Introduction

During the past decades, the role of optimization has steadily increased in such diverse areas as, for example, electrical engineering, operations research, computer science and communication. Linearand non-linear optimization [WAG75], the search for an optimum of a function of continuous variables, has seen major breakthroughs in the fifties and sixties (the best known example being the simplex algorithm to solve linear programming problems, introduced by Dantzig in 1947 [DAN63]). Major results in combinatorial optimization [PAP82], the search for optima of functions of discrete variables, were obtained predominantly in the seventies. For example, for the Travelling Salesman Problem (TSP) [LAW85], probably the best known problem in combinatorial optimization, both approximation and optimization algorithms of high quality became available in the seventies ([LIN73] and [GRÖ77], respectively) and nowadays, TSPs of up to 318 cities are solved to optimality in reasonable computation times (less than 10 minutes CPU time on an IBM 370/168-computer) [CRO80].