The Impact of AI-Driven Prescriptive Analytics on Supply Chain Disruption Management and Operational Performance

Abstract

This study has examined how artificial-intelligence-driven prescriptive analytics affects the management of supply chain disruptions and, through that mechanism, the operational performance of the firms that deploy it, and it has developed and tested a framework linking analytics capability to disruption outcomes across sourcing, planning, production, and distribution. The central problem has been that supply chain disruption management remains largely reactive, judgement-based, and escalation-driven, so that firms detect disruptions only once their operational consequences have become visible, respond through improvised expediting and reallocation decisions taken under time pressure with incomplete information, and absorb the resulting cost through premium freight, excess inventory, and lost service, while the analytic infrastructure now routinely present in these firms remains configured to report what has happened rather than to prescribe what should be done. Guided by five research questions, concerning how prescriptive analytics affects disruption management performance, which disruption categories are most effectively managed, which analytics maturity tiers are most effective across decision criteria, what risks reactive judgement-based regimes carry, and whether disruption management effectiveness transmits analytics capability into operational performance, the study has proposed a framework comprising five contributing constructs: data foundation and real-time visibility; predictive disruption sensing and risk forecasting; prescriptive decision automation and optimisation; human–AI decision governance and explainability; and supply chain reconfigurability and response capability, with disruption management effectiveness as the proximal outcome and operational performance as the distal outcome. A quantitative, cross-sectional design supplemented by a simulation-based evaluation has been used, and data have been collected from supply chain directors, sales and operations planning managers, procurement and sourcing managers, logistics and distribution managers, and analytics leads across manufacturing, retail, pharmaceutical, electronics, and logistics organisations. Out of 341 distributed questionnaires, 277 valid responses have been retained, producing an 81.2% valid response rate. The analysis has included descriptive statistics, reliability and validity testing, correlation analysis, hierarchical multiple regression, bootstrapped mediation analysis, and a simulation evaluation of detection lead time, response optimality, and recovery trajectory across disruption categories. The findings have shown that all five constructs were significantly and positively associated with disruption management effectiveness, that the model was significant, F(5, 271) = 75.42, p < .001, explaining 58.2% of the variance, and that prescriptive decision automation and optimisation and data foundation and real-time visibility were the strongest predictors. Disruption management effectiveness in turn predicted operational performance and partially mediated the relationship between prescriptive analytics and performance, carrying 59.1% of the total effect.