Many mathematical models of freeze-drying have been applied to pharmaceutical freeze-drying, where a product solution is set in a glass vial and sublimed from the solution top by receiving conductive heat from the vial bottom. Sublimation of the frozen solution in the vial progresses rather one-dimensionally from the top to bottom. A simple model thus assumes one-dimensional progress of sublimation. When simulating the curvature of the sublimation interface, well sophisticated computational method is required. Finite element analysis is a successful method, however it could make a difficulty in applying moving boundary to the sublimation interface (Ferguson et al., 1993 and Mascarenhas et al., 1997). When products such as fruits or vegetables placed directly on a shelf are dried, sublimation progresses multi-dimensionally, depending on geometries of the products. Therefore, a modeling approach to calculate the velocity of the sublimation interface based on classical mass and heat transfer equations, as is described above, could be significantly complicated. An empirical and statistical approach using the response surface method would be a solution to obtain parameters to realize optimized operation (Hammami and René, 1997). This is a method to predict optimal solution from the fitting surfaces obtained from trial experiments. This approach does not give the status of products during drying, but gives a predicted property of the resultant dried product. Therefore, the database used for the prediction must be made from the system that we wish to optimize.