Understanding data quality in a data-driven industry context: Insights from the fundamentals

Abstract

The increasing adoption of commercial-off-the-shelf infrastructure components and the rising integration of sensors into assets have led to a notable proliferation of operational data in industrial systems. As a result, a significant portion of investment and risk management decisions now heavily rely on the provenance and quality of heterogeneous data, sourced both internally and externally from specific industrial systems. This paper presents a review that covers three critical aspects of data quality: first, ensuring data quality through deliberate design; second, understanding the dynamic interplay between data and its users within sociotechnical systems; and third, attributing ongoing value to data resources as their roles evolve. These aspects are examined through a lens encompassing both traditional and the state-of-the-art theoretical frameworks for defining data quality. In addition, we incorporate insights from contemporary empirical research and highlight relevant industry standards and best practice guidelines. The synthesised insights serve as a practical foundation and reference for researchers and industry professionals alike, enabling them to refine and advance their understanding of data quality within the landscape of data-driven industries.