A novel feasible task sequence-oriented discrete particle swarm algorithm for simple assembly line balancing problem of type 1

作者：Dou Jianping^*; Li Jun; Su Chun

来源：International Journal of Advanced Manufacturing Technology, 2013, 69(9-12): 2445-2457.

DOI：10.1007/s00170-013-5216-2

• Assembly line balancing
• Particle swarm algorithm
• Feasible task sequence
• Fragment crossover
• Comparative study

To solve the simple assembly line balancing problems of type 1 (SALBP-1), almost all of particle swarm algorithms (PSAs) for SALBP-1 adopt task sequence-oriented solution representation and are limited to the priority-based indirect encoding of feasible task sequence (FTS) so far. In this paper, firstly a novel FTS-oriented particle swarm algorithm (FTSOPSA) that directly records a FTS by a particle, named direct discrete PSA (DDPSA), is proposed to solve SALBP-1. In the DDPSA, a new multi-fragment crossover-based updating mechanism is developed, and the fragment mutation is incorporated into the DDPSA to improve exploration ability. Secondly, a systematic comparison of DDPSA and two existing FTSOPSAs as well as two existing genetic algorithms (GAs) has been presented against a set of instances selected from the literature and 15 randomly generated instances of SALBP-1. Comparisons between the FTSOPSAs and existing GAs show promising higher performance of the proposed DDPSA for SALBP-1, and also show that the direct encoding of FTS seems superior to the priority-based indirect encoding of FTS for solving SALBP-1.

• 出版日期2013-12
• 单位东南大学