Monte carlo simulation based estimations: Case from a global outsourcing company

Effort estimation of various software project life cycle phases have been traditionally performed using deterministic methods. Deterministic methods fail to account for the variability inherent in the software engineering processes deployed in executing the project. We have developed stochastic notations and models that can represent these process variations. We have also constructed a Monte Carlo simulation tool that works using the stochastic model we have developed for software development projects. This tool has been tested in a mid-sized software services outsourcing (SSO) organization. The tool and results are described and reported in this paper.