A fuzzy random conditional value-at-risk-based linear programming (FCVLP) model was proposed in this study for dealing with municipal solid waste (MSW) management problems under uncertainty. FCVLP improves upon the existing fuzzy linear programming and fuzzy random conditional value-at-risk methods by allowing analysis of the risks of violating constraints that contain fuzzy parameters. A long-term MSW management problem was used to illustrate the applicability of FCVLP. The optimal feasibility solutions under various significance risk levels could be generated in order to analysis the trade-offs among the system cost, the feasibility degree of capacity constraints, and the risk level of waste-disposal-demand constraints. The results demonstrated that (1) a lower system cost may lead to a lower feasibility of waste-facility-capacity constraint and a higher risk of waste-disposal-demand constraint; (2) effects on system cost from vague information in incinerator capacity inputs would be greater than those in landfill capacity inputs; (3) the total allowable waste allocation would vary significantly because of the variations of risk levels and feasibility degrees. The proposed FCVLP method could be used to identify optimal waste allocation scenarios associated with a variety of complexities in MSW management systems.
Computing near-optimal Value-at-Risk portfolios using integer programming techniques