Residual Life Prediction of Multistage Manufacturing Processes With Interaction Between Tool Wear and Product Quality Degradation

Multistage manufacturing processes (MMPs) usually exhibit an interactive relationship between tool wear and product quality degradation. On one hand, the tool wear in a stage may result in the quality degradation of the products fabricated on that stage. On the other hand, the quality degradation at a preceding stage may lead to the change of the operational condition and thus affect the tool wear in subsequent stages. This interaction needs to be considered to accurately predict the residual life distribution (RLD) of MMPs, which will benefit condition-based maintenance and tool inventory management. In this paper, we propose an interaction model that utilizes a linear model to represent the impact of tool wear on quality degradation and a stochastic differential equation model to capture the impact of quality degradation on the instantaneous rate of tool wear. We then propose a Bayesian framework that incorporates real-time quality measurements to online update the RLD of MMPs. Our methodology is a generalization of an existing "QR-chain model," which is dedicated into a similar research and application area. We conduct numerical studies to test the performance of our methodology and compare with the QR-chain model. The results show that our methodology outperforms the QR-chain model through capturing the impact of quality degradation on the process of tool wear and incorporating real-time quality measurements. Note to Practitioners-MMPs usually exhibit an interactive relationship between tool wear and product quality degradation, which significantly influences the RLD of MMPs. This paper presents a new methodology that utilizes a stochastic model to capture this complex relationship for better prediction of the RLD. In addition, we propose to utilize a Bayesian updating approach that relies on real-time quality measurements to predict the up-to-date RLD of MMPs. Specifically, our methodology is designed to be implemented in a general MMP, in which the tools used to fabricate products are subject to an accelerated wear due to product quality degradation from previous stages. Common examples of such MMPs include multistage machining processes and multistage stamping processes. This method can be potentially applied in the factory floor and improve the planning of condition-based maintenance and tool inventory management by providing more accurate and up-to-date residual life prediction. Our methodology is based on the following assumptions: 1) tool wear impacts quality degradation in a linear fashion; 2) quality degradation adjusts the instantaneous rate of tool wear based on the natural rate of tool wear; and 3) the natural rate of tool wear exhibits variability among identical tools. Successful implementation of this methodology requires users to focus on the following aspects: (a) Identify the sources of process noise, which may affect product quality degradation, and estimate their variance (b) Understand the physical structure of the MMP, including the layout of the manufacturing system, the locations of quality measurements, etc., in order to estimate the impact of tool wear and process noise on quality degradation. (c) Acquire historical observations of quality degradation through sensor-based quality measuring system, which are commonly used in the industries. (d) Acquire historical observations of tool wear through real-time tool-wear monitoring system that identifies tool condition without interrupting the manufacturing process. Such system has also been widely discussed in the literature. (e) Leverage historical observations of tool wear and quality degradation to estimate model parameters that cannot be determined by engineering knowledge. (f) Incorporate real-time quality measurements captured by online quality monitoring systems to update the RLD of MMPs.