Operations Research in World War II

This html article is produced from an uncorrected text file through optical character recognition. Prior to 1940 articles all text has been corrected, but from 1940 to the present most still remain uncorrected.  Artifacts of the scans are misspellings, out-of-context footnotes and sidebars, and other inconsistencies.  Adjacent to each text file is a PDF of the article, which accurately and fully conveys the content as it appeared in the issue.  The uncorrected text files have been included to enhance the searchability of our content, on our site and in search engines, for our membership, the research community and media organizations. We are working now to provide clean text files for the entire collection.

in

11

Operations Research World War

■lit

Navy, the group was concerned mainly M mine warfare. In 1945, Dr. Johnson had s^e<

Operations research originated in Britain where, during World War II, it was called “Operational Research.” A small group of scientists were the first to practice op­erational research. Their job was to make maximum use of the small amount of radar equipment available. The importance of this task can be understood by considering the numerical superiority of the German Air Force. By careful analysis of the operations in­volved, these scientists were able to set up a nearly perfect plan which included location of the radar sites, communication between stations and fighter bases, as well as methods for upkeep and repair. Their contributions are considered by many as a major force in the victory of the Battle of Britain. Had the old trial-and-error method been used, months would have been wasted. Moreover, large numbers of men and equipment would have been lost.

Just who introduced operational research to the United States during World War II is disputable. James Bready attributes this introduction to James B. Conant, former president of Harvard University. According to Bready, Conant was in Britain in the fall of 1940 and was so impressed with OR as it was used there that he advised authorities in the United States to follow the example of the

British. Charles Kittel, of M.I.T., attribute communication of the effectiveness of OR 'f Britain to Shirley Quimby, of Columbia b>¹¹ versity, while the latter was in London.

There is general agreement, however, tn³ the first U. S. OR group was formed in 19 " under Dr. Ellis Johnson in the Naval nance Laboratory. Dr. Johnson was a g^e°j physicist at the Laboratory who had broug¹ about the “degaussing” system which el^£t trically neutralized ships to protect them fr⁰’¹' magnetic mines. Under the auspices of ^ put into effect a mine blockade which he h³ mapped for Japan’s Inland Sea. The blo^ ade was so effective that it almost compleR: halted Japanese shipping. Moreover, its ⁶ fectiveness was further proved when the LL Navy, equipped with charts, had difficn'¹ sweeping the waters clear after the war.

One of the characteristic features of op^eI ' tions research is the use of the exchange ra^l‘, which is the ratio of output to input f°^r given operation. An example of the use of ^e‘ change rates was the determination of ¹ most effective means of attacking Japan⁶"* shipping. The costs for the three availa^ methods—submarines, aircraft, and mining' were compared to their effectiveness in nn^lJl

°f Mexico to the Bay of Biscay. Simi-

man ^r°uti th.

_D^Crs °f Japanese ships sunk. By these com- ^ls°ns it was found that submarines were Plating most efficiently, the ^Slrn^^ar comparison was made between _a- ^number of flying hours put in by Allied in ^Cra^ ^an<^ the number of U-boat sightings ^a given area. This analysis indicated areas _tr ^ere there was “overflying,” leading to Qui^{S ers} of aircraft, for example, from the

g . p exchange rates were used by the ^r’tish to determine whether to equip their . ^W1y built escort-type vessels for minesweep- thg ^ ^ant*submarine duty. The study showed _Qj. ^lrtlPortance of the latter. The same type _0n^stlldy was used to evaluate the gun crews Merchant vessels and the performance of ^ 'submarine aircraft in bombing U-boat ^asesi escorting threatened convoys, and polling shipping routes, he use of exchange rates was enhanced by f ^t3ln “ effectiveness ratios” which were ⁿo to exist in various operations. For in­_s ^rJ^Ce! the ratio of 60 mines laid per ship Was found in every campaign: German ^aes in British ports, British mines on Ger­routes, and U. S. mines in Japanese ^es- In a parallel study, it was found that t*^c Probability of an aircraft, no matter what P^e> attacking a sighted U-boat, no matter

where, is practically constant. Thus, a con­sistency was seen in certain operations which permitted the application of data from one geographic location to the problem of pre­dicting effectiveness in another area. This con­sistency of effectiveness ratios was extremely useful in planning force requirements and preparing for their movement and main­tenance.

One problem solved by OR men during the war was the ineffectiveness of airplanes in killing submarines. The OR men found that, although depth charges were exploding at a level at which they could do great damage, this level was very much below that reached by a diving submarine in the time between an alarm and the actual explosion. From these findings came the advice that the charges be set to explode nearer the surface of the water. As a result, the number of kills more than doubled. Records captured from the Germans show that it was assumed that the British had developed a new weapon.

