Personal and organisational value systems: How should we treat these in OR studies?

Values that are held by individuals, or jointly-held values that are attributed to organisations, affect all stages of the process of decision. It is often difficult to determine how and whether the value systems of decision-makers and analysts should affect the analysis that is carried out; to know what these values are and how to measure or describe them; and to put numbers to values, or the consequences of values, if these are to be represented in models. This paper does not attempt to define the way in which value systems are created or used. It does not deal with value systems as they are treated by decision analysis and theory, both because these ideas are well-documented, and because it seems to me that the attribution of utilities and subjective probabilities, and an appropriate axiom structure, may be impossible or misleading. Arguments for a different approach are given in Part 1, and it is suggested that something less prescriptive and less precise is needed. Part 2 looks at some of the more hopeful ways in which this is being, or might be, attempted. The philosophy generally is to help decision-makers to use their own values with more insight, rather than for these to be defined and used by the analysts. Directions in which further research should be encouraged are thereby implied. Part 3 looks at a number of associated ideas which might well be seen as further material for debate. The paper concludes with these, but it does not draw conclusions. It is presented as a basis for discussion and is necessarily biassed by my own values.