Securing Efficiency
IN
Railroad Work
Story of an attempt to apply Scientific Management
to some departments
A Lecture delivered at Harvard
November 16th 1910
B7 HARRINGTON EMERSON
30 CHURCH STREET
NEW YORK
ERRATA'.
Page 3, Line 1, N. not W.
Page 25, Line 29, $33.00 not $133.00
Page 31, Line 4, switch not freight
Page 34, Line 18, Kouns not Koons
PREFACE
Acting for Dean E. F. Gay, Mr. J. W. Gunn, lecturer in
charge of the Graduate School of Business Administration of
Harvard University, asked me to arrange to give a lecture on
November 16th, 1910, on the subject of Efficiency, especially in its
application to railroad work.
Prof. William Z. Ripley, in his preface to Trusts, Pools and
Corporations, calls attention to the difficulty of securing data
sufficiently concrete, definite and convenient to form a basis for
analysis, discussion and criticism.
This warning led me to use for illustration a concrete and
definite case with which I was familiar. Dean Gay in addition
called attention to the importance of putting in permanent form
the story of an early attempt on a large scale to apply all the
principles of efficiency to an essential part of the work of a great
and progressive corporation.
The story falls into several parts: first, a brief statement of
the underlying mathematical cause of modern corporate
inefficiency of operation, a cause afflicting industrial plants to
even greater degree; second a relation of the particular conditions
existing which had their influence in inducing Mr. Kendrick,
Vice-President of the A. T. & S. F. Ry. to undertake additional,
burdensome, disquieting and troubling work; third, to follow
the logical development of the work up from the little and
insignificant, undertaken for purposes of demonstration, to the
larger and essential, always using and applying the few essential
principles; finally to show the extraordinary although still far
from complete results that flowed from these initial plans.
The methods tried are not solely interesting to the student
because they succeeded, they would have been interesting if they
had temporarily failed from an economic point of view. The
chief interest lies in the fact that the ever existing practical
problems occurring in railroad operation were attacked in a new
way. The steam locomotive did not immediately replace the
horse, the steamboat has not even yet fully displaced the sailing
ship, but in both cases an entirely new way had been applied to
the solution of an old problem. It was similar in this case:
In the Santa Fè experiment it was insisted that if certain
principles were applied better results would follow than if
definite tasks were attacked. The older methods are analogous
to a charge of explosive behind a projectile in a big gun. If
the big gun is well aimed and fired something definite will
happen. The new methods are analogous to the laying out of
an irrigation ditch which when finished will carry by the silent
force of gravity a tremendously greater weight a much greater
3
Securing Efficiency in Railroad Work
distance, the projectile (water) in the end not smashing a twelve
inch steel target but feeding millions of food bearing plants.
In the case of the gun, the target is the object, in the case of
the ditch it is the stream tapped, and the fall of the land that
count. The ditch is not aimed at specific fields, but fertile fields
spring into existence everywhere below the ditch.
The important fact in the Santa Fè experiment is not that
the principles of efficiency were applied to belts, to shop
machinery and tools, to locomotive and car repairs on a railroad,
but that they were applied at all, and that in their application
so many heads of departments helpfully and actively joined so
that in the end whatever success was attained was Santa Fè
success by Santa Fè men with Santa Fè money on Santa Fè
materials and equipment.
The twelve principles which up to this time had been
scarcely more than academic generalizations, demonstrated their
value as producers of results, under the fostering care of
experienced executives and staff specialists. These principles were
of immediate practical help in a great labor struggle, they were of
help in coping with unexpected increase of work, in expediting
its completion, in cheapening its cost, in paying higher remunera¬
tion. A great parallel road to the north had only the year
before experienced defeat in a similar labor dispute, a great and
parallel railroad to the south, furnished the necessary comparisons
to prove whether the gains on the A. T. & S. F. Ry. were due to
weather, general business, etc. or to the use of different principles.
The sole lesson to be drawn from comparisons of the Santa
Fè records in different years, or from comparisons of the records
for identical years of the Santa Fè and of the Southern Pacific
is that the recently formulated principles, practises and methods
of scientific management have value. The teamster who covered
forty miles a day freighting on the plains showed 100 p.c. efficiency.
Carrying coal on the railroad is a different and by reason of
its difference a more efficient method of moving freight even if
the coal car on a certain coal run only averages seven miles a
day. When the best long-hand copyists in Germany were pitted
against American experts on the typewriter, the latter delivered
three copies of typed matter, each copy manifolded three times.
This proves the superiority of the typewriting machine, not the
superiority of operator. The principle of scientific management
may be inefficiently applied by inefficient men and the result may
be less good than the older methods efficiently applied by
efficient men. The fact therefore that some roads with the old
methods show lower costs is not an argument in favor of the
older methods; the fact that some roads applying newer methods
show higher costs is not an argument against newer methods.
4
Securing Efficiency in Railroad Work
It is certain that the best managed railroad under the old
methods will show great improvement if it applies with equal
skill the principles.
The figures used in the lecture are for the most part taken
from the published and public reports of the railroads and from
other reports made public in the railroad and engineering technical
periodicals.
The methods which were evolved to apply the principles,
are due to the cooperation of very many able and practical
railroad men. y
In giving scientific management 'definite form, in counseling
its application, I was unusually fortunate in being both guided and
inspired by an unusually able and experienced body of young men,
nearly all of whom have since risen to positions of great
responsibility.
Harrington1 Emerson.
30 Church St., New York, N. Y.
New York, November, 1910.
5
Securing Efficiency in Railroad Work
THEORY
" Free labor has the inspiration of hope ; pure slavery has
no hope. The power of hope upon human exertion and
happiness is wonderful. Even the slave master has a
conception of it. the slave whom you cannot drive
with the lash to break seventy-five pounds of hemp a
day will break you a hundred and fifty if you will pay
him for all he does over a hundred. you have substi¬
tuted hope for the rod." abraham lincoln—a frag¬
ment written in 1854.
Railroads have been an incalculable benefit to humanity.
Nevertheless, owing to the complexity of their operations they
are often less convenient in minor ways than more primitive
forms of conveyance. Owing to expansion and complexity de¬
pendent sequences of arithmetical losses are introduced which
culminate in a tremendous final geometrical slump. So serious
are the effects of dependent sequences that Mr. A. M. Fisher,
who has given this subject profound study, has deduced a law,
probably inexorable to the effect that "Revenues increase arith¬
metically but expenses increase geometrically."
If this law is inexorable, if there is no escape from it then
railroads must raise their rates and again and again raise their
rates until they perish from inability to compete. If yet there is
escape, if arithmetical gain in each operation of a dependent
sequence can be made to result in geometrical gain over present
conditions, then the remedy is not acquiescence in progressive
increase of expense, but safety lies in a resolute elimination of
inefficiencies, especially those inefficiencies which are due to
dependent sequences.
Railroad interests are not alone in danger. A great many
instances could be given of the isolated, poverty stricken worker
underbidding the great plant. In bicycle days fifteen years ago
a mechanic whose workshop was his bath room, whose equipment
was a few files, a brazing lamp and a pipe cutter offered to
assemble to order a bicycle for half the price charged by the
regular dcalet. There were only two, at most three, dependent
sequences of profit to pay to this mechanic, there were a dozen
dependent sequences when I bought of the retailer.
Living beings die from two main causes—one, the arithmetical
decay brought about by time; and the other, geometrical destruc¬
tion wrought by indefinitely multiplied poisonous microbes or
bacteria. Industrial organizations may decay because they grow
old. They are much more likely to perish because minute single
inefficiencies multiply into an irresistible geometrical end term.
7
Securing Efficiency in Railroad Work
This theory can be applied to all work.
Work is inefficient for two main reasons:
Separate operations are inefficient.
Separate operations are often connected in dependent
sequence so that the arithmetical inefficiency of each increases
geometrically in the combination.
Separate Operations Are Inefficient
Because conditions for favorable operation are deficient.
Because the operation itself is inefficiently performed.
Separate Operations are Often Connected in
Dependent Sequence.
The old time omnibus was a short vehicle which drove up
to the curb when we hailed it and in six seconds we could get
in. It was succeeded by the street car, twice as long, stopping
in the middle of the road at street intersections only. It took
about two minutes to cover the distance from residence to street
car seat. As progress was made we substituted the elevated train
whose stations were blocks apart up difficult flights of stairs.
New cars were fifty feet long and sometimes it took two minutes
to progress from platform to seat. The whole time of reaching
car from house door had advanced to ten minutes.
Railroad suburban trains were put on. These were started
and arrived at remote central stations; the tail end of long trains
was far within so that after reaching the station several min¬
utes has to be allowed to reach the train. After a while there were
more cars and more trains, sometimes three trains strung out
end to end beyond the gate, each train of many cars, until by
actual measurement the unfortunate passenger has to walk six
hundred yards from station front to car seat. It takes half an
hour to go from office to train seat compared to the six seconds
of more primitive times.
