Artificial Intelligence in Production Planning and Supply Chain Resilience: A Structured Review of Decisions, Enablers, and Risks

Although artificial intelligence (AI) is frequently applied to production planning and supply-chain resilience, the evidential meaning of the published results varies considerably. This structured, non-exhaustive review investigates where AI aids forecasting, scheduling, inventory, supplier risk, disruption sensing, and response or recovery. It also identifies the conditions required for dependable implementation and the risks that constrain its use. Twenty-eight sources were classified by decision area, design, inspection level, evidential role, and outcome type. The specific outcome types analyzed were predictive accuracy, benchmark performance, simulated operational outcome, organizational perception, reported case outcome, observed operational outcome, and financial outcome. Technical studies tend to offer evidence restricted to specific datasets and computational settings, particularly regarding scheduling and supplier-delay prediction. In contrast, proof that model outputs actually improve governed decisions, execution, recovery, or sustained operating performance remains substantially thinner. Across the reviewed literature, several mutually reinforcing conditions repeatedly emerge as necessary for practical implementation: data integration, domain and technical skills, process ownership, human oversight, monitoring, cybersecurity, privacy, cost, and scalability. To synthesize these findings, the review outlines an evidence-grounded operational lifecycle that connects data and context, prediction or diagnosis, human-governed choice, executed intervention or adaptation, and measured outcomes. The paper concludes with decision-specific recommendations, maintaining a strict distinction between a model's demonstrated technical performance and any claimed organizational benefits.

کلیدواژه ها:

Artificial Intelligence ، Production Planning and Control ، Supply-Chain Resilience ، Human Oversight ، AI Governance