Over the past decade, geotechnical engineering professionals have increasingly embraced data management and data science principles to structure data collected from site investigations, laboratory results, and monitoring systems. These principles enable exploratory data analysis (EDA) and uncertainty quantification, addressing the epistemic uncertainties inherent in geotechnical practice. This shift is driven by the digital transformation era, exemplified by methodologies and technologies such as Building Information Modelling (BIM) and digital twins (DT), along with a growing awareness of the need to manage uncertainties in geotechnical engineering.
Today, concepts such as business intelligence, data wrangling, relational databases, Bayesian updating, and machine learning have become integral to the geotechnical engineering lexicon. However, the authors argue that the field is still in an early stage of adopting these areas into standardized protocols or procedures. Despite this, significant efforts have been made by organizations and societies, including the Association of Geotechnical and Geoenvironmental Specialists (AGS), the International Society for Soil Mechanics and Geotechnical Engineering (ISSMGE), the International Organization for Standardization (ISO), Eurocode, the Norwegian Geotechnical Institute (NGI), and Bentley Systems, among others, to develop procedures and tools that facilitate data-driven decision-making and enhance the reliability of geotechnical analyses.
This paper shares the authors' experiences and highlights the main challenges encountered in implementing data management principles and exploratory data analysis. Case studies are presented using tools such as Power BI, Python, and R, with a focus on geotechnical monitoring records and site investigation data for Tailings Storage Facilities (TSFs) and Waste Rock Dumps (WRDs). The authors discuss the practical application of these tools in managing large datasets, performing EDA, while addressing the limitations and opportunities for further integration of data science and data management into geotechnical practice.
