Weighted proximity search

Abstract Proximity search is an iterative method to solve complex mathematical programming problems. At each iteration, the objective function of problem at hand replaced by Hamming distance a given solution, and cutoff constraint added impose that any new obtained solution improves value. A mixed integer solver used find feasible this modified problem, yielding improved original problem. This paper introduces concept weighted allows design called proximity search. In function, low weights are associated with variables whose value in current promising change order while high assigned expected remain unchanged. The help distinguish between alternative solutions neighborhood provide guidance when trying locate solution. Several strategies determine presented, including both static dynamic strategies. proposed compared classic on instances from three optimization problems: p -median set covering stochastic lot-sizing results show suitable choice obtain better solutions, for 75 $$\%$$ % cases, than ones using 96 cases running commercial time limit.