Data-mechanism-driven Product Performance Optimization with Multiple Parameters Under Uncertainties in Manufacturing Automation Systems

High-end equipment with robust and excellent performance is of great significance to manufacturing automation systems. Product parameters are the foundation of performance. Usually, there is a lack of clear physical models to reveal the relationship between parameters and performance. Simulation is an important way to understand the design results of product performance parameters. However, simulations are often particularly time and resource-consuming. To address these issues, this work proposes a data-mechanism-driven product performance optimization method, which introduces Taguchi' method, nonparametric correlation testing, and multiple attribute decision making (MADM), to efficiently obtain the optimal parameter scheme. A novel MADM method is designed, which is combined with the Multi-Objective Optimization on the basis of a Ratio Analysis plus the full MULTIplicative form (MULTIMOORA) and the Variable Neighborhood Search. It has been proven that it performs better than the MULTIMOORA based on the well-known Simulated Annealing and Tabu search. Finally, taking the cutting unit of EBZ200i as a case study, we have successfully obtained the optimal parameter scheme that comprehensively performs better in terms of wear performance, robustness, and environmental economic performance.