man who does
not recognize this fact and act upon it must himself be recognized
as hopeless.
INCOMPLETE REPORTING B Y GOVERNMENT SHOPS.

The accident reports from the Government shops, under the Federal
compensation act of 1908, show uniformly what appears to be an
exceedingly large proportion of injuries terminating in the third week.
In most instances the number reported as terminating in the third
week is greater than the number for the second week. This has
frequently been interpreted as indicating a practice on the part of
injured workers of stretching short-time disabilities into the third
week in order to benefit from the compensation act, which allowed
no compensation for the first two weeks of disability but gave full
wages for all of the time lost if the disability extended over 15 days.
A careful analysis of the accident reports indicates that the exces
sive proportion of injuries reported as terminating in the third week,
as well as other peculiarities in their distribution, can be much more
logically explained on the ground that there was a gross deficiency in
the accident reports for short-time disabilities. This analysis is
briefly as follows:
The distribution of disabilities in the Government shops, according
to week of termination, is shown by numbers in part 1 of Table 28,
and by percentages in part 2 of that table.1 There are also shown
i The Government shop data used as a basis for this discussion are given in B ulletin No. 155, U . S.
Bureau of Labor Statistics (report on operation of the Federal compensation act).

ACCIDENTS IN

M A C H IN E B U IL D IN G .

the corresponding data for the iron and steel industry and the
machine-building industry.
T able

2 8 __ACCIDENT R E PO R T S IN GOVERNM ENT SHO PS A N D IN TH E IRO N A N D ST E EL
A N D TH E M ACH INE-BUILDING IN D U ST R IE S.
P a r t 1.—Number o f d isa b ilities term inating in specified w eek.
Government shops.

Week in which disability
terminated.

Arsenals.

Iron and Machine
building
steel
(1910).
(1912).

N avy yards.

1912

1913

1914

1912

1913

1914

Number of 300-day workers..........

3,992

3,950

4,612

15,608

15,226

15,094

65,147

115,703

First w eek.......................................
Second w eek...................................
Third w eek.....................................
Fourth week...................................
Fifth week......................................
Sixth week and later.....................

89
27
57
57
15
55

138
26
69
52
24
36

197
. 46
89
61
19
57

535
153
339
257
129
320

534
136
432
271
125
304

501
140
362
240
132
321

9,889
4,433
1,915
1,014
807
1,251

7,680
2,048
869
512
272
621

Total......................................

300

345

469

1,733

1,802

1,696

19,309

12,002

P a r t 2.—P ercen tages.
First w eek.......................................
Second w eek...................................
Third w eek.....................................
Fourth week...................................
Fifth week......................................
Sixth week and later.....................
Total...................................... 1

30
9
19
19
5
18

40
8
20
15
7
10

42
10
19
13
4
12

31
9
20
15
7
18

30
8
24
15
7
17

30
8
21
14
8
19

51
23
10
5
4
6

64
17
7
4
2
5

100

P a r t 3.—P ercen tages (excluding all under th e third w eek ).
Third week ...................................
Fourth w eek...................................
Fifth week......................................
Sixth week and later.....................

31
31
8
30

38
29
13
20

39
27
8
25

32
25
12
31

38
24
11
27

34
23
13
30

39
20
16
25

38
23
12
27

Total......................................

100

P a r t 4.—A ccid en t frequency r a te s (per 1,000 300-day w orkers).
First w eek.......................................
Second w eek...................................
Third w eek.....................................
Fourth week................................. .
Fifth week......................................
Sixth week and later.....................

22
7
14
14
4
14

33
7
18
13
6
9

43
10
19
13
4
12

34
10
22
17
8
21

35
9
28
18
8
20

33
9
24
16
9
21

152
68
29
16
12
19

66
18
8
4
2
5

Total......................................

101

112

118

112

236

103

A study of the data of these tables shows some striking facts.
First, it will be noted that the percentage of injuries terminating
in both the first and second weeks is very much smaller for the Gov
ernment shops than for the steel or machine-building industry.
Thus, taking the experience of the navy yards for 1914, it appears
that only 38 per cent of all cases are reported as terminated in the
first two weeks, whereas in the steel and machine-building industries
the percentages were, respectively, 74 and 81. Nor is this the only
striking peculiarity. For injuries terminating in the sixth week and

C H A PT E R I I .---- ACCIDEN T E X PERIEN C E.

later the navy yards show a percentage of 19 as against 6 and 5,
respectively, in the steel and machine-building industries.
These comparisons themselves would indicate probable error in the
reports for the Government shops. The probability becomes even
stronger when the comparisons are based upon the accident reports
for the third week and over, all of those for the first and second
weeks being excluded. This is done in part 3 of the table. It is
there seen that when the first two weeks are excluded the experience
of the Government shops is substantially the same as that of the
steel and machine-building industries. It may be particularly noted
that the excessive percentage of Government shop disabilities ter
minating in the sixth week and over disappears, becoming 30 as
against 25 and 27, respectively, for the steel and machine-building
industries.
This substantial harmony in the distribution of disability periods
for three distinct industrial groups is a strong argument for the basic
accuracy of such distribution. If so, there is nothing abnormal in
the percentages for the Government shops for injuries terminating
in the third and later weeks. For the short-time disabilities, how
ever, the distribution for the Government shops is so abnormal that
it seems impossible to explain it except on the ground of extremely
faulty reporting.
The comparisons so far made have been in the form of percentages.
If, in place of percentages, accident frequency rates are used, the con
clusion as to the incompleteness of reporting becomes even more
evident. Part 4 of the table shows the accident rates distributed
b y week of the termination of disability. Thus, the total accident
rate for navy yards in 1914 was 112 per 1,000 300-day workers. Of
these 112 accidents per 1,000 workers, 33 caused disability of less
than a week, 9 caused disability of between one and two weeks, and
so on. Rates of the same character are shown for the steel and
machine-building industries.
Comparing the data in the last three columns of part 4 of the
table, the most striking fact is that, for disabilities terminating in
the third and later weeks, the accident rates in the navy yards are
practically the same as those for the steel industry, the respective
rates being: For the third week, 24 against 29; for the fourth week,
16 against 16; for the fifth week, 9 against 12; and for the sixth and
later weeks, 21 against 19. This close harmony of experience for the
third and later weeks would suggest, with a reasonable degree of
conclusiveness, that the true accident frequency in Government shops
is about the same as that in the iron and steel industry. If this is
so, then there should be a similar harmony in accident rates for the
first and second weeks, inasmuch as there is nothing in the character

ACCIDENTS IN

M A C H IN E B U IL D IN G .

of the work in the Government shops to warrant radical departure
from the experience of other industries. Examination of the acci
dent rates for the first and second weeks, however, shows extraordi
nary lack of harmony. For the first week the accident rate in navy
yards for 1914 was only 33, according to the reports, as against 152
in the steel industry, and for the second week only 9 as against 68.
Inasmuch as it is known that the accident rates of the steel industry
err, if at all, in the direction of being too low for the early weeks, the
conclusion seems clear that the rates as shown for the navy yards
(as also for the arsenals) are entirely too low, an error that could
only be explained by failure to report short-time disabilities in full.
Estimating the true situation from the data of the table it would
appear that perhaps as many as three-fifths of the accidents having
two weeks and less of disability in Government shops were not
reported.
IN FLU EN C E OF THE W A R ON ACCIDENT R A T E S IN M ACH INE
B U IL D IN G .

One of the most interesting questions at the present time concern
ing accident experience is in regard to the effect of the great inten
sification of industrial effort incident to the war. The review of
this subject recently published for the iron and steel industry in
Bulletin 234 of the United States Bureau of Labor Statistics brings
out many points of interest.
In order to compare the war conditions with those disclosed by
Bulletin 216 for the year 1912, the 194 plants whose experience was
utilized in the preparation of that report were requested to furnish
similar information for 1917. Of the 194 plants which furnished
the 1912 data, 94 declined the preparation of the desired information
on the ground of excessive pressure on their clerical forces. Fortu
nately the 100 who could furnish information were so distributed in
the industry that all important branches were adequately represented.
The equivalent of 94,103 300-day workers were employed in the
100 establishments, or 81 per cent of the number included in the
earlier collection of data. This gives a better idea of the relative
importance of the second survey than it does to compare the number
of plants. It was chiefly the larger plants that reported, and these
had in so many cases materially increased the number employed that
some departments are represented by a greater volume of employ
ment than was the case in the earlier report.

C H A PT E R I I .---- ACCID EN T EXPE R IE N C E .

The following table presents in a form for ready comparison the
figures for the two years 1912 and 1917 classified by products:
T able 2 9 . —COMPARISON

M ACHINE-BUILDING
PRODUCTS.

OF FR E Q U E N C Y A N D SE V E R IT Y ' OF ACCIDENTS IN 194
PL A N T S IN 1912, A N D 100 PL A N T S IN 1917, C L ASSIFIED B Y

Accident
frequency
rates (per 1,000 300day workers).

Number of cases.

Product.

N um
ber of
300-day
work
ers.

Electrical:
1912.................. 35,674
1917.................. 29,695
Locomotives:
1912.................. 31,229
1917.................. 25,322
Machine tools:
1912.................. 24,359
1917.................. 20,185
Ships:
1912..................
6,615
1917..................
5,932
Mining machinery:
1912..................
3,994
1917..................
4,227
Cranes, etc.:
1912..................
4,362
1917..................
1,723
Transmission:
1912..................
2,226
1917..................
1,842
Unclassified:
1912..................
7,244
1917..................
5,177
Total:
1912.................. 115,703
1917.................. 94,103

Accident severity r rates
(days lost per 3 00-day
worker).

Per Tem
Per Tem
Per Tem
ma po
ma po
ma po
rary
nent rary Total. Death. nent rary
Total.
Death. nent
Death.
dis dis
dis dis
dis dis Total.
abil abil
abil abil
abil abil
ity. ity.
ity. ity.
ity. ity.

5
4

100 3,455 3,560
40
989 1,033

0.1
.1

22
17

160 4,348 4,530
80 2,268 2,365

3
7
3
7

99.7
34.7

0.7
.8

1.1
.8

1.0
.6

2.8
2.2

.7
.8

5.1 139.2 145.0
3.2 89.6 93.6

4.2
4.3

2.9
3.5

1.4
1.4

8.5
9.2

68 1,486 1,557
63 2,495 2,565

.1
.3

2.8 61.0 63.9
3.1 123.6 127.0

.7
2.1

.8
2.6

.5
1.3

2.0
6.0

15 1,422 1,440
10
361
378

.5
1.2

2.3 215.0 217.8
1.7 60.9 63.8

2.7
7.1

1.6
2.3

2.3
1.4

6.6
10.8

1.1
.9

1.9
2.1

3.0
3.0

12
6

755
794

767
800

1
2

15
10

813
376

829
388

.2
1.2

9
7

186
176

195
184

3
2

32
35

734
769
967 1,004

411 13,199 13,647

251 8,426 8, 717

2.8
1.3

96.8
33.3

1
i

3.0 189.0 192.0
1.4 171.3 172.7
3.4 186.4 190.0
5.8 218.2 225.2

1.3
7.0

.5
3.2

1.6
2.5

3.4
12.7

4.0
3.8

85.6 89.6
96.5 100.8

3.3

2.1
3.3

1.1
1.2

3.2
7.8

.4
.4

4.4 101.3 106.1
5.7 186.1 192.2

2.5
2.3

2.0
3.8

1.0
2.0

5.5
8.1

3.6 114.1 118.0
2.7 89.5
92.6

2.6

1.9

2.1
2.3

1.1
1.2

5.1
6.1

The products are arranged in the order of the volume of employment.
A comparison of the totals brings out immediately a condition which
prevails in several of the product groups. Accident frequency de
clines from 118.0 cases per 1,000 300-day workers to 92.6 cases.
The severity rate, on the contrary, rises from 5.1 to 6.1 days per
300-day worker, or 20 per cent increase. When the items are ex
amined, it becomes evident that this increase is due to a greater
prevalence of fatality and an increased seriousness of the permanent
disabilities.
This decline in frequency with a rising severity is a very common
phenomenon. It in part arises because of a measure of self-deception based on the use of frequency rates. A company which ob
serves a marked decline in frequency is apt to have a feeling of com
placency which may entirely obscure the fact that the serious element
in the occurrences has so far increased as to more than wipe out the
benefit of the lessened frequency.

ACCID ENTS IN

M A C H IN E B U IL D IN G .

It was pointed out that in 1913 several concerns showed a definite
lessening of severity. It is evident that the stress of war activity not
only overcame the. advantage gained but pushed severity to a point
as high or higher than that of 1912.
There are some exceptions to this general experience which de
serve special mention. In the manufacture of electrical apparatus
the frequency rate declined from 99.7 cases per 1,000 300-day work
ers to 34.7 cases, a 65 per cent decrease. The severity rate declined
from 2.8 days per worker to 2.2 days, or 21 per cent. Here exclusive
attention to the frequency showing might induce greater optimism
than the facts really justify.
In the manufacture of machine tools both frequency and severity
increased. Frequency rose from 63.9 cases to 127.0, an increase of
99 per cent; severity from 2.0 days to 6.0 days, an increase of 200
per cent.
The other products are not represented by a sufficient body of
employment to justify placing emphasis on their showing, although
the rates may be accepted as sufficiently typical to give a fairly
accurate idea of the trend during the highly active year of 1917.
The following table presents the same data classified by depart
ments :
3 0 . —COMPARISON OF F R E Q U E N C Y A N D SE V E R IT Y OF ACCIDENTS IN 194 MA
CHINE-BU ILD IN G P L A N T S IN 1912, A N D IN 100 PL A N T S IN 1917, C L A SSIFIED BY
D E PA R TM E N TS.

Table

A c c i d e n t frequency
rates (per 1,000 300day workers).

Number of cases.

Department.

Machine shops:
1912..................
1917..................
Foundries:
1912..................
1917..................
Erecting:
1912..................
1917..................
Boiler shops:
1912..................
1917..................
Forge shops:
1912..................
1917..................
Woodworking:
1912..................
1917..................
Maintenance:
1912..................
1917..................
Yards:
1912..................
1917..................
Unclassified:
1912..................
1917..................

Accident severity rates
(days lost per 300-day
worker).

Num
ber of
Per Tem
Per Tem
300-day
work
ma po
ma po
ers. Death. nent rary Total. Death. nent rary Total. D eath.
disa disa
disa disa
bil bil
bil bil
ity. ity .
ity. ity.

Per Tem
ma po
nent rary
disa disa Total.
bil bil
ity. ity.

37,595
39,634

7
15

132 3,923 4,062
126 3,998 4,139

0.2
.4

3.5 104.4 .108.1
3.2 100.9 104.5

1.1
2.3

1.4
2.4

0.9
1.1

3.4
5.8

12,307
7,245

4
4

49 1,670 1,723
17 1,211 1,232

.3
.6

4.0 135.7 140.0
2.3 167.1 170.0

1.9
3.3

2.1
2.3

1.4
1.9

5.4
7.5

11,373
4,991

6
3

56 1,992 2,054
15
392
410

.5
.6

4.9 175.2 180.6
3.0 78.5 82.1

3.1
3.6

2.8
4.1

1.9
1.0

7.8
8.7

2,994
2,889

6
5

29
10

636
261

671
276

2.0
1.7

9.7 212.4 224.1
3.5 90.3 95.5

12.0
10.4

6.3
3.2

2.4
3.2

20.7
16.8

2,776
2,869

13
16

439
380

455
396

1.1

4.7 158.1 163.9
5.6 132.5 138.1

6.4

2 .8
5.9

1.7
2.1

10.9
8.0

3,571
2,266

27
13

264
190

292
203

7.6
5.7

73.9
83.8

81.8
89.5

1.7

2.4
3 .4

.9
1.5

5.0
4.9

1
1

137
107

138
108

.7
.9

93.3
94.9

94.0
95.8

.8
.5

1.1
1.3

1.9
1.8

7
2

260
108

270
111

2.5
1.0

14.7
5.7

4.3
2.3

2.6
1.6

2 1.6
9.6

97 3,878 3,982
51 1,779 1,842

.2
.4

1.0
2.2

.5
1.4

11
1.0

2.6
4 .6

1,468
1,128
1,221
1,048

3
1

42,398
32,033

7
12

5.7 212.9 221.1
1.9 103.1 106.0
2.3
1.6

91.5
55.5

94.0
57.5

C H A PT E R I I .---- ACCID EN T E X PERIEN C E.

