Sustainable supply chain design is essential for reducing the environmental and economic impacts of products. Moreover, it is becoming increasingly important to consider not only sustainability but also resilience in supply chain design. This requires proactively identifying potential risks, such as production interruptions, and integrating them into planning, while also accounting for future developments like evolving demand and changes in the energy mix. This article presents a planning approach for resilient and sustainable supply chain design that combines machine learning, quantitative sustainability assessment, and multi-criteria optimization. Machine learning is used to forecast future product demands, while environmental and economic impacts of different locations for raw material extraction, processing, component manufacturing, and final production are assessed using life cycle assessment and life cycle costing. The resulting data, along with characteristics such as production capacities, are used as input for a multi-criteria optimization model. This model minimizes the economic and environmental impacts, as well as the value at risk as a resilience measure, in the context of sustainable and resilient supply chain design. The applicability of the developed approach is demonstrated using a case study on the sustainable and resilient design of a generic supply chain.
Ginster et al. (Thu,) studied this question.