Early Indicators of Project Abandonment in Industry- Academia Collaborations: Developing an Assessment Framework for Industrial Data Science Projects

This paper presents a novel assessment framework aimed at identifying early indicators of potential project abandonment in industry-academia collaborations (IAC) focusing on industrial data science projects. As these collaborations become more integral to leveraging academic research for industrial applications, understanding the early stages of project development is crucial for minimizing resource wastage and ensuring project continuation. Our research involved analyzing 13 industrial data science case studies, with a particular focus on those that were discontinued in their initial phases. By synthesizing findings from these case studies with literature on success factors of IAC, we propose an assessment framework that includes a set of success factors tailored to avoid project abandonment. This assessment framework is grounded in a methodological approach that integrates established project management success factors and direct industry experience, ensuring practical relevance and applicability. The developed assessment framework consists of 14 success factors, each described with specific industry-relevant aspects. This paper not only contributes to academic literature by offering a structured method to anticipate early project abandonment but also serves as a practical guide for researchers in acquiring and executing data science projects within the industry-academia context.