The increase in length and number of cars, of number of
trains has brought about a dependent sequence of delays to the
passenger, of costs to the operating railroad, which in some cases
feel obliged to charge extra for the use of the inconvenient,
palatial terminals.
Lifting a load is an operation, carrying it is another. If a
path is so bad that a good man can only take half the load and
walk only half as fast, his carrying power is reduced to one
quarter, through no fault of his own. 50 p.c. efficiency as to load
and 50 p.c. efficiency as to speed, results in 25 p.c. mile pound
efficiency. If the man is in addition slow and lazy, if under any
conditions he moves at half speed and takes only half reason-
8
Securing Efficiency in Railroad Work
able load he will carry only a quarter as much and walk only one
quarter as fast so that the end efficiency is only 6.25 p.c. or one-
sixteenth of an easier standardized task. Grant a man-efficiency
of 90 p.c. a condition-efficiency of 80 p.c., a combined condition
and operation efficiency of 72 p.c. and operations averaging two¬
fold in dependent sequence, the end result will be 51.84 p.c. When
therefore an investigation shows an end efficiency of 50 p.c. it
does not even follow that the workers are as low as 90 p.c. or
that conditions are as low as 80 p.c., since dependent sequences
may average three or four or even five and six. Assume men
at 100 p.c. and conditions at 90 p.c. in a six-fold sequence, the
end result is only 53.14 p.c. If all the operations in a shop
averaged 95 p.c. the shop as a whole may average anything from
95 p.c. down.
Therefore, one way to ascertain just what the loss, is to
investigate
men,
conditions,
sequences.
Losses are Eliminated.
(1) by establishing standards,
(2) by comparing the actualities attained with the established
standards.
(3) by making the actualities approach or surpass the
standards.
All the different ways, methods and devices for doing these
three things can be grouped under a series of principles of
efficiency which must be applied not only to the whole plant, but
to each operation.
These principles are:
(1) Definite plans and ideas
(2) Supernal common sense
(3) Competent guidance
(4) Discipline
(5) The fair deal
(6) Despatching
(7) Reliable, immediate and adequate records
(8) Determination of standards
(9) Standard practice instructions
(10) Standardized conditions
(11) Standardized operations
(12) Efficiency rewards.
9
Securing Efficiency in Railroad Work
If these principles are intelligently observed, high operating
efficiency is as inevitable as any other high engineering standard,
as a bridge, a tunnel, a road bed, a grade, a curve. If any
single one of these principles is not observed, the highest operat¬
ing efficiency is impossible.
Another way therefore of judging 'of the condition of a
plant is to ascertain the extent to which the twelve principles
of efficiency are in operation.
In civil engineering there are also often two different ways
of estimating such a condition as railroad grades. One is to
look at the profile of the survey and the other is to examine the
line location on a contour map. The profile shows what the
grade actually is, the contour shows what the grade must be.
Actual investigation of shop operations corresponds to a
carefully made profile of the line; investigation of the efficiency
principles in operation corresponds to the examination of loca¬
tion on a contour map. These two methods check and supple¬
ment each other.
The Result of the Application of the Efficiency Principles
is Arithmetically to Improve the Isolated Efficiency of
Each Separate Operation. Inevitably there will be
Geometrically Progressive Improvement in the Total¬
ized Efficiency.
A division of the problem is to locate those particular
operations whose arithmetically progressive advance will yield
the largest geometrical gain. This is necessarily the work of
staff specialists and it is not obvious since
(1) It requires the most minute, reliable, immediate and ade¬
quate knowledge of all the current efficiencies of separate
operations.
(2) It requires a knowledge of all the dependent sequences.
(3) It requires the ability to foretell which inefficiencies will
improve in the least time with the least effort and least expense.
(4) It requires the ability to foretell which inefficiencies will
produce the largest financial or output gain in the least time
with the least effort and the least expense.
After existing conditions have been thoroughly analyzed, after
a plan of efficient campaign has been evolved, the beginning has
only been made, no more a victory as yet, than a plan of war
is a victory over an enemy.
The second great division of the problem is to build up
the general organization, of staff and line. The staff plans, creates
standards, the line executes. Standards for enlistment, the
methods of drill, the details of equipment, the welfare of the
10
Securing Efficiency in Railroad Work
men as to food, clothing, endurance, transportation, maintenance,
arc evolved by staff specialists, but put into effect by line officials.
The third great division of the problem is so to use the
knowledge acquired, the plan of campaign and the competent
army, as to circumvent, vanquish the enemy—inefficiency. This
is line duty.
Officials directly in charge of work are too near to see the
problems in full proportion. A nearby hummock may obscure
a whole mountain range, the point of a lead pencil at arm's
length completely covered Johnstone and his aeroplane when
half way up in his 9,714 foot altitude flight at the international
meet at Belmont Park.
Men in the line often resent line orders; officials in the line
often resent staff counsel.
A man with a painful corn and a tight shoe resents it
angrily if any one steps on his foot. He overlooks the
greater fact that corns are inexcusable, that a combination
of good shoemaker, chiropodist and training will make
walking and even a football scrimmage pleasurable. As a
general proposition the instinct of motherhood, continuously and
intensely alert, is far safer than the routine and general atten¬
tion of doctors and trained nurses, but it is the doctor, not the
mother, who diagnoses diphtheria, discovers and administers anti¬
toxins, and if we are lying on an operating table it is the surgi¬
cally clean hands of the trained nurse we want around, not the
tears and contaminating caresses of sympathetic relatives.
When the three great divisions of the problem—analysis—
organization and action—have been competently handled, the
results are so astounding as to be unbelievable, even as it was
until very recently unbelievable that a single germ in mosquito
saliva could multiply so rapidly, divide and subdivide into a
thousand million in a short time and kill the strongest man. Yet
we have always seen all around us similar phenomena of geome¬
trical progression victory over arithmetical ability to resist.
In the Seward Peninsula in Alaska, the snow lies many feet
deep as late as May.
The arithmetically increasing temperature, the arithmetically
increasing length of the warm days make the snow that fell
arithmetically the long winter through disappear geometrically
and the million tree twigs bud and subdivide geometrically into
eight million leaves, each growing a hundred-fold until in ten
days there is a thousand million times as much leaf surface and
the whole country is a mass of green.
11
Securing Efficiency in Railroad Work
Similarly in railroad and other operations arithmetical advance
in efficiency will result in a geometrical progression of improve¬
ment even as a snowball grows geometrically in weight and bulk
with every revolution.
An average of 50 p.c. in all the separate operations of a two
step sequence advanced to an average of 60 p.c. means a gain
from 25 p.c. to 36 p.c. or a 44 p.c. gain. An average of 60 p.c.
in all the separate operations of a three step sequence advanced to
70 p.c. means a gain from 21.6 to 34.3 or an advance of 60 p.c.
An average of 70 p.c. in all the separate operations of a four
step sequence advanced to 80 p.c. registers a betterment from
24 p.c. to 41 p.c. a gain of 70 p.c. An average of 80 p.c. in all
the separate operations of a five step sequence advanced to 90 p.c.
shows an improvement from 32.8 p.c. to 59 p.c., a gain of 80 p.c.
Finally a sequence of six steps of 90 p.c. efficiencies advanced to
six steps of 100 p.c. efficiencies, registers a gain from 54 p.c. to
100 p.c. an advance of 85 p.c.
These are examples of comparatively slight gains in fairly
high efficiencies. There are however, innumerable examples of
tremendous gains in single operations. The cutting power of steel
on steel has been improved from 8 pounds an hour to 1,600
pounds, from an efficiency of 0.5 p.c. to 100 p.c. Dependent
sequences are often fifty in a series.
The instances cited have with intention not been limited to
the crude and elementary material and physical operations. If
two magnets are stroked against each other the wrong way they
deprive each other of all power after they are separated. If
they stroke each other the right way they make each others
power thany times greater than it was at the start. They have
weakened or strengthened by contact each other. Man influences
man far more than steel magnetizes steel. Napoleon's
enemies believed that his mere presence at a battle was the
equivalent of 20,000 men. In no other way could they explain his
incredible victories. When Italian patriots flocked around Gari¬
baldi in 1848 he told them that if they followed him he would
give them hardship, starvation, rags, wounds, imprisonment and
death—and with a horde of thus collected rough riders and
walkers he conquered kingdoms and helped make Italy free and
united.
Peter the Hermit was fired by the abuses he had witnessed
to preach a crusade and he swept two million men away from
home and country in pursuit of a fantastic ideal, the physical
possession of the land where Christ had lived, taught and died.
