2. the beginnings of or

7/29/2019 2. the Beginnings of Or

1/11

The Beginnings of Operations Research: 1934-1941

Author(s): Joseph F. McCloskeyReviewed work(s):Source: Operations Research, Vol. 35, No. 1 (Jan. - Feb., 1987), pp. 143-152Published by: INFORMSStable URL: http://www.jstor.org/stable/170920 .

Accessed: 04/03/2013 09:01

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

.

INFORMSis collaborating with JSTOR to digitize, preserve and extend access to Operations Research.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=informshttp://www.jstor.org/stable/170920?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/170920?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=informs


2/11

THE BEGINNINGSOF OPERATIONSRESEARCH:1934-1941JOSEPH F. McCLOSKEY

California State University, Dominguez Hills, Carson, California(ReceivedApril 1986;acceptedJuly 1986)

This paper, the first in a series on the historyof operationsresearchand management cience, traces the scatteredbeginnings f operations esearchrom WorldWarI upto the activities n Britainbeforeandduring he earlymonthsofWorldWarII.

Operations researchwas born of radar n the eveof World War II. But its advent was forecastbefore and during World War I in connection withthree technologies introduced during that war: thedreadnought, the aeroplaneand the submarine.The dreadnought (a large battleship armed withheavy caliber guns in turrets) was the product ofstudies by Admiral John Fisher while he was super-intendent of Britain's Portland naval base. With theRoyal Navy finally divorced from wind and sail,Fisher and a small group of associates recognizedthetendency to design warshipsin terms of the availabletechnology ratherthan in terms of strategicand tact-ical requirements. This group designed a new navy,with far fewer ships in commission, with heavy reli-ance on reserveships and personnel, and with vesselsthat met the requirements of warat sea. When Fisherbecame First Sea Lord in 1904, he was in position tomove his "Grand Design" from plan to reality,and itwas essentially Fisher'snavy with which BritainfoughtWorld War I.'The aeroplane interested F. W. Lanchester, apioneer in the motor car industry who had alsomade fundamental contributionsto aeronautical the-ory. He developed his N2 law to describe the outcomeof military actions, relating victory to numerical su-periority, superiorityin firepower, and concentrationof forces. He illustratedhis theory by examples from

This article is the first chapter of A History of OperationsResearch and Management Science, sponsored by The Instituteof Management Science and the Operations Research Society ofAmerica, to be published in the Publications in OperationsResearch Series. Permission of Elsevier Science Publishing Com-pany, Inc., to publish this article is gratefully acknowledged.

conflict on land and sea and predicted that, in futuretimes, the outcome of great battles would be deter-mined by the "efficiency of the aeronautical forces."Lanchester's work had no effect on operations inWorld War II, but it now occupies an honored placein operations research.2In 1915, Lord Tiverton,who in 1921 was to succeedhis father as the Earl of Halsbury, began a detailedstudy of strategic bombing. In September 1917, hesubmitted a report to the Air Board that dealt withtarget selection, methods of navigation, weather, andlogistical problems. He argued in favor of daylightbombing because of the characteristicsof availableaircraft, he trainingof crews,and problemsof main-tenance. He also suggestedthat all available aircraftshould concentrate on a single target on any singleday, thus anticipatingthe rationale that led to 1,000plane bombing raidsduring World War II.3In the realm of antiaircraft operations, A. V. Hillheaded the experimental section of the MunitionsInvention Department of the British Army, and hisunit came to be known as "Hill's Brigands." Theystudiedantiaircraftgunneryand developedtacticsandprocedures that enhanced the effectiveness of antiair-craft fire. We shall meet up with Hill and some of hisBrigands ater in this story.4To complete this brief review of the World War Iprecursors of operations research, we look to the ef-forts of Thomas A. Edison as head of the U.S. NavalConsulting Board. He developed statistics to aid inevasion and destruction of submarines, used a "Tact-ical Game Board" for solving problems of evadingsubmarine attack, and analyzed zigzagging as amethod of nrotecting merchant shinning against

Subject lassification: 01 OR/MS historyOperationsResearch 0030-364X/87/3501-0143$01.25Vol. 35, No. 1, January-February987 143 ? 1987OperationsResearch ocietyof America

This content downloaded on Mon, 4 Mar 2013 09:01:44 AMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


3/11

144 / MCCLOSKEYsubmarines.5Ronald W. Clark credits Edison with asignificant role and characterizes this aspect of hiswork as "nothing less than an early piece of opera-tional research." He points out that Edison studiedsinkings by submarines and that "one startlingdiscov-ery emerged: most ships were still following prewarroutes and it was on these routes that most sinkingsoccurred. Another point was that, while only 6% ofsinkings took place at night, most vessels were stilltraversing he dangerzone in daylight."6During the interwaryears, nothing of significanceto the historyof operations researchwas done in theseareas; at least, nothing has come to light. And thisdespite continued progress n the development of air-craftand submarines, surfacevessels, tanks and othervehicles of land warfare,and radio and telephone. Thedesigners ed, the tacticslagged,and effective counter-measures were virtuallynonexistent.

