Performance prediction in early stages of software systems: Artificial neural network model

Performance is an important non functional aspect to be considered for any software system. Software Performance Engineering (SPE) is an approach to predict the performance of a software system early in the life cycle. In this paper we present a neural network model for the performance prediction of Multi-Agent system at the early stages of development. We used Feed forward back propagation neural network model for the performance prediction. The results are validated and a case study of Multi-Agent System is presented.