MATHEMATICAL MODELS FOR DEVELOPING REGIONAL SOLID WASTE MANAGEMENT POLICIES

Various mathematical programming models are applied to evaluate operational (static one-period) and dynamic investment. (multiperiod) policies of regional solid waste management (SWM). Special attention is paid to the use of mixed integer programming (MIP) models for dynamic investment policies.

The common objective of minimizing the present value of overall investment and/or management costs is extended to deal explicitly with land-use policies by evaluating the sequencing of landfilling operations. The question of abandonment or upgrading of facilities is introduced into the mathematical framework because of its importance in the light of more restrictive standards. The decision as to which process should be installed at what location and at what time, as well as the decision which landfill should be operated and when, is identified in the MIP model.

Considering the limitations of the data base and the model formalism, the question of the usefulness of relying on one optimal solution from a model is discussed. The insensitivity of the ordinary present value criterion is displayed by analyzing substantially different policies which have only slightly different values of the criterion function. Attempts at introducing the notion of regret for the policy selection are presented in the context of Paretian environmental analysis.