The lowly founders of all the great religions have worked through
the souls of men. Behind every unstandardized condition, behind
12
Securing Efficiency in Railroad Work
every unstandardized operation, through and through the links
of every dependent sequence there is, as hidden and as real as
the magnetism in steel, liking or dislike, faith or distrust and this
is why the highest ideals, "give that ye may receive," why the
highest common sense, why openess to counsel, why discipline in
its highest meaning, above all why the fair deal must be main¬
tained. Much, very much, can be effected in physical ways,
because the machine, whether lathe or locomotive, has multiplied
and magnified the power of men; but for that very reason the
faults and virtues of men are also multiplied and magnified and
no machine can do good work when it is the end term of a
sequence whose first two factors are a hasty foreman and an
angered worker.
Considering the Tremendous Effect of Dependent Sequences
the Marvel is not that Average Efficiency is so Low, but
that it is so High.
Problems.
The equipment maintenance conditions were serious in May
1904, on the A. T. & S. F. Ry.
(1) An enormous increase in business was beginning.
Passenger Freight ton Number
miles in miles in of
millions millions locomotives
1903 613 4,706
1904 708 4,682 1433
1905 744). 4,730
1906 865 7M" 5.841
1907 970] " 6,843
1454 1
increase ( increase
46 p.c. 1676 V
17911
(2) Because the annual increase in business could not be fore¬
seen, it was impossible to order new locomotives in time to
take care of it. It was also impossible to build and equip new
shops in time to take care of the increased deterioration of
overloaded old equipment and added new equipment.
So great was the increase between 1904-5 and 1906-7 that
President Ripley states, page 20, report 1906-7: "At times the
increase in volume of traffic was so large that the company's
equipment was overloaded. The necessity of moving traffic with¬
out regard to economy of operation caused a material increase
in operating expenses."
The International Association of Machinists which in 1903 had
won a successful strike on the Union Pacific, thought the time
propitious to force concessions and made proposals to the
A. T. S. F. Ry. President Ripley's words, Report 1904, page 19,
13
Securing Efficiency in Railroad Work
are: "Certain unreasonable demands made by the International
Association of Machinists having been declined, that organization
declared a strike at all shops of the company, effective May ?nd,
1904, which order was quite generally obeyed on the Western
portion of the system, as well as by a large number of employes
on the Eastern divisions."
The shops therefore in May 1904 found themselves quite
generally without employes and with run down and crippled
equipment, the bad conditions being largely due to the hostility
of several years' standing of the employes towards the employer,
a hostility always conducive to neglect and worse.
Enormously Increasing Work, Fewer Men and Facilities to
Cope with it—Summarized the Situation.
Also for a series of years cost of maintenance in general
and of locomotives in particular had been rising alarmingly on
the railroads of the United States. This increase had been
particularly noticeable on the transcontinental lines immediately
north and south of the Santa Fe. As early as 1901 President
Ripley in his annual report, page 10, stated: "It will be observed
that the expenses of maintenance of equipment especially as to
locomotives are excessive, exceeding even the very large cost
of last year."
The increase is best evidenced by the figures from the annual
reports—
Cost per
♦Cost of
Number
New
Cost of re¬
loco, mile
repairs per
of
locos.
pairs per ton
for repairs
loco.
locos.
secured
of loco. wght.
1897
$0,045
1899
O.0558
$2,032
1,083
1900
0.0592
2,096
1,136
1901
O.082
2,858
1,174
52
1902
0.0922
3,156
1,312
169
1903
O.0867
3,042
1,309
130
$38.45
1904
0.1134
3,772
1,433
151
40.23
1905
O.i256
4,165
1,454
36
46.23
533
•These figures include supervision, major part of tool and machinery maintenance
expenses and other charges.
The progressive increase during nine years was primarily due
to growing numbers of dependent sequences aggravated by
diminishing efficiency of separate operations, so that the expected
reduction in repair costs due to the large number of new
locomotives secured did not materialize.
14
Securing Efficiency in Railroad Work
PROBLEMS TO BE SOLVED
The problems to be solved fell into three classes—physical—
financial—moral.
(1) Physical—to repair equipment promptly and adequately in
run-down shops, generally old with most of the mechanical
employes not only striking, but very hostile.
(2) Financial—to check the geometrical increase in expenses.
(3) Moral—to eliminate radically and permanently the strain
and hostility between employe and employer.
PLANS AND ORGANIZATION
Early in 1904 Mr. Kendrick retained Mr. Emerson, Coun¬
seling Engineer as to Standard Practice and Efficiency.
The plans agreed upon by Mr. Kendrick and Mr. Emerson
were:
(1) To create a staff organization capable of investigating
and analyzing all the facts.
(2) To induce newly created staff and old line to cooperate
in applying thoroughly to all the work, from item of operation up,
the twelve principles of efficiency.
A prototype for this work was found in modern railroad
accounting. Railroads in the early days had bookkeepers, but
accounting was not yet applied because it had not yet been
evolved as to railroads. As a science it was not so much a
growth as an importation. English chartered accountants set up
principles, standards and methods which have passed very gener¬
ally into railroad practice. Accurate accounting shows where
and how and for what every dollar is spent, but it cannot show
that an equivalent is obtained for the dollar spent. As to some
bills the purchasing agent certifies that the unit costs are correct,
as to some salaries there is other executive action, but in return
for the great aggregate of annual railroad bills for personal
service, for materials, for equipment, there are as yet no scientific
equivalents. It is the field of modern scientific management, of
standard practice, to determine standard equivalents for dollars
spent and to provide methods and practices whereby the equiva¬
lents can and will be realized.
Mr. Kendrick fully realized that progress would be slow and
that it would be at first largely educational, that it would require
unusual' courage to approve the radical, untried undemonstrated
innovations, that it would require very great patience to persuade
employes and minor officials that the new methods were really
better tools, not reflections on their past ability and fidelity.
IS
Securing Efficiency in Railroad Work
Had it not been for his active and directing support no
progress could have been made, and he advised and counseled,
at every step, the whole of the great railroad organization in
all its departments, contributed time, intelligence and expense
to insure success. All necessary accounting records were main¬
tained by the general auditor's department, valuable assistance
being given by the many accounting staff specialists. Without
reliable, immediate and adequate records the comparative value
of the new and old methods would always have remained in
dispute.
The general purchasing agent, now a Vice-President, did not
hesitate to approve requisitions calling for supplies of better
quality, but higher in price, thus not only aiding the principles of
standardized condition and standardized operation but making it
possible to prove that a better quality will result in so much
less quantity as to lessen the aggregate cost. The general
storekeeper continuously and everywhere threw his own vast and
varied railroad experience, as well as his unusually strong and
well organized department, into the new movement. His advice
and cooperation proved invaluable. The general statistician
furnished innumerable necessary comparative tabulations and
performance records. Much that was new, difficult, annoying,
even revolutionary, was cheerfully tried by all the departments.
Towards the end of the second year enthusiastic adherents
were won over in the mechanical department proper, but
for the first year they were almost without exception skeptical,
believing that the existing difficulties would only be aggravated
by the application of untried methods, that it would have been
more prudent to wait until the patient recovered before calling
in a doctor and trying any new kind of treatment, believing
with intense conviction that they, of life long experience, were
better able to cope with the difficulties than the imported
theorists.
It was to the work of the mechanical department that the
principles were particularly applied, and this could not always
be agreeable, but in the end success was due to the direct work
of the mechanical superintendents 'and the many master
mechanics, general foremen and others.
It was particularly at the great Topeka shops that the
Superintendent applied in the most thorough and scientific
manner the twelve principles, giving them in fact new breadth
and broader application from the wealth of his experience.
The special problem and plans were considered at length and
in detail by Mr. Kendrick. The work was to continue for
16
Securing Efficiency in Railroad Work
several years if progress could be shown, necessary specialists
and practical assistants were to be collected and instructed as
to the new plans and were to be empowered to carry the plans
into effect. A beginning was to be made, not with the employes
but with the conditions, with the machines, and after these were
standardized, after separate operations had been standardized, the
men were to be given opportunity to earn efficiency reward.
Further results insisted on were that there must be progressive
improvement, that locomotive failures must decrease, not increase;
that shop detentions must decrease, not increase; that locomotives
must do more work, not less.
It was not for a minute proposed to allow any academic
theories and principles to give any set back to current work. The
principles were to have a clear field only so long as they could
show practical results superior to the old methods.
Work was at first confined to the main shops at Topeka,
Kansas, employing about 1,000 men, not including the car repair
department.
Mr. Emerson had neither official position nor executive
authority. He investigated, planned and then advised. He taught
the underlying principles, showed the methods of research.
Executive orders were issued and the work required was under¬
taken by the regular shop officials and force. Outside of his own
office assistants., the specialists he selected and recommended
were appointed by the regular officials and owed obedience to
them. If important orders were necessary they were issued by
the Superintendent of Motive Power or by Mr. Kendrick, Vice-
President.
Because the plans were new, untried and because of the great
interests at stake involving safety to life, full control was retained
throughout by the regular officials^
The chief of staff counseled the line, but he also could direct
the staff only through counsel. In this exceedingly difficult situa¬
tion to all concerned the consideration given the many recom¬
mendations was extraordinary, and without it no progress could
have been made.
