The use of approximation functions (also termed surrogate models or metamodels) in lieu of a direct link between optimizer and simulation software is becoming a commonplace in applications of design optimization to industrial problems. Similarly, an FE simulation of the process in conjunction with Monte-

Carlo simulation in the stochastic analysis would involve a prohibitive computing effort. Therefore it is necessary to establish an accurate approximation model that captures essential features of the response produced by the FE simulation but does not require excessive amount of computations so that the Monte-

Carlo simulation of the stochastic process can be performed on the obtained approximation model (Figure 1). In order to build the surrogate model, the response of the system is to be evaluated by running the simulation model at a series of sets of parameters defining a design of experiments (DoE) in the range of variation of these parameters.

## The Author

**Mr. Ashis Acharjee, Dr. Prasun Chakraborti**

NIT, Agartala

Tripura, India

<< Back