It is of interest to observe that the tendency to lowered frequency
rates with increasing severity is found in three of the large depart
ments, namely, machine shops, erecting, and the unclassified group.
In foundries both frequency and severity increase. The smaller
departments show an encouraging situation in that with but two
slight exceptions both frequency and severity decrease.
Machine shops show an increase in severity of from 3.4 days to
5.8 days per worker, or 71 per cent; foundries, from 5.4 days to 7.5
days, or 39 per cent; erecting shops, from 7.8 days to 8.7 days, or
11 per cent. The most favorable showing is made by yards, in
which days lost per 300-day worker declined from 21.6 to 9.6,
or 56 per cent. Unfortunately the volume of employment is not in
either year sufficient to warrant acceptance of these rates as neces
sarily typical, but it may be concluded that the reduced rate repre
sents a real improvement in the handling of the problems of internal
transportation. It is still true that industrial establishments have
not brought their equipment and their methods of operation in trans
portation up to the standards set for railroads under Federal super
vision.
Boiler shops, which showed next to the highest severity rate in
1912, had the highest in 1917, but the rate declined from 20.7 days
to 16.8 days, or 19 per cent.
R A T E S IN M UNITION PLAN TS.

Much interest naturally exists regarding the conditions in the
large munition works where the force has been assembled under
great pressure and the work expedited by all known means. It
has not been possible to bring together the data on a large scale, but
the facts are available for two large concerns and are given in the
following table:
T able

3 1 . —FR E Q U E N C Y A N D S E V E R IT Y OF ACCIDENTS IN TWO M U NITIO N PL A N T S
A N D IN LOCOMOTIVE B U IL D IN G A N D MACHINE SHO PS.

Number of cases.

Department.

A c c i d e n t frequency
rates (per 1,000 300day workers.)

Accident severity rates
(days lost per 300-day
worker).

N um
ber of
Per Tem
Per Tem
Per Tem
300-day
Work
ma po
ma po
ma po
ers. Death. nent rary Total. Death. nent rary
nent rary
Death.
Total.
disar disa
disar disa
disa disa Total.
bil bil
bil- bil
bil- bil
ity. ity.
ity. ity.
ity. ity.

M unitions.............. 4,000
M unitions.............. 10,580
Locom otives.......... 25,322
Machine shops....... 39,634

2
7
17
15

804
818
12
11 1,331 1,349
80 2,268 2,365
126 3,998 4,139

0.5
.7
.8
.4

3.0 201.0 204.5
1.0 125.8 127.5
3.2 89.6 93.6
3-2 100.9 104.5

3.0
4.0
4.3
2.3

3.9
2.8
3.5
2.4

2.3
3.0
1.4
1.1

9.2
9.8
9.2
5.8

What the prewar rates may have been in the munition plants
there is no means of knowing. In fact there can scarcely be said to
have been such rates, since the expansion and the change of work

ACCIDENTS IN

M A C H IN E B U IL D IN G .

entirely transformed the plants. It is evident that the same influ
ences which produced the increased rates in the building of locomo
tives and engines and in machine shops in general were operative in
munition works. The severity rates of 9.8 and 9.2 days in the muni
tion plants are markedly close to the 9.2 days of the locomotive
builders. The lower rate (5.8 days) in machine shops corresponds
to the lighter and less hazardous operations which form a large part
of the work of machine shops in general.
GOVERNM ENT PLANTS.

As has been noted earlier, the United States arsenals and navy yards
are essentially machine-building concerns. For purposes of com
parison the following table is introduced:
T able

3 2 . —ACCIDENT R A TE S IN A R S E N A L S A N D N A V Y Y A R D S COM PARED W ITH
R A T E S IN MACHINE A N D SH IP B U IL D IN G , 1912 TO 1914 A N D 1917 A N D 1918.

Number of cases.

Year.

Num
ber of
300-day
work
ers.

Arsenals:
1912 to 1914............ 12,554
Machine building:
1912........................ 115,7031
1917......................... 94,103
N avy yards:
1912 to 1914............ 45,928
1918......................... 54,026
Shipbuilding:
6,615
1912........................
5,932
1917........................

Per
ma
nent
Death. dis
abil
ity.

Accident frequency rates
(per
1,000 300-day
workers).

Accident severity
rates (days lost
per 300-d a y
worker).

•
Tem
Per Tem
ma po
p o
rary To Death. nent rary To Death. N on- To
dis tal.
dis dis tal.
fatal. tal.
abil
abil abil
ity .1
ity. ity.

37 2,338 2,380

0.4

2.9 186.2 189.5

2.4

3.6

6.0

37
40

411 13,199 13,647
251 8,426 8,717

.3
.4

3.6 114.1 118.0
2.7 89.5 92.6

1.9
2.6

3.2
3.5

5.1
6.1

45
41

89 12,704 12,838
45 5,273 5,359

1.0
.8

1.9 276.6 279.5
.8 97.6 99.2

5.9
4.6

4.2
2.2

10.1
6.8

3
7

15 1,422 1,440
10
361
378

.5
1.2

2.3 215.0 217.8
1.7 60.9 63.8

2.7
7.1

3.9
3.7

6 .6
10.8

i The reports from arsenals and navy yards in 1912 to 1914 were manifestly incomplete for disabilities under
tw o weeks. These weeks were excluded from the tables there given. For th e sake of more satisfactory com
parison, in this table, later information has been applied to estimating the number under tw o weeks, and
this estimated number is included w ith th e reported number in the temporary disabilities.

No dependable information is at hand regarding conditions in the
arsenals later than the three-year period 1912 to 1914. The rise in
severity rates which occurred in the machine-building concerns in
1917 brought them to the level of the arsenals in the three-year period.
It was not possible in the navy yard data to separate with accuracy
the losses of time due to permanent disability and temporary dis
ability. The severity rates are accordingly combined to form a nonfatal severity rate. Navy yards, considered as a unit, made a marked
improvement in both frequency and severity between the three-year
period and 1918. The decline is 65 per cent in frequency ancj. 33 per
cent in severity. The private shipyards for which data were avail
able declined in frequency but increased in severity. The number of

C H APTER I I .---- ACCIDEN T E X PERIEN C E. *

300-day workers covered is not, however, large enough to permit posi
tive conclusions.
The general trend upward of severity rates in machine-building
concerns during the war period, when compared with a declining
severity in the navy yards, is very significant. In both machinebuilding operations and naval construction and repair the period was
one of intense activity. Relatively inexperienced men were con
stantly being employed and much overtime work was being done.
Under these conditions machine building suffered a slight but definite
increase in severity, while the naval establishments made a very
decided decrease. No explanation can be offered for this condition
except the earnest effort of the officers in command, the hearty co
operation of the entire management, and the skill and energy of the
safety engineers in direct charge.
On the organization of the United States Employees’ Compensation
Commission a complete survey of all Government establishments was
undertaken at the instance of Mr. Lew R. Palmer, then president of
the National Safety Council. Under the direction of Mr. Arthur H.
Young, director of the American Museum of Safety, 30 or 40 of the
most experienced safety men in the country conducted this survey.
The suggestions developed by this survey were largely carried out
immediately. With the cooperation of the Bureau of Standards and
the Department of Labor a complete system of safety codes was
evolved. All new construction was, as far as possible, subjected to
critical review from the safety standpoint. The results of this effort
are undoubtedly reflected in the improvement in rates noted. It
should be stated that this improvement is contributed by nearly all
the yards and in a few cases which show a rising rate it is due to the
kind of* calamity against which all precautions are from time to
time ineffective.
This statement should not close without special mention of the
group of safety engineers who were drawn into this service and to
whose unstinted efforts it is fair to attribute a considerable portion of
credit for the really remarkable results secured. There has never
been a time when it was more likely, and in a sense excusable, for acci
dent rates to run riot. That they have not done so in any case so
far covered by the Bureau’s studies, and that in notable instances, in
the face of the difficulties, reductions have been possible is a guaranty
of the essential soundness of accident prevention effort.
Evidence is not wanting in the material considered in this brief
review that some of the engineers who share in the credit for this
excellent showing are still too much impressed by the human factor
and attribute too much to so-called “ carelessness.”
Not until this idea is cast out of the mind of the men directly in
charge will engineers get down to the real fundamentals of accident
prevention, namely, (1) adequate “ engineering revision” and (2)

ACCIDENTS IN

M A C H IN E B U IL D IN G .

proper supervision of men and instruction in safe methods of doing
work.
COMPARISONS W IT H TH E IRON AN D STEEL IN D U S T R Y .

The foregoing statement of facts leads naturally to some compari
sons with the iron and steel industry, in which much more extensive
studies have been made.
*
The following table presents the experience of a portion of the iron
and steel industry in the period 1907 to 1917:
T able

3 3 . —ACCIDENT R A TE S IN A PO R T IO N OF TH E IRO N A N D ST E E L IN D U S T R Y ,
1907 TO 1917.

Number of cases.

Year.

N um
ber of
300-day
work
ers. Death.

1907.................... 27,632
1910.................... 79,486
1911.................... 80,029
1912.................... 93,666
1913.................... 91,107
1914.................... 77,474
1915.................... 79,065
1916.................... 108,994
1917.................... 86,847

61
95
72
84
100
62
55
76
69

Per
ma
nent
dis
abil
ity.

Accident frequency rates
(per
1,000 30<3-day
workers).

Accident severity r a t e s
(days lost per 300-day
worker).

j
Per Tem
Tem
Per Temma po
po
ma
J&
°~
rary To Death. nent rary To Death, nent rary To
dis dis tal.
dis tal.
dis dis tal.
abil
abil abil
abil abil
ity.
ity. ity.
ity. ity.

105 6,529
413 14,124
424 13,837
478 17,450
410 14,103
263 8,406
332 8,420
534 10,365
278 6,687

6,695
14,632
14,333
18,012
14,613
8,731
8,807
10,975
7,034

2.2
1.2
.9
.9
1.1
.8
.7
.7
.8

3.8
5.2
5.3
5.1
4.5
3.4
4.2
4.9
3.2

236.3
177.7
167.6
186.3
150.3
108.5
106.5
95.1
77.0

242.3
184.1
173.8
192.3
155.9
112.7
111.4
100.7
81.0

13.3
7.3
5.2
5.4
6.8
4.6
4.2
4.2
4.6

6.8
3.2
4.7
2.2
4.6 • l . 9
2.2
4.3
1.9
3.8
3.2
1.5
3.2
1.3
3.8
1.3
2.5
1.2

23.3
14.2
11.7
11.9
12.5
9.3
8.7
9.3
8.3

When "le foregoing table is compared with the table giving the
rates for machine building it appears that in five of the years covered
by it the frequency rates are higher then in either year shown for
machine building. In three years the rate is lower than the 1912
rate for machine building, but higher than the 1917 rate. In 1917
this section of the iron and steel industry had a lower frequency
rate than the machine builders.
In all the years the iron and steel industry had a higher severity
rate, but in 1917 the rate was below that of 1912, while in machine
building it was materially higher.
Some of the industries, such as the construction of locomotives
and ships and the manufacture of cranes, exceed the iron and steel
plants in their severity rate. Employment in the construction of
locomotives and engines is sufficient to make this showing significant.
With the others the figures must be considered with caution on
account of the small exposure.
Only two departments of machine building can be compared to
advantage with departments of iron and steel. These are machine
shops and the erecting department in the machine-building industry,
which can be compared, respectively, with the mechanical depart
ment and fabricating shops in the iron and steel industry.

C H A PT E R I I .---- ACCID EN T EX PE R IE N C E .
T a b l e 3 4 . —ACCIDENT

R A TES IN MECHANICAL D E PA R T M E N T S OF T H E IRO N A N D
ST E E L IN D U S T R Y , 1907 TO 1917.

Number of cases.
Num
ber of
Per
300-day
work
ma
ers. Death. nent
disa
bil
ity.

Year.

2,542
7,871
6,7127,122
7,474
5,125
5,693
9,185
8,892

1907....................
1910....................
1911....................
1912....................
1913....................
1914....................
1915................ .
1916....................
1917....................

2
7
5
5
7
3
1
9
1

Accident fretjuency rates
(per
1,0C)0 300-day
workers).

Accident severity rates
(days lost per 300-day
worker).

Per Tem
Tem
Per Tem
po
ma po
ma po
rary Total* Death. nent rary Total. Death. nent rary
disa
disa disa
disa disa Total.
bil bil
bil bil
bil
ity.
ity. ity.
ity. ity.

4
641
635
32 1,148 1,187
878
918
35
38 1,023 1,066
981 1,021
33
510
532
19
27
483
511
752
693
50
37
553
515

0.8
.9
.7
.7
.9
.6
.2
1.0
.1

1.6
4.1
5.2
5.3
4.4
3.7
4.7
5.4
4.2

249.7
145.9
130.8
143.6
131.3
99.5
84.8
75.5
57.9

252.1
150.9
136.7
149.6
136.6
103.8
89.7
81.9
62.2

4.8
5.3
4.5
4.2
5.6
3.5
1.1
5.9
.7

1. 7
3. 7
4.2
5. 0
4. 7
2.9
3.1
5.2
2.8

3. 5
1. 4
1.5
1.7
1. 7
1 2
1.1
1.0
.8

10.0
10. 4
10.2
10.9
12.0
7.6
5.3
12.1
4.3

The average frequency rate shown by the mechanical departments
in the foregoing table from 1912 to 1917 is 102.0 cases per 1,000
300-day workers. This is somewhat below the rates in machine
shops, which were 108.1 in 1912 and 104.5 in 1917. In severity, on
the contrary, the average from 1912 to 1917 is 8.8 days for mechani
cal departments, compared with machine-shop rates of 3.4 days in
1912 and 5.8 days in 1917.
It will be noticed that for the year 1917 the severity rate in
machine shops was 5.8 days, while in mechanical departments it
was 4.3 days; also that mechanical departments made a notable
reduction when 1917 is compared with 1912, while machine shops
went decidedly in the other direction. On the whole, it is evident
that these mechanical departments have the higher normal hazard,
but have managed their campaign of accident prevention with
much‘greater efficiency than have the machine shops of the machinebuilding concerns.
The following table shows the condition in fabricating shops
over a period of years:
T able

3 5 __ ACCIDENT R A TES IN FA B R IC A T IN G SHOPS, 1907 TO 1917.
Number of cases.

Year.

1907....................
1910....................
1911....................
1912....................
1913....................
1914....................
1915....................
1916....................
1917....................

Num
ber of
Per
300-day
work
ma
ers. Death. nent
disa
bil
ity .
2,081
3,935
4,007
5,023
5,313
3,811
2,994
4,465
5,020

6
4
1
5
5
3
3
7
4

12
15
28
32
38
13
13
24
26

Accident frequency rates
(per 1,000 300-day
workers).

Accident severity rates
(days lost per 300-day
worker).

Per Tem
Per Tem
Tem
po
ma po
ma po
rary Total. Death. nent rary Total. Death. nent rary Total.
disa
disa disa
disa disa
bil bil
bil
bil bil
ity.
ity. ity.
ity. ity.
571
589
731
750
660
689
956
993
996 1,039
426
442
366
382
656
625
531
561

2.9
1.0
.3
1.0
.9
.8
1.0
1.6
.8

5.8
3.8
7.0
6.4
6.6
3.4
4.3
5.4
5.2

274.4
185. 8
164.7
190.3
187.5
111.8
122.2
140.0
105.7

283.1
190.6
172.0
197.7
195.0
116.0
127.5
147.0
111.7

17.1
6.0
1.5
6.0
5.9
4.7
6.0
9.4
4.8

8.8
2.3
4.6
4.4
6.9
2.8
2.8
2.2
3.5

2.5
1.8
2.0
2.6
2.3
1.5
2.1
2.6
1.6

28.4
10.1
8.1
13.0
15.1
9.0
10.9
14.2
9.9

ACCIDENTS IN M A C H IN E B U IL D IN G .

Erecting departments show constantly a lower severity rate than
fabricating shops. It appears on consideration of the items that this
is due to higher fatality with average greater severity in permanent
injury in fabricating. When this is followed into the causes it is
found to be due to the hazard of moving heavy and awkward pieces,
such as girders, with the crane.
In fabricating shops both frequency and severity rates declined
from 1912 to 1917, while in erecting departments, though there was
a decline in frequency rates, the severity rates increased from 7.8
days to 8.7 days.
CAUSE OF F LU C TU ATIO N IN ACC ID EN T R ATES.

It is established by all the studies so far made that with every
period of increasing industrial stress the accident rates tend to rise.
If there is not some opposing influence, such as increased accident
prevention effort, this increase may carry the rates above those of
previous stress periods. This apparently happened quite generally
in machine building in 1917. Whether this could have been avoided
by active opposing effort is, of course, a question which it is impossible
to answer. A large section of the iron and steel industry did succeed
in holding down its rates below the previous high point which occurred
in 1913.
The general character of the movement in the iron and steel
industry is compactly shown in the following table from Bulletin 234:
T a b le 3 6

.—FR E Q U E N C Y R A T E S OF FA T A L IT IE S A N D OF A L L ACCIDENTS IN TH E IRO N
A N D ST E E L IN D U S T R Y , 1913 TO 1917.

