Title: Remanufacturing of end-of-life laptop based on remaining useful life prediction and quality grading with random forest and cluster analysis

Authors: Gurunathan Anandh; Shanmugam Prasanna Venkatesan; Sandanam Domnic; Santosh Awaje

Addresses: Department of Production Engineering, National Institute of Technology, Tiruchirappalli, Tamil Nadu, India ' Department of Production Engineering, National Institute of Technology, Tiruchirappalli, Tamil Nadu, India ' Department of Computer Applications, National Institute of Technology, Tiruchirappalli, Tamil Nadu, India ' Department of Production Engineering, National Institute of Technology, Tiruchirappalli, Tamil Nadu, India

Abstract: A laptop remanufacturer typically performs recovery, disassembly, functional testing, grading and repair/replacement of parts. The remaining useful life (RUL) of the EOL laptop parts is evaluated and quality graded with the usage statistics to decide the repair/replacement options. Research on RUL prediction and quality grading of EOL laptop parts deserves research attention. This research aims to develop a decision support tool (DST) in Microsoft Excel interfaced with Python for RUL prediction and quality grading of laptop hard disk drive (HDD) and lithium-ion battery (LiB). Random forest (RF) is used for RUL prediction, and the K-means clustering algorithm is applied for quality grading using sample datasets obtained from the online dataset repositories. Typically, a laptop remanufacturer is unfamiliar with machine learning (ML) algorithms; thus, developing a simple user interface is vital. The RF and clustering analysis results suggest that the predicted and experimental values are highly correlated.

Keywords: laptop remanufacturing; remaining useful life; RUL; quality grading; machine learning; ML; hard disk drive; HDD; lithium-ion battery; LiB.

DOI: 10.1504/IJPMB.2024.138350

International Journal of Process Management and Benchmarking, 2024 Vol.17 No.2, pp.137 - 152

Received: 13 Mar 2023
Accepted: 14 Mar 2023
Published online: 01 May 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article