A Methodology for Organizational Data Science Towards Evidence-based Process Improvement

Organizational data science projects provide organizations with evidence-based business intelligence to improve their business processes (BPs). They require methodological guidance and tool support to deal with the complexity of the socio-technical system that supports the organization's daily operations. This system is usually composed of distributed infrastructures integrating heterogeneous technologies enacting BPs and connecting devices, people, and data. Obtaining knowledge from this context is challenging since it requires a unified view capturing all the pieces of data consistently for applying both process mining and data mining techniques to get a complete understanding of the BPs execution. We have presented the PRICED framework in previous works, which defines a general strategy for performing data science projects. In this paper, we propose a methodology with phases, disciplines, activities, roles, and artifacts, providing guidance and support to navigate from getting the execution data, through its integration and quality assessment, to mining and analyzing it to find improvement opportunities.