摘要:Phytoremediation is a cost-effective, efficient and environment-friendly biological method to reduce petroleumbased contamination in soil. However, limited access to soil samples from various depths duringphytoremediation along with the cost, time and effort required for quantitative measurement of total petroleumhydrocarbons (TPH) necessitates the development of a mathematical model to overcome the existingobstacles. Since fuzzy logic is an appropriate method for meddling systems with inadequate or vague andnon-specific information, the present study sought to determine TPH concentrations during phytoremediationthrough such a technique. Based on previous research and the conditions of areas adjacent to Isfahan OilRefinery (Isfahan, Iran), sorghum and barley were planted in 130 cm long polyvinyl chloride pipes containingcontaminated soil samples from the mentioned area. After 17 weeks, TPH concentrations were measured at25, 50, 75, and 100 cm depths of soil. The percent reduction in TPH concentrations was 23%-35% higherin the presence of sorghum and barley than in unplanted treatments. Fuzzy inference with two inputs (timeand depth) and 10 membership functions was used to quantify TPH concentrations (the output) at differentdepths of planted and unplanted soils during the phytoremediation process. Since the calculated and measuredvalues were consistent, the developed model can be applied in future phytoremediation studies in othercontaminated areas.