No requests were made for new equipment or for other
employes or for a change in officials. During his three years
connection with the work he never asked for the dismissal of any
employe or for the purchase of any new machine.
For Mr Kendrick' two ideals were immediate—to meet the
present requirements and obviate recurrence of trouble.
He realized that the task was long and hard; he authorized
the employment of competent staff specialists; the strike had
17
Securing Efficiency in Railroad Work
regiven him control of the mechanical employes,; he intended to
extend the fair deal. Five principles were thus in operation.
It was requested that the records be made reliable,
immediate and adequate; that schedules be made for all work;
that all work be despatched; that definite standard practice
instructions be reduced to writing; that conditions be standard¬
ized; that operations be standardized and that efficiency rewards
be given. These were the other seven principles.
The essential staflf organization was soon in process of for¬
mation and development. At that time competent efficiency
specialists as to railroad operation did not exist, no scientific
time studies of railroad shop operation had ever been made.
Brilliant and experienced men there were, but they had to be
collected, schooled, trained and in shortest time be sent forth to
direct and train others.
In Topeka the machine shop was large, new and well equipped,
but because the more serious strike difficulties had absorbed the
attention of shop officials what were considered minor matters had
deteriorated.
Shafting was not always true, machines and shafting were not
always in line, belting had been purchased from lowest bidder,
and the machinist and planing mill hands were paid overtime
to repair the belt failures; abrasive wheels were generally oval-
shaped, slow running grindstones; the small tools and machine
tools were not up to modern requirements; high speed steel was
just beginning to make its way into railroad shops; machine tool
builders had just begun to redesign their tools. Everywhere
pulleys and gears were too light, small tools were generally of
carbon steel and of fearful and wonderful variety of shape.
Mr. Kendrick appointed on Mr. Emerson's nomination a
number of specialists of experience and reputation to take practi¬
cal charge of the application of the principles. They were
gathered from various great railroads including the Santa Fè
and from other fields of experience. The railroads whose
advanced mechanical and operating practices were represented
were Union Pacific; Burlington; Northern Pacific; Chicago and
Great Western; New York Central; Delaware; Lackawana and
Western; New York, New Haven and Hartford; Boston and
Maine, as well as Santa Fè, but there were specialists experienced
in other lines than railroading. In selecting a staff the need of
cross fertilization was recognized. These selected men constituted
the staff, the specialists, each expected to know more about his
own line than any one else, specialists whose advice and
experience were available as to all the phases of every operation,
yet they had to be trained as to the wider problems of waste
elimination. Individual enthusiasm, skill and effort had to be
18
Securing Efficiency in Railroad Work
fanned to white heat, but on a foundation of balanced principles.
It was not easy to maintain peace and good will in such a family.
The enthusiast in charge of the application of the fundamental
principle of reliable, immediate and adequate records could not
see that good despatching with deficient records was a more
immediate need than perfect records with poor despatching. The
enthusiast anxious to make a demonstration of the possibilities
of high speed steel was blind with anger against the belt specialist
who objected to overloading belts for any cause.
It was only the acute criticisms and active opposition of the
line that welded the separate staff specialists into a united
organization. It was a man's game. Nothing was allowed to
go by default. Everybody from Vice-President down was from
Missouri.
The work undertaken was so extensive, 1,500 locomotives
with 50,000,000 mileage, 50,000 cars, twenty shops large and small
scattered along 9000 miles of railroad in twelve different states,
12,000 mechanical employes, appropriations were so scant for
what was as yet only an experiment, that it was necessary to
impose on each staff man double duty, functional and territorial.
Each man in his own alloted territory directed the plans as to
all the twelve principles of efficiency, being advised by Mr.
Emerson, each man outside his territory planned as to his own
specialty for all the others, advising Mr. Emerson.
The crying need was to turn out work in shorter time, there¬
fore more work in the same time. Economy was not primarily the
object although it was desired as a by-product. It is however
difficult to separate time reduction from cost reduction because
time reductions are summarized into cost reductions and thus
appear in the official records, output increase and time reduction
finding no place.
In the following statements cost reduction is often quoted to
exemplify time reduction.
PHYSICAL BEGINNING OF THE WORK
Belting is an insignificant item in railroad operation. In the
A. T. & S. F. shops it was nobody's care. Foremen, superintend¬
ents and higher officials had other duties, machinists and planer
hands were paid overtime nominally to repair but actually to
destroy the belts on their machines. The only official who showed
any interest in belts was the claim agent who wanted the shops
to take and pay for a lot of vile belting that had been first singed,
then water-soaked in a wreck and therefore was refused by the
19
Securing Efficiency in Railroad Work
consignee. When the application of the twelve principles, the
elimination of dependent sequences, was started as to machine
shop belts, nobody objected except the drawers of overtime.
The purchasing department cordially cooperated, approving a non¬
competitive supply of the highest grade of belting manufactured,
the manufacturer cooperated, giving most valuable instruction as
to installation, operation and maintenance of good belting. The
belting work was scorned even by some of the selected staff
assistants. But belting is more important than it seems, it is a
link in a chain.
Every belt failure entails a dependent sequence of loss as
the broken belt puts machine and man out of commission, delays
the work in progress, holds locomotives in the shops, prevents
the moving of trains, lessens revenue, in an endless "This is the
house that Jack built" series.
For the year ending June 30th, 1904, the cost of new belting
had averaged $1,000 a month, and belt failures had averaged 300
a month. It is well known that if best belting is used, if its
care is entrusted to a competent man, annual maintenance cost,
labor, materials and supplies ought not to exceed 14 p.c. of first
cost of belts.
In May, 1905, belt maintenance expenses had been reduced to
$163 a month, the standard had been realized, belt failures had
been reduced to 43 as early as November 1904.
The material cost of an operation depends partly on quantity,
partly on price per unit. The equation is—
in which Q stands for quantity and P for price per unit. In the
Topeka shops P as to belting has been kept low with consequent
enormous Q. The real problem is to have Q P a minimum. This
was attained by greatly increasing P and everlastingly watching
Q. Reduction in cost of belts for repairs:
In 1903-4 the quality of belting was poor and therefore broke
often, the installations were poor and would have been destructive
even to good belts, a premium was offered—overtime—for belt
break-downs, the men who repaired the belts had no skill. The
efficiency of belt costs was about 5.25 p.c., the end term of a
five step dependent sequencer
Cost = Q P
1903-4-
1904-5-
P $ 0.40
Q ft. 30.000.
P Q $ 12,000.
$ 0.60
ft. 1.059.
$ 630.
20
Securing Efficiency in Railroad Work
Cheapness Efficiency, 150 p.c.
Quality Efficiency, 30 p.c.
Installation Efficiency, 80 p.c.
Skill of Repairer, 30 p.c.
Care during Operation, 50 p.c.
Final Efficiency, 5.4 p.c.
It was the general program to standardize quality,
operation and maintenance of belting, from this to proceed to
the standardization of quality operation and maintenance of all
shop tools and machinery, to extend the work to the standardiza¬
tion of quality, operation and maintenance of all equipment,
locomotives and cars. A demonstration as to belting was to
become the razor edge of the entering wedge.
As to the fundamentals of mammalian anatomy as much
can be learned from the mouse as from the elephant, more from
the mouse than from the whale, but the mouse is far easier to
catch and to dissect. Belting was the mouse, locomotives the
elephant, all operation the whale in this railroad dissection.
The system that had been tried out as to belts was extended
to maintenance of all shop machinery and tools.
In order to bring out results more plainly the figures for
several years in comparison with those of the Southern Pacific
are added, expense items taken from respective annual reports.
Maintenance of Shop Machinery and Tools*
A. T. & S. F. Ry
(applying new methods after january 1st, i905.)
Expense. Unit Cost.
$487,171 $10.31 *
486,620 . 10.16*
376,106 6.51
315,844 4-89
*Had risen to $11.10 before the principles were applied.
Southern Pacific.
(continuing the old methods)
Year.
Units.
1903-4
47,250
1904-5
47,854
1905-6
57,76o
1906-7
64,628
Year.
Units.
Expense.
Unit Cost.
1903-4
51,003
$487,150
$ 9 55
1904-5
52,037
567,161
10.90
1905-6
57,034
537,3i8
9.42
1906-7
65,076
638,193
9.81
Comparison of old practice, standard set and new practi
17P rl •
realized :
*See preface as to comparisons.
21
Securing Efficiency in Railroad Work
Material. Labor. Total.
Old Practice:
Actual average monthly cost for
year ending Dec. 31, 1904 $25,200 $18,600 $43,800
Efficiency of actual cost compared
to standard 52.6 p.c. 60.8 p.c. 55.8 p.c.
Standard Set:
Monthly allotment made Jan.
