Integrating artificial intelligence and multi-criteria analysis for sustainable hybrid renewable energy in healthcare facilities

• AI-based ANN (LM algorithm) accurately forecasted wind speed and solar irradiance. • Eleven solar–wind–diesel–battery systems were modelled in HOMER software. • BWM and AHP integrated to optimize techno-economic and environmental factors. • Optimal design: 3.52 MW PV, 31 × 20 kW wind, 1.6 MW diesel; 91.95% renewable share. • Achieved LCOE = 0.225 USD/kWh and CO₂ reduction from 4.6 M kg to 0.39 M kg per year. • Demonstrated AI + MCDM integration improves reliability and sustainability. The transition to sustainable and reliable energy systems is a global necessity driven by the depletion of fossil fuels and environmental degradation. This study focuses on the design and optimization of a solar–wind hybrid energy system for a mid-sized off-grid hospital with a daily load of 18000 kWh in Amman, Jordan, utilizing AI-based forecasting and multi-criteria decision-making (MCDM) approaches. Five years of meteorological data (2020–2024) were analyzed to predict wind speed and solar irradiance for 2025 through an Artificial Neural Network (ANN) trained using the Levenberg–Marquardt algorithm. The model achieved high accuracy, with a correlation coefficient (R) of 0.98425 and a minimum mean squared error (MSE) of 0.0073. Forecasted data were used in HOMER software to simulate eleven different hybrid configurations integrating photovoltaic panels, wind turbines, diesel generators, and batteries. Evaluation criteria included Net Present Cost (NPC), Levelized Cost of Energy (LCOE), Renewable Fraction (RF), Simple Payback Period (SPP), Excess Electricity (EE), and CO₂ emissions. Results revealed that the optimal configuration comprised a 3,523.75 kW PV system, 31 of 20 kW wind turbines, and a 1,600 kW diesel generator operating 964 hours annually. This system achieved an NPC of 19.11 million USD, LCOE of 0.225 USD/kWh, RF of 91.95%, and SPP of 5.36 years, reducing CO₂ emissions to 388,504 kg/yr compared to 4.6 million kg/yr in diesel-based systems. The study demonstrates that integrating AI forecasting with optimization tools significantly enhances hybrid system efficiency and sustainability.