As for countering aircraft, Stanley Baldwin, a for-mer and again-to-be BritishPrime Minister, summedup the 1932 position succinctly: "The bombers willalways get through."That estimate was made some-what more comfortable by the "official"assumption(as a basis for military budgets) that there would beno war for 10 years.Work was underway on acoustic devices to provideearly warningof the approachof aircraft,but the rangeof these devices was limited and they were sensitive toground noises. Good work was being conducted onaircraftand on the tactics to be employed by fighteraircraftagainst intruding bombers, but there seemedto be no way around the requirement for standingpatrols. The advantagesof barrageballoons had alsobeen recognized. Nevertheless, Britain'sposition wasbecoming increasingly perilous.7There were voices abroad in the land calling foraction and specifically for greaterreliance upon sci-entists.Foremostamong them was Winston Churchill(not then a member of the Government), promptedno doubt by his friend and personal scientific adviser,Frederick A. Lindemann, later Lord Cherwell. Bothwanted action in support of increased scientific en-deavorbeforeabandoning the field to the bomber.8Their concern may have influenced A. P. Rowe,then assistant for armaments to H. E. Wimperis,Director of Scientific Research in the Air Ministry.He was certainly influenced by his own convictionsabout the value of civilian scientists to the military.In any event, afterwitnessing a demonstration of theacoustic mirrors then being installed on the southcoast of England, Rowe, on his own initiative, wentthrough all 53 files on activities within the Air Min-istry relating to airdefense. He found little to encour-

age him and so wrotea memorandum to his superior.He statedvery bluntly that, unless the scientific com-munity could come up with something new withwhich to counter the bomber, Britainwould lose anywar that started within the next 10 years. This wasJune 1934.9The Tizard Committee and RadarIn November 1934, Wimperis acted. He recom-mended that a Committee for the Scientific Study ofAir Defence be created to "considerhow far recentadvances in technical knowledge can be used tostrengthen the present methods of defence againsthostileaircraft."LordLondonderry,Secretaryof Statefor Air, approved, and the committee was formed.Henry Tizard accepted the chairmanship.The otheroutsidemembers wereA. V. Hill and P. M. S. Blackett.Wimperis represented the Air Ministry, and Roweserved as secretary.Because of its accomplishments,this committee, usually identified by its chairman'sname rather han by its mission, must be ratedas oneof the most unusual and successful bodies of its kindthat has ever existed.Tizard was a chemist, Rector of the ImperialCollegeof Science and Technology, and a World War I testpilot. Hill was a physiologist, a professorat UniversityCollegeand, at the time, the most distinguishedandinternationally best known of the members of thecommittee, having shared the Nobel Prize in physiol-ogy and medicine in 1923. I have alreadytouched onhis World War I servicefor the Army. Blackett, gain-ing recognitionas one of the best of the youngermenin physics, had served in the Royal Navy in WorldWar I. His distinction in operations researchand hisNobel Prize in physics were in the future.Wimperis was an engineer and inventor. He isknown for the Wimperis accelerometerand for thecourse-settingbombsightthat he developedforthe AirService in 1917 (it remained in use until 1939). Hehad been appointed the first Director of ScientificResearch in the Air Ministry in 1924. Rowe, at thetime of his appointmentas secretaryof the committee,wasa civil servant of no particulardistinction.As withBlackett, his great contributions lay in the future, assuperintendent of the Air Ministry Research Stationat Bawdseyand of its successor(afterseveralchangesof name and location), the TelecommunicationsResearch Establishment (TRE), from 1938 until theend of the war.At Hill's suggestion,and before the firstmeeting ofthe Tizard committee in January 1935, WimperisapproachedRobert Watson-Watt, superintendent of



4/11

ORForum / 145the Radio Department of the National Physical Lab-oratory, about the feasibility of "death rays,"a per-ennial favorite of amateur inventors. Watson-Wattand his assistant,A. F. Wilkins, made some calcula-tions and concluded that there then existed no meansof incapacitating a crew or disabling an aircraftbyradiation. But Watson-Watt added a paragraph o hisreport to Wimperis suggesting that reflected radiowaves might be used to locate aircraft.At Wimperis's request, Watson-Wattexpanded onhis idea and it became the principal tem on the agendaof the first meeting of the Tizard committee. Onemonth later, an experiment, conducted by Wilkinsand witnessedby Watson-Wattand Rowe, was carriedout, using a transmitter near Daventry as the sourceof radio waves. The crude receiver that had beenassembled for the test detected a plane, whose pilotwas unaware of the reasons for the course he wasflying. Rowe's report to the committee was favorable,and work on radar was pressedfrom that time on (asRDF, for radio direction finding, until adoption ofthe U.S. name).1?Watson-Wattwas not a memberof the Tizardcom-mittee, but his concept of radar gave the committeepoint and purpose. He had served in World War Iprimarily n meteorological posts. He had participatedin the work that had gone on in the 1920s on themeasurement of the Heaviside layer, and knew ofother work with radio that laid the foundation forradar. There is some controversy about his contribu-tions to radar and operations research,"'but withouthis particular genius, the Tizard committee mighthavebeenjust another committee. On the other hand,without the influence, access, and drive of the com-mittee and especially of its chairman, Watson-Watt'sideas might have suffocated in the machinery ofbureaucracy (as had happened with earlier workon reflected radio waves, and particularly that ofW. A. S. Butement and P. E. Pollard)."2As it is, theTizard committee and Watson-Watt, working to-gether, deserve a significant share of the credit forAllied victory in World War II.The path of the Tizard committee was not com-pletely smooth, however. Shortly after it was formed,Churchillinsisted that Lindemann be made a mem-ber. Lindemann felt very strongly that high priorityshould be given to developing aerial mines and in-frareddevices;he was also criticalof the slownesswithwhich the committee moved. Because the committeemaintained its priorityon radar,Lindemann becameincreasingly critical and shared his criticism withChurchill, who would then air Lindemann's viewsabout the committee and its workingsin meetings of

