This paper (https://lnkd.in/eQz63ZBz) presents a beautiful framework that integrates AI, ML, and Operations Research. After going through this, you can see how you and pick elements of this framework to create your own for different supply chain and operations-related business processes.

The proposed model enables firms to evaluate and select resilience strategies, like multi-sourcing, local production, or long-term contracts, based on financial returns per dollar spent under uncertain conditions such as tariffs, supplier shutdowns, or shipping delays.

AI has been leveraged in four different ways in this proposed solution, and as you go through each, you can see how AI brings more “realism” to the analysis.

1. Predictive modeling of disruptions

AI algorithms are used to simulate uncertain conditions, like supplier shutdowns, tariff hikes, or shipping delays, and to estimate how those events propagate through the supply chain. Machine learning models analyze historical and synthetic data to learn the probability distributions of such disruptions.
This enables scenario generation that goes far beyond static “what-if” analysis.

2. Optimization through operations research + AI

The framework combines operations research optimization (e.g., linear programming, stochastic modeling) with AI’s ability to approximate complex nonlinear relationships. It determines optimal order scheduling, production planning, and inventory levels for different resilience strategies. The AI layer dynamically adjusts parameters based on observed disruptions and outcomes, yielding adaptive decision rules rather than fixed plans.

3. Translating resilience into measurable financial outcomes

Traditional resilience assessments are qualitative (“more flexible,” “more redundant”). Here, AI quantifies:

a. Service-level improvement per dollar spent,
b. Expected cost reduction,
c. Recovery time after disruption.

This lets decision-makers rank strategies, like multi-sourcing vs. consignment inventory, by financial efficiency.

4. Sensitivity and policy analysis

AI also identifies critical vulnerabilities, suppliers, nodes, or transport routes where investment yields the largest resilience gain. That’s valuable for:

a. Corporate risk management (where to allocate resilience budgets), and
b. Public policy (where subsidies or regulations could reduce systemic risk).