Year ending with—

December, 1913..........................
March, 1914.................................
June, 1914...................................
September, 1914.........................
December, 1914..........................
March, 1915.................................
June, 1915...................................
September, 1915.........................
December, 1915..........................
March, 19l6.................................
June, 1916...................................
September, 1916.........................
December, 1916..........................
March, 1917.................................
June, 1917....................................
September, 1917.........................
December, 1917..........................

Number
of 300-day
workers.

153,098
146,522
137,816
128,023
117,214
111,881
111, 794
117,933
133,627
148,221
160,819
168,790
175,013
178,937
182,587
185,445
186,357

Fatalities
per 1,000
300-day
workers.
1.34
1.29
1.09
.81
.70
.63
.65
.85
.86
.96
1.09
1.02
1.11
1.15
1.08
1.11
.98

Total
accidents
per 1,000
300-day
workers.
181.0
168.4
154.7
138.9
130.4
118.0
114.0
118.6
124.5
131.8
134.1
135.5
133.2
128.5
121.6
110.9
103.4

This table shows a decline in the rates from the year 1913 to the
year ending with June, 1915. From that point, with the increasing
industrial activity, the rates increase to about the year ending with
September, 1916. This was the period in which all industry was
adjusting itself to the tasks of war. As that adjustment was accom-

CH A PT E R I I .---- ACCID EN T EXPE R IE N C E .

piished the rates began to decline and continued to do so until the
end of 1917.
It is an interesting and important question, not yet solved, as to
which factor is the more important in causing this fluctuation. If,
as seems to be indicated by all available data, it is due more than
anything else to the introduction into the working force, of relatively
inexperienced men, it adds much emphasis to the idea that insta
bility in industry is exceedingly costly. It is costly to train new men.
If during their period of training they are extremely liable to be num
bered among the casualties, the cost is enhanced to a degree not yet
fully determined.
Of various methods devised for testing the influence of the intro
duction of new men none exhibits the effect more directly than the
chart on page 78.
This chart is a “ ratio” ch a rt1 and exhibits the rate of change in
four related items— employment, product per worker, labor recruit
ing, and accident rates. The chart covers two periods of industrial
stress, one culminating with the year ending May, 1910, and the
other with the year ending in February, 1913. It has not been pos
sible yet to extend this study into the war period, but it i3 the inten
tion to do so as soon as possible.
It is evident on even casual inspection that the curve of accident
frequency follows that of labor recruiting with surprising promptness.
Whenever recruiting becomes active there is a strong tendency for
the accident rate to increase. This is even more true of severe injury
than of minor injury.
The chart shows further that increased activity of the individual,
measured by increased product per worker, is a less controlling factor
than recruiting, since product continues to increase in both periods
after accident rates have begun to fall. Since this accords with
other evidence regarding the influence of the inexperienced man, it
may be regarded as fairly established that this factor is of great, if
not of major, importance.

i See

M onthly L abo r R

e v ie w

March, 1919, p. 28.

Ch a r t

ACCIDENTS
IN
MACHINE
BUILDING,

F.-C O M PA R ISO N OF TH E T R EN D OF SEVERAL RELATED ITEMS B Y REDUCTION TO IN D E X NUM BERS, SMOOTHING THE
CURVES A ND PLOTTING ON A PERCENTAGE SCALE.

CHAPTER HI.—SAFETY ORGANIZATION.
To be effective in. producing the best results the safety movement
must rest uponforces steadily operative within the industrial organism.
As long as the standards are fixed and enforced entirely by outside
agencies, such as governmental bureaus or insurance companies,
the utmost desire to obey the law or follow the suggestions of the
inspector seldom suffices to bring about the hoped-for results. When
there is a disposition to evade the law and to refuse constructive
advice the results, of course, are very much worse.
A motive must be supplied for thorough internal organization of
the plant. This is now furnished in an important degree by the
pressure of compensation acts. When an injury means a certain
cost instead of a possible one there is an insistent reminder of the
importance of accident prevention which can be furnished in no other
way. The marked success of some great corporations in accident
prevention has been due in great part to the establishment of com
pensation plans before any State had enacted a workable measure.
The safety organization, called by various names, is fast becoming,
where it is not so already, as much a department of the business as is
accounting or production. The elements of such an organization
are briefly presented below.
THE INSPECTOR.

Safety activities in machine building have been somewhat different
from those in the iron and steel industry. In the latter industry an
entire system had very often grown up around some efficient inspector.
He had been obliged to devise for himself means and methods. The
results were characterized by a good deal of crudity but nearly always
presented original features of much interest. The inspectors of the
machine-building concerns ha^e had the advantages and the dis
advantages resulting from the fact that they have frequently been
called in to administer a plan already determined by the managers’
offices. Having a definite plan saves many mistakes. It may also
be a serious handicap to a man of original ideas.
There appears a greater tendency in machine-building concerns to
employ as inspectors men of technical training or practical shop
experience. Many safety men in iron and steel came from the legal
department. This was probably due to the manner in which safety
effort originated in that industry.
124566°— 20— Bull. 256------ 6

ACCID EN TS IN M A C H IN E B U IL D IN G .

On the whole the choice of engineers, draftsmen, and shop men by
the machine-building concerns is undoubtedly wise. Many prob
lems of a mechanical character remain to be solved and if this industry
uses largely men with that kind of skill it should hasten the solution
of these problems and contribute to the development of safety organi
zation.
The hostility, often quite plainly marked, on the part of operating
men against the safety man has in a measure disappeared. The
operating man now recognizes his colleague as a cost saver and
welcomes his attention to matters which an active superintendent
of production has little time to work out.
THE SAFETY COMMITTEE.

The safety committee in its various forms is the mobile army in the
attack upon bad working conditions. It can not be too often or too
emphatically s tated that the safety committee sys tem has an influence
upon the conduct of business much beyond its immediate purpose.
In most lines of industrial endeavor there is frequent conflict between
the man employing and the man who works. There is almost no sub
ject which can be discussed between the two interests which may not
at some point develop antagonisms. The safety movement in this
respect occupies a rather unique position inasmuch as there is essen
tial accord of interest.
Committee organization has not yet made great progress among
machine builders, but where it was observed in operation it showed
the same adaptation to conditions which has appeared conspicu
ously in other industries.
The following outline of a committee system is applicable to a large
concern. All that is necessary to adapt it to smaller plants is to
consolidate the elements to meet the less complicated situation.
1. Central committee: Chairman, the general superintendent or
his immediate assistant; secretary, the director of safety; mem
bers, superintendents of departments, changing at intervals so that
each department head serves at some time during the year.
2. Departmental committees: Chairmen may be either the super
intendents or important foremen; members, either foremen or fore
men and workmen. The mixed committee seems, on the whole, most
satisfactory.
3. Areal committees: These are the members of the departmental
committees charged with responsibility for a certain area. A whole
some rivalry can be introduced regarding the maintenance of good
conditions in these restricted areas. The determination of this con
dition should be by the inspection of the director of safety.
4. Special committees: These will be organized from time to time
to study some technical problem or special condition. These special

C H A P T E R II I .— SAFETY O RG AN IZATIO N .

committees are the readiest means by which the director of safety
can keep interest alive. His ingenuity will be tested and his useful
ness measured by his employment of them.
MAINTENANCE OF INTEREST.

In the report upon accidents in the iron and steel industry the
topic of “ Educational work of safety committees” was discussed.1
Further study of the situation leads to the conclusion that the more
fundamental problem is maintenance of interest. If that is done,
the matter of education will almost take care of itself. That is to
say, keeping interest alive will result in education.
This must be accomplished, first of all, among the superintendents.
Under compensation, a constant reminder is furnished by the fact
that cost of accidents, like other manufacturing costs, regularly ap
pears on the departmental cost sheets. It has been the custom of
some companies to treat the cost of safeguarding in the same way.
This is not a reasonable procedure. A superintendent ought not to
be penalized for the imperfections of machines and conditions fur
nished him. The cost of remedy for these imperfections should rest
on the business as a whole rather than upon an individual department.
Departmental committees may be held up to their responsibilities
by close supervision from the office of the director of safety, particu
larly by what the railway men call “ surprise tests.’ ’
The maintenance of interest among the men is a rather difficult
matter. In the outset of the safety movement it had the aspects of
a crusade and appealed strongly to the crusading spirit. Ultimately
it must settle down to a regular element of ordinary business life.
The illuminated sign at plant entrances has been a very useful
device. An extension of this method is the safety bulletin board,
for which the National Safety Council has been furnishing its members
a weekly supply of material. These bulletin boards appeal to the
eye and are quickly and easily apprehended. Safety maxims on pay
envelopes have been used with good effect.
It has, however, become obvious that all these measures gradually
lose their force as their novelty declines. As a result, foresighted
managers have been looking for something which could be relied
upon as a more permanent influence.
One important steel company has been experimenting with a bonus
system to foremen. Neither the strength nor the weakness of this
plan has had time to develop.
One very ingenious plan of this nature was found in operation in
one of the plants covered. The account of this method is given
practically as formulated by the company, with some abbreviation
and a few minor changes.
i Conditions of E m ploym ent in the Iron and Steel Industry in the United States (S. Doc. No. 110,
32d Cong., 1st sess.), Vol. IV, pp. 183-185.

ACCID EN TS IN M A C H IN E B U IL D IN G .

The accident-prevention score board stands just inside the main gate of the factory.
I t is 24 feet long, and on it are shown the departments, foremen, percentages for t i e
month, and rank of the various competing divisions.
The starting point i& 1,000 both for year and for month. Each division is penalized
according to its accidents—minor accidents of less than one day’s absence not as yet
being considered. Each day’s absence bears a percentage charge in proportion to
the total number of “ m en-days” per month per division.
There are 26 divisions in the competition of various degrees of natural hazard and
of wide variation in numbers of men. The degree of hazard is disregarded in our
business, which covers the same general subject throughout the plant, the differentia
tion being considered as equalized in the choice or selection of men with reference to
their ability and fitness for their respective class of work. As to the variation in the
sizes and groups of workers, we meet this b y establishing a differential charge per man
per day for time off, which is computed b y reducing each division to men-days for
each month and using a multiplier of 10 to raise the figures to a more workable and
understandable basis.
A division working 50 men for 25 days per month amounts to 1,250, and m ultiply the
result b y 10 equals 8 points for each man off one day on account of accident in that
division. W ide variations noticed in a year’s competition in the different divisions
should be the basis of an adjustment of this penalty charge, which adjustment should
not have to be made during a month.
In this manner large and small divisions are equal as to their penalties. In the
fourth column of the score board will be noticed the figures which represent the deduc
tions for absence in that division.
We disregard small accidents that do not entail appreciable loss of time, and we do
not penalize for the remainder of the day on which the accident occurs. It is possible
b y this provision to insure the prompt report of all accidents, however small, so that we
may be sure of proper attendance and avoid, as far as possible, such suffering as may
be otherwise charged to secrecy on the part of either men or division superintendents.
A t the end of 12 months the employees of the divisions scoring 1,000 receive two
days’ extra pay, or such part of that amount as their time and employment bears to
the full year. If none score 1,000, then the highest ranking department receives two
days’ extra pay and the second highest one day’s extra pay. General foremen of any
division under them earning these premiums also participate on the same basis, but
may earn but one prize if other divisions under them score perfect.
The original plan was to distribute $25 in cash each month to all foremen of divisions
earning perfect scores; but, due to the relative importance and efforts of the foremen,
with a widely varying number of men to deal with, we were obliged, in fairness, to
change this arrangement so that one-half of each prize is paid on a flat basis and onehalf distributed according to the number of men overseen. Thus, a foreman in charge
of 50 men will get a proportionately larger premium than one in charge of 10 men.
It may be noticed that the cash prize is rather small, and to some might be even con
sidered trivial; but to such there has not come the meaning of the spirit back of the
accident-preventing board as it prevails in our factory. It is the difference between
success and loss that counts, and men who work at the lathe, the forge, or the cupola
have the same aspirations to participate in the winning spirit that inspires any team
or organization, however or whenever formed. Several efforts have been made b y
psychologists visiting our plant to analyze the mental attitude which these men
must carry, and it has been the unanimous opinion that departmental loyalty is the
first stone, the great foundation, upon which stands the success and cooperation of this
idea. It is the aim of each division to head the list, and they must feel that they have
a chance of winning tnroughout the year. This interest is fostered b y making up
the yearly basis out of the monthly average. The great thought is then concentrated
on the yearly contest, and the discouragement of any unfavorable monthly showing is

C H A P T E R II I.-----SAFETY O RG AN IZATIO N .

avoided, because any other division may have a sufficient penalty in some months
throughout the year to equalize these unfavorable periodical conditions.
We have found that this system is a matter of personal interest to both foremen and
employees, and so intense has the Competition become at times that an unforeseen
condition arises which must be met b y extreme diplomacy, and that is the ill-feeling
that may be occasioned against a worker who has been responsible for causes which
might have been controlled. Careful investigation and study has shown that the
personal interest manifests itself, and the feeling that the loss must be minimized is
responsible to a great extent toward urging them to get back to work as quickly as
possible. The foremen of the various divisions of the factory are members of a safety
committee which meets at regular intervals under the direction of the general super
intendent. A board of governors of five looks after the details of inspections, reports,
investigates complaints, and approves the monthly penalty charges. This has served
as an admirable promotion toward the further education of foremen in matters per
taining to accident prevention, as well as sanitation, cleanliness, and fire prevention,
etc.
For the year closing September, 1913, 10 divisions out of 26 showed perfect scores.
The division ranking 16, the lowest, has a penalty of but 51 points. Included in the
perfect scores is the south foundry, the division in which our heaviest work is prch
duced, making single castings up to 50 tons in weight and generally classed as a haz^
ardous occupation. An analysis of the year ending September, 1913, shows 161
accidents— 17 applying on foot, 77 on eye, 45 on hands or fingers, 45 on scalps or face
6 on bum s or scalds, 5 miscellaneous. The total expense of first aid was $308.50;
hospital service, $31.50; claims, $50; a total of $390. Tim e lost was figured at 218
hours; thus the average cost per accident was $2.42.
For the 12 months ending September, 1914, 11 departments of the 26 showed an
improvement over their record for year ending 1913. Nine departments of the 26
showed a decline. Six departments maintain their averages of the previous year,
and five of these six have now presented perfect scores of 1,000 for two years. W ith
one exception, all hazardous departments show a gain.

The following table shows the results of the methods described,
over a series of years:
Table

37.—COST OF ACCIDENTS COM PARED W ITH PA Y ROLL, A N D TIME LOST COM
P A R E D W ITH TIME W O R K E D .
[From a bulletin issued by the National Safety Council.]
1910

Total cost of accidents for each $100 of annual pay ro ll2........... . $0,503
Time lost due to accidents beyond the fraction of the first day
(per cent)........................................................................................
(3)

1911

19121

19131

1914 1

$0,228

$0,112

$0,079

$0,070

( 3)

0.394

0.192

0.116

1 In th is year th e score board and wage bonus were in use.
2 Including first aid, hospital bills, and claims, if any.
3 N o record kept.

SURGICAL CARE.

Since many of the machine-building companies have no need for
extensive premises they are frequently located in the heart of busi
ness districts in cities. This location, with its proximity to hospitals
and dispensaries, has brought it about that there are comparatively
few emergency hospitals and emergency rooms in the plants them

ACCIDENTS IN M A C H IN E B U IL D IN G .

selves. The large companies have, as a rule, such conveniences, and
an extension of this plan is practicaLly certain when the possibilities
of the service become more fully realized.
An emphatic word of caution is needed regarding the use of “ firstaid equipments.” They are useful if confined to first aid under close
supervision. If their use is relied upon as final and not subjected
to rather prompt scrutiny, preferably of a physician, or at least of
some one with the training of a nurse, it is dangerous.

CHAPTER IV.— DIRECT SAFEGUARDING
MACHINE BUILDING.

METHODS IN

Since factory conditions have much to do with safe production, it
is desirable at the outset to describe briefly some instances of the
best recent construction. No attempt will be made in this or in
other descriptive matter to give the technical details. It is the aim
to present broadly the features which appeal to the nontechnical
observer as bearing on the matter under consideration.
The important departmental units concerned in machine building
are the foundry, the machine shop, and the erecting or assembling
shop. In very many cases erecting is so intimately associated with
machine-shop operations that no line can be drawn between them.
This being the case, it must be understood that throughout this report
many processes strictly belonging under erecting are discussed, for
convenience, in the machine-shop section.
SHOP CONDITIONS IN FOUNDRIES.