the Subcommittee on Air Defence of the Committeeof Imperial Defence, of which Tizard was also amember.By September1936, this situationreached the pointwhere the public membersresignedand the committeewas dissolved-only to be reconstitutedimmediatelywith its originalmembership, plus EdwardA. Apple-ton, Britain's oremostauthorityon radio and a futureNobel laureatein physics (1947).13That Lindemannremained opposed to radar, as some have claimed,14is belied by his support of Watson-Watt and by hissubsequent strong interest in Bomber Command ofthe Royal Air Force (R.A.F.) and its need for radar.Iwill have more to say on this subject.Despite the problems within the committee, workwas moving forwardon radar.An "Ionospheric Re-search Station" had been established at Orfordness,on the east coast of Britain. It was staffedby Watson-Watt on a part-time basis, by Wilkins and E. G.Bowen, all transferredfrom the Radio Department,and by a few assistants. Within a few months, theorganizationwas moved to BawdseyManor, nearby,and was renamed the BawdseyResearchStation, withWatson-Watt, now transferredo the Air Ministry, assuperintendent.By this time, construction of a chainof radarstations to be locatedalongthe east and southcoasts had been approved, with the first one to belocated at Bawdsey.Radar and Fighter ControlAs the possibilitiesof a radar-aideddefense began totake shape, Tizard recognizedthat radarhad createda whole new set of problems in fighterdirection andcontrol.Accordingly,he began a series of experimentsat the R.A.F. Fighter Command stationat BigginHill,starting n the latterpartof 1936 and lastingfornearly2 years. Inasmuch as the first radarstation was notoperational when these trials began, artificial infor-mation of the type that radar was expected to supplywas passed to the control center at Biggin Hill sothat new methods of interception could be devised.This work was under the immediate supervision ofB. G. Dickins, an engineer from the Royal AircraftEstablishmentat Farnborough.The Biggin Hill experiments integrated radar intothe early warning systems, which included the Ob-serverCorps, and into fighterdirection and control.They obviated the need for standardpatrolsof fighteraircraft and the coincident constant alert status ofantiaircraftgun crews. And they started operationsresearch.5While the Biggin Hill experimentswere in progress,



5/11

146 / MCCLOSKEYthe radarstation at Bawdseywas finished (1937) andthe complete radarchain was readyby the summer of1939. Meanwhile,earlyin 1938, Rowe was transferredto Bawdsey as assistant superintendent and, whenWatson-Wattwas transferredto the Air Ministry asDirector of Telecommunications Research, Rowebecame superintendent.The Recognition of "Operational Research"One of Rowe's first acts aftermoving to Bawdseywasto assurethe continuation of the work that was con-cludingat BigginHill. He assignedE. C. WilliamsandG. A. Roberts to do "operationalresearches"on thecontrol room and the communication system, respec-tively.Williamswas the first scientistfroma universityto join the Bawdsey staff. Roberts was a telephoneengineer.Their assignmentsrepresentthe firstknownuse of the term "operationalresearch."'6In a sense, the task assigned to Williams was themore significantin terms of the development of OR,for he analyzedand comparedthe performanceof theseveral control centers that had been established inorderto identify factors that could explaindifferences.Thus, there was no direct developmental objective inthe work; ratherit was a case of "doing the best youcan with what you have."Roberts,on the other hand,became involved in development and createdan elec-trical converter that simplified the transmission ofradardata to the control centers.'7During this same period, Tizard interested JohnCockcroft, a professor of physics at Cambridge, inwhat was going on at Bawdsey.Some visits involvingscientistsfrom Cambridgewere arranged,but secrecywas a problem and nearly a year was to pass beforethe decision was made to reveal the secret of radartoscientists outside the government. By that time, sci-entists not only at Cambridge but also at Oxford,Manchester,Birmingham, and London had becomeinvolved. As a result, some 80 scientists reported toBawdsey on September 1, 1939, to spend a longvacation on the radarsites.'8 For some, the vacationprovedlong indeed!Shortly before the arrangementsfor the universityscientists to spend time on the radar sites had beencompleted,Rowe took anotherstep in his sponsorshipof operations research. He entered into agreementwith Raymund G. Hart, a Royal Air Force squadronleader, to assignsome of his staff to FighterCommandheadquartersat Stanmoreimmediately upon the out-breakof war. As it happened, Williams and Robertswere already therefor the 1939 summer air exercises,continuing their work on the control room and on

