Task allocation optimization model in mechanical product development based on Bayesian network and ant colony algorithm

Abstract

A task allocation optimization model in mechanical product development was proposed. First, Bayesian network was introduced and used to describe quality characteristic (QC) relations of mechanical products. The quality characteristic importance was then analyzed based on the Bayesian network. Weights were assigned to QCs according to the QCs importance. QCs are generally manufactured by different types of manufacturing resources, such as machine tools, cutting tools, measuring tools, and fixtures. For convenience, these manufacturing resources were encapsulated as manufacturing cells. QCs were considered as tasks to be completed by manufacturing cells. Numerous tasks and manufacturing cells were involved in mechanical product development; therefore, the core problem was to allocate these tasks to manufacturing cells in an optimal way to achieve better product quality. Time and cost were considered as the main constraints. The ant colony algorithm was used to solve the task allocation optimization problem. A numerical case study was also presented to indicate the effectiveness of the proposed model.