Data Governance for Data-Driven and Intelligent Transformation: Overcoming Data Challenges in Operations and Supply Chain Management

In today’s rapidly evolving business environment, the advancements in technology have pushed companies toward data-driven decision-making. In the Operations and Supply Chain (O&SC) context, where complex networks, real-time decisions, and stakeholder coordination are critical, data plays a pivotal role. However, many organizations struggle to fully capitalize on their data due to persistent issues such as poor data quality, fragmented systems, and ineffective data sharing. These challenges have become a major bottleneck in adopting advanced technologies.

This PhD research is motivated by the urgent need to understand and systematically address these data-related challenges through robust, contextaware data governance (DG) strategies and practices. DG provides the structural foundation, e.g., policies and standards, roles and responsiblities, and technical engagement, needed to ensure data is accurate, secure, and valuable. The objective of this research is to investigate the current landscape of data usage and governance in Operations and Supply Chain Management (OSCM) field, uncover key data challenges, and offer actionable insights and guidance for designing effective DG solutions.

Through a mixed-methods approach involving a quantitative survey, modeling, literature research, and qualitative case studies, this research addresses five core areas: (1) the critical role of data in business and operations during digital technology adoption; (2) the state of DG research in OSCM; (3) the specific data issues organizations face in data-driven transformation; (4) the systematic approaches to resolving these data issues; and (5) how DG strategies and practices can be designed and implemented to unlock data value and drive intelligent, future-ready operations. This PhD research consists of four interconnected studies.

Study 1 aims to understand how important data is to digital technology adoption (DTA) It investgates the impact of data collection strategy and data quality on companies’ operational performance and digital technologies’ implementation performance. Using survey data analyzed through Partial Least Squares Structure Equation Modeling (PLS-SEM), the study finds that: Both data collection strategy and data quality significantly influence operational performance and implementation performance, meanwhile, data collection strategy plays a crucial role in shaping data quality.

After proving the importance of data, Study 2 conducts a systematic literature review to synthesize existing research on DG in the OSCM field. This study maps current research and identifies key challenges and gaps in DG literature. It also proposes future research directions that align DG strategies with the dynamic needs of production, manufacturing, and supply chain environments.

After conducting the first two studies, it became evident that despite the growing trend of adopting advanced technologies such as AI, their actual implementation remains less mature than commonly perceived. A clear bottleneck hindering progress is the persistent data issues. How to address these data issues stands as a significant concern confronting both industry and academia. Thus, two further studies were conducted, focusing on identifying and mitigating data issues from both academic and industrial perspectives.

Study 3, from an academic perspective, reviews data issues and solutions in industrial AI projects. It identifies 82 data issues across seven stages of the data lifecycle, and extends the focus area beyond historical data to real-time sensor data and expert domain knowledge. The study introduces a modelaware data preparation approach that alignsdata characteristics with AI model requirements, thereby enhancing both data usability and algorithm performance. Additionally, it develops a conceptual framework that integrates managerial and technical perspectives to systematically address data challenges. This study highlights a shift from Model-Centric AI (MCAI) to Data-Centric AI (DCAI), underscoring the central role of high-quality, well-governed data in enabling sustainable AI adoption.

Finally, Study 4, from the parctical perspective, analyzes three case studies to investigate real-world data challenges and DG practices in the O&SC context, highlighting three key areas for successful DG implementation: 1) policy and standard development, 2) role-based governance, and 3) technical enablement. Building on these insights, the study proposes a business centric roadmap to guide the design and implementation of DG solutions that support digital and intelligent transformation.

By synthesizing these findings, this PhD research proposes a three-pronged DG pathway tailored to the OSCM context, integrating top-down oversight, bottom-up role clarity, and technical support mechanisms to enable effective and actionable DG strategies. It also outlines how DG researchers can stay relevant in the face of rapid technological change, by prioritizing domainspecific research and adopting more adaptive, interdisciplinary approaches.

All together, this PhD research contributessignificantly to both the theoretical advancement and practical application of DG within OSCM.

From a theoretical standpoint, this research makes several important contributions: 1) It strengthens the academic foundation of DG by extending research beyond its traditional Information Systems focus, offering a contextspecific approach tailored to O&SC environments. 2) It advances theoretical understanding of DG by positioning it as a key enabler of digital transformation, particularly in complex, data-rich industrial settings. 3) This research explores practical methods to mitigate data issues, enriching the literature on data quality and master data management with concrete, actionable insights. 4) It also bridges the gap between MCAI and DCAI approaches, supporting the development of more effective and robust AI solutions. 5) By highlighting the interdependence between policy and standard development, organizational roles, and technical integration, this research encourages a shift from reactive data management toward proactive, business-centric DG. 6) Finally, it outlines future research directions that reflect the dynamic and interdisciplinary demands of DG in modern industrial environments.

From a practical perspective, this PhD research makes four key contributions. 1) By systematically identifying and mapping data-related issues across all stages of the data lifecycle, the study enables organizations to diagnose problems more precisely and anticipate potential challenges before they disrupt operations. This structured approach helps shift DG from reactive problem-solving to proactive planning, allowing for more effective resource allocation and targeted interventions. 2) It offers actionable guidelines for overcoming persistent data-related barriers. These guidelines help practitioners achieve data excellence and facilitate the effective adoption of advanced technologies such as AI. 3) Companies are advised to adopt a step-by-step roadmap. This staged approach allows firms to resolve data issues in a manageable way, and gradually achieve enterprise-wide, sustainable DG that supports digital transformation and long-term value creation. 4) It suggests that companies should consider DG from three levels: management, operational, and technical; each with distinct areas of focus. This pathway offers a structured, high-level view of DG that supports organizations in establishing company-wide DG strategies. Together, these contributions advance both the strategic planning and operational execution of DG, enabling organizations to integrate DG into daily workflows and scale solutions to support data-driven and intelligent transformation.