The steady tendency in foundry architecture in recent years has
been toward an increased height of walls with larger window area
and better disposition of the artificial lighting.
Two general types of building may be noted. One, which may be
called the standard type, consists of a central bay with louvered roof.
In this bay the floor molds for large castings are built. On each side
of the central space is a side aisle where smaller molds may be pre
pared and poured, the core making done, and other accessory opera
tions carried out. An excellent example of this standard type is the
new foundry at the Washington Naval Gun Factory. A brief descrip
tion of this foundry follows:
The main axis extends north and south. On the west is an area
for the reception and distribution of raw materials. Along the high
brick wall which forms the boundary of the yard are a series of bins
in which are stored pig iron, coke, limestone, etc. Parallel to this line
of bins are the railway tracks over which the materials are brought.
Between the bins and the track is an elevated structure carrying the
outer end o f the bridge of a traveling crane, whose inner end is car
ried on tjie wall of the foundry. Between the tracks and this wall
are two sand houses with hatches in the roofs, by which sand taken
from the cars may be received directly from the crane buckets. Out
side the foundry wall at the height of the charging floors and com

ACCIDENTS IN M A C H IN E B U IL D IN G .

municating with them by large doors are two platforms upon which
materials may be delivered directly from the crane. These arrange
ments permit the most prompt and satisfactory handling of all raw
materials.
Entering the foundry at the north end there is at once the impres
sion of ample space, good wall height, and satisfactory lighting. In
the side walls there are two tiers of windows occupying nearly all the
wall area. In the end walls there are three tiers in the central sec
tion, above which extends the louvered roof. These windows are
glazed with a ribbed glass which produces a very uniform distribu
tion of the light.
The central section of the foundry is served by two 25-ton traveling
cranes and one 50-ton crane. These ran upon tracks located at the
point where the walls of the central roof spring from the roofs of the
lateral sections. As is usual, this central section is utilized for the
placing of large molds used in making heavy castings.
Down the western side on the right of the entrance the lateral sec
tion is occupied by theapparatusforproducingmoltenmetal. This con
sists of a 5-ton, open-hearth furnace, two small converters for making
Bessemer steel, and four cupolas in which iron is melted. The eastern
lateral portion is utilized for the making of small molds and for core
making. It is served by two small overhead cranes. A t about the
middle of its length are the ovens in which molds and cores are dried.
Such a foundry may be regarded as the direct evolution of the lowwalled, dismal, and dirty buildings which were earlier considered good
enough for a foundry. Three causes have been operative in produc
ing this evolution: First, the introduction of the overhead crane as
a means of transportation; second, the demand for better lighting;
and, third, willingness to improve the general working conditions.
The other type of foundry is illustrated by one recently put in use
by the National Brake & Electric Co. of Milwaukee. (Plate 1.)
Here the main floor is surmounted by a saw-tooth roof. This method
of securing uniform and well-distributed light has been long in use
for the weaving rooms of textile establishments but is a rather recent
innovation for foundries; in fact, but one other was noticed in the
course of this study, which extended to all the important machinebuilding centers.
Since the methods of handling raw material and molten metal in
this foundry, while having many interesting features, are not essen
tially different from those in use elsewhere, space will not be taken
for their description.1 Some features of internal arrangement will be
presented farther on in connection with foundry safeguarding.
i For a full description of this plant see “ Foundry” for February, 1914.

Bull. 256— Labor.

P L A T E 1.— E X T E R IO R OF F O U N D R Y ,

CHAPTER. IV .---- D IRECT SAFEGUARDING M ETH O D S.

It was formerly assumed that being necessarily a dirty trade the
provision of facilities for cleanliness was unnecessary. It might seem
at first glance that the installation of toilet facilities has no significance
to those interested primarily in safety. Nothing has been more
impressive during the course of this study than the cumulative
force of the evidence that nothing which bears upon the health and
comfort of the workers is without its relation to the safety problem.
This is the reason for the expanding field of the safety director. If
he follows the natural leadings of his office he will become a sanitarian
and ultimately will be obliged to go outside the bounds of his plant
into the problems of community life. A study of the needs of men
in industry can not stop with the industrial field. It must of neces
sity reach all their relations as men.
SAFEGUARDING IN FOUNDRIES.

Two sources of hazard at once suggest themselves when attention
is directed to the foundry. There is considerable transportation of
objects and there is the problem of handling molten metal. The
machines used in molding present the usual features of belts and
gears and other moving parts requiring covering and fencing. They
do not, however, constitute a very large factor in the accident occur
rence.
A primary necessity is for the safe movement of material. The
character of the work done in a foundry renders it very easy for it to
fall into a disorderly condition. The passageways become choked
with debris and with apparatus awaiting its turn for use. With large
castings particularly there is a considerable amount of material partly
usable and partly useless derived from each operation. Without
rather rigid rules and close supervision a very chaotic condition soon
prevails.
Not a few foremen insist that this condition is inevitable where
work is being turned out rapidly, and it is undoubtedly true that in
many cases disorderly conditions and large output go together. The
fact remains, however, that while no exact figures have been obtain
able, the investigation has failed to produce a single instance in which
the development of more orderly methods has not been accompanied
by greater output. So uniform has been this experience as to cast
grave doubt upon the insistence of disorderly foundries that they
could not afford better order because it would decrease production.
The maintenance of clear aisles and the orderly disposition of appa
ratus have a direct bearing upon safety. In foundries where it is
not the rule the men are endangered every time they attempt to
pass from one part of the foundry to another, and since on every
portion of the floor work is carried on, crane loads can not be car
ried without at some point passing above the workers.

ACCIDENTS IN M A C H IN E B U IL D IN G .

The preparation of the metal for casting usually involves the use
of the cupola, though some recent installation provides for the melt
ing in a furnace much like that used for puddling, while for steel
castings open-hearth furnaces, both stationary and tilting, are util
ized, and in a few cases steel is made and cast from a form of Bessemer
converter. The cupola is a cylindrical shell of steel lined with refrac
tory material. The pig iron, coke, and flux are introduced from a
charging floor at a level about half the height of the cupola. The
details of hoists, charging cars, and barrows all require the same care
called for in all machinery as to covering gears, the use of safety
gates, and other precautions for safety.
Plate 2 shows a foundry interior with cupolas in the foreground
on the right. The ladle is in place upon the car ready to receive
the hot metal from the runner. As the metal comes down the run
ner sparking is apt to occur, and as the metal falls into the ladle
these sparks are particularly likely to be projected forward in the
direction of the workman. To avoid this one foundry has adopted
runners with a turn near the end at nearly right angles. As a result,
the metal when falling into the ladle is not moving toward the tapper,
who must stand opposite the tapping hole when opening it.
In addition to the ladle on the car, other ladles appear in the pic
ture which have bails for carrying by the crane. Since all these are
suspended at a point but slightly above the center of gravity appro
priate means of locking must be utilized to prevent oversetting and
due care exercised not to overfill.
The crane service in the picture deserves a moment’s notice. It
consists of an ordinary crane on an upper track. Its bridge is seen
above in the background. In the center of the picture is a wall
crane.- It serves the area next the wall to which it is attached for
objects whose weight does not demand the service of the larger crane.
The safeguards possible in these operations are mainly those of
good construction and careful operation. When it comes to the
actual transfer of the metal to the molds two precautions become
important. The sparking of the metal above mentioned suggests
some protection for the eyes. Since the time during which this
protection is required is quite brief it may be afforded by inexpensive
glasses which would not be usable for the prolonged wearing neces
sary in the cleaning room. Eye injuries from flying sparks are suffi
ciently numerous and serious to demand this attention.
Among the most frequent and serious injuries in foundry operations
are burns of the feet occurring during the pouring of castings.
At times serious explosions and breakouts may occur. Foundry
fatalities are sometimes due to such causes, but they are fortunately
rare and the losses are much less than those arising from the frequent
recurrence of minor burns.

Bull. 256— Labor.

PL A T E 2.— I N T E R I O R OF F O U N D R Y .

CH A PT E R IV .---- DIRECT SAFEGUARDING M ETH O D S.

For the majority of foot burns there is no excuse. If the employer
provides proper means of protection and insists upon their use these
injuries can be practically eliminated. It is altogether a question of
proper shoes and the use of leggins during the operation. What is
coming to be known as the foundry man’s shoe is of specially prepared
leather not easily affected by heat and having rubber cloth gores
at the sides so that they can be readily removed. Many foundry
owners are now securing and keeping on hand a supply of the proper
type of shoe for the benefit of their employees since such shoes can
not be easily obtained in the general market. If with this shoe a
suitable canvas leggin is worn when pouring the foot burn becomes
exceedingly rare.
The cleaning room has two dangers, dust and flying metal. The
removal of the sand and the cores by the usual methods is exceed
ingly dusty. In many cases this is so serious a menace to health
that either respirators should be worn or a dustless method of clean
ing adopted. One foundry observed has introduced such a method.
It consists of a high pressure stream of water. One man using this
method cleans a greater number of castings than several by the
old method. The quantity of water used makes necessary special
means for its disposal and this together with the difficulties inci
dental to cold weather is an obstacle to the use of this method, but its
speed and thoroughness, and the absence of dust, render it worth
consideration.
Few, if any, castings come from the mold in condition to be imme
diately machined. The gates by which the metal enters must be
removed and at each junction of parts there is a fin of metal to be
taken off. This work is done by the chipper, using chisel and ham
mer or a pneumatic chipping tool. It is obvious that the process
must be accompanied by flying fragments liable to cause injury,
especially to the eyes, both to the chipper himself and his associates.
The remedy for eye injuries is the use of proper goggles. Emphasis
is placed upon “ proper,” since in the early attempts to meet the
situation there was much complaint of the refusal of the workmen
to use the protection when furnished. They were fully justified in
their antipathy. Ordinary glass has uneven surfaces and when used
in glasses gives the effect of a number of irregular prisms before the
eyes. A few minutes’ wear of such a glass produces discomfort
and prolonged wear causes serious pain. If persistently used such
glasses result in grave injury to the eyes.
Proper goggles must have three qualifications: First, clearness and
accurate surfaces; second, a mounting which can be adjusted, either
by selection or changed shape, to the individual wearing the goggles;
third, sufficient strength to withstand the blows of flying particles or if
the glass yields it must do so without itself giving rise to flying pieces.
An employer should not be satisfied with the assurance of the dealer

ACCID ENTS IN M A C H IN E B U IL D IN G .

regarding this last point. He should insist upon tests. Several
large buyers have devised testing apparatus to which they subject
each lot purchased, rejecting the lot if it fails to come up to the
required standard.
The apparatus of the American Car & Foundry Co. and their
requirements of the goggles they purchase are here described.
The sliding carrier shown in plate 3 holds a steel ball of five-eighths
of an inch in diameter, weighing not less than 16 grams. The
goggles are supported as shown and the ball drops upon pressing the
button in the base.
The specifications which determine what the goggles must with
stand are given below.
It may be said that not very long ago there was not in the market
a goggle which would come even approximately near to meeting
these standards. Now they may be obtained without difficulty.
S p e c if ic a t io n s

for

T e s t in g G o g g l e s .

1. Drop-test machine.— The drop-test machine to consist of a support for the goggles
made of white pine; the support being designed to accomodate the goggles under test
in such a manner that the frame of the goggle is given proper bearing on the rubber
and cotton composition strips without permitting the lens itself to rest on the support
or to receive any “ backing,” other than that naturally given by its own frame. A
representative form of support is shown in plate 3.
2. Height o f drop.— Twenty-one inches from bottom of ball to the surface of the
goggle lens.
3. Size and weight o f ball.— The ball to be made of steel and hardened. The diameter
will be five-eighths of an inch and the weight not less than 16 grams. When released
the ball must fall freely without any initial momentum.
4. Extreme variations allowed in the thickness o f a single lens.— Five millimeters. The
measurement of the thickness to be taken with a standard gauge used b y opticians
at five points.
5. Extreme variation in thickness allowed between two lenses o f the same pair o f goggles.—
Ten millimeters. The measurement of the thickness of each lens to be taken as de
scribed in paragraph 4.
6. Number o f blows.— The maximum number of blows to be given is 15; the blows
to be given consecutively on the center of the lenses and on the surface of the lens
which is exposed to flying matter.
7. Number o f tests.— On each shipment of one gross of goggles, one dozen of the gross
must be tested as described in paragraphs 4, 5, and 6, and of this dozen 25 per cent,
or three pairs of goggles, must stand 15 blows without breaking or cracking. This
means that both lenses of at least three pairs of goggles in the test dozen must be
intact after 15 blows. If less than 25 per cent stand the test the entire gross will be
rejected.
8. Flying chips o f glass.— If in the test dozen of goggles any goggles break under the
drop in such a way that glass will fly from the inside surface of the lens— meaning the
surface which is next to the eyes of the men wearing the goggles— then the entire gross
will be rejected, even if three pairs of goggles have stood the number of blows required
in paragraph 7.

With the growth of oxyacetylene and arc welding and other indus
trial operations of similar character another method of eye protection

Bull. 256— Labor.

P L A T E 3.— M A C H I N E FOR T E S T I N G G O G G L E S .

Bull. 256—Labor.

P L A T E 4.—T U M B L E R A N D S A N D - B L A S T I N G M ACH I N ES. W I T H E X H A U S T S .

C H A P T E R IV .---- DIRECT SAFEGUARDING M ETH O D S.

becomes of extreme importance— the use of suitably colored glasses.
This has received small attention until recently, it being assumed
that any coloring of the glass which reduced intensity served the pur
pose. With the advent of the intense heat and light of the arc
this impression is being rapidly destroyed.
A study of the conditions shows that the harmful effect is not due
to the intensity of the luminous rays of these operations but is due
to the accompanying increase in the invisible rays known as the ultra
violet and the infra-red. An illustration of the power of the ultra
violet rays is seen in their capacity to destroy germs, which power is
utilized in the water sterilizers lately put upon the market. The
forms of colored glass hitherto utilized for eye protection afforded
practically no defense against these harmful rays.
Since the fragments thrown off by chipping often fly with force
enough to make a serious flesh wound, screens of burlap placed in the
line of usual flight are a useful protection to those who work or must
pass in the vicinity.
Further cleaning of the castings is accomplished by tumblers or
sand blasts, both of which are shown on plate 4. In the tumbler
or tumbling barrel in the rear at the left of the picture the small
castings are placed in a horizontal cylinder, which is then revolved,
the pieces being cleaned and even polished by attrition against one
another or introduced substances. Formerly this process was very
dusty but now in many cases, as shown in the picture, exhausts
are applied and the dust effectively removed. In sand blasting a f
stream of air projects sand with much force against the surface,
smoothing it and removing scale and rust. In the plate two com
pletely inclosed machines for this purpose, which are provided with
exhaust arrangements, are shown in the rear at the right of the picture.
Sometimes it is impossible to use such a machine, as, for example,
with very large castings. Two safeguarding methods are possible:
(1) The object to be cleaned may be inclosed in a chamber and the
sand applied through guarded slits in the walls, or (2) the workers
obliged to be in the room with the casting may be provided with
suitable helmets.
CONDITIONS IN MACHINE SHOPS.

Typical machine shops for the handling of large work are those of
the Allis-Chalmers Co., at West Allis, Wis. Each shop consists of a
central portion, u<nder the louvered roof, in which are located the
large machine tools served by traveling cranes, with side aisles where
smaller machines are located. These side aisles have small traveling
cranes, a monorail hoist, or other means of transportation. Above
the side aisles are a varying number of gallery floors occupied by
automatic machines, small lathes, and benches in front of the
windows.

ACCID EN TS IN M A C H IN E BU IL D IN G .

