Local search methods for type I mixed-model two-sided assembly line balancing problems

Two-sided assembly lines are widely utilized to assemble large-sized products such as cars and trucks. Recently, these types of assembly lines have been applied to assemble different types of products due to a large variety of customer demands and strong market competition. This paper presents two simple local search methods, the iterated greedy algorithm and iterated local search algorithm, to deal with type I mixed-model two-sided assembly line balancing problems. These two algorithms utilize new precedence-based local search functions with referenced permutation and two neighborhood structures to emphasize intensification while preserving high search speed. Additionally, these local search methods are enhanced by utilizing the best decoding scheme amongst nine candidates and a new station-oriented evaluation to guide the search direction. New lower bound calculations are also presented to check the optimality of the achieved solutions. Eleven recent and high-performing metaheuristic algorithms are re-implemented to test the performance of the proposed algorithms. A comprehensive study on a set of benchmark problems demonstrates the advantages of the improvements and the superiority of the two proposed methods. Experimental results show that the proposed algorithms obtain 23 new upper bounds compared with two recently published algorithms, among which 19 cases are proven to be optimal for the first time.

Automated summary

Two simple local search methods, iterated greedy algorithm and iterated local search algorithm, are introduced to deal with type I mixed-model two-sided assembly line balancing problems. These methods utilize new precedence-based local search functions with referenced permutation and two neighborhood structures to emphasize intensification while preserving high search speed, resulting in obtaining 23 new upper bounds compared to recently published algorithms.