The Role of Affect in Decision Making for Agile Processes

软件生产变得越来越复杂。到2017年底，全球IT支出预计将增长到近3.5万亿美元。软件开发项目现在通常可以与大型制造项目相比较，并且具有可比较的预算。

Software production has become increasingly more complex. Global spending on IT will foreseeably grow to nearly 3.5 trillion US dollars by the end of 2017. Software development projects are now often comparable to large manufacturing projects and have comparable budgets. In the last decade, agile processes have been developed to address the intangible nature of software development that creates software in a more iterative fashion, one piece at a time, and is guided by customer needs. Industry today claims to use agile processes, or parts thereof, for a sizeable part of their development efforts. Human beings are at the centre of this creative effort, and thus emotions and moods may affect important decisions such as requirement prioritization, architectural and design choices, and implementation strategies. Making a challenging decision can result in conflict among team members, with consequences that may hamper or even prevent a project's successful completion. Yet, the influence of these conflicts on software development has hardly been considered to date. We propose to build an affect model for software production and test whether or not emotions and mood affect software development in a systematic way, and by how much. We will do this by conducting rigorous studies in Industry. Selected industry partners include both large organizations (IBM and Siemens) and medium and small enterprises (EduCode, Serenova, and BlueSpurs). These studies will rely on both quantitative and qualitative measures to ensure we capture the contribution of affect in the most comprehensive way possible. Our results, whether positive or negative, will shine a new light on the human aspect of software development, one that has been traditionally very difficult to qualify. The expertise gained in these studies will directly involve more 77 graduate students and benefit more than 1,100 students in five years through directed courses and participation in graduate and undergraduate research. Furthermore, if the results warrant further involvement from industry partners, we will create a centre of excellence where research results can be transferred to the industry as technology (tools) and dedicated instruction (methods). Finally, our results will be published in conferences, journals, book chapters, and books; and the study data will be made available on the University's Open Access archive. This will be crucial to guide future researchers on new discovery paths for software development and have the potential to provide billions of dollars in benefits if adopted by even a very small percentage of global Industry.