1, 1905 $13,250 $11,250 $24,500
New Practice:
Actual average monthly cost for
year ending March 31, 1907. .$14,116 $11,009 $25,125
Efficiency of actual cost compared
to Standard 93.5 P-c. 102.3 p.c. 97.6 p.c.
The allotment made January 1905 had been averaged for a
whole year by April 1st, 1907, in spite of the fact that volume of
machine shop work had increased 35 p.c.
Also in other respects comparisons were instructive. From
former statements it is to be presumed that the Southern Pacific
was as well equipped with shop machinery and tools as the A. T.
& S. F. Nevertheless the A. T. & S. F. maintained its existing
tools during the three years of the test for 78 p.c. of the cost
on the Southern Pacific, the A. T. & S. F. bought 52 p.c. less
new tools and maintained its more complex locomotives for 27 p.c.
less per ton of locomotive weight.
comparisons, 1904-5 to 1906-7.
Southern
Pacific. A. T. S. F.
New shop machinery charged to capital..$ 765,169 $ 363,968
Maintenance shop machinery and tools... .$1,742,671 $1,169,938
Total $2,507,840 $1,533,906
Difference $ 973,934
Maintenance of Locomotive Per Ton of Locomotive
Weight*
s. p. a. t. s. f.
1904-5 $52.66 $46.23
1905-6 $52.28 $34-10
1906-7 $49.86 $33-13
It is not claimed either as to the item belting or as to the
whole account Maintenance of Shop Machinery and Tools that the
figures attained are remarkable or the ultimate. In fact the
expenses for both belting and tool maintenance were markedly
bettered in subsequent years, the tool unit cost dropping for
♦Sec preface as to comparisons.
22
Securing Efficiency in Railroad Work
1909-10 to $3.24, a total of $236,696. With an increase of work
of 60 p.c. maintenance costs dropped 51.4 p.c. If the last figures
are of xoo efficiency, then the condition on a unit basis in 1903-4
was 31.4 p.c. The Southern Pacific has also not been idle, reducing
its total in 1908-9 to $298,856. This reduction was undoubtedly
stimulated by the constant unfavorable comparisons with the
A. T. & S. F. and merely illustrated the psychological effect of
emulation. A man who in the absence of comparisons considers
an outrage the suggestion of a unit cost of $9.00 will succeed
in reaching $3.00 if thereby he beats a rival's record and proves
himself the better man.
If $3.25 had been a standard unit cost on these two railroads
between 1903-4 and 1908-9, if it had been attained the Southern
Pacific would have been $1,760,000 better off, the A. T. & S. F.
$1,120,000 better off. These savings could have been expected if
the matter had been given competent attention and the cost of
competent attention and organization would not have exceeded
10 p.c. of the possible economies. What is remarkable about the
earlier efforts on the A. T. & S. F. is not the low unit cost first
attained, but that the innovation of staff supervision was given a
full trial and also that against a steady upward drift in similar
expenses on the only comparative road, this account receded to
a predetermined allotment.
It was not care of belting, not supervision of shop machinery
and tools, not planning and despatching the repairs of locomo¬
tives and cars that produced the results. Improvements in other
directions were also going on, revision of grades, water purifica¬
tion, new designs of locomotives and cars, a pension system,
welfare work, but these also on this road and others would have
been powerless either to attain, maintain or surpass the earlier
results unless the whole plan was in balance, unless sufficient
individual operations, whatever they might be, were so improved
as to overcome the losses due to increasing dependent sequences
whether physical, mental, moral, or all three at once.
BELIEF IN THE MEN
In carrying out the principles of Standard Practice Instruction
a pamphlet was printed in an edition of 10,000 and distributed to
all the mechanical employes.*
The motto was:
Fairness not Favoritism
Efficiency not Drudgery
Individuality not Subserviency.
•Shop Betterment and the Individual Effort Method of Profit Sharing, 1905.
23
Securing Efficiency in Railroad Work
The opening sentences were:
"The employe wants as high wages as he can get.
The employer wants his output to be as cheap as that of his
competitors. Both desires are reasonable and the problem» Is to
reconcile them without injustice to either party.
An absolutely clear understanding of the problem by both
parties is necessary.
The worker cannot be expected to work for one employer for
less pay than is paid under similar conditions for the same work
by another employer. The wage payer will not pay higher wages
than the current rate or than the business conditions permit.
There may be, however, quite a gap between the wages paid by
competitors and the higher wages the employer would be willing to
pay if it can be proved to him that it is to his advantage to do this.
Wages above current rate should result from individual effort."
In the Topeka shops while the tools were being put in order
machine and man operations were gradually standardized and the
efficiency on all of them arithmetically advanced.
At the end of a year scheduling as to cost and despatching
as to time could be applied to locomotive repairs. On March
20th, 1905, when this work was started an enumeration was
made of all locomotives and the sum of their detentions in the
Topeka shop, a total of 1,735 days for 56 locomotives. A year
later, a larger number of locomotives being repaired monthly,
the sum of the detentions had sunk to 254 days for 33 locomotives.
Here again was a great improvement in a very important depen¬
dent sequence. Fewer locomotives in the shop at once and each
detained a much shorter time, the average detention having
dropped to less than half the number of days. Similar improve¬
ment was being effected elsewhere over the system.
Mr. Kendrick had sent a letter to Mr Emerson calling
attention to the high average locomotive repair cost in 1904-5
which amounted to $4,165 (President Ripley's Annual Report
1904-5 p. 10) and expressed the hope that the efforts being made
by line and staff would succeed in reducing this cost to $3,165. In
1905-6 the cost was reduced to $3,101 for slightly larger locomo¬
tives (Annual Report, p. u). In the next year the repair cost
dropped to $3,037 (Annual Report, p. 10).
Gross earnings divided by locomotive owned had dropped
from $47,640 for 1902-3 to $47,028 in 1904-5 but in 1906-7 the figure
jumped to $52,308 because the locomotives were more days in
service and because they hauled more tons, about 2,000,000 more
each in the year.
24
Securing Efficiency in Railroad Work
A. T.
& S. F.
Rv.
Maintenance of
Per
Maintenance of
way and structures
loco¬
locos, per ton
per mile of road
motive
of loco, weight
1901
$2,730
$33.80
1901-2
$ 781
2,830
40.45
1902-3
1,168
2,644
38.45
1903-4
1,121
3,189
40.34
1904-5
1,371
3-519
46.23
1905-6
1,479
2,695
34-10
1906-7
1,648
2,623
33-13
1907-8
$1,506
$3,083
$37-14
1908-9
1,315
2,541
30.62
1909-10
1,796
3,216
38.44
Dependent sequence improvements beginning with shop opera¬
tions, extending up through locomotive detentions greatly aided
by the greater power of the additional locomotives purchased
finally culminated in an increased earning power for locomotive.
Comparative Table.
Southern
Pacific
*
*
♦
*
$52.66
52.28
49.86
*
$44.44
*
♦Figures not obtained by author. Their omission has no planned signifiance.
Between 1901 and 1907 prices of labor and material steadily
increased.
This last year was the one of which President Ripley states
(Annual Report 1906-7 page 20), "At times the increase in volume
of traffic was so large that the Company's equipment was over¬
loaded, that traffic was moved speedily without regard to economy
of operations.
It is not asserted that for repairs $2,623 per locomotive or
$13313 per ton is an economical or ultimate figure, as lower
figures were subsequently attained, but Mr. Kendrick had speci¬
fied as satisfactory for the moment a reduction of 25 p.c. An actual
reduction per ton cost of over 26 p.c. was attained and there
was other work requiring attention, detentions, coal wastes,
etc. One of the special assistants made a unique and tre¬
mendous study covering five years individual operation of every
locomotive owned by the A. T. & S. F., nearly 1,800 different
locomotives. The unusually complete records started by the
general statistician at Mr. Kendrick's request very greatly aided
this work. This study (Report to Mr. J. Kendrick, Vice-President,
by H. Emerson, Sept. 23, 1907), conclusively showed that A. T. &
S. F. locomotives, under better standard conditions, could be main¬
tained for an average of $1,900 per locomotive or per ton $22.50
or, for a better unit, per road unit* $50. In 1906, on many whole
divisions this figure was averaged, as to all locomotives, on other
divisions it was reached as to some locomotives. It was also
•The road unit is locomotive miles multiplied by weight on drivers in pounds
divided by 100,000,000.
25
Securing Efficiency in Railroad Work
shown that record low maintenances costs went hand in hand
with highest ton miles per locomotive.
On almost any railroad locomotives can be properly main¬
tained as a whole for $50 for road unit, a figure realized on the
Northern Pacific (by Mr. D. Van Alstyne); on the A. T. & S. F.
it had been approached in 1906 on the following divisions.
Freight Service.
Southern Kansas Division $47-50
Albuquerque Division $56.20
Passenger Service.