communications. Harold Larnderwas sent to Stan-more immediately after September 3, 1939, to headthe group, to include Williams and Roberts as wellas J. C. Bower, I. H. Cole, and W. E. Egner, all ofwhom were destined to head OR sections in variouscommands.'9Because this agreement between Rowe and Hartrepresents a milestone in the development of opera-tions research,Rowe's observationsare of interest:By this agreement hopedto give and to get. I hoped togive FighterCommand the servicesof men who had anintimateknowledge f the performance f the radar hainand who had, as scientists, been trainedto use theiranalytical aculties; ndI hopedto getfrommy detachedmembersof staff he true facts concerning he operationsagainst he enemy, andso enablethe Bawdsey taff o bepermeatedwith an intimateknowledgeof the needs ofFighterCommand.20Larnder'sgroup remained a part of Rowe's organiza-tion for nearly a year, operating as the StanmoreResearch Section. During that time, the calls on itsservices began to move away from the original exclu-sive concern with radar and into differentand moregeneral problems. Thus, during the Battle of France,British and French losses in aircraft were heavy, butthe French kept asking for more and more fightersupport. The Stanmore Research Section was askedto evaluate the situation. The analysis showed thatadditional transferswould involve attrition that couldnot be made good and that Fighter Command wouldbe weakened beyond recoveryin the face of the like-lihood of a German attempt to invade Britain.2' Ofthis situation,Churchillhas written:The hard question of how much we could send fromBritain without leaving ourselvesdefencelessand thuslosing the power to continue the war presseditselfhenceforward ponus. Our own naturalpromptings ndmany weightymilitaryargumentsent force to the inces-sant,vehementFrenchappeals.On the otherhand,therewasa limit, and that limit if transgressedwould cost usour life.Atthistime all these ssueswerediscussedbythewholeWarCabinet,which met several imes a day. Air ChiefMarshalDowding,at theheadof ourmetropolitan ightercommand, had declared to me that with twenty-fivesquadronsof fightershe coulddefend the islandagainstthe wholemightof the GermanAirForce,but that withlesshe would be overpowered.Thiswould haveentailednot only the destructionof all our airfieldsand our airpower,but of the aircraft actorieson which our wholefuturehung. My colleaguesand I were resolvedto runall risks for the sake of the battleup to that limit-andthoseriskswerevery great-but not to go beyond it, nomatterwhat the consequencesmightbe.22



6/11

OR Forum / 147The supportgiven to Dowding by Larnder and hispeople "was a notable milestone in the progressof theidea of operationalresearch"23nd led to the decisionto transfer the group formally to Fighter Commandand to designate it the OperationalResearchSection

(ORS) Fighter Command. (It would appear thatRowe's original use of the term operational researchhad dropped out of use and that it was revived byLamder.24)

Scientific Analysis in WarfareWhile operations researchwas thus developing andfinding a place within the structureof the Royal AirForce, other developments destined to have direct orindirect bearing on OR were taking place. Thus, in1937, B. H. Liddell-Hart,one of Britain'smost influ-ential militarythinkers,submitteda memorandum tothe Secretaryof State for War that merits quotationat length:The more one examines he courseof pastwars he moreone is impressedby the frequencywith which militarypolicy and preparationshave taken the wrong turning.And this abnormalpercentageof errorcan be tracedtothe habit of basing policy and preparationson an as-sumption,withoutadequateverification.The way thatdecisionsare reachedon questionsof strategy, actics,organization, tc., is lamentablyunscientific. t is due inpartto the difficultyof developinga trulycriticalhabitof mind underthe conditionsof militarysubordination,and in part to the lack of any staff organ devoted toresearch.... At present he investigationof problems spushedonto officerswho areoccupiedwith currentmil-itarybusiness.That task oughtto be givento a body ofofficerswhocandevotetheirwhole time to exploring hedata on record,collecting t from outside,and workingout the conclusionsin a free atmosphere.Such a bodyshould be composed of the best intellectsin the armywitha good blendof practical xperience, nd in selectingthem particular ttentionshould be given to originalityof thoughtor criticalpowers.It is alsoverydesirable hatthey should be supplementedby a permanentnucleusconsistingof some first-rateUniversitymen who havebeen trained n the processesof scientificenquiry; hesemight take Territorialcommissions to give them anacquaintancewithmilitarypractice.25Thereis no indication that the army took any stepstoimplement Liddell-Hart's deas, but he certainly rec-ognized the problem and recognized that scientifictraining would require reinforcement by an under-standingof the militaryand of military operations.Another approachto bringingscientific talent intothe service of the nation was started in 1938 byG. T. R. Hill and CharlesF. Goodeve, who wrote to

the secretariesof the Royal Society on "PeacetimeOrganization for Voluntary Training of ScientificWorkersin the Event of a National Emergency." Atthat time, A. V. Hill was Biological Secretary andA. C. Egerton was Physical Secretary of the RoyalSociety, which may account for the fact that thesuggestion fell on fertile ground. The result was theestablishment of a Central Registry of scientists,engineers, and other specialists.26Events bore in on the nation before the Registrycould be fully organized, however, and so much ofthe success in bringing the "right"people into radar,and from there into operations research, must go toTizard, Watson-Watt,Cockcroft and Rowe. Addition-ally, many of those who found their way into opera-tions research,whether for the Army, Navy, or AirForce, did so because of the efforts of FrederickBrundrettof the Admiralty.27

When war broke out, Rowe's organization wasmoved from Bawdseyto Dundee in Scotlandbecauseof the former's vulnerableposition on Britain's eastcoast. At Dundee, the groupbecame the Air MinistryResearch Establishment (AMRE), but the locationproved unsatisfactory and so the organization wasmoved to WorthMatraverson the south coast in May1940 and then expanded into the Swanage area. Whenthe Ministry of Aircraft Production was formed,AMRE became MAPRE for a short time but in No-vember 1940 it was renamed the TelecommunicationsResearch Establishment (TRE), under which name itcame to be recognizedas one of the greatlaboratoriesof World War II. Finally, in May 1942, TRE wasmoved to Malvern, where it still exists as the RoyalRadar Establishment.Amid all these changes, it was at Worth Matraversthat an institution conceived by Rowe reached frui-tion. These were meetings held in Rowe's office onSundays. They were open to anyone interested inradar(and with access to the secrets) whether govern-mental, civilian, or military, and regardlessof rank.Thus Lindemann (Lord Cherwell), Tizard, Watson-Watt, members of the TRE staff, and serving officersof the Royal AirForcewerewelcome to airtheirviewson ways in which the uses of radarcould be improvedand on requirements that might be filled by radar.These Sunday Soviets, as they came to be called,proved of great value to the operational research sec-tions of the various commands of the Royal AirForce.28At this point, we must clarify the chronology andterminologyinvolved in the Battle of Britain and theBlitz. The former was Hitler's effort to destroy theRoyal Air Force, and specifically Fighter Command,



