Tolerance allocation: Balancing quality, cost, and waste through production rate optimization

Abstract Dimensional tolerance allocation is a very important and difficult task that traditionally seeks to balance cost/productivity and quality. Common tolerance allocation models have two shortcomings: i) they are overly reliant on models focused on minimizing cost and tend to ignore waste, and ii) they fail to connect to the root cause of many quality issues: process variation. This paper proposes a tolerance allocation model that addresses these shortcomings. The proposed model considers both product design (tolerance selection) and operation planning (or production rate selection). Relations among production rate, production cost, processing precision, and waste are considered. A gradient-based optimization method is proposed to minimize the cost and waste. A clutch assembly case study is analyzed to evaluate the method. Monte Carlo simulations are employed to validate the accuracy of the proposed cost model. The proposed method is compared with a heuristic method from the literature. The proposed method produced more satisfactory products at a lower cost while producing less waste. For the case study, it is found that when the precision of a process is high, it is not necessary from an economic standpoint to inspect the quality of individual components. For poor precision processes, inspecting the quality of individual components is the preferred approach from a cost/throughput standpoint.