Planning quality and maintenance activities in a closed-loop-serial-multi-stage manufacturing system under different deterioration scenarios

Abstract

This thesis investigates joint part quality inspection and condition-based maintenance strategies for a closed-loop-serial-multi-Stage manufacturing system under three different deterioration scenarios. Each scenario comprises a serial-forward manufacturing system and a serial-reverse re-manufacturing system; both share a production line of five working machines that produce one product type throughout three labor shifts. Constant wear, tear, fatigue, and aging processes lead closed-loop-serial-multi-stage manufacturing to experience three diverse deterioration scenarios. In the first scenario, a serial manufacturing production line suffers three different deterioration levels in all machines based on their operational complexity (The more complex the operation, the higher the deterioration factor). Rework and repair operations are considered for items classified as non-conforming; however, after failing the rework and repair operations, they will scrap. In the second scenario, reverse re-manufacturing production is fed by collected items at the end of their useful life (EOL) from the same family type as those produced in the forward manufacturing system. The three deterioration levels (stable, incremental, and random) are presented in each working machine and quality inspection error types I and II. Due to the high risk of product re-manufacturing, machine deterioration and inspection errors, repair and rework operations are not allowed to reduce the chances of a high fraction of non-conforming items shipping. The third scenario presents a more intricate situation where each machine of a closed-loop - serial-multi-stage manufacturing system experiences the three deterioration levels simultaneously in both forward and reverse echelons and quality inspection errors type I and II. However, to reduce the chances of non-conforming items shipping and increase the manufacturing systems surplus, scrap items, as well as EOL collected items, are re-utilized as raw materials in the re-manufacturing process. The study's primary aim is to determine the optimal number and location for allocating part quality inspection and condition-based maintenance activities in each scenario to minimize the total cost associated with inspection, production, maintenance, scrap and penalties for nonconforming item production. The proposed tools rely on a Mixed-Integer-Linear Programming (MILP) model that incorporates production, quality and maintenance constraints, facilitating simulating the closed-loop-serial-multi-stage manufacturing system's behavior. To demonstrate the effectiveness of our model, we conducted a series of different numerical experiments to find the optimal solution based on the quality of the proposed results.

Date	4 Aug 2025
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Jean-Pierre Kenné (Supervisor)