Thabo Jena
Civil engineering enters the second quarter of the twenty-first century carrying a conceptual inheritance it has rarely interrogated directly. This paper advances a single, falsifiable thesis: the defining failure of twentieth-century civil engineering was not a failure of materials, computation, or design method, but a category error in the treatment of infrastructure as a class of physical assets to be optimized rather than as a class of adaptive, decision-making systems operating under irreducible uncertainty. We call this inheritance the Asset Paradigm and argue that it persists because it is embedded in codes, procurement, liability law, curricula, and professional identity, not because it remains effective. Building on, and formally sharpening, the emerging literature on infrastructure intelligence and resilience, we contribute: (i) necessary and sufficient conditions that distinguish Infrastructure Intelligence from smart infrastructure, digital twins, and cyber-physical systems, together with a quantitative formalization; (ii) a five-stage Infrastructure Evolution Model, classifying real systems by the locus of run-time decision authority and justified against maturity-model precedents from other engineering domains; (iii) four empirically testable propositions, each with explicit variables, candidate datasets, and validation methods; (iv) an illustrative bridge structural-health-monitoring application; (v) a phased, institutionally feasible transition pathway; and (vi) a ten-item Grand Research Agenda for 2025–2050. We position the framework against complex adaptive systems, socio-technical, resilience, adaptive-governance, and deep-uncertainty theory, and identify precisely where each breaks down under conditions of artificial intelligence, climate non-stationarity, and machine-speed autonomy. We conclude with a set of concrete claims and institutional commitments.