The first thousand-plane raid, made by the R.A.F. in 1942, was the result of an operations research study. Statistics of the losses of R.A.F. Bomber Command aircraft over certain Ger­man cities were collected, and analysis re­vealed that the percentage of aircraft lost decreased as the number of participating air-

1”

craft increased. These results suggested that the German defenses could reach a saturation limit beyond which the losses would remain constant. The 1942 raid as well as later large air attacks confirmed this. A similar effect was noted in an analysis of North Atlantic merchant-vessel convoys by Navy operations analysts under Philip M. Morse. The number of ships sunk was constant and independent of the convoy’s size. As a result, in 1943, the convoy runs were made larger and at less fre­quent intervals. Greatly decreased losses were witnessed with this tactic.

The problem of training and operational distribution was also studied by OR men. Squadrons of B-29s in the Marianas were able to fly a fairly constant number of hours per month. This time had to be distributed be­tween training and operational missions, since low “scores” were made when no time was allowed for training. The original dis­tribution was approximately 4 per cent for training and 96 per cent for missions. How­ever, analyses showed that maximum effec­tiveness occurred when 10 per cent of the time was used for training and 90 per cent for missions. This maximum as much as doubled the bomb weight on the target of the original distribution.

A U. S. Navy OR group studied the ef­fectiveness of creating co-ordinated submarine groups. Clearly, a submarine patrolling by itself can attack only those contacts it makes. If, on the other hand, a group of submarines is working together, each submarine can at­tack its own contacts as well as those of the other group members, thus increasing the effectiveness of the patrol. What, then, should be the maximum size of such submarine “wolf packs?” Since total destruction of ³ convoy was very rare and would usually quire tremendous numbers, the actual upp^er limit was not the most practical size. Anoth^cr consideration was the loss of possible ^c0"- tacts in other routes if large numbers ⁰ submarines were concentrated on one roth^e’ The antisubmarine devices used by Japanese and the ability of the U. S. marines to keep in touch with each other al^s° had to be considered. The OR men esta^ lished predictions as to the amount of ^11,1 provement in relation to the size of the pa^c^' In agreement with these predictions, the U-¹' Navy most effectively used groups of th^|tl submarines in their attacks on Japan^eSe shipping.

Another OR study suggested ways of protec¹ ing maneuvering ships from kamikaze plane⁵' Strangely enough the recommended ted’ nique for small ships was exactly the oppos’^ of that for large ships. These methods '^vC,t able to net a 50 per cent decrease in losses-

The delay in the introduction of the use ⁰ artillery rockets as opposed to guns can ^ attributed to the customary stress on ^a° curacy. It became recognized, however, th‘ for barrage purposes, this lower accura^c' was actually advantageous. Because ¹ broader barrage could be made, there no need to retrain the rocket launchers.

addition, accuracy was useless when the es^a position of the target was unknown. Tj’¹ realization was put to use in connection vvh an anti-aircraft gun. This decrease in ^aC curacy led to a greater percentage of kn¹ ’ since the accurate gun had consistently sh° at the wrong point as determined by the g¹’¹ director. In short, when considering the pr°

^nternational News Photo

p j^na^^s'^s divulged that the number of ships sunk in convoys was constant and inde- ^ent of the convoy’s size; thus, larger convoys resulted in smaller losses.

le

^n,s terms of cost per kill, it became obvi-

j" *'                             ^r

^{)<: Cor}>sidered.

l^{Us t}^^lat other criteria besides accuracy had to ^e Coi '

As ;

₀j. ^a result of an investigation into methods ^Ca'ⁿ°uflaging aircraft, OR men were re- sible for the development of an enamel t which was extremely effective for night ° ^ts- It had a high gloss which also served ^reduce skin friction with the air. Previously

to

^a dull

"'as

^coat had been put on the planes, but it yj Unsuccessful for camouflage purposes. ^ orcover, the glossy coat, because of its re­cod friction, caused air speeds to be in­^Creased.

