On the potentialities of AI to deal with the combinatorial complexity of a telecommunication network planning problem

Abstract The mathematical modeling of many real world problems is a complex and eventually an impracticable task when it leads to a combinatorial explosion. Previous papers have described the structure of a decision support system (DSS) for rural telephone network planning 1 and the structure of a knowledge based system, integrated in the DSS, to support a post-optimal analysis of the results obtained using heuristic techniques. 2 This paper attempts to show the potentialities of AI knowledge-representation techniques for improving heuristic approaches to combinatorially complex problems arising in a rural telephone network planning model.