ABSTRACT
This chapter focuses on the optimisation in multi-stage decision processes. The optimisation approach that may be used in analysing such problems was developed about thirty years ago. It was named 'dynamic programming' and this name is universally used. Dynamic programming, in contrast, is a mathematical theory of multi-stage decision processes. Although the basic idea of dynamic programming is a simple one, its execution in any given context can require a lengthy chain of reasoning and a great deal of formal mathematical notations. In its application to this context, the fundamental dynamic programming principle yields the following proposition: whatever the commodity amount the firm decides to sell in any one market, the remaining amount must be divided optimally between the other two markets. Optimisation over time is a field in which there is a particularly large number of successful applications of the optimality principle of dynamic programming.
