Integration of Machine Learning Tools with Lean Six Sigma Activities in Forensic Engineering: A Literature Review

Purpose – This paper addresses challenges in forensic engineering, particularly in the detection and analysis of material failures. It explores the integration of Machine Learning (ML) techniques, specifically Convolutional Neural Networks (CNN), within the Lean Six Sigma (LSS) DMAIC methodology to streamline and automate the process of material failure detection. It aims to contribute towards leaner, more efficient, and cost-effective methods for forensic engineering investigations.
Methodology – The methodology employed in this study entailed a thorough literature search,
involving the selection of relevant articles based on predefined criteria, their subsequent
categorization into distinct sections, a careful analysis of key insights, and the identification of significant research gaps. Utilizing the VOSviewer software, bibliometric analysis was conducted to examine the collected data. This paper offers a comprehensive review of the existing literature, conducts a rigorous analysis, outlines research gaps, and proposes future avenues for exploration in this interdisciplinary domain.
Findings – The paper identified various difficulties within forensic engineering, including challenges such as expert time loss and complex processes, alongside emphasizing ML’s promise in detecting material failures and proving for its combination with LSS to enhance operational efficiency.
Practical implications– Forensic engineers can leverage the identified synergy to enhance efficiency in their practices, while lean experts may optimize processes by integrating ML technology for automation and process improvement.
Originality/value – The originality lies in the interdisciplinary fusion of forensic engineering, ML,and LSS, offering fresh insights to improve material failure detection processes.

Forensic engineering, Lean Six Sigma, Convolutional Neural Networks, image recognition.