T h e I n s t i t u t i o n

VOL. 38 NO. 7

o f P r o d u c t i o n E n g i n e e r s J o u r n a l

JULY, 1959

PRODUCTION ENGINEERING

AT IMPERIAL COLLEGE

A Report prepared for

the Journal by

Professor O. A. SAUNDERS, F.R.S.,

Head of the Department of Mechanical

Engineering and Dean of City and

Guilds College;

Dr. J. M. ALEXANDER,

Reader in Plasticity;

and R. C. BREWER,

Lecturer in

Mechanical Engineering

THE question of how production engineeringshould be taught at a University is a difficult oneand the answer is probably different for each

University. Imperial College has been consideringthis problem for a number of years. It is clear thata technological institution with a longstanding recordof contribution to technology and science should beable to make a substantial contribution to the educa-tion of production engineers in this country. Themain difficulty is to decide how this may best bedone, bearing in mind the diversity of problemswhich normally confront the production engineer.The Imperial College has perhaps a certain advantagein the very diverse activities within its organisation,ranging from the very theoretical, such as nuclearreactor physics, to the very practical, such as ore andmineral dressing, with subjects such as statistics,plasticity and engineering sciences somewhere inbetween. Moreover, the level of these subjects ranges,in general, from first year undergraduate to post-graduate, which includes the Higher Degrees of M.Sc.and Ph.D. of the University of London, of whichthe Imperial College of Science and Technology ispart.

Within the Imperial College, the study ofengineering is carried out in the City and GuildsCollege which has five Departments, namely Civil,Mechanical, Electrical, Chemical and Aeronautical.It would be possible to introduce productionengineering as a separate department, but thispresents certain difficulties. One of these is the factthat production engineering differs from the above-mentioned branches of engineering, both as regardsits field of interest and the extent to which its fieldof interest overlaps other disciplines. The five existingdepartments have fields of interest in which a sub-stantial proportion of the whole is relevant mainlyto that particular branch; production engineering hasa much smaller area of its own and a far larger area

351

where it overlaps other disciplines; moreover, it over-laps a whole range of disciplines with which theother branches of engineering are not normally incontact, e.g., the field of Human Relations involvingPsychology and Sociology and the new fields whichhave developed from statistical mathematics. Thesolution has been to incorporate the activities withinthe Department of Mechanical Engineering, but toregard the service as available to and involving allinterested departments. At the moment the Depart-ments of Mining and Electrical Engineering are theonly ones using this facility, but it has the advantageof ensuring continuity in lecturing, as compared witha programme given by several outside lecturers.

Although the Production Engineering Course atImperial College is only a few years old, the Collegehas made many contributions, in the past, in thefield of research into production engineering problemsof a technological nature. These were notably inthe field of metal working and plasticity under thesupervision of Professor Hugh Ford who holds theChair of Applied Mechanics (l)-(7). It is natural,therefore, that the present activity should include afairly complete consideration of the technologicalaspects of production. The course is thus rather broadin conception and may be conveniently divided intotechnological and non-technological sections.

A full-time postgraduate course is offered whichlasts for one academic year. Qualification for entryis a University degree and two or three years'experience in industry although the degree require-ment may be waived in the case of candidates whocan show that they would benefit from the course.During the time the course has been running theapplications have considerably exceeded the numberof places available, which is of the order of 10- 12,this being felt to be the maximum number ofstudents who can be adequately supervised underpresent conditions owing to the re-building activities.

award of the D.I.C.

Satisfactory completion of the course leads to theaward of the Diploma of the Imperial College(D.I.C). There is a minimum of formal examinationsand such examinations as exist are very muchdesigned to test the candidate's ability to think andapply rational principles and this has led to a formof examination rather different from that usuallyaccepted in the branches of engineering more akin tothe field of natural philosophy. One of the principalfactors in assessing a candidate's suitability for theaward of this postgraduate diploma is his ability toconsider some particular problem, technological orotherwise, and solve it under the conditions whichwould normally pertain in industry. For this to be areal test a period of about three months is required;to this end the course has been designed so that theformal lectures are given in the first two terms ofthe year, thus leaving the third term free for theproblem on which the student will submit his thesis.Evidently if the student, with the agreement of hissupervisor, chooses a subject outside the technological

field, it is generally not possible for him to conducthis investigation within the college, due to thedifficulty in providing continuous production facilities,although a problem in a field such as office organisa-tion could perhaps be studied. The College is mostgrateful to several industrial companies who haveassisted in making such investigations possible,particularly in view of the fact that the problemmust be such as to enable the student to exercise hiscritical abilities.

