Cost estimation in software development projects with Monte Carlo simulation

The realization of cost estimates on projects is a process of fundamental importance to ensure the economic viability of these endeavors. Specifically in software development projects where typically the focus of management is on project execution processes that add value and generate customer satisfaction, the estimation of costs should be done quickly, but with accuracy. In this type of environment, an approach that can be used is the parametric estimate, which bases its calculations on historical data to establish the project\´s costs. In order perform this approach, a method that can be used with great advantage is the Monte Carlo simulation (MCS). The objective of this study is to understand, through a case study in a software development company, as this method can be performed in this type of project pointing their advantages and disadvantages. Among the main results showed that MCS provides, in a rapid way, a baseline for managing project costs which supports the decisions made. Moreover, associating with a graph of "S" curve with simulated values, it generated visual control ("traffic light") which facilitates the cost control during the project