Data management and quality control

Abstract

It is often said that data gathering is much more expensive than data-management software (such as geographic information system). Indeed, data are perhaps the most important element of any asset management approach. In this comprehensive chapter, we embark on a journey through digital era, highlighting the pivotal role of data in our contemporary world, emphasizing the importance of data in today’s landscape. We delve into the critical question of which data to collect, providing insights into the strategic selection of data based on a cost–benefit approach and the significance of anticipation in data collection. A three-layer approach, encompassing object, system, and urban fabric levels, is proposed as a structure to organize data, elucidating the diverse information requirements at each layer, from descriptive data to performance assessments and requirements. A substantial portion of this chapter is devoted to data models and bias, elucidating the complexities of modeling sewer pipe deterioration and addressing issues such as selective survival and recruitment bias. Quality control emerges as a pivotal concern, clarifying the requirements for data quality, methods to assess completeness, and handling issues such as incompleteness, timeliness, uncertainty, and imprecision. Questions related to data quantity are explored, discussing the data-loop problem, reconstruction methods, and the implications of big data. Practical considerations related to data access and storage are also addressed. The chapter concludes by three enlightening case studies illustrating real-world applications of data models.