This study presents an integrated Water-Energy-Food-Ecosystem (WEFE) engineering education framework that combines risk-aware optimization, machine learning surrogates, explainable artificial intelligence (XAI), digital twin, and virtual reality (VR). A synthetic WEFE scenario library of 10,200 operating conditions and 20,000 intervention policies was generated to support solver-based labeling, XAI training, and immersive VR labs without using real learner data. Benchmark experiments show that the optimization core reduces aggregate water and food deficits by 25–45% while lowering WEFE loss tail risk (CVaR) by 15–30% across baseline, stress, and extreme regimes. Surrogate policies achieve 86–92% fidelity to the optimizer, with objective-performance gaps below 5%, enabling real-time VR interaction at frame rates above 60 fps for more than 500 concurrent “what-if” evaluations per session. The framework is instantiated across 6 program streams at the Hijjawi Faculty for Engineering Technology, with 12–16 contact hours per course allocated to WEFE-XR labs aligned with ABET-related outcomes in systems design, sustainability, data-driven decision making, and ethical responsibility. Analytics from more than 1,000 synthetic scenario playbacks indicate consistent selection of risk-aware strategies, with simulated policy trajectories showing, on average, 18–25% lower tail risk than naive baselines. By quantitatively evaluating the technical performance of each layer, the study shows that integrated WEFE technologies can be embedded in engineering curricula as a reproducible and measurable instructional infrastructure rather than as a standalone enrichment tool.
Shehadeh et al. (Sun,) studied this question.