A Strategy Based on Multiple Decision Criteria to Support Technical Debt Management

Abstract

Technical debt (TD) refers to likely long-term costs associated with software development shortcuts taken by programmers to achieve short-term business benefits. If a development team does not manage TD, it can cause significant long-term problems such as high maintenance costs. Management strategies monitor debt items and evaluate when and if they should be paid. In order to effectively support this task, the systematic use of decision criteria can be decisive. In this context, this paper presents a strategy for TD management that uses multiple decision criteria to decide when to pay debt items off. In addition, it presents a case study that assessed the feasibility of the proposed strategy regarding its usefulness, ease of use and self-predicted future use. The results provided positive evidence on the use of the proposed strategy, indicating (i) that it can be useful in supporting TD management activities and (ii) that it can bring gains in terms of productivity, performance, and efficacy when evaluating the desirable time to pay debt items off.