7/11

148 / MCCLOSKEYso that the invasion of Britain could go forward;itstartedin August and ended in September, 1940, andinvolved mostly daylight attacks. The latter was Hit-ler's revenge when he lost the Battle of Britain. Itconcentrated on the bombing of British cities, andespecially London. It began in September 1940,peaked duringthe autumn months, and ended in May1941. Most of the attacks were at night.Blackett's CircusAt the height of the Battle of Britain, Tizard and Roweboth urged General FrederickPile, the Army's Anti-aircraftCommander,to use scientists. This wasa quitenaturalcrossing of service lines inasmuch as AA Com-mand was under the operational control of FighterCommand. Hill introduced Pile to the physicistP. M.S. Blackett who, when asked, became Pile's scientificadviser. Blackett's first recommendation was thattechnicians be trainedto put and keep the gun-layingradars in order and that they live on the sites. ByNovember 1940, a school for radar technicians hadbeen establishedat Petersham under J. A. Ratcliffe.Blackett next proceeded to recruit scientists whohad no particular background in radio or radar;hehad seen that much more than radar was involved inthe problems of AA Command, especially since therehad been no equivalentof the BigginHill experimentsto integrate radarinto the overall operations of thecommand. He called this new unit the AntiaircraftCommand Research Group, but it soon becameknown as Blackett'sCircus.29I shalldiscuss some of the work of the Circusin thenext installment, on wartime OR in Britain. But be-fore we leave it, it is interestingto speculate on thereasons for its composition, which gave OR the con-cept of the mixed team. Blackett may have beeninfluenced in the direction of physiologists by hisfriendshipand respectfor Hill, whose son, D. K. Hill,also a physiologist, became a member of the Circus.Or Blackett may have recognized that the man-centeredproblemsof an armyarequite different fromthe ship-centered problems of a navy or the plane-centered problems of an air force and that scientistsaccustomed to working with individual differencesinvolving statistics and probability might have a spe-cial contribution to make. Or it may be merely thatBlackettentered the market for scientists too late andthat most of the "hard"scientists had been absorbedinto the war effort elsewhere. Whatever the reasons,the Circus consisted of Blackett, a physicist, then oftwo physiologists, two mathematical physicists, anastrophysicist, a surveyorwith a backgroundin anti-

aircraft,and lateradded another physiologist and twomathematicians.30The Spread of OR WorkThe next milestone in the development of OR alsoinvolves Blackett-and Rowe. One of Rowe's greatconcerns was that, given a successful outcome in theBattle of Britain,the U.K. still could not hope to winthe war unless the submarine menace to its lifelinescould be countered. This meant not only a need forradarequipment, but also the kind of scientific sup-port Larnder and his group were giving to FighterCommand. As Rowe puts it:... we did not know whatwas happening n the battlesagainst hesubmarines,argelybecause herewasnothingat CoastalCommandcorrespondingo the OperationalResearchSection at Fighter Command. We wanted toknow thehoursflownpersubmarine ightingby dayandby night,the averagerangesof radar ocationsand, lessobviouslyour business, he resultsof attackson subma-rines.I thereforewent to AirMarshallSirWilfredFree-man, Vice-Chiefof the Air Staff,who was everready olisten,and urged hat P. M. S. Blackettshould be askedto form an OperationalResearchSection at CoastalCommand.Thiswasarrangedwithin a fewdays.31

Thus, after little more than six eventful months atAntiaircraftCommand, Blackett moved to CoastalCommand in March 1941. C. E. Bayliss replacedBlackett, but the Circuswas soon mergedinto an AirDefence Research and Development Operational Re-search Group under Ratcliffe. Ratcliffe in turn waslured away by Rowe to head up the Post-DesignService as part of TRE that Rowe felt was needed tofill the gap created both by the growth of radar andby the broadeningof operations research away fromits initial association with radar. Ratcliffe was idealfor the assignment because of the experience he hadgained in training radar technicians for AA Com-mand. His placethere was takenby B. F. J. Schonland,a physicist from South Africa, who expanded theorganization into the Army Operational ResearchGroup (AORG), servingthe whole Army rather thanjust AntiaircraftCommand.When Blackett arrived at Coastal Command, hefound there a Radio Operational Research Officernamed John C. Kendrew.His title was one that Roweused when he assigned TRE staff members to anorganizationthat did not yet have a formaloperationsresearchgroup to serve as liaison between the com-mand and TRE. Althoughhe was in a sensedisplacedby Blackett, Kendrew was destined to make uniquecontributions to wartimeoperationsresearchand was