The general arrangement of the Allis-Chalmers shops deserves some
notice, since the entire group was built as a unit, with the purpose of
maintaining a steady progression of the material through all the
processes. Many, perhaps most, shops have developed by a process
of accretion which involves more or less difficulty in the transpor
tation problem. The movement of material from place to place is a
very serious factor in the causation of accidents. Evidently then
arrangements which reduce the need of moving material are impor
tant safety measures.
The arrangement of the Allis-Chalmers shops is as follows: Front
ing upon the street is a building several stories in height devoted to
the general offices and to pattern storage. At the back of the build
ing is a one-story extension devoted to the pattern shop. Between
the pattern shop and the foundry is a yard space where flasks and
other foundry apparatus which can be kept out of doors are stored.
The foundry extends parallel with this pattern shop and pattern
warehouse. Beyond the foundry are the machine shops, six in
number, their axes at right angles to that of the buildings already
mentioned. The space between the foundry and the shops and
between the individual shops is utilized for outdoor storage and is
served by traveling cranes and railway tracks which make every part
readily accessible. The shops open directly at their extreme end
into an erecting shop extending across the entire group of six shops.
In this the products of the several shops are brought together and
assembled into the completed machines. Railway tracks enter this
building, so that the completed product may often be sent directly
to its destination.
It requires no more than this statement to show that if the processes
are organized to fit the plans of the buildings, it should be possible to
conduct them with a minimum of transportation, and so greatly
reduce that element of hazard.
Of shops intended for smaller work such as turret lathes and similar
products two recent types of construction are particularly noticeable.
The first of these is illustrated by the shop of Bardons & Oliver in
Cleveland, makers of turret machinery. This shop exemplifies the
sort of plan likely to be adopted in a city location where the ground
values are high and it is therefore necessary to secure floor space by
vertical instead of lateral extension.
The present bililding has a groiind plan like a reversed letter “ L .”
The stem of the “ L ” contains the shops, while the foot is devoted to
stairways, elevators, and locker rooms for the workmen. The
arrangements of these locker rooms, one on each floor, present fea
tures deserving attention. The entrance for workmen is at the
extreme of the foot of the UL .” The stairway is immediately acces
sible. The workman, on reaching his floor, finds an entrance to the

C H A P T E R IV .---- DIRECT SAFEGUARDING M ETH O D S.

locker room adjoining the stairway. Down the middle of the room
are the washing facilities, while along the wall opposite the entrance
are the individual steel lockers. These stand upon a low concrete
base with no space beneath for dust to accumulate. The base is
given a curve down to the floor so that it may be easily washed
without danger of splashing on or into the lockers. After putting on
his work clothes the workman passes to the shop by a door next the
shop section, saving any retracing of his steps or interference with
his fellows who may be coming from the stairs. Entering the shop
the worker finds on his right the rack for the clock cards and on his
left a bubbling fountain.
Two features of the shops attract particular attention. A grad
ual transformation to individual motor drive of the machines is in
progress. The position of machines throughout the shops was care
fully determined when preparing the plans. Proper conduits for the
reception of the electric wires were placed, with the result that as
new machines are installed a connection is available from which
electricity may be obtained with a minimum of difficulty and with
practically no exposure of the wiring.
It is becoming more and more a settled feature of such construc
tion that it shall be fireproof. The building in question is reenforced
concrete, brick, and tile throughout. It has been found by expe
rience, however, that concrete floors are exceedingly tiresome to the
workman who must stand upon them all day. To avoid this and at
the same time keep down the fire hazard this expedient was adopted:
At the location of each machine a depression in the concrete was
formed to a depth equal to the thickness of the flooring desired.
This depression was filled with a carefully laid wooden floor. The floor,
having no air space under it, would burn very slowly in case it
took fire, and since the area about each machine is isolated from
other areas by broad strips of concrete, the spread of fire is rendered
nearly impossible.
The second type of shop is illustrated by the building of the Cin
cinnati Milling Machine Co. It may be called the suburban type.
Some years ago a number of machine-tool builders removed from
their urban location in the city of Cincinnati to the suburb of Oakley.
With the lesser cost of ground area they were able to secure adequate
space without resort to many-storied structures. When, however,
a large ground area is covered, lighting becomes a matter of consid
eration. It is sometimes solved by long and rather narrow build
ings, in which the center of the room is not so far removed from the
walls as to reduce the light unduly. In the case of this machine
shop the result was secured by the use of the saw-tooth roof. The
part of the building fronting on the street is several stories in height

ACCID EN TS IN M A C H IN E B U IL D IN G .

and serves for office purposes and for some of the shop operations.
In the rear is the large one-story structure in which the adequate
distribution of light is secured by the form of roof mentioned. A
building of this sort standing by itself presents a rather hopeless
problem from the architect’s point of view, but from an operative
standpoint it has many things strongly in its favor.
Erecting does not demand any special features in the buildings in
which it is done, except in the case of locomotives. With them, of
course, a large area of ground space is essential, and lighting by
some overhead device, either the ordinary louver or saw-tooth con
struction, is essential. One of the most impressive industrial spec
tacles to be seen anywhere is the erecting shop of the Baldwin Loco
motive Works at Eddystone, which will accommodate over a hun
dred large locomotives at one time.
In general it may be said that the machine-building concerns con
sidered in this report are well housed and that some of the buildings
of recent construction represent the best so far attained in safety
from fire risk, sanitary convenience, and adaptation to efficient oper
ation. Unfortunately it will appear in the course of the report that
while doing considerable for the safety of their operations, it can not
be said that these firms have been leaders in safeguarding except in
individual instances.
SAFEGUARDING IN MACHINE SHOPS.

A later section under machine design is devoted to machine-shop
equipment. In view of this fact only those forms of equipment will
be here discussed which do not appear in the discussion of design.
No single change in shop equipment has more strikingly modified
conditions than the introduction of electrical drive. It has repeat
edly been assumed that this introduction adds to the dangers of the
shop. The argument practically is that there is now added to the
existing hazards the chance of burns and shocks from the electric
current. This takes no account of the dangers removed. A moment’s
consideration will show that these are many and that they far out
weigh those added.
Plates 5 and 6 give an idea of the different appearance of shops
belt driven and motor driven. Inspection is enough to show the
elimination of shafts and belts, generally recognized as a frequent
source of injury. A considerable amount of severe injury has con
stantly occurred in connection with the oiling of shafts and the adjust
ment of belts. Group electric drive reduces this hazard, and indi
vidual drive eliminates it altogether. More important than the
elimination of shaft and belt hazard is the control over the machine.
A very frequent cause of injury in the old-type shop was the unex-

Bull. 256— Labor.

PL A T E 5.— B E L T - D R I V E N SC R E W M A C H I N E S .

Bull. 256— Labor.

PL A T E 6.— L A T H E S D R IV E N BY I N D I V I D U A L M O T O R S .

Bull. 256— Labor.

P L A T E 8.— S T A N D A R D T E S T I N G S W I T C H B O A R D .

C H A P T E R IV .---- DIRECT SAFEGUARDING M ETH O D S.

pected starting up of the machine. This may occur when a shaft,
stopped for some cause, begins to revolve or the belt of the particu
lar machine may creep from the loose pulley to the tight one. Such
unexpected starting is practically impossible when the machine is set
in motion by the closing of a switch.
Not only does the workman have perfect control of his machine,
but it is perfectly feasible to extend this control so that it can be
exercised at several points. The stoppage of a machine promptly
from several points in its vicinity may make the difference between
serious, even fatal, injury and complete escape.
Another element of safety is the ready and exact adjustment of
the amount of power to the demands of the work in hand. This is
accomplished with mechanical drive by means of cone pulleys, gears,
clutches, and other devices, but in no case can it be done with the
precision possible by the use of electricity.
It is still a matter of discussion whether the use of individual
motors is an economy of power. The claim is made that the opera
tion of a given group of machines can be conducted with an expend
iture of about 80 per cent of the power required by mechanical
drive. Whether this saving would offset the higher cost of electric
installation would be a question to be settled only by a study of the
individual case.
Although less serious in results than in the chipping of castings,
there are several machine-shop operations in which fragments are
apt to fly.
Grinding wheels are an important adjunct in machine-shop opera
tions. Two items are worthy of attention in the safeguarding of
these: First, proper mountings, and, second, hoods inclosing the
wheel, so that if it explodes the pieces will not fly.
SAFEGUARDING IN ELECTRICAL MANUFACTURE.

In the production of electrical apparatus the foundry and the
machine shop play an important part. These have already been dis
cussed. In addition, there are two operations in the production of
electrical apparatus which require particular attention, namely,
testing and the use of power presses.
The testing department of every important electrical manufac
turer has at some time presented cases of severe injury and in most
cases some fatalities. This occurrence of fatal burns and shocks has,
with the growth of the industry, led to a revolutionary modification
of testing apparatus and testing methods. The subject is too tech
nical for discussion in any complete fashion in this connection.
Plate 7 shows the earlier form of testing switchboard and plate
8 that now regarded as standard. The most important differences
are the bringing of the cables to the latter form at a higher level and
124566°— 20— Bull. 256------ 7

ACCIDENTS IN M A C H IN E B U IL D IN G .

in a more substantial manner and the method of making the
contacts. In the older switchboard the live parts are carried on
the front of the rack, where the hands of the operator may easily
touch them. In the newer form these contacts are all behind the
slate boards. When it is desired to make a particular connection a
rod is thrust into the opening in the board and the current com
pleted at a point some distance from the hands of the operator and
which is perfectly screened.
One large manufacturer of electrical apparatus had been having
one or more fatalities annually in testing operations prior to the
installation of improved apparatus. Since the change no fatalities
have occurred in strictly testing operations.
In the production of electrical apparatus there are a great number
of parts blanked out and formed upon the punch press. A previous
study 1 brought out the excessive danger of this operation and the
figures now available show that in spite of much improved methods
it still remains hazardous.
Two devices designed and in use by one large electrical company
deserve mention. On very heavy presses two buttons are provided
at the edge of the press table. Both of these must be pressed at
the same time to release the press. Both hands of the operator
being occupied at a position out of the danger zone, injury is impos
sible. The second device is a suction handle, by which small parts
may be adjusted under the press. At the end of the handle is
a disk, to the center of which a tube running through the handle
extends. The handle is connected by a flexible tube with an exhaust
pump. In use the disk is applied flat upon the surface of the piece
of metal and a small knob on the handle is pressed by the thumb,
opening a valve. The suction causes the metal to adhere, when it
can be placed upon the machine and the suction released. The
machine is then tripped in the ordinary manner by a foot treadle. One
edge of the disk is prolonged into a hook or horn, by which the formed
part may be pried up and thrown from the machine. Both of these
devices are effective in keeping the hands away from danger.
Other devices operate by preventing the action of the press when
ever the hands are in a dangerous position. These devices may be
combined with others releasing the press upon a positive removal of
the hands.
An important feature of electrical manufacture is the winding and
forming of coils and the application of insulating material, but the
hazards of these operations are not sufficiently serious to justify
special reference to the safeguarding methods observed.
1 Condition of W oman and Child Wag© Earners in the U nited States (S. Doc. No. 645, 61st Cong.. 2d
sess.), Vol. X I.

Bull. 256— Labor.

P L A T E 9 . - R I P S A W G U A R D — P O S I T I O N W H E N SA W IS US ED .

C H A P T E R IV .---- DIRECT SAFEGUARDING M ETH O D S.

SAFEGUARDING IN WOODWORKING SHOPS.

These shops contribute but little directly to the products now under
consideration, but their indirect contribution in the making of pat
terns, templets, etc., is of such considerable importance that a brief
statement of safeguarding devices observed in them is necessary.
Plate 9 presents a view of a saw guard which has some unique
features. This guard consists of two curved arms, mounted so that
they revolve about the same center as the saw. Upon pushing any
piece of lumber of whatever thickness against the free end of the arm
above the saw it revolves, the free end rising until it rests upon the
surface of the lumber. Plate 9 shows two things: First, how the arm
in front of the saw comes into action, maintaining at all times a com
plete screen in front of the saw teeth; second, that the arm above the
saw not only serves as a guard above the teeth, but acts as a splitter
behind the saw, so preventing the “ kick back” which is one of the
most serious happenings with a saw. If the lumber isthicker than the
exposed portion of the saw, and it is desired to saw a groove in it, the
arm above the saw simply retreats entirely and disappears beneath
the table. The arms are actuated and kept in proper position by
counterweights.
The features of this guard may be summarized as follows: (1) It
can not be removed by the operator; (2) every operation possible with
an unguarded saw is possible with the guard; (3) the guard does not
interfere with any necessary view of the work during operation;
(4) a constant screen is maintained over every part of the saw edge;
(5) it effectually prevents “ kick backs.”

The band saw quite eariy came in for attention, since, when the saw
ran unscreened, very ugly accidents would occur when the saw broke
and the free end came flying out into the surrounding space.

CHAPTER V.— MACHINE DESIGN AS A FACTOR OF SAFETY.
The machine-building plants covered by this report are engaged
in the production of machines for use in various manufacturing and
transportation industries—such machines as cranes and hoists, en
gines, boilers, dynamos, locomotives, and machine tools. The extent
to which these machines will be a source of danger to those who later
have to operate them is dependent, in considerable degree, upon the
manner of their construction in the machine-building shop.
In response to the demand for safer machines the builders have
done two things. They have (1) applied safeguards to their existing
designs similar to those developed by the users of their machines, and
in other cases they have (2) undertaken a radical revision of their
designs in order to secure the desired result. It is difficult to draw
a line between these two methods and for practical purposes it is
unnecessary. All changes made by the maker which result in safer
operation will therefore be considered without reference to the dis
tinction mentioned above.
MACHINERY FOR THE STEEL INDUSTRY.

In the course of this investigation the building of ore unloaders, of
rolling-mill equipment, and wire-mill machinery came under partic
ular scrutiny.
In the matter of ore-unloading apparatus the chief modifications
have been directed to increased efficiency, and their effect upon acci
dents has been due to a reduction of the shoveling crew who in the
earlier types were exposed to material falling from the grab buckets
and were endangered by the swing of the buckets themselves. Nearly
as important has been the modification in the structure of the orecarrying boats, making them more accessible and requiring fewer
men in the operations.
The degree of change in ore-handling equipment is shown by the
fact that in a large ore yard 289 men were required in 1905 and 93 in
1910, although the quantity of ore handled was much larger in 1910.1
The only machine used in distinctively steel works’ operations which
came particularly under observation was the open-hearth charging
machine. The manufacturers are now providing fenders for the
truck wheels, covers for the gears, and other similar devices, bringing
1 Conditions of Employm ent in the Iron and Steel Industry in the United States (S. Doc. No. 110,
62d Cong., 1st sess.), Vol. IV, p. 132.

100

ACCIDENTS IN M A C H IN E B U IL D IN G .

their apparatus up to the standard set by the steel mills. Probably
the most important modifications concern the electrical portion of
the machine. They involve improvements in switches and controllers,
better arrangements for carrying the feed wires, and more reliable
and better inclosed motors. It is impossible to point out precisely
the modifications which contribute directly and intentionally to
greater safety, but the general truth that the more efficient machine
may easily be made also the safer machine is clearly evidenced.
In rolling-mill machinery certain changes, designed primarily to
secure greater tonnage, have benefited the working conditions. This
is well illustrated by the case of the sheet mills. In 1892 when this
department of the industry was gaining a foothold in this country,
the roll housings in use weighed about 5 tons each. The rolls were
22 inches in diameter, with 16-inch necks. Those seen in process of
manufacture have rolls not less than 28 inches, and in some cases 32
inches. This increase in size of rolls and an accompanying change
in housings have materially reduced breakage of rolls which was a fre
quent cause of accidents. Breakage has been still further reduced by
the introduction of a device by which a uniform temperature of the
rolls may be maintained.
The peculiar hazard of wire mills is in connection with the wire
drawing process. The wire-drawing bench had undergone no material
change for many years until the demands for safety directed atten
tion to it. In the iron and steel report1 is shown an automatic
stop. This was connected below the floor to the treadle by which
the revolving block was stopped. This did not prove entirely
satisfactory since the force necessary to apply was so considerable
that a tangle might be pulled into or through the handle of the stop
strongly enough to do considerable damage. Accordingly the stop
was modified by carrying a rope up from the handle over two pulleys
and down through the top of the bench. Below the bench top it is
attached to a strap collar around the shaft which drives the blocks.
A comparatively slight pressure on the handle or upon the cord at
any point of its course will tighten this collar and cause it to revolve
with the shaft. The collar is connected with the treadle and easily
exerts sufficient pressure to pull it down and stop the block. In this
modified form the stop facilitates operation since the wire drawer
can use it to stop his machine at a distance from the block. Such a
stop is important, since it can be applied to old-type installations
where the machinery is still so useful as to make it undesirable to re
place it.
Plate 10 shows an old-style drawing bench with fixed die boxes and
positive clutches. There are no stops. With this type of equipment
i Conditions of Em ploym ent in the Iron and Steel Industry in the U nited States (S. Doc. No. 110, 62d
Cong., 1st sess.) Vol. IV , plate 36.

Bull. 256— Labor.

P L A T E 10.—O L D S T Y L E W I R E - D R A W I N G B E N C H .

Bull. 256— Labor.

P L A T E 11.— M O D E R N W I R E - D R A W I N G B E N C H .

CH A PT E R V .---- M A C H IN E DESIGN AS A FACTOR OF SAFETY.

101

it happens very frequently that the operator is caught by hand
or foot in a tangle of wire and drawn against the die box. When
this occurs, the loss of the member can scarcely be avoided. Plate 11
shows a drawing bench in which the entire mechanism has been modi
fied. The changes affect materially the efficiency of the machines,
but nearly all of them have bearing upon safety. A conspicuous
feature of the design is the automatic stop built as part of the ma
chine. On each of three blocks the loop through which the wire is
carried appears and upon that by which the workman stands the wire
is shown in place. The clutch which operates the block is thrown in
or out by a double-ended treadle. The workman is depressing the
starting end. By depressing the other extremity in the same
manner the block is stopped. The automatic stop operates as
follows: A tangle of sufficient size to endanger the worker would
pull the loop through which it passes toward the block, as seen
at the right of the picture. This would act upon a plunger, seen
below the loop, forcing it downward. This acts upon the stopping
end of the treadle and releases the clutch. The necessary force is
determined by a spring whose tension can be modified to suit
conditions. The position of the stop is so convenient and its action
so satisfactory that it becomes the usual means of stopping the block
for ordinary purposes. Other features of the design, such as the
arrangement by which the block stops automatically if the wire
breaks, add to the safety of its operation, but these features are not
apparent in the illustration and therefore description is not attempted.
CRANES AND HOISTS.

