Aims: The aim of this study was to evaluate which of the three mathematical models (Weibull, Geeraerd and Logistic) makes the best approximation for Hypericin-based photosensitization bacterial inactivation. Methods and Results: The inactivation of common food-borne pathogens Listeria monocytogenes and Bacillus cereus has been analysed. Three models, Weibull, Logistic and Geeraerd, have been used for different treatment conditions (10(-7) mol l(-1) and 10(-8) mol l(-1) Hypericin concentrations and 2 and 60 min incubation time). The Logistic method showed better correspondence (estimated R-2 = 0 >= 98, RMSE = 0 <= 41), than the Weibull (estimated R-2 = 0 >= 91, RMSE = 0 <= 90) and the Geeraerd model (R-2 = 0 >= 83, RMSE = 1 <= 27). After evaluation of best fitting model for the dependence of bacterial population reduction on irradiation time, the parameters describing treatment efficiency have been calculated. Conclusions: Our results indicate that the Logistic model described in the best way the inactivation of B. cereus and L. monocytogenes by Hypericin-based photosensitization. Significance and Impact of the Study: This study identified a best model for describing bacterial inactivation by hypericin-based photosensitization, which provides a way to evaluate the efficacy of different treatment conditions for this novel decontamination technique.