Albuquerque Division $36.10
Oklahoma Division $37.20
Illinois Division $45-io
Missouri Division $54-30
Southern Kansas Division $55-40
Albuquerque, Illinois and Missouri are main through-line
divisions, Southern Kansas and Oklahoma, branch-line divisions.
It was shown that variations in cost were not due primarily
to unavoidable conditions of fuel, water, grade and service, but
to difference in design and above all to different personalities.
Costs Per Road Unit.
New Mexico Division Freight Service.
Some Classes More Expensive Than Others.
Repairs. Total Operation.
Class 789, 180,440 lbs. on drivers $109.10 $325.00
Class 900, 234,580 lbs. on drivers $ 74.70 $242.00
Southern Division, freight service, oil burners. Larger
weight on drivers, lower road unit costs.
Class 151 96,500 on drivers $103 $331
Class 769 161,650 on drivers $ 54.20 $221
Personality.
Class 825, coal-burners, freight service.
Albuquerque Division $55-6o $210
Colorado Division $110.35 $287
Cost of locomotive supplies per road unit on several divisions
1906.
Illinois $1.00
Albuquerque $1.40
Western Division $3.80
26
Securing Efficiency in Railroad Work
Similar locomotives were running on Illinois and on Western,
both are dead level main-line divisions. Locomotive supplies
are not affected by anything except the discipline and efficiency of
the division.
Differences in operation are very astonishing. Mr. F. W.
Taylor was led to the discovery of high speed steel operation
by first noticing and then ascertaining why one tool cut better
than an apparently similar tool.
As a result of this discovery the machine shop practice of
the world has been revolutionized.
If the differences in locomotive operation were followed up
with the same analytical care that Mr. Taylor gave to shop
tools, similar revolutionary change in practice would probably
ensue.
Certain passenger locomotives on the A. T. and S. F. have
given 160,000 to 231,000 miles between the periods of general
repair. Locomotive 1420* of the Albuquerque division up to
June 10th, 1910, had been in service four years without shopping,
had made 232,894 miles, was not due for shopping for two months,
had an allotment for repairs of $0,059 a mile and an unexpended
credit of $1500. Others of the same type show a performance
of only 34,000 miles. Can anyone doubt that this average unsatis¬
factory performance is due to causes ultimately if not immediately
preventable?
On any railroad the cost of locomotive repairs can be stand¬
ardized and the predetermined standards can be realized, not by
main strength, but solely through a combination of local
experience and a plan applying all the efficiency principles.
A special study of the losses in railroad operation in the
United States for the year 1908 (losses due to failure to realize
attainable standards in isolated operations and to inefficient
dependent sequences) indicates that the total exceeds in amount
$360,000,000.
By the shop operation and maintenance work, by the locomo¬
tive maintenance work some progress, just a little had been
made. In the old shops with the old equipment a very great
increase of work, 44 p.c. in two years had been taken care of
at a 20 p.c. to 25 p.c. reduction in repair cost. For the moment
this was satisfactory. Work of this kind, however, rarely pro¬
gresses without relapses. The task is similar to that of rowing
and poling a boat up a rapid stream to a placid lake above. The
♦Railroad Age Gazette, July, 1910
27
Securing Efficiency in Railroad Work
strain is prodigious, the progress small, but if effort is relaxed,
if an eddy is mistaken for smooth water ahead, the boat may lose
in a minute what it took hours to accomplish.
The full value of principles as distinguished from apparent
practical results is not always realized, a temporary improvement
is mistaken for a permanent cure, the ultimate is lost sight of in
satisfaction over the immediate.
There was a back set on the Sante Fe to the great joy of
skeptical scoffers. Average efficiency which had increased from
55 p.c. in 1905 to 91 p.c. in 1907, sank back to 87 p.c. in 1908,
but triumphantly ran to 92 p.c. in 1909.
If $50.00 is an ultimately attainable standard for locomotive
repair costs per unit, then between 1903-4 and 1909-10 we have
the following table:
Effi- Actual Effi¬
ciency R. U. Cost ciency
100 p.c. $ 96.72 51.7
106.72 46.7
" 76.20 65.6
7546 66.3
91.47 54-6
76.35 65.4
84.30 59.3
Road
Standard
Date
Units
R. U. Cost
1903-4
47,250
$50
1904-5
47,854
(i
1905-6
57,760
U
1906-7
64,628
H
1907-8
64,326
11
1908-9
62,325
li
1909-10
73,034
417,177
$ 87.42 57.2
It makes an important difference in expense whether stand¬
ards are realized or not. In the seven years from 1903-1910 the
total number of roads units was 417.177. At 100 p.c. efficiency
the standard cost would have been $50 per road unit or a total
of $20,858,850. As compared to the actual costs various hypo¬
thetical comparisons can be made.
Locomotive Repair Costs.
Actual Records for 1903-10, Seven Years.
Remarks R. U. Costs Efficiency Total Repair Cost
7 years' average. $ 87.42 57.2 $36,469,706
Hypothetical Records.
On basis of 1903-4 $ 96.72 517 $40,349,359
" " 1904-5 $106.72 46.7 $44,522,129
" 1908-9 $ 76.35 65.4 $31,851,464
" " standard $ 50.00 100. $20,858,850
Between 1905 and 1910 labor efficiency as to 74.5P.c. of the
active payroll had increased from less than 60 p.c. to 93 p.c. The
28
Securing Efficiency in Railroad W o.rt'
fact that locomotive repair efficiency is still down to about 60 p.c.
is solely due to the great and increasing number of dependent
sequences occurring in separate operations each unusually high
in efficiency. It is manifest that even 93 p.c. will not bring down
the road unit cost of a locomotive of unfortunate design, or one
whose mileage is curtailed from any cause.
Let it not for a moment be supposed that the A. T. & S. F.
was particularly bad. Its main transcontinental competitors
south and north were worse, some eastern roads were worse.
The A. T. & S. F. in 1903-4 reflected general conditions which
still almost universally exist. That locomotive repair costs have
not reached the standard of $50 per R. U. is due solely to the
continued existence of dependent sequences whose elements are
of low efficiency.
No on can with authority state that locomotives ought to be
maintained for $50.00 per road unit. It is however possible to
point out inefficient single operations and inefficiency dependent
sequences and further to show that if these are eliminated road
unit costs will come down to $50.00. This method differs
radically in principle from the two usual methods, one hopeless
leniency, the other senseless comparisons. Under the first high
tire repair costs are excused because the road has many curves.
Curves are only one of the elements in tire wear. Soft tires,
bad design, slipping of driving wheels, poor handling by engineer,
excessive removal of tread in shop, each one of these causes
may contribute to such a degree that a careful road with bad
curves might make a better tire record than a careless road
without curves.
The other usual method is to claim that because one locomo¬
tive attained a record for repairs of $50 per road unit therefore
each locomotive ought to attain it. High repair costs may
result from sequences for which neither operating nor mainte¬
nance departments are responsible. Low repair costs may not
be due to virtue but to fortunate circumstances. If, however,
we can set up all the operations and their efficiencies, if we are
in position to analyze each string of dependent sequences, we
first find out where losses occur and we are next able to eliminate
them, not by bigger and more expensive tools, not by driving
men beyond reason, but by taking a little thought, in fact by
applying generally and specifically the twelve principles of
efficiency. If these principles are applied one locomotive may
legitimately cost for repairs $100 per road unit, another cost
only $25. It is highly improbable that any road ought to have
a higher average than $50, some ought to be much lower.
29
Securing Efficiency in Railroad Work
Car Repairs.
Car repairs in 1903-5 were on piece work. At Cleburne,
Texas, prior to the introduction of efficiency principles as to
separate operations there were continually from 600 to 700 freight
cars on the repair tracks, with a force of over 260 men. and an
average monthly payroll of $11,750.
Since the introduction of efficiency principles, cars detained
for repairs have averaged 200 a day, the force has been greatly
decreased and the payroll reduced to $10,300, although each
individual worker earns very much more, the bonus plan having
replaced piece rates.
At the Topeka shop planing mill bonus was introduced Janu¬
ary 1st, 1908. For the three months ending May, 1910, the reduction
in cost of work was 11.9 p.c.
Comparisons at Richmond, California. Monthly average car
repairs before and after bonus.
Last 6 mths. First 6 mths. Per cent.
1906. 1909.
Cars repaired .... 1,572 2,130 35.8 increase
Pay roll $5,716 $5,583 2.3 decrease
Decrease in cost per car repaired 28 p.c.
Examples of early gains and of subsequent improvement
could be multiplied indefinitely.
No railroad shops have gone so far, so thoroughly into stand¬
ardizing individual operations as the Santa Fè since 1907. Some
of the recent standard practice instructions are marvels of
research, diligence, skill and style.
Fuel.
Fuel is the largest single item of expense on most railroads.
On the Santa Fè equated* fuel paid for amounted to an average
of 250 lbs. for each 1000 tons weight of freight trains per mile.
