Cost tolerance modelling and optimisation of machining tolerance design through intelligent techniques

Abstract

Tolerance is one of the most important parameters in design and manufacturing. Tolerance design has impact on machining cost, product quality and quality loss function. The purpose of tolerance design in product components is to produce a product with least machining cost possible, while meeting all functional requirements of the product. Usually, the cost tolerance model is constructed by a linear or nonlinear simple regression analysis based on the data of the cost?tolerance experiments and a correlation is derived between the two. Though, these correlation curves show the relationship between machining cost and tolerances, fitting error is inevitable. In this study, higher order polynomial regression model is developed based upon the empirical cost?tolerance data of typical production processes. The introduced tolerance design model provides more reliable result for tolerance design. In this work, the optimisation of tolerance allocation of over running clutch assembly is taken for analysis. The optimisation model is to minimise the sum of machining cost and quality loss under the constraint. This multiobjective, non linear, constrained optimisation model is solved with the Simple Genetic Algorithm (SGA) and Particle Swarm Optimisation (PSO). The results are compared with simple regression model using SGA, PSO and Stochastic Integer Programming (SIP) and the performances are analysed.