Data driven decision-making methodology : a study of repair decision in remanufacturing

• Main Author: Passaporn, Kanchanasri
• Other Authors: Moon Seung Ki
• Format: Theses and Dissertations
• Language: English
• Published: 2019
• Subjects: Engineering::Industrial engineering::Operations research
• Online Access: https://hdl.handle.net/10356/90309; http://hdl.handle.net/10220/49932
• Institution: Nanyang Technological University

Decisions on recovering a low volume product such as a High Pressure Compressor Stage 3 Stator Vane (HPC Stg 3) could make a significant difference in cost saving. By a traditional assessment method considering only physical conditions of End-Of-Life parts, the cost of repairing the end-of-life parts is costly when comparing to replacement with spare parts due to dynamic resource constraints, human and machine limitations. This research is focused on investigating and developing an approach to solving the difficulty of repair decisions under various constraints. The main objective of this work is to provide quantitative decision rules for justification of the repairing of the returned parts. The proposed method consists of a data driven decision-making model that, given a set of parameters, determines the optimal damage point. The application of the system to possible scenarios is demonstrated through numerical studies. By comparing execution time, the proposed model solving by the linear programming provides a better optimal solution than the evolutionary approach. Taking more execution time, the evolution approach can also provide the optimal solution as good as by the linear programming. The numerical study demonstrates how to apply the proposed decision-making model to assess the acceptability of End-Of-Life parts and enable identification of the optimal solution. The optimal solution shows the significant cost saving when comparing repair and replacement options. In this research, there are three major accomplishments which are (1) an investigation of non-trivial decision-making factors, (2) a simulation model of End-Of-Life part repair decision constructed by using ARENA® to examine the concept of part sentencing in remanufacturing and (3) the development of a decision-making model to minimize repair and inventory costs by integrating the concept of data driven decision-making and optimization techniques i.e. the linear programming and the evolutionary approach. This work addresses the repair decision of returned parts. The research concept can be deployed to various industrial applications which include (1) multinational original equipment manufacturers in areas such as aerospace, automotive and marine regarding restoration of components to serviceable condition and remanufacturing of products, (2) small and medium enterprises about improvements of remanufacturing operational process and (3) optimization of remanufacturing for implementation in various industries.