Actual measured consumption on test runs of freight trains proved
to be about 80 lbs. instead of 250.
According to the Erie Railroad Magazine September 1910,
the average amount of coal used per mile by its passenger
locomotives is 108 lbs. One division was selected for special
supervision and care and its average was soon brought down to
79.5 lbs. per mile. One locomotive of four of same class on this
division was selected for attention, and during six days' test its
coal consumption was reduced to 35.1 lbs. per mile.
A switch locomotive is reported by Mr. W. C. Hayes, Super¬
intendent locomotive operation work of the Erie Railroad, whose
*60 lbs. of oil counted as 100 lbs. of coal.
30
Securing Efficiency in Railroad Work
fuel consumption on test with more work was reduced per hour
from 2,088 lbs. to 720 lbs.
These three examples cover locomotives in freight service, in
passenger service and in freight service, in each case test con¬
sumption being about one-third of average consumption. The great
difference is due to the ten links of dependent sequence between
coal mine and ash pit. The ten links are—
Coal charged by mine, but not delivered to car.
Coal shrinkage in transit.
Coal shrinkage in unloading.
Coal shrinkage in bunker.
Coal shrinkage in loading.
Coal wasted in round house before locomotive takes train.
Coal lost through wasteful firing.
Coal lost in wasteful running.
Coal burned, waiting on side tracks.
Coal lost to ash dump, especially when fire is pulled.
The writer knows personally of great losses that have occurred
in each of these steps, from million dollar swindles at the mine to
diversion of whole cars from bunker. Even in this test, Erie
record of 35.x lbs. per mile, out of 143 hours the locomotive waited
under steam 53.5 hours, remade the fire and heated cold water three
times.
If we assume an average efficiency of 90 p.c. for each of the
ten links, our end efficiency is only 34.83 p.c., not much more than
the 32 p.c. realized on A. T. & S. F. under average conditions and
operations as compared to standard.
The Erie passenger record and the switch engine record are
not as good since the former included at most six of the depen¬
dent sequence links and the latter included only two, wasteful
firing and wasteful running.
It does not seem open to question that fuel consumption on
American railroads is twice as high as it would be if—
Conditions were standardized,
Operations were standardized.
Some of the sequence links eliminated.
On the A. T. & S. F. the few studies made, the intermittent
attention given seemed to reduce average coal consumption from
260 lbs. to 239 lbs. for a whole year, or about 9 p.c., but the
records were not reliable and therefore had questionable value.
MORAL GAIN.
The second ideal, that of restoring harmonious relations
between employer and employe was realized even more success¬
fully, during the progress of reductions. In the first official
31
Securing Efficiency in Railroad Work
conference with Mr. Kendrick he authorized progressive efficiency
reward based on demonstrated attainment. It was the staff's
task to work out the details.
It is a fact that an arithmetical advance from 80 p.c. to
90 p.c. in four different operations will effect a geometrical im¬
provement from 41 p.c. to 65 p.c. It is the highest kind of
economy to offer an efficiency reward sufficiently high to secure
the workers' earnest cooperation. It is also evident that stan¬
dardized conditions and standardized operations will enable any
worker, without undue strain, to do better work in less time. It
is easier to ride 20 miles over a level, smooth and well lighted
but shady road on a good bicycle than to walk S miles over a
bad, hilly, dusty road. In the former case conditions and tools
are standardized, in the latter they are not.
It is a profound physiological truth that a dirge is depressing,
that a quick step is exhilarating, that pleasureable dance music is
fast, that it is intensely exhausting to drag along slowly in a
crowd. There is a certain determinable speed of work that can
be maintained with least effort and depression. It is this speed
that ought to be set in standardized operation and while it is
much slower than spurt performances it is much faster than the
average of unstandardized performance. Who has the desirable
and crack run, the engineer on the 60 mile an hour flier or
the novice on the switch engine or on the slow local freight?
It is also evident that if a worker earns $75 a month and
has to spend $70, his net profit is $5.00, a sum so low as to make
a man of advancing years and with a growing family and burdens
incline to anarchy. If, however, this worker can earn an extra
efficiency reward of $15 to $25, his net earnings become $20 to $30.
In commercial shops the manufacturer's upper limit is the
selling price. A lessening in cost and an increased output will
often increase the profit manifold. The worker's lower limit is
the reasonable minimum cost of living. A larger earning
increases his net profit, the difference between cost of living and
earnings, 400 p.c. to 600 p.c. and he becomes an incipient capitalist.
His position is also much more secure since a scientifically
managed shop would no more permit a foreman to discharge a worker
of high efficiency than it would permit him to take a sledge
hammer and destroy a costly machine. If either valuable man or
valuable machine is out of order, he or it must be nursed back
to standard condition.
Just as in the material equation—
Cost = Q P
it was found that Q was far more important than P, so in the
labor equation—
Cost = T W
32
Securing Efficiency in Railroad Work
(T, time; W, wage rate per hour) W can be advantageously
increased if thereby T is lowered. It has been demonstrated
again and again by all the great time and motion study specialists
that T can be lowered to one half or even one quarter by stand¬
ardizing conditions and operations and by securing the coopera¬
tion of the worker.
In a scientifically standardized shop the question of workers'
earnings ceases to be a problem. Such a shop can always afford
to pay more than any other shop, can always secure the very
best men, who require far less supervision, who make fewer
mistakes, who play the game as others will not and cannot.
The efficiency reward system proposed by Mr. Emerson and
approved by Mr. Kendrick was to be based on preliminary work
as follows:
Conditions to be first standardized.
Operations to be standardized.
Every operation to be scheduled on the basis of carefully
made time studies.
All work to be despatched.
Instructions to be definite and clear.
Records to be reliable, immediate and adequate.
There are thus six efficiency principles leading up to the
just and equitable application of the final principle of
EFFICIENCY REWARD.
The standardized, equitable conditions are that:
(1) Each man shall be engaged at a definite and equitable
hourly rate, this rate being the current local rate for his trade
or kind of work. This rate will be paid for every hour he works,
irrespective of his efficiency.
(2) Definite time unit equivalents shall be determined and
stated in advance for each operation assigned. This time unit
equivalent is the mathematical expression of the fair deal, a fair
hour's work for a fair hour's pay. It is an extension of the idea
that a bushel of wheat is 60 pounds, not an uncertain basket
quantity.
(3) The carefully defined fair hour's work for the fair hour's
pay is called 100 p.c. efficiency and for attaining this efficiency
the worker is paid a bonus of 20 p.c. The bonus diminishes as
efficiency diminishes. At 90 p.c. efficiency the bonus is 10 p.c., at
80 p.c. efficiency the bonus is 3.25 p.c., at 67 p.c. efficiency or two-
thirds of a fair hour's work the bonus stops.
(4) If a worker, owing to his own skill or strength, delivers
more than 100 p.c. efficiency he is paid at full rate for all the time he
saves and in addition he is given wages and 20 p.c. bonus for the
time he works. The whole of the gain in time is his own.
33
Securing Efficiency in Railroad Work
Foremen are paid bonus on the basis of the average efficiency
of the men under them, superintendents are paid bonus on the
basis of the foremen's efficiency. The method could be extended
up to the president.
This was the plan of the efficiency reward approved for the
A T. & S. F. It contributed its share to arithmetical improve¬
ment in separate operations and this resulted in geometrical gain
in dependent sequences.
Through the application of the various principles, including
increased interest of the men, average efficiency at the largest
shop was brought up from 60 p.c. to 100 p.c. and the plan has
been found so satisfactory that it has been extended from its
small beginnings until now in the fiscal year 1909-10 about
$1,250,000 dollars a year was paid in bonus and its administration to
the Sante Fe employes over and above the current wages of their
craft and locality.
The testimony of Mr. Kendrick, Vice-President, of General
Manager Koons of the Eastern Grand Division, of Superintend¬
ent of shops J. Purcell at Topeka, of Mechanical Superintendents
and Master Mechanics, of Mechanical Accountant as well as of
the disinterested investigators Mr. Chas G. Fry and Mr. R. V.
Wright of Railway Age Gazette and of Mr. Chas. B. Going of the
Engineering Magazine are unanimous in their endorsement of the
moral and material gain.
A few opinions as to the effects of this plan of efficiency
reward must suffice.
Mr. W. F. Buck, Supt. of Motive Power, Feb. 13th, 1908.
"Since 1904 I had continuous opportunity to watch the effect of
this plan and I cordially endorse it as being of great value in
systematizing work, increasing output and encouraging vigorously
diligence and fidelity on the part of all employes." Mr. D. E.
Barton, Gen. Foreman of the Topeka shops in an address to the
Railroad Foreman's Association: "The Efficiency System is
distinctly cooperative. It changes the men from half-hearted,
listless, idle, indifferent workmen to striving, alert, active,
intelligent, honest, self-respecting workers, who take an intense
interest in the work at hand and are willing to do what¬
soever their hand finds to do with all their might. This system
also stimulates the judgment of the foremen to select the proper
men, the proper machines for each job, (because the foreman's
efficiency consists in the efficiency of the men under him.)