8/11

OR Forum / 149then to join the impressively long list of wartime ORpersonnelwho won Nobel Prizes(chemistry, 1962).32After some 9 monthsat CoastalCommand,Blackettmoved again,this time to his firstlove, the Admiralty.He became Scientific Adviser and Director of NavalOperational Research. There he established a smallcentralizedgroup, and he remained at this post untilthe end of the war.This move represented ittle breakin continuity for Blackett inasmuch as Coastal Com-mand was under the operational control of the Ad-miralty. Thus, his efforts continued to be focused onthe problem of ridding the sea lanes of the submarinemenace so that the suppliesneeded forthe prosecutionof the war could reach Britain.Perhapsmore important to the history of operationsresearch than his pioneeringwork in AA Command,CoastalCommand, and the Navy was the note Black-ett prepared on the occasion of his move to theAdmiralty.This was his "Scientists at the OperationalLevel," "written n December 1941 in order to informthe Admiralty of some of the developmentswhich hadoccurred n the OperationalResearch Sectionsalreadyestablishedat Fighter,Anti-Aircraftand CoastalCom-mands." This note was accompanied by a seconddocument, "A Note on Certain Aspects of the Meth-odology of Operational Research," "originated in1941 as an attempt to set out, for the benefit of newscientific recruitsto the operationalresearchsections,some of the principlesthat had been found to underliethe work of the first two yearsof the war."33In "Scientists at the Operational Level," Blackettcrystallized he ideasthat led him to develop hisCircusas something more than a group dedicated to makingradar work. He makes a case for operations researchas a distinct and identifiable activity requiring theefforts of highly competent, scientifically trained in-dividuals, working at the operational level, and pro-viding commanders with analyses of problems whileservingasa communication link betweenthe technicalestablishmentsand the operating commands. As heput it, the use of operation research "can help avoidrunning the war by gusts of emotion." It could alsoserve as a corrective to his conclusion "thatrelativelytoo much scientific effort has been expended hithertoin the production of new devices and too little in theproper use of what we have got." Together with hisintroduction of operations research into AntiaircraftCommand, Coastal Command, and the Admiralty,"Scientists at the OperationalLevel" provides amplesupportfor those who regardP. M. S. Blackett as the"father"of operationsresearch.Blackett's move to the Admiralty completed theintroduction of OR into Britain's major commands

inasmuch as a BomberCommand group had been setup in September 1941.There,bomber-lossstudies hadbeeninitiatedunder A. E. Woodward-Nuttabout June1940. After a few months, B. G. Dickins, then withthe Ministry of AircraftProduction, took over thesestudies.Later,Lindemann initiatedstudiesof bomberaccuracy. These showed that Bomber Command, re-lying on sextant navigation, was achieving accuraciesone-third to one-fourth of what had been expected.Crews that claimed to have found the target were,two-thirds of the time, not within 5 miles of it.34During the summer of 1941, A. 0. Rankine wasassigned from TRE as a Radio OperationalResearchOfficer and was succeeded by G. A. Roberts fromORS Fighter Command. Finally, the decision wasmade to establish an Operational Research Sectionwith Dickins as its head and Roberts as his principalassistant.35

In addition to the military OR groups, there wasone very significant nonmilitaryeffort. The Ministryof Home Securityestablisheda researchunit at PrincesRisborough, Buckinghamshire, shortly after the out-breakof war,withtwo principal objectives:The designof equipment and proceduresfor air raid protectionand analysis of the effects of enemy bombing. Thiswork was under the direction of Reginald Stradlingand provided the channel throughwhich J. D. Bernal,Solly Zuckerman, and John Baker, among others,became involved in operationsresearch.36In the next installment, I shall review the work ofall of these groups.But first,some recapitulation s inorder.The ActorsInasmuch as my primaryconcern at this point is withthe people involved rather han with specifictechnicalaccomplishments,I shall divide those whom we havemet so farinto a cast of leading and supportingplayersin the unfoldingdrama of operationsresearch.The supporting players include Wimperis, who hasnot been given adequatecreditin these pages, primar-ily because he retiredin 1937. He had played a mostsignificant role in advancing radar, but he left thestage just as operations research was making itsappearance.Tizard'sawarenessof operationalrequirementsandparticularly his initiation of the Biggin Hill experi-ments generated an atmosphere in which OR, onceborn, could thrive. Cockcroft, another of the postwarNobel laureates(physics, 1951), was a major catalystin bringing civilian scientists into effective workingrelationship with service scientists. Watson-Watt's



9/11

150 / MCCLOSKEYperception of the potential of reflected radio wavesearns him a respected place in the history of OR, asdoes his presumably approving, as superintendantatBawdsey, of the "operationalresearches"assigned byRowe to Williams and Roberts.

