Optimum Tolerance Design Using Constraint Networks and Relative Sensitivity Ratio Algorithm

Tolerance design plays a significant role in the relationship between the performance and the manufacture of a product. It is important to maintain valid tolerances when product design is constrained by the relationship between the dimensions of entities of a component and the functional requirement of the design. Increasing the tolerance may decrease the manufacturing cost, but will also worsen the performance. In this paper, an efficient algorithm is proposed for computerized optimal allocation of tolerance among the components of a complex assembly with a large number of constraints and entities. Basic concepts of hierarchical interval constraint networks have been used in combination with an iterative relative sensitivity analysis procedure for modeling and solving the tolerance allocation problem. The proposed algorithm can handle practically any number of constraints and entities with the required level of accuracy. It can also accept any type of cost-tolerance relationship for modeling. Examples have been discussed and the results of tolerance design obtained using the proposed method are presented.