Designing a bi-level waste disposal network by integrating environmental protection and sustainable development

Effective management of construction waste (CW) is not only essential for the rational utilization of resources, but also plays a crucial role in environmental protection and sustainable development. This study proposes a novel bi-level distributionally robust optimization (DRO) model with probabilistic guarantees to address the uncertain construction and treatment costs in the waste disposal network design. Methodologically, we construct an ambiguity set with a sub-Gaussian structure to describe the uncertain parameters. Leveraging the problem’s structural properties and the robust counterpart approximation (RCA) method, our proposed model can be transformed into a computationally tractable mixed integer linear programming (MILP) model. A tailored Benders decomposition (BD) algorithm with two acceleration strategies is designed to solve the resulting MILP model. Our proposed method is validated through a real CW disposal case in Hong Kong. The computational results demonstrate that (i) our model can effectively mitigate the impact of uncertain construction and treatment costs, while incurring a robustness price of approximately 10.75%; (ii) under extremely negative cost fluctuations, our DRO model achieves a 2.29%–9.61% cost advantage over the nominal model; (iii) the accelerated BD algorithm reduces the solution time by 26%–35% compared to the standard BD approach. Besides, this study offers valuable managerial insights for decision-makers in CW management, supporting the development of the optimal waste management strategies that promote sustainable growth in the construction industry.