to

Th

^e effects of OR’s work were not limited

, Machinery and maneuvers. OR dealt with V human problems too. Medicine made ^^Cat Use of OR’s ability to determine treat- ■ ts which were most effective. The Amer- _ch*« Soldier in World War II had greater _gri^^nces for survival, lower risk of infection, ^ ^lnore speed in recovery from wounds and P ^Scases than any soldier in previous history. f_a°.^ncern for the high incidence of combat ip *^^{Ue an<}^ other such casualties which occur _tl| ⁿ'°dern warfare led psychiatrists to make p_r '^Crous case studies and develop improved cedures to cope with these problems. ^r'S^adier General S. L. A. Marshall found gjj * ^^ar was a cause of intense fatigue. The ^a,1°n was so poor that many soldiers were a ^a% too tired to fight or fire their rifles as

a that

^c°nsequence of fear. Moreover, he found

ea_r

^lhe amount of such fatigue-producing

v ^c°'ild be measured by the load a man ^was * ⁷

[jU 4.                                *                                 -

‘U only 12 to 25 per cent of all the armed

tj^^akle to carry. Studies by OR men showed ^CQttib_a

^Jat soldiers who were placed in a position

where they could fire at the enemy were able to pull the trigger. Consideration of human elements such as fatigue and fear was essen­tial for the OR men who calculated force re­quirements for given operations.

Another human problem tackled by OR men was the care of combat soldiers’ feet. Obviously, soldiers could not march well if they had the severe foot trouble that was so common during the war. The operations re­searchers studied socks and boots. Their ad­vice was that each soldier be given two extra pairs of socks. This conclusion was based on the finding that the previously smaller supply of socks had forced the soldiers to wear either clean but wet socks, or dirty socks. In either case, the effect was quite harmful to the feet. With the suggested increment, a soldier could wash his dirty socks and still have clean, dry socks to wear. A similar recommendation was for jackets to replace coats. The coats were found to be so heavy as to tire the soldiers and cause lower travelling speeds for march­ing units.

As the result of a time and motion study carried on by one OR group, the Army dis­tributed posters of GI Joe shaving “without the unnecessary details.” The poster included an instruction sheet which prescribed a method of shaving. If this method were used, it was estimated that one could save as much as five minutes a day. This saving would mean about 1,400 hours of extra sleep for a soldier before he reached the age of 65. Further calculations showed that if the entire shaving population of the armed forces used this method, approximately 304,166,666 hours would be saved in a year. Not to mention 833,333 extra hours for a leap year.

82 U. S. Naval Institute Proceedings, May 1968

Allied OR men under the auspices of the Navy conjectured that the Germans had nothing similar to operations research for planning their naval maneuvers. A study of their activities seemed to indicate that the Germans would have avoided many of their mistakes if they had had a similar analytical unit to organize their military operations. The conclusion that there was no German naval counterpart was later confirmed by Konteradmiral Godt who had taken over Admiral Doenitz’ position as Commander of Submarines. In ground fighting, however, the Germans made extensive use of map maneu­vers. A map maneuver involves the use of topographic maps. The prominence of map maneuvers in German war plans was seen in the Ardennes invasion, the Ukraine invasion, and the proposed invasion of England. Further information on the German use of OR is not readily available. In addition, no mention of the use of OR by the Japanese is made in the accessible literature.

It appears, then, that operations research made its formal debut in World War II al­though some of its methods and principles had been used before that time. For example, F. W. Lanchester, the British aeronautical pioneer, was one of the first men to apply quantitative reasoning to military strategy. His original writings appeared in 1914-15. His most important contribution is known as the “N² Law.” From it are derived the follow­ing principles:

(a)     Numerical superiority may be relatively more important than superiority in wea­pon performance; and

(b)     it is of the highest importance to deploy available combatant units in a single large force and to endeavor to split up the enemy force (strategic principle of concen­tration).

Another instance of use of OR in World War I was in U. S. naval maneuvers. Thomas A. Edison, as chairman of the Naval Con­sulting board during World War I, used a “tactical game board” to plot courses and to analyze results of zig-zagging and other tech­niques as used by merchant ships trying to evade enemy submarines.

After World War I, operations research had

A recipient of both a bach^ lor’s and a master’s degt; from The Johns Hopkins L’ⁿ‘ versity, Mr. Milkman p^artlCl pated in and directed ^stu | j in the operations research fie for the General Electric Co¹® pany Center for Advance Studies and the Planning search Corporation bet⁰¹ joining Peat, Marwick, Living ston & Company’s Operations Research Group ^aS senior consultant. He has conducted analyses for * Office of the Secretary of Defense, NASA, and bo¹ the Navy’s Weapon Systems Analysis Office and ¹ Special Projects Office. He is also a part-time facu> member at Johns Hopkins.

other applications. One area in which it '^vaS effectively, though not extensively, used ^v(^aS business. During the 1930s, Horace C. Levi¹' son undertook the application of scientifi^c analysis to merchandising problems. In l¹*⁵ work for L. Bamberger and Company, l^lC studied customer buying habits, the respond to advertising and the relationship betweej’ environment and the types of articles sold' Previously, Levinson had done a study fo^{r a}, mail order house on the effect on gross sales ⁰ quick shipment of C.O.D. deliveries. Becai^|Sl of his success in predicting public reacting from the study of masses of data, he later i®| tiated a study for Bamberger on the desirabi ity of keeping department stores open at nigh¹' He based his decision on the amount of ^1,1 creased sales as opposed to the expense ⁰ maintaining later hours.

