Concurrent design for nominal and tolerance analysis and allocation of mechanical assemblies using DE and NSGA-II

Abstract

Tolerance analysis of mechanical assemblies promotes concurrent engineering by bringing engineering requirement, manufacturing cost and quality together in a common model. To improve the tolerance design, the designer could first try to decrease the sensitive components by moving the nominal values to a less sensitivity portion. Second, the cost-based optimal machining tolerances are allocated. Differential Evolution (DE) and Non-dominated Sorting Genetic Algorithm (NSGA-II) are used for minimising the critical dimensions deviation and manufacturing cost with optimal machining tolerances. A numerical example (Stacked blocks assembly) shows the superior nature of DE and NSGA-II algorithms.