Innovative local search heuristics for uncapacitated facility location problem

This paper presents four different local search heuristics, abbreviated as PSC, PSTC, PTSC and PFSC, for handling the UFLP. It compares their outcome with that of a previously proposed equivalent heuristic called add-swap neighbourhood search (ASNS). One of its main focus is to reduce processing time since ASNS heuristic's computational time is quite long and continues to increase in an exponential order with magnitude of the problem. The proposed heuristics use a two-phase approach wherein the first phase ranks the potential facilities' locations and selects an initial solution from the ranking; and the second phase keeps on adding other locations from the ranking list, one at a time, until the stop criterion is achieved. Its results show that the proposed heuristics reduce computational time dramatically when compared with the ASNS. Furthermore, the PFSC heuristic provides better performance in both computational time and solution quality when dealing with large size problems.