The purpose of this work was to determine the effect of previously designed test cell thickness and environmental heating temperature on the accuracy of estimated D- and z-values in low-moisture foods. The D- and z-values of Clostridium sporogenes were determined isothermally by 1-step and 2-step regression methods. The more accurate D- and z-values were obtained by 1-step regression compared with 2-step regression. The non-isothermal and 1-step estimation method was applied to determine the microbial inactivation parameters of Clostridium sporogenes. The D- and z-values estimated by that method had the same level of accuracy as those from the isothermal and 1-step method. The novel non-isothermal and 1-step estimation method of D- and z-values had the advantage of reduced experimental effort. Using literature values of D, z and thermophysical properties, and the exact solution for transient heat conductions, the optimum design of test cell for determination of D and z-vales of Clostridium sporogenes and Salmonella in low-moisture food was presented.