Cherwell must be recognized for his interest inBomber Command and his support of OR activitiesthere. Moreover,because of his key position as scien-tific adviser to Churchill, he was in position toadvance-or obstruct-this use of scientific talent.Hill and Rowe proved so influential in so manyways that one is strongly tempted to classify them asforegroundcharacters n the OR drama. Both appearto have had the knack of being in the right place atthe right time-and doing the rightthing-to advancethe development of OR.Of the leading players, the scientists and engineersdirectly involved with OR in the formative days, thenumbers grow largevery quickly. In our story so far,full recognition must go to Larnder,E. C. Williams,and Roberts as the pioneers at Fighter Command; toDickins for his work at BigginHill and then at BomberCommand; and, above all, to Blackett for his percep-tion of operations researchas a new and potentiallypowerful scientific activity in its own right and notonly as an adjunct of radar,as well for his pioneeringwork at AA Command, Coastal Command, and theAdmiralty. And we will hear more of Bernal, Zuck-erman, and Baker in connection with civil defense,and of Bernal and Zuckermanin military OR.And to whom do we credit origins?Watson-Wattmarried radio to the military usage we call radar.Tizard saw that radar changed fighter direction andcontrol and so laid on the Biggin Hill experiments.Rowe saw continuing value in having scientists studyoperations. Larnderheaded the firstformal OR groupand set standards of organization and reporting.37Blackett "universalized" OR in the British servicesand defined the function and scientific credentials ofOR.Perhapsthe best judgment is that of E. G. Bowen,who became known as the "father of AirborneRadar,"when asked about credit: ".... the point isthat this is the way many of us were thinking at thattime, . . . taking a hard quantitativelook at an estab-lished service procedureand suggestingnew methodsand in some casesdevelopingnew equipment to carryit out. This is one reason why the real origins arehardto pinpoint."38Notes

1. John E. Tashjean,Arlington,Virginia, opened hisfiles on Fisher to me. Includedwere extracts from

P. K. Kemp (ed.), The Papersof Admiral Sir JohnFisher (London: Navy Records Society, 1960). Healso made availablea manuscript opy of his paper,"AdmiralFisher, ProfessorPareto, and Social Sys-tems Theory,"since published as "Contributiallosviluppodellateoriadei sistemisociali: 'ammiraglioFishere il prof.Pareto,"n CharlesR. Dechert(ed.),Sistemi Paradigmi Societa (Milan: Franco AngeliEditore,1978),pp. 171-183.2. FrederickWilliamLanchester,Aircraft n Warfare,the Dawnof the FourthArm(London:Constable&Co., 1916).The sectionon the developmentof theN2law is availableas "Mathematicsn Warfare"nJamesR. Newman(ed.), The WorldofMathematics(New York:Simon & Schuster, 1956), Vol. 4, pp.2138-2157. See also Joseph F. McCloskey, "OfHorselessCarriages,Flying Machines,and Opera-tions Research:A Tribute to Frederick WilliamLanchester 1868-1948)," Opns. Res., 4, 141-147(1956).3. NevilleJones, TheOriginsof StrategicBombing.AStudy of the Developmentof BritishAir StrategicThoughtand Practiceup to 1918 (London:WilliamKimber, 1973). (Excerptssupplied by Ronald W.Shephard.)LeeKennett,AHistoryofStrategicBombing NewYork: Charles Scribner'sSons, 1982), p. 87, aftercommentingon visitsby Billy Mitchell, states that"otherAmericanairofficersworkedwith the RNAS[RoyalNaval Air Service] and so became familiarwith the ideas of ViscountTiverton . . on the valueof precisestrikeson critical ndustrialobjectives."4. Noel Falconer,Universityof Manchester,brought omy attentionthe seminal role played by Hill (letterto author, August 17, 1976). See also C. P. Snow,Science and Government Cambridge:UniversityPress,1960) p. 86, note 20.5. William F. Whitmore, "Edison and OperationsResearch,"Opns.Res. 1, 83-85 (1953).6. Ronald W. Clark,Edison (New York: G. P. Put-nam'sSons, 1977), p. 227, and letter,Edisonto SirEricGeddes,November21, 1917.7. A. P. Rowe, One Story of Radar(Cambridge:Uni-versityPress,1948), pp. 3-4.8. Ronald W. Clark,Tizard(Cambridge:MIT Press,1965), pp. 107-108, and The Earl of Birkenhead,The Professor and the Prime Minister (Boston:HoughtonMiffin Co., 1961),p. 110.9. Rowe,OneStory,pp. 4-5.10. Clark, Tizard, pp. 108-18; Rowe, One Story,pp. 5-9; Robert A. Watson-Watt, ThreeSteps toVictory(London: Oldhams Press Limited, 1957),pp. 78-83, 107-112, and, for the memorandum othe TizardCommittee,pp. 470-74.11. Part of the controversyover Watson-Watt's ole inthe developmentof radarstems from his less thangenerousattitude toward the contributionsof hispredecessors nd contemporaries. o far as his role



