Machine learning and artificial intelligence methods and applications for post-crisis supply chain resiliency and recovery

Abstract

Resilient and adaptive strategies for recovery have been underscored by supply chain disruptions induced by natural disasters, pandemics, and wars. Supply chain resilience protects enterprises, communities, and humanitarian activities during pandemics and wars. This study investigates the utilization of artificial intelligence and machine learning methodologies to enhance supply chain resilience and recovery in the aftermath of these crises. Leveraging data-driven methodologies, these technologies provide opportunities to improve the overall resilience of the supply chain, optimize resource allocation, and enhance decision-making. Proposed newer measures to protect economies, national security, lives, and a more resilient future are discussed in this study. Machine learning and artificial intelligence can process vast amounts of data quickly to provide real-time insights into the state of the supply chain, including damage assessments, demand fluctuations, and disruptions to transportation routes. Machine learning and artificial intelligence in supply chain management have reduced demand forecasting errors by 10–20 % and enhanced disruption reaction times by 20–30 %. The delivery reliability was also enhanced by 10–20 % as the artificial intelligence can forecast the delays and recommend alternate routes. Machine learning and artificial intelligence provide insights, automation, and agility to rebuild and enhance supply chains after challenging circumstances. This work is unique in showing how to improve supply chain resilience at critical moments by combining technologies and adopting hybrid methodologies.