The evaluation and control of research and development projects

Abstract

In recent years the funds spent on research and development (R & D) have grown considerably. An indication of the extent of the growth in the U.K. was given by Hart (1) who noted that in 1900 approximately 0.05% of the gross national product was spent on research. This percentage increased to 0.25% in 1938, 1.6% in 1954 and 2.7% in 1962. Villiers (2) quotes a similar growth in the U.S., where research expenditure grew from <1% of gross national product in 1947, to about 3% in 1962. (In the U.K. it appears, from some statistics produced by the Ministry of Technology (3), that research expenditure has remained at about 2.7% of GNP over the period 1962-1967). The allocation of these resources poses a number of challenging questions in governmental, industrial and academic spheres. At a national level the kind of questions that might be asked are (a) what proportion of the gross national product should be devoted to government sponsored research, or (b) how should funds be divided between the claims of the aerospace, computer, or machine tool industries, or (c) how should funds be divided between the competing claims of the nuclear physicists and marine biologists. The large industrial concern is faced with similar problems though the resources involved are smaller. ICI for example spent about £30M on R & D in 1968, and during the later 1960's, the growth rate was about 8% per year. The Company must decide on the total amount to be spent on R & D and how it is to be allocated between different Divisions of the Company and different research categories. At lower levels of management two of the questions arising are (a) which projects shall be selected, and (b) how should the flow of resources to projects be controlled. It is now generally accepted that there is a need for techniques for assisting in the management of R & D. Jones (4) summed up the situation well when he wrote "It is not surprising that there is an increasing amount of discussion on the management of R & D for profit. Business becomes increasingly competitive and R & D activities, just as those of production and marketing must be examined to see how they can best play their part." Already a large number of relevant papers have been published, but as yet no significant breakthrough has been achieved. An important feature of the literature has been the concentration on theoretical models as a means of assisting research managers: reports of new methodology considerably out-number reports of practical testing of the methods in research laboratories. Throughout the author's research the opposite bias, that is to say towards a practical rather than a theoretical approach has been maintained. This was facilitated by the author completing most of his research in the R & D Department of the Mond Division of ICI (of which he is a member). The research presented in this thesis began with the very general objective of examining and developing methods for the allocation of resources (capital and manpower) to R & D and so Chapter 1 discusses some relevant methods that have been proposed in the literature. It was later decided to concentrate on the development of an improved system of project evaluation and control. Chapter 2 analyses an established system in this field, and looks at past projects to demonstrate some of the problems such a system should accept. Later chapters present the system that was developed during the research and record experience of testing the various procedures on a number of Mond Division R & D projects. As these are either still in progress or are only recently completed it has been necessary, for reasons of security, to limit descriptive detail and to normalize numerical data. Such normalization has been made in a manner that preserves the essential financial characteristics of the project. It is well perhaps, in the Introduction, to distinguish between the terms research and development. Following Baines, Bradbury and Suckling ( (5), page (2) ) process definition will be the term used to cover the steps required to take exploratory production activities from laboratory scale to full-scale. Development will refer to the problems of opening up a business area with a new product and will include economic assessment and marketing activities. For the most part these activities are closely linked to research activities and are usually performed by members of the same project team. The convention followed in the thesis will be to use the term 'research' to refer to all the activities of the project team and to assume that these also include some development activities as defined above. Only when discussing the work of others who have used the term R & D, or when there is a reason to emphasise the commercial exploitation content of a project will the word development be used. The work to be described is submitted to meet the requirements of the Ph.D degree in Technological Economics at the University of Stirling. (One academic year of the author's research programme was spent in the Department of Industrial Science of the above University). For permission to publish the results of his research and for generous support throughout the author is indebted to the Directors of Mond Division and of ICI Limited. He also wishes to record his thanks to his colleagues at Stirling University and Mond Division for numerous helpful comments and discussions. Most of all he wishes to thank his academic and industrial supervisors Professors F.R. Bradbury and C.W. Suckling for their patience, guidance and encouragement during a most stimulating piece of research.