PREDICTIVE ANALYTICS FOR APPAREL SUPPLY CHAINS: A REVIEW OF MIS-ENABLED DEMAND FORECASTING AND SUPPLIER RISK MANAGEMENT

Abstract

The global apparel industry operates within a highly volatile and competitive environment marked by rapidly shifting consumer preferences, abbreviated product life cycles, and increasingly fragmented global supply chains. In response to these complexities, apparel companies are progressively adopting advanced predictive analytics techniques integrated with Management Information Systems (MIS) to enhance supply chain visibility, responsiveness, and decision-making accuracy. This systematic review explores the current state and strategic applications of MIS-enabled predictive analytics, with a focused examination of two pivotal domains: demand forecasting and supplier risk management. Drawing from a wide spectrum of peer-reviewed literature and empirical studies, the paper synthesizes the evolution of data-driven forecasting models, particularly those powered by machine learning and artificial intelligence, to illustrate how predictive analytics contributes to anticipating customer demand with higher precision and aligning production accordingly. Moreover, it examines the growing utilization of predictive tools in identifying, assessing, and mitigating supplier-related risks through real-time monitoring, risk scoring, and scenario analysis frameworks. The review underscores the critical role of integrated MIS platforms in consolidating internal and external data, supporting the operationalization of predictive insights, and fostering agile, data-informed supply chain strategies. It further identifies persistent challenges hindering the optimal deployment of predictive analytics, including issues related to data quality, system interoperability, lack of standardized protocols, organizational resistance to technological adoption, and ethical concerns surrounding data privacy and algorithmic bias. The review concludes by highlighting significant gaps in existing research, particularly the underrepresentation of empirical studies in small and medium-sized apparel enterprises, limited cross-functional integration frameworks, and insufficient attention to regulatory and ethical implications in global predictive ecosystems. Accordingly, the paper proposes directions for future studies, advocating for the development of sector-specific, ethically grounded, and contextually adaptive predictive frameworks that align with the digital transformation trajectory of apparel supply chains.