Analytical models (math or computer simulation models) are typically built on the basis of many assumptions and simplifications and hence model prediction could be inaccurate in intended applications. Model validation is thus critical to quantify and improve the degree of accuracy of these models. So far, little work considers model validation for various design configurations so that model prediction is accurate in the intended design space. Furthermore, there is a lack of effective approaches that can be used to quantify model accuracy considering different number of experimental data. To overcome these limitations, objective of this paper is to develop a model validation approach for various design configurations with a reference metric for model accuracy check considering different number of experimental data. Three technical components are proposed to accomplish this goal including: 1) a validation metric using the Bhattacharya distance (B-distance) for model accuracy check with different number of experimental data; 2) the Maximum Entropy Principle (MEP) method for accurate characterization of the model bias; and 3) an approach for constructing response surface of the model bias and accurately predicting system performances in the intended design space. Two examples including a modified vehicle side crash problem and a thermal challenging problem are used to demonstrate the effectiveness of the proposed approach.