AGILE AND SUSTAINABLE SUPPLY CHAIN MANAGEMENT THROUGH AI-BASED PREDICTIVE ANALYTICS AND DIGITAL TWIN SIMULATION

Abstract

This study addresses the practical problem of how global supply chains can become both agile and sustainable while adopting advanced digital technologies such as AI based predictive analytics and digital twin simulation. The purpose is to examine whether and how these capabilities improve supply chain agility, sustainable supply chain performance and overall supply chain performance in real cloud enabled and enterprise level cases. A quantitative, cross sectional, case-based survey design was applied, using Likert 5-point scales with data from 214 professionals (82.3 percent usable response rate) drawn from supply chain intensive manufacturing, logistics and retail organizations. Key variables included AI based predictive analytics capability, digital twin simulation capability, supply chain agility, sustainable supply chain performance and overall supply chain performance. The analysis plan comprised descriptive statistics, reliability testing, correlation analysis, multiple regression and bootstrapped mediation tests. Results show that AI capability (β = 0.42, p < 0.001) and digital twin capability (β = 0.29, p < 0.001) together explain 46.3 percent of the variance in agility, and 41.7 percent of variance in sustainable performance, while the full performance model explains 58.9 percent of the variance in overall supply chain performance. Agility (β = 0.31) and sustainability (β = 0.28) partially mediate the effects of AI and digital twin capabilities, confirming that digital investments translate into superior performance largely through agile and sustainability-oriented practices. The findings imply that managers should treat AI analytics and digital twins as strategic, end to end capabilities for building agile, sustainable and competitively resilient digital supply chains.