At a higher level, it is also possible for studentsto proceed to the M.Sc. Degree of the Universityof London, the residential requirement for which istwo academic years. Agreement has been reachedwith the University that students in ProductionEngineering may need to be away from the College,since an industrial company is often, in effect, theirlaboratory. By this agreement students are allowedto be away from College for two terms and sincethese are normally the Autumn and Spring terms ofthe second year, a period of some nine or ten monthsis available including vacations. The students inproduction engineering at Imperial College havehad very diverse backgrounds, having come from, forexample, various branches of engineering, science ornaval architecture. These students have rarely hadany extensive training in production engineering andit is thus necessary that students reading for the M.Sc.Degree should be offered some basic lectures onproduction engineering; this is normally achieved byarranging for the student to attend certain of thelecture courses for the postgraduate diploma inproduction engineering.

At the undergraduate level, third-year mechanicalstudents may take production engineering as anoptional subject. This subject is divided into twoapproximately equal parts, the first concerned withmachine tools and the machining process and thesecond with industrial engineering. The first sectionis complementary to the compulsory first year subject" Mechanics of Manufacturing Processes " and servesto complete the consideration of the processes ofproduction.

The academic arrangement of the postgraduatecourse is broadly as follows. All students are requiredto take three compulsory subjects :-

1. Industrial Engineering,2. Management and Industrial Administration,3. Statistics and Statistical Methods.

The Industrial Engineering lectures and tutorialsare given entirely by the staff of the ProductionEngineering Group but, for the other two subjects,use is made of the specialist facilities of other Collegeswithin the University. Management and IndustrialAdministration is taught at the London School ofEconomics, which is a School of the University andspecialises in Economics and Political Science.Imperial College has a Department of Mathematicswhich includes a Professor of Statistics and severalmembers of the staff specialising in statistical methodsand operations research; it thus is natural that the

352

third compulsory subject should be taught in thatdepartment.

In addition the student is required to attendsufficient of the following list of optional lecturecourses as to give a minimum of 50 lecture-hours.The optional lecture courses are :-

1. Production Processes I,2. Production Processes II,3. Dimensional Metrology,4. Gauging, Inspection and Testing,5. Industrial Sociology,6. Principles and Practice of Machining,7. Economics of Production Processes,8. Non-Metallic Materials,9. Automatic and Numerical Control of Produc-

tion Processes,10. Operations Research,11. Plasticity of Production Processes.

The majority of these lecture courses consist of aone-hour lecture each week for one term and thuscount as 10 lecture-hours.

In selecting the subjects included in this list, theirchoice has been influenced by the following concepts.It is felt that the science of engineering productionshould not be limited solely to considerations of" industrial engineering ". The technological aspectsof various processes and their inter-relation with thematerials used in manufacture deserve equal emphasiswith industrial engineering, since a radical improve-ment in a manufacturing technique will usually out-weigh previously optimised procedures, so that a newappraisal must be/made of the situation.