In the machine-building industry cranes and hoists stand third
in accident frequency and fourth in severity. In the iron and
steel industry they are fourth in frequency and second in severity.
It is probably true that faulty methods of operation have been
more responsible for this record than has faulty construction. The
degree to which construction has needed modification is clearly
indicated by the specifications adopted by the iron and steel
electrical engineers.
The gantry crane has been responsible for a great many injuries
from the fact that the track on which it moves is at the ground level.
A wheel guard is now being incorporated in the design.
The chief implement of transportation in these shops is the over
head traveling crane. Its importance is so great that although its
structure and operation are thoroughly known to everyone familiar
with shops, it is desirable to present a brief description for the nonin
dustrial reader.
The use of the traveling crane has brought about a marked change
in shop-building construction. Since the walls must bear the weight

102

ACCIDENTS IN

M A C H IN E B U IL D IN G .

of the apparatus and of the loads conveyed by it they must be made
sufficiently strong and also given height enough to permit the transit
of materials above any obstructions which may be upon the floor.
Indeed, in some erecting shops where large work is handled cranes
may travel at two levels one above the other.
The elements of an ordinary traveling crane are the bridge and the
trolley. The bridge consist^ of two girders strongly fastened parallel
to each other, the size and distance apart being governed by the
weights intended to be lifted and carried. At each end of the bridge
are truck wheels which rest upon rails securely fastened to the walls
of the building. In the case of outdoor cranes the walls of buildings
are used, when convenient, otherwise special structures are provided.
The truck wheels are operated by a shaft running the length of the
bridge, to which is geared the bridge motor. By the action of this
motor conveyed through the shaft the bridge is propelled along the
rails in either direction.
The trolley (plates 12 and 13) moves from end to end of the bridge.
Upon it are mounted commonly two hoisting drums, a main hoist
used for weights up to the capacity of the crane, and an auxiliary
hoist applied to lesser weights and frequently used in conjunction
with the main hoist in the manipulation of the loads.
A moment's consideration will make it clear that this combination
of movements, the translation of the entire crane through the length
of the building and the transverse movement of the trolley, make it
possible to reach every point upon the floor below.
The operation of the crane is accomplished from the crane cage
suspended below the girders. In this cage are usually four controllers
by which the craneman starts and stops and determines the amount
of power applied to each of the four motors, namely, the bridge
motor moving the entire crane, the trolley motor actuating the
trolley transversely, the main hoist motor revolving the main drum,
and the auxiliary hoist motor operating the secondary hoist.
It is desirable to present in general terms some of the dangers and
their remedies. Beginning with the craneman the first essential is
safe access to his crane cage. One steel company presents three
pictures as illustrating the stages of their evolution in the matter of
access to the cage. In the first is shown a craneman climbing a
column; in the second he is seen going up a ladder; and the third
shows him carrying a basket of tools up a flight of stairs with
apparent ease. The elements of safety in this matter of ap
proach are stairs, a well placed and properly railed landing platform,
and a railed extension of the cage floor from which there is easy access
to the landing platform. The cage should be of noninflammable con
struction, and the electrical devices should be so inclosed as to mini-

Bull. 256— Labor.

P L A T E 13.—C R A N E T R O L L E Y , G U A R D E D .

C H A PT E R V .---- M A C H IN E DESIGN AS A FACTOR OF SAFETY.

103

mize the chance of burn or shock. In cranes used for the constant
transportation of large quantities of hot metal either provision should
be made for the escape of the craneman to an outside gallery if nec
essary, or, perhaps better, a fireproof chamber should be provided
into which the craneman can easily enter in case of emergency and
from which he can manage his crane. Such provisions are impera
tive in steel mills and may properly be considered in foundries where
large quantities of metal are handled.
Safety in inspection and repair should be the next consideration.
In the older type of crane these essential processes could be carried
out only by climbing on the girders and making way precariously to
the place of work. All cranes should be provided with adequate
foot walks. With some small cranes to which such walks could with
difficulty be applied repair platforms should be constructed at the
ends of the run. To the danger of falling during the process of repair
was added in the old type crane the chance that the machine might
be unexpectedly started from the crane cage. To guard against this
either a lock removable only by the repairman could be used on the
switch in the cage, or a special switch which could be thrown open
when repairs began could be installed upon the top of the crane.
What the chances were of being caught when such starting oc
curred may be judged by reference to plate 12. Not only did the
gears of this trolley menace the man who was working upon them
but the overhung gears would from time to time break and the fall
ing fragments have not infrequently caused fatal injury to workmen
below.
Plate 13 illustrates the thorough revision which has taken place in
the design of this piece of machinery. The two pictures are of ma
chines produced by the same company at different times. The
present construction not only safeguards the man who must work
upon the machines, but there is nothing which can fall in case of
breakage, and the inclosed gears, protected from grit and dust and
running in oil baths, operate more smoothly and have a longer life.
It might seem that with these improvements in design and with
high standards of quality in the materials and workmanship enforced
the chief difficulties of the overhead-crane menace had been met.
This, however, is far from being the case. These structural improve
ments ward off many serious accidents, but the larger number of those
occurring are due to faulty method, to swinging loads, falling loads,
to sudden or unexpected raising and lowering, and to improper man
agement of the chains and slings by which loads are carried. One large
steel company was led to make a study of the signaling system in
use in their works only to find that there was no system. The same
signal w&s being used for different operations and different signals

104

ACCIDENTS IN

M A C H IN E B U IL D IN G .

for the same operation. The result of the inquiry was the develop
ment and introduction of a definite code which became the standard
practice in all the company’s mills.
Beside the cranes above described which serve large areas and
move heavy weights there are many types of small hoists which have
in recent years been much improved and which greatly facilitate and
make safer the handling and transportation of less weighty objects.
ELECTRICAL APPARATUS.

It is somewhat difficult to decide whether the turbo-generator
should be considered as an electrical machine or under the heading
of engines. But in view of the fact that the manufacture of these
generators is largely in the hands of electrical companies, they will be
here considered.
Plates 14 and 15 present two power houses, one equipped with
a reciprocating engine belted to the generator, the other having an
installation of turbo-generators. It is not possible to convey by the
pictures a complete impression of the contrast presented in the matter
of moving parts which constitute the source of danger in power
houses. In all types of turbo-generators the number of exposed m ov
ing parts is very few, and in some nothing moving can be seen.
Beyond question then, the substitution of this form of prime mover
for the older form with its flywheel and belt reduces power-house
hazards.
The number of power-house and boiler-house accidents is compara
tively small. It is easy to draw the conclusion that they are not
important. Chart C, page 31, shows that in frequency of injury in
relation to the number employed they stand seventh, while in severity
they are third among the departments. It is clear, therefore, that
the reduction of their dangers demands serious consideration.
Beside the moving parts which are dangerous there is another point
in the majority of power-house installations which deserves con
sideration. This is the flywheel. As a moving object its hazard may
be very nearly eliminated by proper fencing.
Such precautions, however, would do nothing to prevent the “ rac
ing” of the engine, which might cause the flywheel to explode. This
subject clearly belongs in the section on engines, but it may be here
discussed.
A t times the mechanism controlling the speed of the engine becomes
deranged, permitting the flywheel to revolve at a dangerous speed.
The strain on the wheel may become so great that it breaks, and the
flying fragments may wreck the adjacent machinery and the build
ing. A recent case will serve to illustrate the destructive possibili
ties of such an event. In a rolling mill, while the attention of the
engineer of one of the engines was momentarily directed elsewhere,

Bull. 256— Labor.

PLATE

14.— P O W E R

HOUSE,

W IT H R E C IP R O C A T IN G
GENERATOR.

E N G IN E

BELTED

P L A T E 15.— P O W E R H O U S E , W I T H T U R B O - G E N E R A T O R S .

C H A P T E R V .---- M A C H IN E DESIGN AS A FACTOR OF SAFETY.

105

the speed of the engine rose, and before he could apply the stop pro
vided the wheel had exploded. The rim was broken into several
pieces and the spokes of the wheel detached from the rim and also
from the hub. The surrounding structures were badly shattered, but
more serious than this the controlling apparatus of another engine
was disabled and its flywheel also burst. A fragment of this second
wheel in its flight through the mill severed the lower part of the
roof trusses, allowing the roof to collapse for some 200 feet upon the
wreckage below.
Manifestly, two procedures would afford some protection against
such an occurrence. First, the generation of electric energy by means
of a turbo-generator and its application to the rolls by means of
motors. Second, the use of some effective engine stop which upon
increase of speed would automatically shut off the steam and so
bring the engine to rest. These stops will be further discussed at a
later point.
There are serious difficulties with each of these plans for the
particular mill in which the accident occurred. It is not necessary
to elaborate upon these since the occurrence is described simply to
give point to the assertion of this kind of danger and to emphasize
the need of adequate attention.
Deep-seated flaws in the wheel are sometimes the underlying
cause of explosion. Inspection is the only safeguard.
In generators and motors it is noticeable that recent patterns
inclose and screen the moving parts much more perfectly than was
formerly the case. In fact accidental contact with such parts is
almost impossible in the cases which came under observation.
The proper installation of the wiring by which the electric current
is conveyed to the machines is important both from the standpoint
of fire hazard and because where open wiring is used workmen
engaged in other operations may come in contact with wires upon
which the insulation has become worn or otherwise imperfect. The
general use of metal conduits for all machine feed wires greatly
reduces these risks.
Danger to machine operators and others controlling the ultimate
application of the current arises almost entirely in the manipulation
of switches and controller handles. These are now usually com
pletely inclosed.
Another element of safe operation resulting from improved electrical
construction is impossible to present by illustration. Its character
is indicated by a case occurring in a mill whose product is tubes.
At one step of the process the tubes were conveyed by an electrically
actuated transfer. Occasionally this transfer would move too far
and cause the load of pipes to spill into a passageway. The solution
was found in the introduction of an improved electrical equipment

106

ACCIDENTS IN

M A C H IN E B U IL D IN G .

which could be more promptly and exactly controlled. The installa
tion was made with no other idea than that of removing the risk
incident to the operation of the original apparatus. When it was
put into action it was discovered that it permitted an increased
production.
LOCOMOTIVES.

From the manufacturing standpoint locomotives are essentially
like other machines. Their construction involves the same opera
tions and shop problems. Since these are necessarily on a compara
tively large scale the establishments devoted to this form of con
struction afforded excellent opportunities for prosecuting the inquiries
essential to this study.
When the locomotive is considered from the operative standpoint
it seems to be in a class by itself. The engineer and fireman are
exposed to comparatively few of the sort of dangers which appear in
the operation of other machines. With the single exception of the
valve-shifting levers, the moving parts are so placed that the engineer
is not menaced thereby. His danger arises almost entirely in the
peculiar purpose for which his machine is designed. In the liability to
explosion of the boiler the engineer and fireman share a hazard to which
others in power-house operations are exposed. The breaking of a
driving rod which in its revolution destroys the side of the cab and
kills the engineer might be compared with the explosion of a flywheel
as heretofore described. Consideration of these elements of danger
common to locomotive operation and the workings of other machines
serve rather to emphasize than otherwise the essential dissimilarities.
While the dangers, which have been more or less modified by changes
in design, lie in the field of transportation, and accordingly rather
outside the scope of the present study, it is desirable to indicate some
of them.1
The increase in size has produced for the engineer an increasing
physical strain accompanied with some chance of physical injury in
the operation of his reversing lever. The size and weight of the
valves to be shifted have necessarily kept pace with the other in
creases in dimensions. To offset this there has been reintroduced a
steam reverse. This was tried out some years ago, but the ad
vantages under the then existing conditions did not suffice to keep
it in use. A common form used at present has a small lever whose
movements are precisely similar to those of the hand reversing lever.
The engineer has only to set this lever in the proper position and the
mechanism does the rest. This may be regarded as a change in
design which both relieves the man of undue stress and contributes
in some measure to his safety.
i The facts presented are largely drawn from a paper presented to the Franklin Institute on
“ Recent development of the locomotive,” by Mr. Geo. R . Henderson, consulting engineer of the Baldwin
Locomotive Works.

C H A PT E R Y .---- M A C H IN E DESIGN AS A FACTOR OF SAFETY.

107

The development of mechanical stokers and the use of oil as fuel,
which requires only the proper adjustment of valves, have proceeded
under the impulse of the necessity to keep the firing operations within
the compass of the activity of a single fireman. These changes have
either improved the working conditions or prevented them from
becoming so bad as to be intolerable. They have not greatly modified
the liability to such accidents as were likely to happen under earlier
conditions.
Some structural features recently incorporated have a bearing upon
accidents due to breakdown of the machine. For example, until
recently the use of what is known as the Stephenson valve movement
has been almost universal in this country. This involves the use of
eccentrics on the axles and the adjustment of certain working parts
in the space below the boiler and between the driving wheels. The
utilization of this space for this purpose rendered it impossible to
brace the frames as securely as their increasing size demanded.
Coincident with this need for greater strength the Walschaerts valve
motion was being adopted. Since its working parts are entirely out
side the frames it left clear a space for extra bracing and so provided
a needed factor of safety in the development of the locomotive.
The replacement of cast iron and to a considerable extent of
forgings by cast steel has contributed more than any other single
item to the maintenance of a factor of safety commensurate with the
increasing size. Prior to the development of the open-hearth process
steel of a suitable quality for castings was obtainable only from cru
cible furnaces. The small quantity thus produced necessarily limited
steel castings to parts of small size. Now locomotive driving wheels,
frames, saddles, and other large parts are made in cast steel, with a
great saving in weight and increase in strength. Further progress in
this direction is likely to occur from the introduction in locomotive
construction of the alloy steels which have a still higher degree of
strength.
As before suggested, the dangers to workmen who operate locomo
tives lie in the field of transportation and their consideration would
lead too far from the subject of the present study. What has been
given above serves only to indicate some of the changes which have
tended to keep down the accident hazard.
OTHER PRIME MOVERS.

Certain of the dangers attendant upon the operation of the com
mon reciprocating engine have been already presented in the section
on electrical apparatus and need not be further considered. Plate 14
shows the need of certain precautions in an engine room where the
prime mover is of this character.

108

ACCID EN TS IN M A C H IN E B U IL D IN G .

Going back a step it is pertinent at this point to consider a few of
the methods adopted by manufacturers of boilers which tend to the
safety of those employed about them. At the present time there is
in progress an important effort to determine standards of construc
tion suitable for general adoption. In this effort the American
Society of Mechanical Engineers has had a most prominent part.
Installations for the production of industrial power have been
undergoing a development similar to that, already outlined, in the
case of the locomotive. The demand for increased power more
economically produced has led to great increases in boiler capacity,
which in turn has led to the use of stronger materials and the use of
more reliable structural methods. On the whole these changes have
kept pace with or gone beyond the increase in size. As a result the
modern boiler room is much less likely to be the scene of a wholesale
disaster from explosion than was the case some years ago, while the
introduction of mechanical stokers and fuel-handling devices has
appreciably reduced the danger of minor injury.
The effect of the introduction of the turbo-generator in reducing
almost to the vanishing point the exposed moving parts of engines
has already been sufficiently indicated.
It remains to speak briefly of the effect upon safety of the rapid
development of internal combustion engines, both gas and oil. In
general it may be said that such engines present fewer exposed mov
ing parts which the attendant is required by his duties to approach
than was the case with former types. There is one added hazard which
experience proves must be considered. The gas used or produced in
the operation of these engines is frequently noxious and has caused a
number of deaths from asphyxia. It will not do to depend upon
what are called natural means for preventing the accumulation of
these harmful emanations. Mechanical means of ventilating and
a rigorous enforcement of a rule that men shall not go alone into
places where the gas may possibly accumulate are necessary. This
matter has not received as much attention as it deserves.
Steam pumps have the features characteristic of prime movers
and require similar safeguards.
MACHINE TOOLS.

The term “ machine tools” applies to a great variety of apparatus
used in the machine shop. Since the modifications in design are
more conspicuous than in any other of the groups in which product
has been considered it is desirable to present them in greater detail.
Accordingly the forms will be briefly described, their particular
hazards noted, and the methods of protection indicated.

C H A PT E R V .---- M A C H IN E DESIGN AS A FACTOR OF SAFETY.

109

METAL PLA N ERS.

