Developing reliable software estimates continues to elude many organizations. Recent studies suggest that over one‐third of
software projects suffer from schedule and/or cost over‐runs severe enough to warrant the projects as failures. One reason for
such poor industry performance is that progress in this area has been relatively slow compared to the tremendous advances in
technology and software methods over the last three decades. Traditional industry techniques for estimating software projects
(e.g. COCOMO II, SLIM) remain highly complex, require non‐intuitive inputs, and are geared toward large waterfall projects. The
advent of Agile and other progressive‐elaboration software methods has seen a trend towards empirical estimation techniques
that focus on relatively short iterations, but fail in their attempt to accurately estimate entire software projects. In the face of
these difficulties, many organizations continue to rely on naive “seat‐of‐the‐pants” estimation approaches. In response to these
issues, Quadrus has developed an easily accessible and accurate estimation methodology that successfully models real‐world
software projects through the use of simulation and Monte Carlo methods ‐ the next generation in software project estimation.