Systematic Mapping Study of Dealing with Error in Software Development Effort Estimation

Over the last decades, the software engineering community has investigated new techniques for software development effort estimation. Unfortunately, the estimates were not always accurate. Error approaches are then, an interesting track for improving the projects running performances and their financial profitability. The aim of this systematic mapping study is to summarize and synthesize the existing studies dealing with effort estimation error and uncertainty and to classify them based on research approaches, contribution types, accuracy criteria, datasets, error approaches and effort estimation techniques used. In total 19 papers published between 1990 and 2015 were selected. We observed a balance between the managerial approaches and the technical ones. Furthermore, the proposed error techniques and frameworks improve in general the accuracy of effort estimation techniques. Fuzzy logic, bootstrapping and risk analysis are promising avenues that could be combined with various estimation techniques.