This system has increased output because it has secured the
confidence and enlisted the cooperation of the worker.
It has brought about a genuine reconciliation between those
who work and those who direct. Grumblings have ceased and
34
Securing Efficiency in Railroad Work
instead one hears that this man or that one is leaving the
service because he is not allowed to work on the efficiency plan,
(because it has not yet been extended to his department.)
If anyone has misgivings about material being spoiled, come
to Topeka, look over a carload ready for shipment, take gauges,
templates, rule and callipers, see and judge for yourselves."
Aside from these expressions from line officials there is also
the testimony of the records.
Comparison in conditions at the Needles, Arizona Division:
April 1902. 1909.
Passenger locomotives turned daily 6 7
Freight locomotives turned daily 18 13
Gross tons handled 77,742,800 152,430,860
Mechanics' wages per hour $o-375 $0.42
Average bonus per hour $0,043
Pay roll $16,813 $17,813
Bonus paid $1,849
Number of men 260 223
Locomotive failures 57 n
Miles between failures (all) 4,377
Miles between failures (passenger trains) ... 17,683
Miles between failures (freight trains) 29,995
As to this improvement Master Mechanic Wall, then at Needles,
now in charge of the San Bernardino shops, is quoted as follows:
"You wish to know what effect the bonus system has had
on improving conditions. The first and very important is that it
has provided one of the best systems for checking the work of
round house repairs I have ever experienced or handled in and
around round houses during the past fourteen years. It brings
the men and foremen closer together. It excites the interest of
the individual because it increases the pay check. It has also
increased the efficiency of the individual workman, has guided us
in weeding out poor men, in keeping and encouraging the better
men. It has assisted us in making more quickly our heavy repairs.
Before the system was installed it was common practice for a
machinist and his helper to spend twenty-five hours swinging a
cylinder on the 900 class engines and completing the job, but at
present we are doing this work in eleven to twelve hours or
while the boiler is washed out and other repairs made, thus getting
the engine out in her turn, and saving us from heavier assign¬
ment of power to protect us against protracted repairs. We
obtain more miles from each engine assigned."
Previous to the introduction of this efficiency reward plan, labor
difficulties had been chronic on the A. T. & S. F. as on other
railroads.
35
Securing Efficiency in Railroad Work
In 1903, President Ripley stated (Annual Report p. 12) : Con¬
trary to the expectations of your officers the expenses of mainte¬
nance of equipment have not materially diminished. Among the
principal causes were the increased cost of material and of labor
and the decreased efficiency of shop labor.
Report 1904, p. 11: The Maintenance charges include the
extraordinary expenses due to the machinists' strike.
On the A. T. & S. F. the very carefully planned and insistently
carried out method of efficiency reward was evolved to remove
once for all the constant clash between employer and employe, to
eliminate inefficiency and carelessness and to adjust wages auto¬
matically in proportion to efficiency.
Since 1904 there have been no labor difficulties among the
workers under the efficiency reward plan.
On other roads, Eastern and Western, labor difficulties have
been chronic and labor increases the rule. I need only refer to the
protracted Erie strikes, to the strike on the Denver & Rio
Grande; to the March, 1908, defiance of the management, by the
N. Y., New Haven & Hartford shopmen, and to the March, 1908,
report of President A. B. Stickney, of the Chicago Great Western,
in which he states, "More than half of the 12 p.c. increase in
expense was in maintenance of equipment. The road shops and all
its round houses were in the grasp of men who disregarded the
interests of the company. The rate of wages of all men engaged
in the maintenance of equipment was increased about 10 p.c. in the
autumn of 1906. The larger part, however, of the increased cost of
maintenance must be ascribed largely to the general insubordination
and disorganization of the forces employed in making repairs."
"When the company tried to get rid of inefficient strike¬
breakers, it had as much trouble as with the union men."
Of an Efficiency Check
The moral effect does not, however, apply to wage-earners
alone.
Where there is no positive efficiency check it is very common
for foremen and other officials to pad the expenses in various
ways, some careless, some corrupt, as the following occurrences
show;
"Guilty of Stuffing Road's Pay Roll."
"William B. Blauch, twenty years of age, of Richmond Hill, a
former time-keeper of the Long Island Railroad, was convicted
before County Judge Humphreys at Long Island City on a charge
of conspiracy and larceny. He will be sentenced. Blauch was
accused of putting the names of five "dummies" on the pay roll
and with collecting the pay that was coming to them."
36
Securing Efficiency in Railroad Work
In connection with car repairs on the Illinois Central a number
of high men were found to have cheated the road out of more than
one million dollars.
But a short time before there was a defalcation on one of the
lines of the New York Central system of over $600,000.
These frauds are impossible when standards of expense are
established and actual expenses are comparatively checked.
In a higher way the principle of efficiency reward has empha¬
sized the principle of the fair deal. In extensive investigations
covering the whole of the United States I have only found one
plant in which the piece rates were based on fairness. If con¬
ditions and operations are unstandardized piece rates to be
effective or attractive must be high, far higher than they ought
to be under standards. There is also no incentive to the
management to better the conditions but if individual men
better the conditions or if they show individual rapidity the rate
is cut as soon as it can be done as to all the men.
The bonus method of efficiency reward rests on a wholly
different moral basis. On the Santa Fe it is not the mechanical
employes who are clamoring for higher pay, it is a Vice-President
who is reported to have said that in spite of the fact that over
one million dollars was paid annually in bonus above regular
rates, he would favor a readjustment upwards of the Santa Fè
basic rate so as to maintain an attractive differential, if other
roads in Santa Fè territory should increase rates of pay to
employes so as to equal Santa Fè rates, for it would be unreason¬
able to expect men to perform a greater amount of work for
the same or lower rates than are paid by other companies for the
performance of a less amount of work.
The A. T. & S. F. Ry. has applied efficiency principles and
the graduated bonus efficiency reward plan on the largest scale
that this or any other efficiency reward plan was ever attempted,
but even there, owing to the magnitude of the task, the novelty
of the whole plan, the imperfect comprehension, when it was
started of the underlying principles, by all concerned, they have
been only partially and unevenly applied. The comparatively few
men engaged in the work have accomplished wonders, but much
more remains to be done, especially in the elimination of depen¬
dent sequences.
The same principles have been applied elsewhere in industrial
companies of the first magnitude with varying success, depending
wholly on the conviction with which the principles were accepted
and put into effect.
There is at the present time and in the present wealth of
experience but little more difficulty about the attainment in any
37
Securing Efficiency in Railroad Work
shop or among any employes of 100 p.c. efficiency than there is in
civil engineering in realizing a St. Gotthard or Simplon tunnel,
in marine engineering in realizing a Mauretania or Lusitania.
Scientific management can be applied to industrial production,
railroad operation and maintenance with absolute certainty as
to results.
It is not a question of reducing, costs, since costs apply to money
only, but of increasing efficiency, and efficiency applies to time and
output, to reliability and well being, all more important than cost.
But costs are not lost sight of. One railroad Vice-President
who for six years had watched the effects flowing from efficiency
states with emphasis:
"I have yet to know a case in which with extra compensa¬
tion as a motive any given undertaking has not been greatly
cheapened to the Company, in which revenue had not been
increased to the worker."
But how do costs fare under the old plan, if costs are taken
as the test of methods?
Another Vice-President of a great railway system who
deliberately and scornfully refused even to consider efficiency
principles, holding them impractical and useless, states October
15th, 1910, before the Interstate Commerce Commission that in
his belief an increase in wages to employes is from the railroad
point of view absolutely a total loss, since the men do not do as
much work as before.
This essay on the causes and dangers of inefficiency has been
illustrated by examples taken from railroad practice solely because
the writer's experience has lain extensively in that direction and
because railroad operations are among the most difficult to
standardize. But inefficiency exists everywhere. Inefficiency is a
prevalent disease, but by reason of the dependent sequences of
modern complex operations it has become violently dangerous.
On the germ free western prairies in the hands of a hunter
a dirty hunting-knife was a safer instrument for a capital surgical
operation than the best surgical instruments in the hands of
skilled surgeons a few years ago in the infected wards of the
seemingly perfect hospitals. Modern surgery has attained its
victory over progressive decay by making all surroundings
efficiently clean. Modern industry must attain its victory over
progressive loss by making all operations efficiently economical.
38
This book is a preservation facsimile
produced for the Northwestern University Library.
It is made in compliance with copyright law
and produced on acid-free archival
60# book weight paper
which meets the requirements of
ANSI/NISO Z39.48-1992 (permanence of paper)
Preservation facsimile printing and binding
by
Acme Bookbinding
Charlestown, Massachusetts
2012