Hybrid Cost-Tolerance Allocation and Production Strategy Selection for Complex Mechanisms: Simulation and Surrogate Built-In Optimization Models

Abstract

In manufacturing companies, assembly is an essential process to obtain the final product. The life cycle of an assembly product depends on various production strategies, e.g., resource allocation, rework decision, selection strategy, etc. In this regard, achieving a reliable assembly product commence with engineering a comprehensive design plan which can mitigate various uncertainties a company can face. The counteraction of uncertainties can be altered by introducing a set of tolerances into the design of the components. Tolerances define a practical margin on components design without downgrading the required performance of products. Thus, producers are confronted with high-quality requirements, cost pressure, and a rising number of demands. On these bases, this paper aims at modeling a statistical framework for a set of production strategies, including resource allocation (as a decision to assign practical resources to components) and reworking decision (as a decision to improve components’ conformity rate). Moreover, a generic simulation and surrogate approach are established to evaluate the performance of the assembled product. Within this approach, simulation and surrogate models can be used to investigate a variety of deviations over components’ geometries within the process deviation domain and deploy reworking decision. Ultimately, a modular costing system is developed, and a genetic algorithm is adapted to locate optimal solutions. In addition, the applicability of the statistical model is studied on an assembly product.