The COVID-19 pandemic has highlighted the critical need for supply chain resilience in the face of unforeseen disruptions. This research investigates the application of machine learning (ML) algorithms to enhance supply chain resilience during the COVID-19 crisis. The authors evaluated several ML algorithms, including decision trees, random forests, naïve bayes, and LSTM. They explored using the SPIN COVID-19 RMRIO dataset to develop a proactive and data-driven approach to mitigate disruptions and improve supply chain performance. The ML model worked with and without feature selection. With chi-square feature selection, the long short-term memory (LSTM) performed well and achieved the highest accuracy, 96.74%, with an F1 score of 91.01%. Without feature selection, random forest outperformed, which provided an accuracy of 96.21% with an F1 score of 81.25%.
Top1. Introduction
In the wake of the unprecedented global upheaval caused by the Corona Virus Disease 2019 (COVID-19) pandemic, supply chains (Pandey, D. et al., 2021) worldwide have encountered unparalleled disruptions across multiple dimensions (Sheng et al., 2020). These disruptions have reverberated through every facet of supply chain management, encompassing individual transportation, raw material procurement, as well as the distribution and manufacturing of products. The emergence of this novel reality has imposed extraordinary challenges on supply chain management professionals, compelling them to seek innovative solutions to navigate these intricate problems. What sets the COVID-19 pandemic apart as a distinct instance of supply chain disruption is its unique global impact on diverse sectors, distinguishing it from other disruptions traditionally studied in operations and supply chain management (Zhang et al., 2020). The COVID-19 (Pandey, D. et al., 2021) pandemic has exposed the fragility of contemporary supply chains and underscored the critical importance of supply chain resilience. In response to these challenges, this research paper explores the integration of ML (Pandey, B. K., & Pandey, D., 2023) algorithms into supply chain management strategies. By harnessing historical data and predictive analytics, we aim to develop a more adaptable and resilient supply chain system (Malhotra, P. et al., 2021) capable of effectively mitigating unforeseen disruptions. While the existing body of literature offers valuable insights into forecasting, managing, and responding to various disruptions (Craighead et al., 2020), the COVID-19 pandemic has magnified the necessity for efficient risk management strategies.
Consequently, the examination and evaluation of Supply Chain Resilience (SCR) within the context of the COVID-19 (Parthiban, K. et al., 2021) pandemic, coupled with developing strategies to enhance this resilience to withstand external disruptions, have emerged as a critical and imperative field necessitating further research and analysis. SCR represents a pivotal capability that empowers supply chains to respond to interruptions and promptly restore their original functionality (Mena et al., 2020). As a strategic imperative, resilience equips supply chains with the ability to anticipate, respond, recover, and extract invaluable insights from unforeseen events, thereby reducing the inherent risks associated with supply chain operations.
The significance of resilience has reached unprecedented levels during the COVID-19 pandemic, underscored by the substantial research volume produced since its onset. In light of this, it is advisable to compile a comprehensive synthesis of the extensive investigations conducted, building upon prior research in this field to provide a unified perspective and lay a robust foundation for subsequent, more impactful research initiatives. To address the challenges posed by this scenario, artificial intelligence (AI) and ML have emerged as potent tools. As articulated by (Baryannis et al., 2019), AI plays a pivotal role in facilitating environmental scanning processes and bolstering the resilience of supply chains on a broader scale. This perspective is depicted in Figure 1, which offers a two-dimensional view of the symbiotic relationship between AI, ML, and supply chain resilience. In the following sections, we delve into the methodologies, findings, and implications of our research, offering a comprehensive exploration of the integration of ML into supply chain management strategies to enhance resilience in the COVID-19 pandemic's disruptions (Lelisho, M. E. et al., 2023).