A stochastic bilevel model to manage symbiotic flows in an Industrial Symbiosis network under demand uncertainty

Industrial symbiosis (IS) networks offer a valuable means of reducing environmental impacts by optimizing by-products utilization and minimizing resource consumption. However, managing these networks remains challenging due to uncertainties in supply and demand, as well as the hierarchical decision-making structures within eco-industrial parks (EIPs). This study presents a novel stochastic bilevel programming model to coordinate by-product flows in an IS network. The model balances the EIP authority's objective of minimizing raw material use with companies' objectives of maximizing profits under uncertain demand. Unlike single-objective models, our bilevel approach uncovers intricate relationships between conflicting objectives. Applied to a theoretical IS network in the forest products industry, the model shows that demand fluctuations disrupt material flows and increase raw material consumption, worsening the EIP authority's environmental objectives, however, paradoxically, boost company profits by encouraging higher production. This interesting trade-off between the hierarchical objectives along with the model's ability to maintain feasible operations under varying demand conditions demonstrates the network's resilience.