The main goal of the Software Effort Estimation Special Interest Group is to support researchers, practitioners and educators in software cost and effort estimation through identification of relevant and useful knowledge (papers, books, journals, conferences, tools, etc). The long-term goal is to summarize all the knowledge on software cost and effort estimation as principles (guidelines, advices, etc).
Software development effort estimation is a branch of forecasting that has received increased interested in academia, application domains and media. Efficient development of software requires accurate estimates. Unfortunately, software development effort estimates are notorious for being too optimistic, and there seems to have been no substantial improvement in estimation accuracy over the years. Inaccurate software estimates cause trouble in business processes related to software development such as project feasibility analyses, bidding, budgeting and planning.
It is unrealistic to expect very accurate effort estimates of software development effort because of the inherent uncertainty in software development projects, and the complex and dynamic interaction of factors that impact software development effort use. Still, it is likely that estimates can be improved because software development effort estimates are systematically overoptimistic and very inconsistent. Even small improvements will be valuable because of the large scale of software development.
Software researchers have addressed the problems of effort estimation for software development projects since at least the 1960s. Research is found in areas such as:
- Creation and evaluation of estimation methods. Describes work on the creation and evaluation of estimation methods, such as methods based on expert judgment, structured group processes, regression-based models, simulations and neural networks.
- Calibration of estimation models. Tailoring a model to a particular context (calibration) has been found to be difficult in practice. Problematic issues are related to, among others, when, how and how much local calibration of the models that are beneficial.
- Software system size measures. The main input to estimation models is the size of the software to be developed. It has been proposed many size measures, for example based on the amount of functionality that is described in the requirement specification.
- Uncertainty assessments. Software developers are typically over-confident in the accuracy of their effort estimates. Realistic uncertainty assessments are important in order to enable proper software project budgets and plans.
- Measurement and analysis of estimation error. Proper accuracy measurement is essential when evaluating estimation methods, and identifying causes of estimation error.
- Organizational issues related to estimation. Organizational issues such as processes to control the cost and scope of the project, may have a large impact on estimation accuracy.
What you would like to see on this website? What should be changed? Please send us suggestions and corrections!