An official website of the United States government.
<OL> <LI> To formulate a set of non-linear and mechanistic stochastic models based on Markov processes for disinfection and sterilization of food and foodstuff. <LI>To derive the master, i.e., governing, equations, for each of the models formulated. <LI> To solve analytically the governing equations, whenever possible. <LI> To solve numerically the governing equations. <LI>To simulate the models by the Monte Carlo method. <LI> To validate the results of solution and simulation of the models in light of the available experimental data. <LI>To develop a user-friendly computer software to execute the solution and simulation of the models formulated.
NON-TECHNICAL SUMMARY: Currently, food safety and security are of utmost societal concern; naturally, the thorough disinfection or sterilization of food and foodstuff is of keen public interest. In fact, the complete or near-complete disinfection or sterilization of pathogenic populations in food and foodstuff is frequently required to meet the regulatory constraints imposed for their consumption by humans and animals. Hitherto, no sufficiently robust and efficient means is available to quantity the fluctuating number concentration of bacteria during disinfection or sterilization of food and foodstuff; hence, it is highly desirable or even essential that a means to quantify the pathogen populations in disinfection or sterilization be made available, especially in a user-friendly and computer-aided format. This proposed research is concerned with establishing a comprehensive approach for quantitatively analyzing, modeling, and simulating the dynamic behavior of populations of pathogenic cells or bacteria being disinfected or sterilized from food and foodstuff by a variety of methods including chemical, thermal, mechanical and radiative methods or any combination thereof. Such analysis, modeling, and simulation will be carried out by resorting to a stochastic paradigm.
<P>
APPROACH: The methods and approaches to be deployed in this proposed research include: i. Stochastic modeling based on Markov processes with mechanistically derived intensity functions. ii. Derivation of the governing, i.e., master, equations of the resultant Markov models. iii. Analytical solution of the governing equations of linear Markov models by resorting to the conventional techniques of differential equation or integral calculus, and rational approximate solution of those of the non-linear Markov models. iv. Numerical solution of the governing equations by means of finite difference and/or finite element methods. v. Direct simulation of the Markov model by the Monte Carlo method. vi. Validation of the models in light of the available experimental data by means of the method of least-squares and the method of the maximum likelihood estimation. vii. Development of a user-friendly computer software specified with the Unified Modeling Language (UML) and implemented with an object oriented programming language such as C++ or Java.