A knowledge management-driven and DevOps-based method for situational method engineering

Razieh Dehghani,Raman Ramsin
DOI: https://doi.org/10.1007/s10799-023-00386-y
2023-01-19
Information Technology and Management
Abstract:Earlier software development processes (SDPs), such as waterfall processes, were mainly focused on process steps and did not address people- and product-related issues. Emergence of Software development methodologies (SDM) has created a new paradigm for developing software systems. A SDM is a special kind of technically engineered framework for organizing SDPs; this framework is expected to specify three main interwoven elements, namely people, products, and process. It has since become evident that it is impossible to provide a general-purpose SDM for developing all the various kinds of software systems, and it has thus become essential to construct the most appropriate methodology for the system development situation in hand, a practice commonly called Situational Method Engineering (SME). The problem with existing SME methods is lack of adequate attention to the role of people who might seek or possess valuable knowledge about the project situation. This knowledge can be tacit information that is hidden in the developer's mind, or it might be explicitly available. This paper proposes a knowledge management (KM)-driven and DevOps-based SME method as a new integrated multi-view methodological paradigm that satisfies the need for sharing human experience in engineering SDMs. The method has been proposed by reusing general SME practices and complementing them by embedding appropriate KM and DevOps practices to alleviate the weaknesses of previous SME methods. Furthermore, the proposed method has been evaluated through four case studies and also by conducting a criteria-based comparison with eight prominent SME methods.
What problem does this paper attempt to address?