During the war, this new science developed through the application of the scientif'¹ method to problems of a broad nature such ^a* were presented by the strategical and tactic³ operations of warfare. Teams of civilian scief tists were employed in several Allied ^C0N mands to work on all aspects of military sta* problems:   planning, intelligence, deploy

ment, training, and maintenance. Initial- these groups were particularly active und^ef the British.

The main purpose of operations reseafC' in the war was to find ways for making M best use of the military forces and weap⁰¹’’ available at the time. These same technique*’ which were so successful during the "^,ar' found many uses in peacetime. National d^e

fens •

in ’ ^ra^^roa<i operations, highway develop- and traffic control, city planning, and -^nbly-l.^ne operations could all be handled (jj* ^lhose principles of OR developed during ^trion'^Var ^ ^t^^lese situations have in com- ⁿ the fact that their components can be ^can ^S|^^re<^ ^^or effectiveness and conclusions , oe drawn on a statistical or analytic _tor ; ^Jnce again the British have been initia- _ay.^{S ln} the application of OR, using it in civil Co ^l°ⁿ’ housing, the steel industry, and ^uirnerce. In the United States, one of the J* Promising fields for OR—industry—is pli^ .Sinning to take advantage of its ap- _u ,^atlon- The postwar business boom, the ^ ^etlned nature of business problems, and sc' ^SUsP*^ci°ⁿ of businessmen towards this new ; ^tl*i^c technique and towards its costs were °IOR^rtant ^^actors *ⁿ th^e deferral of acceptance

‘^^{n ar}ticle, written in 1949, reported:

It is a new and mysterious science: Opera­tes Research. So new it is that only one j^Uerican university [M.I.T.] offers a course , ^ll> ^and only one textbook has been written

°^ut it. So mysterious is it, because of mili-

^e very existence of some of the country’s ^eading

^e atentioned more than fleetingly in print.

rtior ^rat^°^nS I^^esearcI¹ (^OR) h^as since become jr ^e widespread in its use and exposure. Al- kn°^U^^ ^ *^{s s}lHI ^{not a} subject of common _Sl)|^. ^edge, several universities now teach the ha ^^{eCt an<}I many of its military applications Pof⁶ *^{3een ma<}Ie known. The scope and its ^Cnt*^aI °I operations research were seen in ^successful ^use during World War II, and ,_nd^Sently, many progressive federal agencies business organizations have established

Prii ^atlons research programs ^ ^Clples to peacetime objectives.

^Cati^CCaUSe *^ts y°^{uth an<}^ dynamic appli- ^var'°^nS’ °P^eraff°^ns research is subject to _Wr.^l0us forms of definition. James Bready, ^ltlng for the Baltimore Sun defined it in the

to extend its

abi

^ry security regulations, that until recently ’^ery existence of some of the country’s ag operations research centers could not

0

Operations Research in World War II 83

following way:

Operations research can be called the scientific analysis of problems involving any form of action with a view of making that ac­tion more efficient.

Thomas Saaty is more whimsical in his defini­tion:

Operations research is the art of giving bad answers to problems to which otherwise worse answers are given.

Professor P. M. S. Blackett, a British pioneer of operations research, aptly de­scribed the great procedure:

Past operations are studied to determine the facts; theories are elaborated to explain the facts; and finally the facts and theories are used to make predictions about future opera­tions.

Dr. Ellis A. Johnson, director of The Johns Hopkins University Operations Research Office, had demanding requirements in mind when he opined on the nature of operations research men. His view was that ordinarily a man cannot absorb the scientific background and the methodology of inquiry and analysis needed for operations research before he is at least 30 years old. In his opinion, the opera­tions researcher can specialize in any field ex­cept engineering. His explanation of this ex­ception is that “it is not the making of a better gadget, but the better using of an existing gadget” that is the concern of the operations research man. He deals with facts and must have practical skills. Most important, how­ever, is his ability to formulate and organize generalizations.

Expanded university programs and pro­fessional societies have helped to overcome some of the antipathy toward the use of op­erations research. Today, however, more and more business concerns are making good use of those principles developed during World War II to help them ensure maximum pro­ductivity and efficiency.