The first optional subject, Production Processes I,is intended largely for students who have not had apredominantly engineering training and is a verygeneral course on the various production processes.In the lecture course, Production Processes II, thetechnological aspects of these processes are consideredin greater detail. The remainder of the optionallecture courses are virtually self-contained, inasmuchas they assume no particular previous knowledge ofthe subject. Consequently the first lecture or twoof several of these lecture courses are devoted to theformulation of the problem, a consideration of how itwill be tackled and a brief introduction to any new.techniques or disciplines which may be required; inthe case of Plasticity of Production Processes, a wholeterm of introductory work is undertaken and it hasbeen found that very few students can omit thisintroductory work satisfactorily.

weekly seminarsOne feature of the course is the weekly seminar

which occupies a period of two hours, the first hourof which is devoted to a lecture, often given by oneof the students, and the remaining hour to questionsand discussion. This seminar period serves a two-foldpurpose; firstly, since each student is required togive one lecture during the year, he acquires someexperience in the marshalling of facts into a formsuitable for presentation as a lecture including the

very valuable experience of subjecting himself toquestions and discussion. Secondly it enables anumber of subjects to be dealt with which are eitherisolated or highly specialised and consequentlydifficult to incorporate into the formal lecturecourses. These latter lectures are for the most partgiven by outside experts and on some occasions arecoupled with a visit to the industrial company withwhich the lecturer is associated.

The accusation has often been levelled thatengineering courses, in both technical colleges anduniversities, tend to place too much emphasis on whatmight be called the " blackboard solution of pro-blems ". The works visit is often regarded as oneway in which this can be partially remedied, butit is felt that very often the traditional half-dayvisit to a works is more a test of the physical endur-ance of the students than a process of industrialenlightenment. With this in mind, three or fourvisits are arranged during the year to differentindustrial companies, each visit lasting two or threedays and involving lectures given by the staff of thecompanies. The students are conducted round theworks or selected parts of the works in a manneraimed at giving them the best appreciation of theproblems facing the company and the way in whichthey have been or are being solved. It will be appre-ciated that visits of this nature are not easy to arrangeand, indeed, would be quite impossible but for thevery generous co-operation of several of our moreprogressive companies.

The aim throughout the course has been torecognise that differences exist in the educationalapproach to production engineering problems andevery effort has been made to allow the studies toproceed in the best possible way and not be hamperedby preconceived notions based upon many years'experience of a general engineering education.

research work in production engineeringIn this account of the research work which is in

progress in the Imperial College, no mention will bemade specifically of research into forming processeswhich is undertaken as part of the plasticity pro-gramme within the field of Applied Mechanics. Theresearch work undertaken within the ProductionEngineering Group falls broadly into three cate-gories : (a) technological, (b) industrial engineering, (c)human relations. These will be considered in turn.

(a) technological:In view of the previous remarks about the research

in plasticity of forming processes, it is not surprisingthat the bulk of the technological work has centredon the machining process. Two studies have beenmade on the machining process at low speeds; inone of these, an experimental method was used toderive a slip line field. This was in contrast to, andin some ways complementary to, the work done byOxley at Leeds University (8). Work is still pro-ceeding on the other study, in which the mechanismof deformation is being investigated metallurgically.This work is rather more qualitative than the former

353

study but it covers a wider range of variables andhas a two-fold object:- to discover the extent of thezone of plastic deformation and the way in whichthis changes with strain rate. An experimentalinvestigation has just been concluded on the heatbalance of the shaping process. Up to this time, mostexperimental programmes of heat balances formachining processes have centred on the continuousprocesses, such as turning and drilling which hasrestricted the theoretical analyses virtually to theorthogonal turning operation. It is hoped that thislatest work will give some information on the effectsof the intermittent heat generation in processes suchas shaping.

A further investigation which is rather on theborder-line between technological and industrialengineering interests, has been concerned with theutilisation of banks of automatic screwing machines.The allocation of a number of automatics to a setterand a feeder is obviously a problem in machineinterference but is not a problem for which a readysolution exists. With the co-operation of two industrialcompanies, an experimental programme was under-taken involving between 100,000 and 150,000observations. The analysis of these observations isproceeding in parallel with theoretical work based onstatistical concepts and it is hoped that a significantcontribution may be made to this particular problemin the field of interference.

(b) industrial engineering:There are three main industrial engineering prob-

lems being studied. The first is concerned with anoptimum procedure for inspection in a factoryengaged in batch production. Up to now thisproblem appears to have been solved in industry ona rather haphazard basis, often using the concept ofthe travelling inspector. Little regard seems to havebeen given to the optimisation of inspection pro-cedures in batch production factories and the experi-mental programme has revealed to date that incidenceof wasted time due to interference is higher eventhan was expected. As part of the optimisation,queuing theory is being used to assess the number ofinspectors required in relation to either the numberof operators or the inspection work.

