An IIM Indore framework combines machine learning, generative AI and optimisation to identify logistics hubs and routes better able to withstand conflict, sanctions and cyberattacks.
Researchers at the Indian Institute of Management Indore have built an artificial intelligence-based framework designed to help semiconductor firms withstand wars, cyberattacks and wider geopolitical shocks that can break fragile global supply chains. The study, led by Professor Tanmoy Kundu and published in IEEE Transactions on Engineering Management, combines machine learning, generative AI and mathematical optimisation to identify more resilient logistics hubs and transport routes across Asia.
The timing reflects a broader industry concern. Semiconductor production is concentrated in a small number of countries and facilities, leaving the sector exposed to conflict, sanctions, cyber threats and trade restrictions. Deloitte has warned that geopolitical tensions and tighter trade barriers are already making chip supply chains more brittle, while recent academic work has argued that geopolitical risk must be taken into account when valuing and allocating AI and semiconductor infrastructure. The IIM Indore study adds to that debate by offering a practical decision framework rather than a purely descriptive risk assessment.
Its model works in three stages. First, machine-learning tools score conflict-related risk using historical data together with economic, geographic and cyber indicators. In that analysis, population size, proximity to power plants and closeness to international borders emerged as the strongest predictors of conflict risk, while cybersecurity weaknesses, income levels and dependence on oil exports also shaped the results. Second, generative AI is used to simulate possible war, sanctions and cyberwarfare scenarios, estimating how likely they are to occur, how severe they might be and how long they may last. Finally, an optimisation layer translates those risk estimates into route and hub choices that seek to preserve supply coverage while limiting disruption.
The framework was tested on semiconductor manufacturing sites and ports across Asia, with particular attention to Taiwan, South Korea and Japan, alongside other important production and assembly centres such as Malaysia, Vietnam, Thailand and the Philippines. The study suggests that a port or transport corridor that looks efficient on a map may not be the safest choice once geopolitical exposure, cyber risk and network effects are considered together. It also finds that prolonged disruptions do more damage than short, sharp shocks, reinforcing the need for contingency planning that anticipates drawn-out interruptions rather than only temporary delays.
For companies, the approach could support sourcing, route selection, risk monitoring and emergency planning. For policymakers, it may help identify where infrastructure is most exposed and where resilience spending would have the greatest value. The authors note that the model does not yet capture every variable, including natural hazards, shifting trade rules, financial costs and the difficulty of coordinating multiple stakeholders. Even so, the research points to a broader shift in semiconductor logistics: resilience is increasingly becoming a data problem as much as a physical one.
Disclaimer: This content is intended for informational purposes only. Readers are advised to exercise their own judgement, conduct due diligence, or consult a qualified expert before acting on any information provided.
