Predictive Modeling of Microbial Growth in Stored Fresh-Cut Green Beans

Abstract

Microbial growth is a critical factor influencing the quality, safety, and shelf life of fresh-cut vegetables, including green beans. This study aimed to model the microbial growth dynamics of fresh-cut green beans during storage using two nonlinear predictive models: the Modified Gompertz model and the Logistic model. Microbial counts (log CFU/g) were measured over a 21-hour storage period with triplicate observations. Model parameters were estimated using non-linear least squares (NLS) in R, and model performance was evaluated based on residual standard error (RSE), parameter significance, convergence behavior, biological plausibility of parameters, and visual agreement between observed data and fitted curves.

The Logistic model produced stable parameter estimates with significant growth-related coefficients and an RSE of 0.6186, whereas the Modified Gompertz model exhibited unstable parameter estimation for μ and λ despite a comparable RSE (0.617). Visual inspection also showed the Logistic model provided a smoother and biologically consistent fit to the observed microbial growth pattern. These results indicate that the Logistic model is more suitable for describing microbial dynamics in fresh-cut green beans under the tested conditions.

Overall, this study demonstrates the importance of selecting an appropriate predictive model to accurately characterize microbial behavior in fresh-cut produce. The established model can be applied in shelf-life prediction, risk assessment, and optimization of storage conditions for minimally processed vegetables.