The second study is concerned with the problemof a " feeder'" division in a medium-sized manu-facturing company. The idea behind such a divisionis that it should act, in effect, as a sub-contractor tothe main manufacturing divisions and would be ableto operate efficiently because it can guarantee a betterutilisation of facilities than any one of the maindivisions. This is particularly true of expensive specialpurpose machinery. The problem, however, is notas simple as this. To mention one thing only, itmight be economically more profitable to sub-contractto an outside company. Another point which maybe equally important is the question of delaysoccasioned by the fact that the feeder division has togear its' progress of work to the other manufacturingdivisions and it may not always be able to do this ina satisfactory way. This can lead to a manufacturingdivision having to wait for a sub-contract to be

completed by the feeder division, under circumstancesin which the work could have been done immediatelyhad it been placed on external sub-contract. Thisgeneral problem is being studied for one industrialcompany and although the particular results will beunique to that company, it is hoped that many broadconclusions may be drawn regarding the usefulness offeeder divisions.

The third industrial engineering study is concernedwith a basic cybernetic problem, namely the flowof information through a system and the successfulfeed-back of information which is necessary to ensurethat adequate control can be exerted. It is conceivedlargely as a control problem (speaking technologically)but obviously there are many differences between thefunctioning of a physical closed loop system and thatof a complex industrial organisation. This is apressing industrial problem which has been solvedon a largely ad hoc basis and it is becomingincreasingly apparent that the natural process ofmaking control more complex is proceeding at afaster rate than man's steps to simplify it. This typeof problem is not easy to formulate but the fact thatit is being studied at all, inevitably leads to someimprovement within the organisation concerned. Itis hoped that this study will at least contribute some-thing, to our fundamental knowledge of transmissionof information in a complex organisation.

(c) human relationships:

The research undertaken recently in South EastEssex under the direction of Miss Joan Woodwardindicated that similarities in structure between manu-facturing companies are based, not so much on theirsize or type of industry, but on the production systemoperating in them. It also suggested that the natureof the production spstem has a marked effect on bothmanagement behaviour and relationships.

This study was rather limited in nature and itwas impossible to do much more than compare threebroad categories of production, viz., unit, mass andprocess. The report (9) aroused considerable interestin many circles and the Production EngineeringGroup at Imperial College decided to carry the workfurther. Initially, this is involving the considerationof a system for classifying manufacturing concerns, alltypes of production and production control beingcovered. This work is being undertaken mainly bypost-graduate students.

With the co-operation of Miss Woodward, it ishoped to extend the work in two further stages; thefirst of these is fairly well-defined and will consist ofsurveying recent work done in the Human Sciencesfield related to the system of classification referredto above. The second stage would consist of a studyof the human behaviour and relationship associatedwith the different types of production control. Sincethis is connected closely with the outcome of theprevious work it is not possible to say much morethan that the study would concern itself with threedifferent levels, viz., operator behaviour, supervisoryrelationships and management structure.

354

work done for D.I.C. thesesIt has been mentioned previously, that one of the

conditions for the award of the D.I.C. is that thestudent should undertake some investigation orresearch and submit an account of it in the form ofeither a thesis or a dissertation.*

The work undertaken for this purpose may formpart of the general programme of the ProductionEngineering Group or may be quite independent ofit. Typical of the former are the investigation intothe economics of numerically-controlled machinetools and the work mentioned above, on classificationof production systems. Recent individual studies havebeen concerned with the following topics: theestablishment of a quality control system for a factoryworking to exceptionally close tolerances; determina-tion of set-up times for automatics using syntheticmethods; an investigation into the possibility of usingspot-welding in shipyards for the construction ofdivisional bulk-heads.

