The research goal of this grant is to provide a scientific foundation for the economic implications of software modularization activities during software development and maintenance. A difficult question during software development is when to perform software code refactoring to restructure the code to achieve better modularity for the purpose of improving reducing complexity, improving code readability, and achieving other, so-called nonfunctional properties. The costs of modularization activities such as refactoring is significant and immediate, and the benefits are largely invisible, intangible, and long-term. However, delaying refactoring causes modularity decay, which is blamed for significantly reduced productivity, operational failures and even canceled projects. Knowing when to refactor is a long-standing, important and difficult question. This research will conduct extensive empirical studies to establish quantitative relationships between software project parameters and software modularity decisions. Based on the empirical data, the project will develop modularity violation detection techniques to locate modularity debts and evaluate a proposed modularization activity by combining Monte Carlo simulation based on Datar-Mathews option valuation method. These scientific foundations for economic decision-making will lead to a better understanding of the costs and benefits of modularization activities, with a goal of providing tools that will better support such decisions. The research will be performed by an interdisciplinary team of researchers with expertise in software engineering and economics.
