Simultaneous control of the production, maintenance, and inspection strategies for a failure-prone manufacturing system with quality-based financial penalties/incentives

Abstract

This thesis investigates the simultaneous production planning, maintenance, replacement, and inspection control strategies for a system in three different scenarios. In the general scenario, the system consists of an unreliable manufacturing machine capable of producing one type of part to satisfy the demand. In the second scenario, to increase the efficiency of the system, a rectifying machine is added to the system to recover the rectifiable parts of defective products. In the third and the most complete scenario, a remanufacturing machine helps the system to recover the non-rectifiable parts along with the products returned from the market. Also, a recovery stock room is added to the system to store the defective parts for remanufacturing processes. The aging process and imperfect repair activities lead to the deterioration of the manufacturing machine. This deterioration affects the reliability of the manufacturing machine and the quality of produced parts. To ensure the continuous fulfillment of the demand and to palliate the undesirable effects of deterioration, several maintenance activities are applied. The maintenance activities consist of preventive maintenance and repair/replacement switching options. A quality control step is also carried out to conduct the inspection process in which the detected defective parts are sent to be recovered either by the rectifying machine or the remanufacturing machine. The main objective of this study is to jointly determine the cost-optimized production policy for the manufacturing and remanufacturing machines, maintenance planning in terms of PM and repair/replacement switching activities, and inspection strategy over a finite planning horizon. The proposed model is subject to a quality constraint on the average outgoing quality limit (AOQL) which is specified by the market, where a financial penalty or incentive is considered for providing products with a quality lower or higher than AOQL, respectively. Furthermore, numerical methods are adopted to find the optimal control solution. Finally, numerical examples are provided as well as the sensitivity analysis to verify the accuracy of the proposed model.

Date	21 Dec 2021
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Jean-Pierre Kenné (Supervisor)