Investigation into GA and GSOA optimisation approaches for solving assembly sequence problems

Abstract

Assembly sequence planning (ASP) is a vital part in reduction of cost and lead time of a product that needs to be assembled. It involves a determination of assembly process often coupled with constraints that need also to be addressed. In order to resolve ASP optimisation problems, it was reported that genetic algorithms (GA) were used for gaining an optimal solution for sequence-dependent or non-sequence-dependent job scheduling of product assembly in order to maximise production volume and minimise production delay. A latest development through a literature review indicates that glowworm swarm optimisation algorithm (GSOA) can also be used effectively and efficiently for solving system engineering optimisation problems in terms of such as non-linear equation scheduling. This thesis presents an investigation of using the GA and the GSOA approaches, respectively to seek an optimal solution from possible assembly sequences of a car engine pump valve and a ball pen as a case studies. The research work was conducted based on a comparative result of minimal assembly time by searching an optimal assembly sequence using these two algorithms, which were implemented in a JAVA program. The research outcomes show that the GSOA outperforms the GA in generating an optimal assembly sequence with a minimal assembly time. It also demonstrates that the GSOA can be a useful decision-making tool for searching an optimal or near-optimal assembly sequence of a product for product designers

Date of Award	Jul 2020
Original language	English
Awarding Institution	University of Portsmouth
Supervisor	Qian Wang (Supervisor), David Sanders (Supervisor) & Nicholas George Bennett (Supervisor)