A simulation-based fuzzy multi-criteria decision analysis (SFMCDA) method is developed for supporting the selection of remediation strategies for petroleum contaminated sites. SFMCDA integrates process modeling (using BIOPLUME III) and fuzzy ranking (based on fuzzy TOPSIS) into a general management framework, and can compare various remediation alternatives, in light of both cost-risk tradeoffs and uncertainty impacts. The proposed method is applied to a hypothetical contaminated site suffering from a benzene leakage problem. Six remediation alternatives are taken into consideration, including natural attenuation (NA), pump-and-treat (PAT), enhanced natural attenuation (ENA), and a number of their combinations. Six fuzzy criteria, including both cost and risk information, are used to compare different alternatives through fuzzy TOPSIS. The results demonstrates that the proposed method can help systematically analyze fuzzy inputs from contaminant transport modeling, cost implications and stakeholders’ preferences, and provide useful ranking information covering a variety of decision-relevant remediation options for decision makers.