A metal planer consists of a horizontal bed upon which a platform
travels back and forth. The upper surface of this platform has
slots and openings by which the work to be machined may be fastened
securely in place. The tool or tools are carried upon an arm, or may
be attached to a crosspiece supported by uprights on either side
of the bed. Elaborate means are supplied for raising and lowering
the tool-carrying bars and for shifting the tools laterally. The size
of the castings which can be machined is limited only by the width
of the platform and the height to which the tool can be raised.
The application is mainly to rather heavy work which requires
modification in the form of cuts in one direction.
The dangers in operating these machines are: First, one which
pertains in varying degrees to all machining processes, namely, flying
fragments projected from the tool. These may be thrown off with
sufficient force to penetrate the flesh but more frequently are harm
less except when the eye is struck; second, contact with exposed
belts and gears; third, being caught by the tool when inspecting,
measuring, or brushing off chips; fourth, and most serious, being
caught by the moving platform or the work thereon.
The use of goggles by machine operators has already been dis
cussed. Against the third danger no provision can be made except
the care and caution of the workman. The openings in the planer
bed offer a serious menace. Illustrations will best show its nature.
A shop foreman was standing on the moving platform observing the
progress of the work being done on the machine. As he stepped
backward a nut or some other fragment upon the platform caused
an unexpected disturbance of his footing and he stepped down in
front of this moving platform, losing his leg. In another case the
workman had some tools in the end compartment of the bed. Reach
ing in, his foot slipped and he fell head forward into the space and
was fatally crushed. A third case happened during the shop studies
made preliminary to this report and the writer had an opportunity
of personal inspection. The workman in this case was entirely alone
and so there is no evidence on some points. Apparently he leaned
forward to take a measurement and slipped, falling in front of the
casting. He was drawn against the upright and killed.
The second and third illustrative cases call attention to a hazard
which undoubtedly underlies cases of accident which appear as due
to other causes, namely, insecure footing. The oily floors in the
vicinity of machines may probably have caused serious accidents
which here appeared mysterious. The magazine 1 ‘ Safety Engineering 7 ’
has recently been doing good service in centering attention upon

110

ACCIDENTS IN

M A C H IN E B U IL D IN G .

this possibility. The safeguard may be in proper construction of the
floor, or in the shoes worn, or both. This cause of injury deserves
more serious attention than it has yet received.
BORING MILLS.

The name of these machines does not convey an exact idea of
their function to those not acquainted with them. A boring mill
may be described briefly but quite properly as a planer whose plat
form rotates instead of moving back and forth. Plate 17 shows
the essential features. The tools are carried in a manner exactly
similar to that adopted in the planer and the arrangements for
clamping the work to the platform are the same. Whereas the
planer permits extended longitudinal cuts the boring mill allows
those of a circular form, the size being determined by the diameter
of the platform. As originally designed the machine was intended
for producing or machining out cylindrical openings in the cast
ings, hence the name. Its function has now been so much amplified
as to render this name not fully descriptive.
The most important changes are due to the application of elec
tricity as a motive power (plates 16 and 17) and to the introduction
of high-speed steels making possible deeper cuts. Both of these
items have influenced the demands regarding strength and rigidity
of construction.
The dangers are (1) flying fragments, (2) gearings, (3) injury when
inspecting or cleaning work, (4) a possibility of falling on the plat
form and being bruised or crushed between work and uprights.
Inspection of the plates will show the provisions made in the later
types against the danger of being caught in gears. Crushing injury
is mentioned as a possibility although no case of such injury has
come under observation. It is evident that such occurrences are
not so likely to happen with this machine as with the planer.
Plates 16 and 17 show the rear of the machine and those modifica
tions of the application of the motive power which make for greater
safety.
It will be noticed in plate 17 that the gearing at the top of the mill
is fully covered. It is sometimes urged that this is needless. There
are three justifications: (1) An oiler must sometimes approach
these gears when in motion; (2) where electric lamps are used about
the machine at the end of long leaders these sometimes become en
tangled in open gears, and the machine may be seriously damaged;
(3) gears so covered wear longer and run better than when not covered.
In even a very clean shop there is a good deal of dust and grit in
the air. The exclusion of this is a distinct advantage.
It is sometimes urged that counterweights such as appear in these
plates should be arranged so that in case the chain breaks their fall

Bull. 256—Labor.

P L A T E 16.— B O R I N G M I L L , W I T H U N G U A R D E D GEARS.

P L A T E 17.— B O R I N G M I L L , W I T H G U A R D E D G EA R S.

Bull. 256— Labor.

P L A T E 18.— LA R G E GU N L A T H E , W I T H U N G U A R D E D G EA R S.

Bull. 256— Labor.

PL AT E 19.— L ARG E GU N L A T H E , W I T H G U A R D E D GEARS.

CHAPTER V .---- M ACH IN E DESIGN AS A FACTOR OF SAFETY.

H I

will not endanger the workman. Careful scrutiny of the records
fails to disclose instances of such injury in the operations of tools of
this kind. The case is different where the weight is very frequently
shifted and a strain thus put on the chain likely to cause deterioration
and final breakage. However, some cases have been observed
where a pit was made below the floor and the counterweight chain
extended so that the weight moved in the pit. This removed even
the remote danger due to the breakage of the chain.
LATHES.

The essential elements of a lathe are the bed, the head stock, the
tail stock, and the tool holder. The tail stock and tool holder are
movably mounted upon the bed so that they can be adjusted to
different points of its length. In general the lathe is used to produce
changes in shape of cylindrical objects, such as shafts, rolls for rolling
mills, and similar parts.
The head stock contains the mechanisms by which the speed of
rotation is determined and the motion of the tool holder along the
bed is accomplished. The tool is moved up to the work either
manually by the workman or by automatic devices.
A modification of the ordinary lathe, much used where a large
number of pieces of the same form are to be produced, is the turret
lathe. Upon the tail stock is mounted a revolving turret bearing a
number of different tools which in succession perform operations
necessary to the formation of the required part. The turret is
revolved by a handle operated by the workman. From this type
of lathe has been evolved a number of machines which perform all
the operations automatically. The hazard of their operation is so
small that no attention has been given to them in this study.
For special processes suitable types of lathe have been produced.
Plates 18 and 19 illustrate lathes for the turning of large harbor
defense and naval guns. They were made by the same firm. Plate
19 represents a lathe now in use in the gun shop at the Washington
Navy Yard.
Comment upon the differences between these pieces of apparatus
is scarcely necessary. It may be pointed out, however, that plate 19
represents a more radical revision of the entire design than anything
heretofore presented.
The dangers arising in the operation of lathes are those already
enumerated and in addition one not hitherto encountered. With all
revolving cylindrical objects there is danger that some loose portion of
the workman's clothing may be caught. This is not very serious with
metal-working lathes since even in the most rapid operation the rate
of revolution is not high enough to be seriously dangerous. When,
however, there are projections on the revolving part injuries may
12456C)°— 20— Bull. 256------ 8

112

ACCIDENTS IN M A CH IN E BUILDING.

occur. These have been most frequent from the set screw of the
dog used to hold the work in place against the stress of the tool.
Several forms of dog in which either no set screw was used or
the set screw was securely covered have been developed in the shops,
but apparently no lathe manufacturer had given the matter serious
thought at the time when the field work for this study was in progress.
DRILLS.

A number of different forms of apparatus for carrying these tools
are in use. Plates 20 and 21 present one of these in an early and
one in a recent condition of development.
In the earlier form the familiar cone pulley will be noticed in a
position liable to cause injury sometimes even when shifting of the
belt is not occurring. The many exposed gears at various points
are obvious. In the modern form delineated the application of power
is by a constant speed pulley, variations in transmitted speed being
accomplished by gears in the box shown. The almost complete
invisibility of the various actuating gears is noteworthy.
When all these precautions have been taken there remains a source
of danger scarcely possible to provide against in the drill itself. This
may be illustrated by a case or two. A workman operating a hori
zontal drill of rather large size drew the drill back from the work to
make some measurement. As ha bent forward the point of the drill
touched his overalls just below the hip. Instantly he was snatched
from his feet and whirled around on the drill. Before the machine
could be stopped by his fellows his head was repeatedly struck against
the iron platform to which the work was attached, with fatal results.
The foreman in charge of this work afterwards devised a drill chuck
in which the drill ran free except when the drill was pressed against
the resisting work. Some difficulties developed in the operation of this
chuck and so far as known it has not been successfully applied in
practice.
In another instance a young apprentice reached around an upright
drill and was caught by the sleeve. The twist of the drill gathered in
the cloth, tearing it as it did so, until his garments were entirely
removed except his shoes.
Almost the only possible safeguards appear to be the wearing of
close-fitting clothing, and extra care on the part of the operator.
Safeguards thus far tried have either proved ineffective or have so
seriously interfered with the usefulness of the machines as to be
impracticable.
M ILLING M ACHINES.

The development of this machine has been of great importance in
the recent history of shop practice. Its essential element is the
toothed cutter. This may be regarded as a special form of saw

B ull. 256— Labor.

P L A T E 20.—O L D T Y P E OF D R I L L , W I T H U N G U A R D E D G E A R S .

B u ll. 256— Labor.

P L A T E 21.— N E W T Y P E OF D R IL L , W IT H G U A R D E D G E A R S .

Bull. 256—Labor.

P L A T E 22.— M I L L I N G M A C H I N E , W I T H U N G U A R D E D CU T TE R .

P L A T E 23.— M I L L I N G M A C H I N E , W I T H G U A R D E D C U T T E R .

B u ll. 256— Labor.

P L A T E 25.— B E L T S H I F T I N G W I T H S H I F T E R .

CHAPTER V .---- M ACH IN E DESIGN AS A FACTOR OF SAFETY.

113

since the teeth of the revolving cutter are brought into successive
relation with the work in precisely the same manner as the teeth
of a saw. It is evident that the possible speed of work in cases to
which the machine is adapted would be materially greater than that
with the methods of reduction already described.
Plates 22 and 23 illustrate a case in which a development secured
for improvement of production has in part solved a safety problem.
Plate 22 shows a machine with a spiral cutter as usually operated.
Plate 23 shows the same machine with a hood over the cutter,
the lubricant delivered through a pipe being confined by the hood
and made completely to flood the cutter. Very important advan
tages are secured in rapidity and smoothness of action by this
method of lubrication. It will appear at once that this covering of
the cutters by a hood lessens in a material degree the chances of
accidental contact and injury.
MACHINE TOOL ACCESSORIES.

Loose pulleys on the driving shaft, with a fork by which the belt
may be shifted from the loose to the tight pulley and reverse in
starting and stopping the machine, have now been in use so long that
the fact of hand adjustment is almost forgotten. Early safeguarding
laws, by requiring such appliances, bear witness to the fact that
machines were once operated without them.
Changes of speed by the cone pulley are still secured by the method
shown in plate 24. The dangers of the operation are twofold. Any
roughness on the belt will by its swift motion be liable to cause
laceration of the hand, and in case of motion in certain direction
the workman runs a risk of having his hand drawn between belt
and pulley. He would be fortunate if the injury were confined to
his hand.
Plate 25 shows a compact and simple shifter by which both upper
and lower cones are cared for by rotating a handle. A careful test
of this shifter on work which required somewhat frequent shifts
showed a very considerable increase in output.
TRANSMISSION GEARING.

The great and rapidly increasing use of electrical distribution and
application of power has ajpeady been pointed out. In spite of this
development there remains an immense field in which mechanical
transmission will for a long time, if not permanently, be of great
importance.
In conveying power from the prime mover to the first point of
application some form of belt was formerly the exclusive method.
With increasing size of engines it became difficult to produce belts
in proportion. This led in one direction to direct connection for the

114

ACCIDENTS IN M A CH IN E BUILDING.

generation of electricity and in the other to the substitution of rope
drive.
Such a drive requires complete housing to be safe. This instance
is on record: A strand of the drive broke, the loose end wrapped
around a railing, tore it from its fastenings, and, carrying it to the
far end of the drive, struck a man standing there, inflicting serious
injury. Since the safeguards in this case can not be made an integral
part of the machine the manufacturer’s duty becomes that of giving
proper advice to his customer rather than that of modifying his
design.
Following the distribution of power still farther, the line shaft is
reached. The primary danger from this is to the oiler who must
approach the swiftly revolving shaft in the prosecution of his duties.
This involves danger in any case, but it is aggravated when the shaft
has projecting set screws. This danger is obviated when these screws
are of the safety pattern or are properly covered. A still more
effective safety device is the use of self-lubricating bearings. These
when properly constructed need no attention for considerable periods
and any necessary attention can easily be given when the mill is not
running. These self-oiling bearings can be attached not only to line
shafting but to countershafts as well, and thus entirely do away with
a very dangerous practice.
The secondary danger of shafting arises in connection with the belts
by which power is finally brought down to the machines. If a work
man is caught by such a belt it becomes of the utmost importance
that he or his fellows be able promptly to stop the movement of the
shaft. If to do this requires communication with the power house the
stoppage will probably be useful only for the removal of the injured
man. Both for safety and for economy in the use of power the sub
division of the power transmission into small units which can be
independently brought to rest is of the highest importance. This
control of reasonably small portions of the transmission equipment
is made possible by the use of friction clutches. Arrangements,
mechanical or electrical, can be installed for the operation of these
at various points so that in case a man is caught some one will be able
promptly to disconnect the portion of the transmission involved from
the source of power.
£>uch clutches have been in use for a long time, but the present
demand for safe operation has stimulated improvement in the appa
ratus and has led to a marked extension of their use.
OTHER PRODUCTS.

The divisions of mining machinery and shipbuilding need not be
considered from the standpoint of machine design, this being some
what outside the primary purpose of this study.

CHAPTER V .---- M ACH IN E DESIGN AS A FACTOR OF SAFETY.

115

In conclusion, it may be said that, in every particular, American
machinery has made a notable advance since the publication in 1880
of Volume X X I I of the Tenth Census, which has been used in this
chapter as a point of reference.
The manufacturers of machines are now ready to furnish machines
safe so far as mechanical device can make them. It remains for users
of machines to put into their shops only those which embody these
safeguards. Unsafe machines can still be bought at some saving in
price. A single accident might wipe out years of such saving.

116

ACCIDENTS IN M A CH IN E BUILDING.

APPENDIX A.
R E SU LT S OF ACCIDENTS IN 194 M ACH INE-BUILDING
Accidents resulting in
Permanent injury.
B

1 joint of finger
or fingers.

1 joint of thumb.

Thumb.

bi)
S

Both legs.

Both feet.

1 joint of other
toe or toes.

153
fc

Other toe or toes.

Great toe.

Foot.

Loss of-

1 ,joint of great
toe.

Departments of plants and
products.

DEPARTMENT OF PLANT.

2,994
20,144
11,373
2,776
717
12,307
37,595
1,468
877
3,571
1,221
20,660

6
4
6
3

4
7

2
2

2
1
3
1

1
1

Total.................................... 115,703

Machinery for the steel industry. 2,692
Cranes and hoists......................... 4,362
Generators and motors................ 35,674
Engines......................................... 31,229
Machine tools............................... 24,359
Mining machines.......................... 3,994
2,226
Transmission................................
6,615
Ships..............................* .............
Unclassified................................... 4,552

1
1
5
22
3

2
4

1
7

Total.................................... 115,703

Boiler shops...................................
Electric shops...............................
Erecting shops..............................
Forge shops...................................
Foundries (brass).........................
Foundries (iron)...........................
Machine shops..............................
Maintenance.................................
Power............................................
W ood working...............................
Yards.............................................
Unclassified...................................

1
6
1
1

8
32
20
5

3
4

19
73

1
1
i

9
3
16

186

3
4
2
1

12 i

1
1

PRODUCT.

3
2

7
15
42
50
47
6
5
4
10

186

8
8

117

APPENDIX.
APPENDIX A.
PLA N T S IN 1912, B Y D E PA R TM E M TS A N D B Y PRODUCTS.
Accidents resulting in —
Temporary disability (loss of 1 day or over)
terminating in—

Permanent injury.

Loss ofM

o>
©
s* .

o £ a*
©
-t-3

d
Q
Ph

a
EH

I
4
16
7

362
929
1,148
248
24
926
2,627
71
36
148
136
1,025

4
18

112

263
330
71

290
537

19
37
44
316

81
131
4
5

13
60

7,680 2,048

512

24
15
13
3

271
498
1,983
2,597
1,037
456

19
137
195
61
30

112

596
130

7,680

636
97 1,579
92 1,992
28
439
52
11
153 1,670
233 3,923
14
137
12
87
264
18
260
19
2,160

30
6789

272

1,197 13,199 13,647

438
148
813
193 3,455
301 4,348
46 1,486
...... 755
186
380 1,422
97
296

657
196
137
32
209
28

671
1,648
2,054
455
52
1,723
4,062
138
91
292
270
2,191

540

455
829
3,560
4,530
1,557
767
195
1,440
314

1,197 13,199 13,647

118

ACCIDENTS IN M A CH IN E BUILDING.

APPENDIX B.
R E SU L T S OF ACCDJENTS IN A M A C H IN E -B U IL D IN G
Accidents resulting in
Permanent injury.

S
-4
©

1 joint of finger
or fingers.