short coursesIn common with other activities in the Department

of Mechanical Engineering, a course of three weeksduration is given each year, mainly to cater for theneeds of industry. This year the title of the course is" Plasticity of Production Processes" and techno-logical aspects of production engineering havereceived emphasis. In brief, the first week of thecourse is devoted to the fundamentals of plasticitytheory, the second week to large scale processes suchas rolling, extrusion, and forging, the third week tofinal shaping processes such as deep drawing, wiredrawing, impact extrusion and machining. In otheryears, the emphasis will be placed on other aspectsof engineering production. Candidates for thesecourses are in general responsible engineers fromindustry, and in order to keep them abreast ofcurrent development, experts who are eminent in

* The University of London makes a definite distinctionbetween a thesis and a dissertation; the former is concernedwith original research whilst the latter is an " ordered andcritical exposition of existing knowledge ".

their own particular field are invited to contributelectures.

In conclusion, it is hoped that this somewhat briefreview of the activity in production engineering atImperial College will have given a broad idea of theaims and present achievements in this field. The stafffeel that the target of providing a well balancedtraining as between technological and non-techno-logical subjects has been attained, whilst allowing thestudents a significant degree of choice in their detailedcourses.

Mention should be made here of the manyindustrial companies who have assisted the Collegein its production engineering activities, either byproviding facilities for research (both long and shortterm), by arranging visits to their works or in pro-viding lecturers with contemporary knowledge ofspecialised aspects of production engineering. Itwould not be invidious to make special mention ofProduction-Engineering Ltd., who have used theirconsiderable facilities to foster interest in the courseand to establish a liaison between the College andmany of the companies mentioned below :-

Albert E. Reed and Co. Ltd., Larkfield.Alexander Stephen and Sons Ltd., Glasgow.Barclay, Curie and Co. Ltd., Glasgow.Broom and Wade Ltd., High Wycombe.Brush Electrical Engineering Co. Ltd., Loughborough.Bryce Berger Ltd., Staines.C.A.V. Ltd., Acton.English Electric Go. Ltd., Stevenage.Ford Motor Co. Ltd., Dagenham.Hoover Ltd., Greenford.Igranic Electric Co. Ltd., Bedford.International Computers and Tabulators Ltd.,

London.Joseph Lucas Ltd., Birmingham.Production-Engineering Ltd., London.Quasi-Arc Ltd., Bilston.Siemens, Edison, Swan Ltd., Woolwich.Vauxhall Motors Ltd., Luton.William Denny and Brothers Ltd., Dumbarton.

GENERAL REFERENCES

1. P. W. Whitton and H. Ford. "Surface Friction andLubrication in Cold Strip Rolling." Proc.I.Mech.E.,Vol. 169, No. 5, pages 123 - 140.

2. G. Lianis and H. Ford. " Control Equations of Multi-stand Cold Rolling Mills." Proc.I.Mech.E., Vol. 171,No. 26, pages 757 - 776.

3. H. Ford, F. Ellis and D. R. Bland. " Cold Rolling withStrip Tension." Journal of the Iron and Steel Institute,Part I, Vol. 168, pages 57-75; and Parts II and III,Vol. 171, pages 239-249.

4. A. B. Watts and H. Ford. "On the Basic Yield StressCurve for a Metal." Proc.I.Mech.E., Vol. 169, No. 58,pages 1,141 - 1,156.

5. J. M. Alexander and H. Ford. " Experimental Investiga-tion of the Process of Expanding Boiler Tubes."Proc.I.Mech.E., Vol. 171, pages 351-381.

6. L. E. Culver and H. Ford. " An Experimental Study ofSome Variables of the Tube-expanding Process." To bepublished in Proc.I.Mech.E.

7. J. M. Alexander. " Deformation Modes in MetalForming Processes." Proc. of the Conference on Tech-nology of Engineering Manufacture, London, 1958.

8. P. L. B. Oxley. Ph.D. Thesis, University of Leeds, 1957.

9. Joan Woodward. " Management and Technology."D.S.I.R. Report, H.M. Stationery Office.

{A list of recent publications by the staff of the Production Engineering Groupat Imperial College appears in Library Additions.)

355