Combined Hyperheuristics for Searching of Series of Non-Dominated Service System Designs

The development of various heuristic algorithms for finding a set of mutually non-dominated solutions to the location problem is generally a consequence of the extremely time-consuming exact methods, which are not practically usable. Therefore, the goal of experts in this field is to create such algorithms that could approximate the Pareto front in sufficient quality with the smallest possible demands on computational time. In this paper, we suggest and study the algorithm combining tabu search approach with a hyperheuristic based on random selection of the subordinate heuristics according to their rank. The reported computational study will be aimed at evaluating the efficiency of suggested approach applied to common benchmarks.