1 joint of thumb.

Thumb.

hi>
<D
Hi

Both legs.

Both feet.

[ Foot.

1 joint of other
toe or toes.

Other toe or toes.

Great toe.

a
3
£

1 joint

of great
toe.

Loss of-

Occupations.

PRODUCTIVE.

Bench, and vise hands.................
2,937
4,350
Blacksmiths and helpers............
Boiler makers and helpers
2,413
Calkers and chippers...................
1,769
Drillers and helpers.....................
3,269
Erectors and helpers.................... 2,360
Machinists and helpers................ 18,534
Machine hands.............................. 1,290
Reamers, riveters, and helpers.. 2,734
Sheet-iron workers....................... 1,946
Occupations w ith less than 1,000
exposures................................... 7,035

1
1
2
5
1
3
3

Total.................................... 48,637

1
1
1

2
1
1

1
2

1
1
2
1

1
4

1
2

11
4
22
1
5

3
1

1
1
3

1
7
3

24
4

1
1

2
3

4
5

NONPRODUCTIVE.

Carpenters ................................... 1,074
920
Core makers .............................
Cranemen...................................... 1,011
Laborers........................................ 10,035
Pattern makers............................
1,246
Occupations w ith less than 1,000
exposures................................... 2,377
Unclassified................................... 7,236

3
14
1

6
13

1
2

T otal...................................

16,663

Grand total........................

72,536

101

119

APPENDIX.

APPENDIX B.
PL A N T, 1907 TO 1913, B Y OCCUPATIONS.
Accidents resulting in—
Temporary disability (loss of 1 day or over) ter
minating in—

Permanent injury.
Loss of—

d 03

B
CD ’S
d
pr-*

I
2

'i*
3

109
187
354
214
268
256
816
136
291
32

18
36
75
39
44
-30
147

165

45
7
34
715
18

159
147

17
4
3
3

160

342 3,953

334

352

4,643

120

114

55
7
260 1,278

61
1,365
49

3
136
5

755

351

133

54
74

978

178
339
620
321
452
436
1,371
234
511
48

63
234

24
174
318
36
78
607
35
311
62
429
408
69
196 1,301
50
225
67
490
2
46

191

291
292

310
321

344 1, 761

1,911

10 1,100 6,696

"7T094

INDEX.
Accident:
Page*
Definition of..............................................................................................................................................
15
Important causes o f..................................................................................................................................
11
Accident causes:
B y plant groups and by departm ents.................................................................. ................................46-48
In a steel plant, 1905-1913........................................................................................................................ 49,50
In five machine-building plants, 1907-1912............................................................................................
48
In four machine-building plants, 1910-1913........................................................................................... 50,51
Necessity of rates for the measurement of............................................................................................. 51,52
Over a series of years................................................................................................................................48-51
Accident experience:
In a large steel plant, 1910-1913...............................................................................................................22,23
Of machine-building plan ts.................................................................................................................... 27-78
Accident frequency rates:
For a large steel plant.........................................................................................................................21-23,49
iPor a portion of the iron and steel industry, 1907 to 1917..................................................................74,75
For electrical assembly shops............................................................................................................ 37,47,54
For fabricating shops..............................................................................................................................
75
For fatalities in engine building.............................................................................................................
36
For forge shops..........................................................................................................................................
38
For foundries.................................................................................................................................. 38,39,47,54
For locom otive building.........................................................................................................................
71
For machine building...................................... 22,27,30,34,36,41,43,47,48,50,53,54,57-61,63,66,69,70,72
For machine shops....................................................................................................... ........... 39,40,47,54,71
For m unition plants............................................................................................................................... 71,72
For private shipbuilding.........................................................................................................................63,72
For the iron and steel industry, 1910.....................................................................................................
66
For United States arsenals and. navy yards..................................................................................... 63,66,72
For woodworking shops...........................................................................................................................
40
Meaning of, and method of determ ination...................................................................................... 8,16-18
Accident occurrence, effect of safety system s upon............................................................................... 10,40-42
Accident prevention:
Importance of nature of injury from standpoint of..............................................................................11,12
Methods of.................................................................................................................................................13,14
Accident rates:
B y departments......................................................................................................................... 9,29-33,37-40
Course of, over a series of years............................................................................................................... 9,10
Day and nigh t.......................................................................................................................................... 57-60
Day and night, for machine shops in machine building and for mechanical department of iron
and steel industry, comparison of....................................................................................................
60
59
Day and night, in a large steel plant, 1905-1913, comparison of, by departments (chart)..............
Day and night, in machine-building plant, comparison o f.................................................................
58
Fluctuation in, cause of........................................................................................................................... 76,77
For a large machine-building establishment, by occupations.............................................................10,11
For machine-building industry in 1912, by character of product.......................................................27-29
For machine-building industry in 1912, by departm ents....................................................................29-33
For the industry as a w hole...................................................................................................................
9
Higher at n ig h t.........................................................................................................................................
12
Industrial, discussion o f.......................................................................................................................... 15-25
In steel manufacture and in machine building, comparison of...........................................................
22
Methods of determining........................................... ..............................................................................
16
Occupational, discussion of..................................................................................................................... 42-46
Over a series of years................................................................................................................................34-36
War, influence of the, on accident rates in machine building............................................................ 68-78
(See also Accident frequency rates; Accident severity rates.)
Accident reports in Government shops and in the iron and steel and the machine-building industries. 65-68
Accident severity rates:
Advantage over frequency rates.............................................................................................................
25
B y character of product...........................................................................................................................
9
For a large steel plan t........................................................................................................................ 21-23,49
For a portion of the iron and steel industry, 1907 to 1917................................................................ 74,75
For electrical assembly shops............................................................................................................ 37,47,54
For fabricating shops.............................................................................................................................
75
For forge shops.........................................................................................................................................
38
For foundries.................................................................................................................................. 38,39,47,54
For locomotive building........................................................................................................................
71
For machine building................................................ 22,27,30,34,36,41,43,47,48,50,53,54,57,63,69,70,72
For machine shops.................................................................................................................. 39,40,47,54,71
For m unition plan ts............................................................................................................................... 71,72
For private shipbuilding...................................................................................................................
63,72
For U nited States arsenals and navy yards......................................................................................... 63,72
For woodworking shops...........................................................................................................................
40
Illustrations of use of................................................................................................................................ 21-25
Meaning of, and method of determ ination........................................................................................8,18-25
Scale of tim e losses, used in .....................................................................................................................
20
Accident studies, purpose of..........................................................................................................................
15
Accidents:
Cost of, compared w ith pay roll, under bonus system of safety organization...................................
83
Distribution of, by months................................................................................................................ 12,61,62
Inability to speak English as related to ......................................................................................
12,55-57
Method of measuring seriousness of.......................................................................................................
8
R esults of, in a machine-building plant, 1907-1913, by occupations...............................................118,119
Results of, in 194 machine-building plants in 1912, by departments and by products................116, 113k

121

122

INDEX.

Accidents, frequency and severity of:
Page.
Among American and foreign-born workmen in a machine-building plant, 1910-1913...................
57
In a machine-building plant, 1907-1913, by occupations......... ........................................................... 43-46
In five machine-building plants, 1907-1912............................................................................................ 34-36
In four machine-building plants, 1910-1913...........................................................................................
36
In three groups of machine-building plants, b y character o f safety system s....................................
41
In U nited States arsenals and n avy yards, 1912-1914, and in machine and ship building, 1912 63,66,72
In 194 machine-building plants in 1912, b y departm ents.................................................................... 30,31
In 194 machine-building plantain 1912, b y products...........................................................................27-29
American-born and foreign-born workers, comparison o f accident rates of.............................................
57
Arsenals, Government. (See U nited States arsenals and navy yards.)
Boiler shops, high accident rates in...............................................................................................................
9
Bonus systems. (See Safety organizations.)
Boring mills, description, dangers of operation, and methods of protection........................................ 110, i l l
Causes of accidents. (See Accident causes.)
Cranemen, high severity rate of..................................................................................................................... 10,43
Cranes and hoists:
Dangers of operation and methods of protection.............................................................................. 101-104
Improvements in design of, making for safety.................................................................................. 103,104
Relative standing as to accident frequency and accident severity in machine-building and
iron and steel industries......................................................................................................................
101
Crane, traveling, description of...................................................................................................................101-104
Day and night accident rates:
In a machine-building plant, 1907 and 1910 combined, comparison of..............................................
60
In a machine-building plant, 1913, comparison of...................................... ! .......................................
58
In machine building................................................................................................................................ 57-60
D ays lost:
Definition and use o f term .......................................................................................................................
23
In seven machine-building plants, 1907-1913, as result of injuries to workers..................................
53
Used as a measure in severity rating.....................................................................................................
19 .
Departments of plants:
Accident rates for 1912 in ......................................................................................................................9,29-33
Causes of accidents i n ...............................................................................................................................46-52
Course of accident rates for......................................................................................................................37-40
D ay and night accident frequency rates for a machine-building plant, 1907 and 1910 combined, by
60
Direct safeguarding m ethods in machine building..................................................................................... 85-97
Disabilities, number and per cen t of, term inating in specified weeks......................................................
66
Drills, description, dangers of operation and methods of protection........................................................
112
Electric wiring, dangers from, and methods of protection.........................................................................
105
Electrical apparatus, improvements in, making for safety.................................................................... 104-106
Electrical assembly shops:
Accident frequency rates for six plants, 1907-1913................................................................................ 47,54
Accident rates in .......................................................................................................................................
37
Accident severity rates for six plants, 1907-1913................................................................................... 47,54
Eectrical manufacture, safeguarding in ....................................................................................................... 95,96
Engine building, fatal accidents i n ...............................................................................................................
36
English, inability to speak:
As related to accidents........................................................................................................................12,55-57
56
As related to accidents, experience of a large steel plant, 1906-1913 (chart).....................................
Fatalities:
In engine building, 1902-1913..................................................................................................................
36
' Tim e losses for injuries causing, how determ ined................................................................................ 19,20
Flywheel, dangers in operation of, and methods of protection...................... ....................................... 104,105
Foreign-born and American-born workers, comparison of accident rates of............................................
57
Forge shops, accident rates in .......................................................................................................................
38
Foundries:
Accident frequency rates for five plants, 1907-1913..............................................................................
47
Accident frequency rates for six plants, 1907-1913................................................................................
54
Accident rates in .......................................................................................................................................38,39
Accident severity rates for five plants, 1907-1913..................................................................................
47
Accident severity rates for six plants, 1907-1913...................................................................................
54
Description of types of buildings for...................................................................................................... 85,86
Goggles for use in, necessity for suitably colored glasses......................................................................89-91
Goggles for use in, qualifications of an a specifications for testin g...................................................... 89,90
Safeguarding i n .........................................................................................................................................87-91
Shop conditions i n .................................................. ................................................................................. 85-87
Full-time worker, definitions o f..................................................................................................................... 7,17
Goggles. (See Foundries.)
Government shops. (See U nited States arsenals and navy yards.)
Industrial accident rates, discussion of.........................................................................................................15-25
58
Injuries in a machine-building plant, 1913, distribution through the day and n ig h t..................... .
Injury, nature of:
Classification o f.........................................................................................................................................
53
Importance from standpoint of accident prevention...........................................................................11,12
In seven machine-building plants, 1907-1913........................................................................................
53
Injury, permanent results o f......................................................................................................................... 54,55
Iron and steel industry, accident rates in a portion of, 1907 to 1917......................................................... 74-77
Lathes, description, dangers of operation, and methods of operation.................................................. I ll, 112
Locomotives:
Changes in construction making for safety........................................................................................ 106,107
High accident severity rate in construction o f.....................................................................................
9
Machine-building and steel manufacture, comparison of accident rates for............................................
22
Mft^iT ip-hnilding industry, accident rates in, influence of the war o n .................................................... 68-78
Machine-building plants, number covered by investigation...................................................................... 7,27 .
Machine design as a factor of safety........................ * ..................................................................................99-115
Machine shops:
% Accident frequency rates for six plants, 1907-1913....................................................... -...................... 47,54
Accident rates i n ...................................................................................................................................... 39,40
Accident severity rates for six plants, 1907-1913................................................................................... 47,54
Conditions i n ............................................................................................................................................ 91-94
Safeguarding i n .........................................................................................................................................94,95

INDEX.

123

Machine tool accessories, dangers in operation and methods of protection..............................................
113
Machine tools, description, hazards, and methods of protection........................................................... 108-113
Machinery for th e steel industry, improvements in design making for safety.......................................99-101
Man-hours, definition of..................................................................................................................................
17
Mechanical safeguards, need o f.......................................................................................................................
14
Metal planers, description, dangers of operation, and methods of protection....................................... 109,110
Milling machines, description, dangers of operation, and methods of protection................................ 112,113
Munition plants, accident rates in ........................................................... ..................................................... 71,72
Nature of mjury. {See Injury, nature of.)
N avy yards. Government. (<See U nited States arsenals and navy yards.)
N ight and day accident rates. (See Day and night accident rates.)
Occupations, accident rates for...................................................................................................................... 42-46
Open-hearth charging machine, improvements in design contributing to greater safety...................... 99,100
Ore-unloading and ore-handling apparatus, improvements resulting in safer operation.....................
99
Permanent injuries, frequency and severity of, in specified groups of plan ts........................................ 54,55
Permanent disabilities, tim e losses for injuries causing, how determ ined............................................... 20,21
Prime movers (boilers, intemal-combustion engines, etc.), improvements in design of, making for
safety.......................................................................................................................................................... 107,108
Purpose of accident studies.............................................................................................................................
15
Purpose of th is investigation.......................................................................................................................... 7,40
Safeguarding:
In electrical manufacture........................................................................................................................ 95,96
In foundries................................................................................................................................................ 87-91
In machine shops......................................................................................................................................94,95
In woodworking shops.............................................................................................................................
97
Safeguarding methods, direct, in machine building.................................................................................... 85-97
Safety committee............................................................................................................................................. 80,81
Safety inspectors— ....................................................................................................................................... 79,80
Safety measures, necessity of, in the machine-building industry.............................................................
14
Safety organization:
Cost of accidents compared w ith pay roll, and tim e lost compared w ith tim e worked, in five years
under a bonus system ...........................................................................................................................
83
Description of bonus plan of a steel company....................................................................................... 81-83
Elements of................................................................................................................................................ 79-84
Maintenance of interest necessary...........................................................................................................81-83
Outline of committee system ...................................................................................................................80,81
Requisites of a good..................................................................................................................................
41
Safety bulletin board as a means of maintaining interest...................................................................
81
Surgical care as part o f............................................................................................................................. 83,84
Safety systems, effect of, upon accident occurrence............................................................................... 10,40-42
Scope of report..................................................................................................................................................
7,8
Severity rates. (See Accident severity rates.)
Sheet mills, changes in design, resulting in greater safety.........................................................................
100
Shipbuilding, private, frequency and severity rates in, 1912.....................................................................
63
Shop conditions in foundries.......................................................................................................................... 85-87
Steel industry, improvements in design of machinery, making for safety............................................. 99-101
Steel manufacture and machine building, comparison of accident rates for............................................
22
Surgical care. (See Safety organization.)
Tabulatable accidents, diseases, and injuries, definition of.. * ......................................................... ..—
15
Time losses:
For death and permanent disabilities, schedule of.............................................................................. 20,21
For injuries, how determined.................................................................................................................. 19-23
For results of accidents in one p la n t......................................................................................................
21
Time lost compared w ith tim e worked, under bonus system of safety organization,............................
83
Transmission gearing, dangers in operation and methods of protection............................................... 113,114
Turbo-generator, use of, as a means of reduction of power-house hazard............................................. 104,105
U nited States arsenals and navy yards:
Accident experience o f........................................................................................................................ 62-68,72
Accident frequency rates of.....................................................................................................................
13
Accident severity rates o f........................................................................................................................
13
Incomplete reporting b y .......................................................................................................................... 65-68
Reason for excessive proportion of accidents reported as terminating in third w eek.....................
65
Reason for low accident rate in ...............................................................................................................
64
War, influence of the, on accident rates in machine building..................................................................... 68-78
Wire-drawing bench:
Description of old and new sty le......................................................................................................... 100,101
Modifications in construction, to provide greater safety......................................................................
100
Woodworking shops:
Accident rates in .......................................................................................................................................
40
Safeguarding i n .........................................................................................................................................
97
Worker, 300-day, definitions o f...................................................................................................................... 7,17
Workers, number of 300-day, in plants covered by investigation............................................................. 7,27

A D D IT IO N A L COPIES
OF THIS PUBLICATION MAY BE PROCURED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
WASHINGTON, D. C.
AT

15 CENTS P E R COPY
A

Full text of Accidents and Accident Prevention in Machine Building. Revision of Bulletin 216 : Bulletin of the United States Bureau of Labor Statistics, No. 256

FRASER