Self-adaptive randomized constructive heuristics for the multi-item capacitated lot sizing problem

The Capacitated Lot-Sizing Problem (CLSP) and its variants are important challenging optimization problems. Constructive heuristics known to be the most intuitive fastest methods for finding good feasible solutions CLSPs therefore often used as a subroutine in building more sophisticated exact or metaheuristic approaches. Classical constructive heuristics, such period-by-period lot elimination widely by researchers. This paper introduces four perturbation strategies further improve solution quality. We propose new procedure automatically adjust parameters of randomized (RPP) heuristics. is proved offer better with reduced computation times improving time-consuming parameter tuning phase. Combinations self-adaptive RPP Tabu search tested effective. Computational experiments provided high-quality 0.88% average optimality gap on benchmark instances 12 periods items, an within 1.2% 24 items.