Flow shop scheduling with no-wait flexible lot streaming using an adaptive genetic algorithm

Abstract

In this paper, we propose a flow shop scheduling problem with no-wait flexible lot streaming. The problem involves the splitting of order quantities of different products into sublots and considers alternative machines with different processing times. Sublots of a particular product are not allowed to intermingle; that is, sublots of different products must be non-preemptive. The objective of the problem is the minimization of the makespan. An adaptive genetic algorithm (GA) is proposed, which is composed of three main steps. The first step is a position-based crossover of products and local-search-based mutation for better offspring. The second step is an iterative hill-climbing algorithm to improve the current generation. The last step is the adaptive regulation of the crossover and mutation rates. Experimental results are presented for various sizes of problems to describe the performance of the proposed algorithm and they show promising outcome compared to traditional GAs.