Ad-a017 935 Capsule View of Future Developments on Large Scale Network and Network-related Problems

The primary purpose of this paper is to identify recent major accomplishments and prophesy (suggest) future trends in the solution, modeling, and human engineering aspects of network and network-related problems. In particular, the suggested future trends consist of an integrated sequence of modeling, theoretical and computational developments which may allow mathematical programming to attain its envisioned goal of being a major vehicle for solving pressing industrial and world problems. Network and network-related problems include such inathematical problems as assignment, transportation, transshipment, multi-commodity networks, generalized networks, set covering and partitioning problems, matroicl problems, plant location problems, fixed charge networks, constrained networks, general linear programs with GUB and network substructure, constrained spanning tree problems, traveling salesman problems, and combinations of these. In fact, it hafj recently been shown that any 0-1 integer programming problems are equivalent to 0-1 generalized network problems. Network-related problems that arise "naturally" (i.e., those whose formulations make conspicuous reference to a network structure) cover a diverse spectrum of practical applications. These applications include problems in telecommunication, microdata set merging, cash flow,