Figure 1
The diagram is organized into three main vertical sections connected by arrows. On the left, the section titled “PHYSICAL SYSTEM” contains three stacked boxes: “OPERATIONAL LOADING”, “ENVIRONMENTAL EXPOSURE”, and “SPARSE MEASUREMENTS (SENSORS, INSPECTIONS)”. Arrows from these boxes point toward the central section. A note at the bottom reads, “Relevant to S D G s 7, 9, 11, 12, 13”. The central section is labeled “PHYSICS-INFORMED DIGITAL TWIN CORE”. At the top is a box labeled “SUSTAINABILITY slash DEGRADATION STATE S (t)”. Below it is a larger box titled “IRREVERSIBLE KINETICS and PHYSICS-BASED CONSTRAINTS”, which includes a small schematic curve labeled “Monotonic, Bounded Evolution”. Beneath this is another box labeled “PHYSICS-DOMINANT LEARNING and DATA ASSIMILATION (e. g., PINNs, Hybrid Models)”. Arrows connect these boxes vertically upwards, indicating flow from learning to constraints to the state. A side arrow labeled “State Feedback or Control” loops from the state back into the learning block. On the right, the section titled “SUSTAINABILITY-AWARE DECISION SUPPORT” contains four stacked boxes: “ENERGY EFFICIENCY OPTIMIZATION”, “MATERIAL WASTE REDUCTION”, “MAINTENANCE PLANNING”, and “ENVIRONMENTAL IMPACT ASSESSMENT”. Arrows from the central section point to each of these outputs, showing how the digital twin informs decision-making.Conceptual architecture of the physics-informed digital twin for sustainability. Schematic representation of the proposed physics-informed digital twin framework, showing the interaction between the physical system and the digital twin, the physics-governed evolution of an internal sustainability state, and the translation of this state into sustainability-aware decision support. The framework emphasizes physical consistency, data efficiency, and interpretability rather than high-fidelity numerical simulation