A method to estimate software strategic indicators in software development: An industrial application

Abstract

Context: Exploiting software development related data from software-development intensive organizations to support tactical and strategic decision making is a challenge. Combining data-driven approaches with expert knowledge has been highlighted as a sensible approach for leading software-development intensive organizations to rightful decision-making improvements. However, most of the existing proposals lack of important aspects that hinders their industrial uptake such as: customization guidelines to fit the proposals to other contexts and/or automatic or semi-automatic data collection support for putting them forward in a real organization. As a result, existing proposals are rarely used in the industrial context.

Objective: Support software-development intensive organizations with guidance and tools for exploiting software development related data and expert knowledge to improve their decision making.

Method: We have developed a novel method called SESSI (Specification and Estimation of Software Strategic Indicators) that was articulated from industrial experiences with Nokia, Bittium, Softeam and iTTi in the context of Q-Rapids European project following a design science approach. As part of the industrial summative evaluation, we performed the first case study focused on the application of the method.

Results: We detail the phases and steps of the SESSI method and illustrate its application in the development of ModelioNG, a software product of Modeliosoft development firm.

Conclusion: The application of the SESSI method in the context of ModelioNG case study has provided us with useful feedback to improve the method and has evidenced that applying the method was feasible in this context.