Factors that influence software project cost and schedule estimation

摘要

Software Project Management is a core topic in software engineering courses because it teaches how software projects planned, implemented, controlled, monitored, and evaluated. The development of theories in software metrics and prediction models builds on the broader project management field but also attempt to overcome the difficulties inherent in measuring an intangible object like software. This paper is situated within research into the factors that influence cost and time estimation for software projects that continue to challenge software development organizations. The study described in this paper explored technical and non-technical factors seen by Sudanese software practitioners as critical in estimation, and if not managed, can result in cost and time overrun or in some cases lead to project failure. Using a mixed-method approach, the research project was first informed through a qualitative study that explored the kinds of problems that face the estimation process from the perspectives of different staff levels. This part of the study revealed a number of factors that can be broadly categorized as technical factors, e.g. the skills of those involved in the estimation process, and non-technical factors such as the high level of uncertainty in the local business environment. The second part of the study focused on one of the leading factors, software project staff training and experience, using the survey method to examine how well the software engineering curriculum is aligned with skills required in the software market, especially those related to estimation. The recommendations this study produced on reducing estimation errors, whether geared towards companies or academia, are preliminary and may only reflect the local setting. However, they also drew upon the vast literature on cost estimation techniques and case studies in similar and more advanced settings. The problem of software effort prediction and estimation models has been a thorny issue in the software engineering field since the concept of “software crisis” and the field itself, as a response to the crisis, emerged in the late 1960s. It still seems to some that “After forty years of currency the phrase `software engineering' still denotes no more than a vague and largely unfulfilled aspiration” [2]. This study develops our understanding of problems facing one of the young professions in the country, as well as contributes to the global body of research on developing techniques to manage the intricacy of software engineering compared to more established engineering disciplines.