10/11

OR Forum / 151in the developmentof OR is concerned, here is noquestionbutthat he was superintendent f the Bawd-sey ResearchStationwhen the BigginHill experi-ments and the first "operationalresearches"wereconducted.Beyondthat, there is little support,es-peciallyamong his contemporaries t Bawdsey, orhis statement that "If I was not the first and trueinventor of Operational Research ... I was at thevery least a first and true inventor with one or twoof my early staff at Bawdsey." ThreeSteps, p. 200).Similarly, support is lacking for his claim that"In 1940 I coined for the subject and for thegroups the specific names 'OperationalResearch'and 'OperationalResearchSection"' (ThreeSteps,p. 203).12. Clark,Tizard, pp. 114-115. These two patentedadevice thatwas intended o beusedforthedetectionof ships, from shore or from anothership; neitherthe War Office nor the Admiraltywas interested.This wasin 1931. During he 1920sand 1930s,mostclues to whateventuallybecameradarwereregardedas annoying interferenceby scientistsconductingradio experiments.13. Clark, Tizard,pp. 128-148; Birkenhead,Professor,pp. 188-216. See alsoRonaldW. Clark,SirEdwardAppleton (Oxford: Pergamon Press, 1971), pp.96-103.14. Birkenhead,Professor,pp. 202-14.15. Clark, Tizard, pp. 149-56; see also Birkenhead,Professor, p. 201, quoting R. V. Jones, wartimehead of scientificintelligence n the Air Ministry,on Tizard'scontribution.16. E. C. Williams, "Reflectionson OperationalRe-search,"Opns.Res. 2, 441-443 (1954), states,"Theterm OperationalResearch was specificallycoined,by A. P. Rowe, to describe he activitiesof a smallsection of the Air Ministry Research Station atBawdseyn the years1937-39."17. Rowe, OneStory, p. 26.18. Clark, Tizard, pp. 171-172; Watson-Watt, ThreeSteps, pp. 211-212.19. Rowe, OneStory, p. 52. Hart had been assigned oBawdseyin February 1937 to establish a trainingschool forR.A.F.personnel.He "soondemonstratedthat radarwas not merely a toy for scientists butcouldbe operatedby Servicepersonnel" p. 23). Seealso Air Ministry,TheOriginsand DevelopmentofOperational Research in the Royal Air Force(London:HMSO, 1963), pp. 7-8.20. Rowe, OneStory, p. 52.21. Air Ministry, Origins, p. 24. Larndercredits E. C.Williamswith makingthe calculationsand, muchlater,admittedthat it was he, Larnder,who trans-lated the columns of figures o graphic orm. Dowd-ing'sbiographer,n his turn, credits he graphswith"turning he trick" n convincing Churchill o denythe Frenchrequest.See HaroldLarnder,"TheOriginof Operational Research," Opns. Res., 32,

465-475 (1984).Formore on Rowe'searlyrole,seeW. Peyton Cunningham, Denys Freeman andJoseph F. McCloskey, "Of Radarand OperationsResearch:An Appreciationof A. P. Rowe (1898-1976),"Opns.Res., 32, 958-967 (1984).22. Winston S. Churchill, The Second World War,Vol.2, TheirFinestHour(Boston:HoughtonMiffinCo., 1949), pp. 41-42.23. AirMinistry,Origins,p. 24.24. John W. Abrams,"The Crystallization f MilitaryOperationalResearch into an Accepted Role-1939-45," paper delivered at the TIMS/ORSAJoint National Meeting, San Francisco, May 10,1977.25. Basil Henry Liddell-Hart,The Defense of Britain(NewYork:RandomHouse, 1939), pp. 346-347.26. Ronald W. Clark, TheRise of theBoffins London:PhoenixHouse Ltd., 1962), pp. 56-57.27. Clark,Boffins,pp. 72-73.28. Rowe, OneStory, pp. 83-86; AirMinistry,Origins,p. 39.29. J. G. Crowtherand R. Whiddington, Science atWar (New York: Philosophical Library, 1948),pp. 95-96.30. Clark,Boffins,pp. 142-145. One of the otherphys-iologistswas A. F. Huxley, who sharedthe NobelPrizeforphysiologyand medicine n 1963.31. Rowe, OneStory, p. 105.32. AirMinistry,Origins,p. 105.33. P. M. S. Blackett,Studies of War, Nuclear andConventionalNew York:Hill & Wang, 1962),pp.171-198. Blackett's wo papers,whileveryinfluen-tial in thedevelopmentof operations esearch,werenot the firstpublicationon the subject.That honorgoes to a PenguinSpecial, publishedanonymouslyin the summerof 1940, underthe title, Science inWar. Solly Zuckerman was the principalauthor,along with J. D. Bernaland J. G. Crowther,with"snippets"from many others, including C. H.Waddington,all of whom weremembers of an in-formaldinner club that met once a month. At ameeting during the Battle of France,Allen Lane,founder of PenguinBooks, a guest, suggested hatthe dinnerdiscussion would make a good Penguinbook. The anonymousauthors had copy ready inless than2 weeks,andcopiesof the finishedproductwere availablein another 2 weeks. AccordingtoZuckerman, he book "dealtwith the problemsofspeedingthe translationof scientificdiscovery ntomattersof practical value for the war." See SollyZuckerman,Scientistsand War New York:Harper& Row, 1967), pp. 147-148, and From Apes toWarlords (New York: Harper & Row, 1978),pp. 111-112 and 398-399.34. Clark, Boffins,pp. 220-21; Birkenhead,Professor,pp. 231-232. The terminologyused here is a para-phrasefrom the reporton these studies, as quotedby Birkenhead p. 232), and presumablyrefersto



11/11

152 / MCCLOSKEYthe percentageof bombs on target, but may referto the frequency with which any bombs hit thetarget.35. Air Ministry,Origins, p. 44. Of interest s the factthat two of the original membersof ORS BomberCommand were the MissesK. M. M. Goggin andHilaryLang-Brown, he latter of whom had beenwith Roberts from the early days of the StanmoreResearchSection.No woman achievedthe statusof

analyst in the various U.S. operations research unitsduring World War II.36. Air Ministry, Origins, pp. 41-2; Lord Baker ofWindrush,EnterpriseversusBureaucracyOxford:Pergamon Press, 1978), pp. 1-8.37. Air Ministry, Origins, pp. 194-198, reprints a paperon "O.R.S. Reports" prepared by Larnder in June1943.38. Letter, Bowen to author, July 18, 1977.

Thi d l d d M 4 M 2013 09 01 44 AM
http://www.jstor.org/page/info/about/policies/terms.jsp

2. the beginnings of or

Documents