Quantification and modeling of disinfectant inactivation of Listeria monocytogenes biofilms on stainless steel surfaces

Abstract

The present study evaluates and models the efficacy of three different disinfectants-sodium hypochlorite (SH), peracetic acid (PAA), and chlorine dioxide (ClO2) solution in inactivating biofilm formation of Listeria monocytogenes on stainless steel (SS) surfaces. The nonlinear survival curves of L. monocytogenes biofilm cells treated with different disinfectants were fitted using the Weibull model and four supervised machine learning models (Artificial Neural Networks (ANN), Random Forest (RF), Support Vector Regression (SVR), and Gradient Boosting Decision Tree (GBDT)). RF model exhibited superior predictive ability compared to other methods (RMSE: 0.12, MAE: 0.10, MAPE: 2.38, and R2: 0.99). Both RF and ANN showed better predictive effectiveness than the Weibull model. This research quantifies and models the efficacy of three disinfectants on L. monocytogenes biofilm on SS surfaces, providing valuable insights into the disinfection process of equipment surfaces in food processing.