Optimizing Engineering Change Management Through ECR Data Analysis and Mining Techniques

In the current field of industrial engineering, Engineering Change Management (ECM) has become a vital factor for preserving product quality and encouraging new ideas. This paper aims to understand how the management of the ECM processes is done at a global automobile firm particularly in the case of Engineering Change Requests (ECR). This study reveals that over 70 new ECRs are created daily, and in the past 10 years, over 120,000 ECRs have been recorded; it stresses the need for better Information Systems (IS) to tackle this problem. This research will employ the Knowledge Discovery in Databases (KDD) framework and text mining to analyze historical ECR data in a bid to make recommendations that should assist to enhance ECM effectiveness and decision making. The findings showed that text mining process consists of data extraction and data pre-processing, document segmentation and tokenization, removal of stopwords, stemming of words and word vector generation yielding 5,783 unique terms or keywords. Through cluster analysis, it was possible to identify common problem areas with some damage in the ECRs. We included keywords frequencies and clusters for Projects A, B and C; which aided in identifying specific project issues and trends with each project. For instance, in Project A, the term frequency and potential issues were pointed out; however, Projects B and C had differences in the term frequency distribution, which helped to pinpoint the common problems and direct further ECM enhancements.