A METHOD FOR USING OR-MODELS TO SOLVE ECO NOMIC-PROBLEMS WITH UNCERTAIN-INFORMATION USING THE EXAMPLE OF SHORT-TERM PRODUCTION PROGRAM-PLANNING IN DAIRIES

To overcome problems encountered in dairies it appears appropriate to use mathematical methods for providing decision-making aids. The tradi tional deterministic methods are, indeed, absolutely unrealistic, beca use economic planning, being related to future, must necessarily be ba sed on uncertain data. Consideration of data uncertainty in decision m odels creates degrees of freedom in choosing an appropriate admissibil ity and optimality notion. It is shown which alternatives are existing and which models of stochastic- and Fuzzy-Sets-optimization as a func tion of this choice are available. In this work emphasis is placed on the representation of the scenario optimization method, which starts f rom the deterministic approach; based on it, the future is described b y different scenari excluding each other. Starting from the solutions of these scenario problems it allows, in a second step by a coordinati on model, an altogether optimal solution to be obtained, which can, de pending on its configuration, cover the range from intelligent averagi ng of the individual results up to data equivalent to the stochastic a nd Fuzzy-Sets-models. The procedure is described using the example of short-term production program planning in a dairy; in this case it is possible to formulate also the coordination model within the linear st ructure of the individual models, however, with additional mixed/integ er variables. It appears that by explicit consideration of data uncert ainty the economic success can be improved compared with the determini stic approach. This work is completed by a comparison with alternative procedures and indications relating to open questions.