Abstract-Cost estimation is a vital task in most important software project decisions such as resource allocation and bidding. Analogy-based cost estimation is particularly transparent, as it relies on historical information from similar past projects, whereby similarities are determined by comparing the projects' key attributes and features. However, one crucial aspect of the analogy-based method is not yet fully accounted for: the different impact or weighting of a project's various features. Current approaches either try to find the dominant features or require experts to weight the features. Neither of these yields optimal estimation performance. Therefore, we propose to allocate separate weights to each project feature and to find the optimal weights by extensive search. We test this approach on several real-world data sets and measure the improvements with commonly used quality metrics. We find that this method 1) increases estimation accuracy and reliability, 2) reduces the model's volatility and, thus, is likely to increase its acceptance in practice, and 3) indicates upper limits for analogy-based estimation quality as measured by standard metrics.

Tool support in practice is an important success factor for a sustained metrics program. A number of commercial metric tools claim to support the collection and analysis of software metrics.However, these tools vary widely in their ability to seamlessly integrate with the many additional software components and tools required in a typical software development process. While the metric tools often provide powerful data analysis features, data input and output is usually restricted to a few formats and often requires manual user interaction. This paper points out key aspects of tool automation for a metric program-like standard interfaces and automation support-and evaluates several prominent commercial software metric tools with regard to these criteria.Main results are: tight integration of metric tools in heterogeneous development environments is still not a standard practice/feature and can thus be a substantial cost burden in a metric program due to restriced interfacing and automation capabilities.

No abstract

Software cost estimation is a crucial task in software project portfolio decisions like start scheduling, resource allocation, or bidding. A variety of estimation methods have been proposed to support estimators. Especially the analogy-based approach-based on a project's similarities with past projects-has been reported as both efficient and relatively transparent. However, its performance was typically measured automatically and the effect of human estimators' sanity checks was neglected. Thus, this paper proposes the visualization of high-dimensional software project portfolio data using multidimensional scaling (MDS). We (i) propose data preparation steps for an MDS visualization of software portfolio data, (ii) visualize several real-world industry project portfolio data sets and quantify the achieved approximation quality to assess the feasibility, and (iii) outline the expected benefits referring to the visualized portfolios' properties. This approach offers several promising benefits by enhancing portfolio data understanding and by providing intuitive means for estimators to assess an estimate's plausibility.

No abstract