Analyzing Trends and Patterns in Supply Chain Case Studies Using LDA Topic Modelling

Abstract

This research project applies Latent Dirichlet Allocation (LDA) topic modelling to a corpus of supply chain case study articles with the main objective of uncovering themes in the literature to improve understanding of dominant trends, challenges, and innovations in supply chain management. By using LDA, we systematically analyse various case studies collected from the Scopus database, identifying key topics and their evolution over time. Unlike previous research that typically relies on article abstracts, this project extracts full article content using advanced text extraction tools, allowing for a deeper exploration of themes. Our methodology includes comprehensive text data pre-processing methods widely used in natural language processing (NLP) today. The findings reveal critical insights into areas such as logistics optimization, risk management, sustainability practices, and technological advancement. This research demonstrates that LDA is an effective tool for analysing large volumes of textual data, providing insights into complex supply chain issues. The approach not only aids researchers in studying supply chain management but also equips practitioners with valuable information for strategic decision-making. This research emphasizes the potential of topic modelling to significantly contribute to academic discourse and practical applications in the dynamic field of supply chain management.