A Differential Evolution-Based Algorithm to Schedule Flexible Assembly Lines

Scheduling is key towards improving the performance of a Flexible Assembly Line (FAL). In this paper, a Bilevel Differential Evolution (BiDE) algorithm to solve a FAL scheduling problem is proposed. The BiDE algorithm optimizes the performance of the FAL with respect to two criteria: the weighted sum of Earliness/Tardiness (E/T) penalties and the balance of the FAL. The performance of BiDE is evaluated using the data sets available in the literature and an evolutionary heuristic algorithm published earlier called BiGA. The experimental results show that the BiDE algorithm can solve the FAL scheduling problem effectively and exhibits a superior performance over BiGA.Note to Practitioners-This paper was motivated by the problem of allocating time and equipment resources efficiently in a flexible assembly line to improve its performance. Among the different methods to solve this problem, heuristic search techniques are becoming more popular. Current work in literature have proposed a heuristic search algorithm to solve this problem. However, there is ambiguity in the model. This paper aims to clarify the ambiguity in the proposed flexible assembly line model, as well as applying a different heuristic search algorithm for comparison purposes. Experimental results show that the proposed technique can solve the problem effectively.