Identification and analysis of the elements required to manage technical debt by means of a systematic mapping study

Abstract

Technical debt, a metaphor for the long-term consequences of weak software development, must be managed to keep it under control. The main goal of this article is to identify and analyze the elements required to manage technical debt.The research method used to identify the elements is a systematic mapping, including a synthesis step to synthesize the elements definitions. Our perspective differs from previous literature reviews because it focused on the elements required to manage technical debt and not on the phenomenon of technical debt or the activities used in performing technical debt management. Additionally, the rigor and relevance for industry of the current techniques used to manage technical debt are studied. The elements were classified into three groups (basic decision-making factors, cost estimation techniques, practices and techniques for decision-making) and mapped according three stakeholders' points of view (engineering, engineering management, and business-organizational management).The definitions, classification, and analysis of the elements provide a framework that can be deployed to help in the development of models that are adapted to the specific stakeholders' interests to assist the decision-making required in

Tracy: A Business-Driven Technical Debt Prioritization Framework

“…Several secondary [3], [4], [10], [11] and tertiary [12] studies analyze technical debt research. With regard to technical debt prioritization, it is a common finding that the criteria, tools, and approaches used to prioritize technical debt lack a business perspective.…”

An Open Tool for Assisting in Technical Debt Management

“…The available tools are focused on implementing metrics, specific techniques for TDM, or both of them. Techniques and tools for TD have been analyzed in literature reviews previously [4], [1], [3]. The goal of this tools is not to be a quality tool that implements metrics to keep the code quality.…”

“…To monitor TD, it is necessary to analyze the change of TD indicators over the software evolution. For TDM, it is also needed to take into accout several requirements, including different points of view [2], [3]. This implies to obtain data from several sources and to apply several techniques.…”

“…This variability of techniques is extended to the approaches for TDM [5]. Therefore, it is compelling to have means to combine metrics, techniques, and management approaches implemented by different tools, considering also that current research in TD [3] is resulting in new metrics and techniques, that should have to be considered when they become available. Also, there are some tools that provide different TD indexes [6].…”

“…Existing tools are mainly focused on obtaining metrics, implementing some specific techniques for TDM, or providing their own models for TDM [4] and, as described in [3], each individual tool does not support all the required elements for TDM and, consequently, it is necessary to integrate them. As discussed in Section III, current tools are often standalone and not devised to integrate other tools.…”