The role of data in the evolution of geotechnical design|Il ruolo dei dati nell'evoluzione della progettazione geotecnica

Like all engineering disciplines, albeit with perhaps greater difficulty due to the trade’s distinctive exposure to the complexity of the natural environment, geotechnical engineering is experiencing a continuous shift in paradigm towards data-centric and non-deterministic approaches in which data is used to model, process, and report uncertainties explicitly. Data plays – or should play – an essential role in enabling engineers to acknowledge, model, and communicate uncertainty and in correctly applying prevalent current design codes such as the Eurocodes. However, the partnership between engineers and data is seldom exploited to its full potential. Why is this so? How could the geotechnical community accept and leverage data to its advantage? This paper provides an insight into such issues and into the current and possible future roles of data in the evolution of geotechnical design. Geotech nical codes which are currently relevant at international level, are examined in the perspective of their conceptual and operational approach towards uncertainty and data, with a specific focus on Eurocode 7. Objective endogenous features and inter-subjective exogenous perceptions, which hinder the full exploitation of the potential of geotechnical data, are presented in the light of the specificities of the latter. In a rationally optimistic spirit, induced by the availability of scientific research results and regulatory implementation, the paper presents selected examples of virtuous applications of statistical and probabilistic frequentist and Baye sian methods for the interpretation of geotechnical data aimed at quantitative uncertainty modeling. Salient benefits and merits of Bayesian approaches are further discussed, highlighting their appropriateness for the geotechnical discipline. The paper also specu lates on future regulatory trajectories, in which data, jointly processed by conscientious and humans (endowed with conscience and ethical sense) and algorithms (lacking conscience but endowed with higher intelligence), may assume an increasingly central role. This apparently dystopic scenario, which may induce discomfort and skepticism in our current views of the world, is discussed and framed in the hopefully reassuring light of the continuous trend towards an increasingly data-centric disposition which has so far characterized the evolution of human knowledge.