Transforming Enterprise Sales Order Management Through AI-Driven Automation: A Framework for Supply Chain Resilience
Abstract
Aim: This study aimed to develop a framework for AI-enabled Resilient Sales Order Management (AI-RSOM), examine the potential of AI-enabled ESOM to enhance supply chain resilience, explain the mechanisms through which AI-enabled ESOM may influence resilience, and identify the conditions under which higher levels of automation may increase operational fragility. Methods: The study employed a critical integrative review methodology and conceptual theory synthesis. A total of 41 verified scholarly sources, including journal articles, books, and relevant documents published between January 2020 and June 2026, together with selected seminal earlier works, were reviewed. The sources were classified according to lifecycle phase, AI function, decision rights, resilience mechanisms, outcomes, boundary conditions, and implementation risks. Results: The synthesis identified five AI-RSOM capability stages namely integrate and sense, interpret, orchestrate, execute and learn, and govern. The findings suggest that these capabilities may influence supply chain resilience through improved order visibility, decision velocity, response flexibility, and organizational learning. The framework further outlines six theoretical propositions for future empirical testing, a process-level measurement architecture, and an evidence-gated implementation pathway. The findings are conceptual and theoretical therefore should not be interpreted as empirical estimates of AI effects on supply chain resilience. Conclusion: The study concludes that AI-enabled ESOM has the potential to strengthen supply chain resilience when AI capabilities are integrated across the sales-order lifecycle and supported by appropriate decision rights, governance mechanisms, and organizational learning processes. Recommendation: Evidence-gated implementation should be used to progressively expand automation only where performance and governance conditions are adequately demonstrated.