A Bayesian Network-Based Analysis of Changing Impact of Organizational Goals on Software Development

Changes in organizational goals in companies affect various metric values related to software development; however, few methods are available to evaluate the impact of such changes. Therefore, we propose a method to divide the collected metric values before and after a change in organizational goals, construct a Bayesian network before and after the change to analyze the causal relationship between the metric values, and compare and evaluate the results before and after the change. In this study, we evaluate changes from 2016 to 2020 in organizational goals and the Bayesian network for person-hours, number of bugs, etc. in projects related to the development of cloud services that support the real estate industry, using data provided by e-Seikatsu Co., Ltd.