METHOD FOR SELECTING A SOFTWARE DEVELOPMENT METHODOLOGY TAKING INTO ACCOUNT PROJECT CHARACTERISTICS

Abstract

Context. With the development of the software industry, the number of applied methodologies and hybrid approaches based on them constantly increases, that is why, the choice of the most suitable/optimal methodology for the project is an urgent problem of software engineering, since the selection process is poorly formalized, requires sufficient experience of the person who will make this decision and depends on many related factors. Objective. The support of decision-making in the process of choosing a methodology for the software project development and increase of the level of adequacy of the above choice. Method. Based on the previously developed algorithm by the author, a generalized method for selecting the best software development methodology is proposed, which consists of 14 steps and takes into account the characteristics of the project, based on the multi-criteria analysis approaches, taking into consideration the opinions of experts for a more reasonable choice of the most suitable methodology for this project. The method uses the aggregated expert evaluation. It was decided to use the AHP to calculate the criteria weights. Based on the established values of criteria, their weights and expert evaluation, the score is calculated for each methodology using the weighted sum and TOPSIS methods. Results. The application of the developed method to the data of actual projects showed a match in 83% of cases (in five out of six cases, the application of the method resulted in the selection of methodology that corresponded to the one actually used in an existing project). In cases when the methodology chosen by the proposed method differed from the one that was actually used, the application of the proposed method recommends going to the stage of the initial determination of criteria and their weights, which will allow making a more adequate choice of methodology. Conclusions. The proposed method can be applied in practice by software project managers to support the decision-making process, and will allow reducing time spent on project management.