Multi-objective optimization of passive material-based and active operational-based parameters in PCM-enhanced building envelopes

• Material–control co-design: PCM grade/thickness + thermostat setpoints. • ANN surrogate (R 2 > 0.91) speeds multi-objective search of design space. • Weighted-sum optimum: 10 cm PCM, Tm 27 °C; heat 17 °C, cool ≈25 °C. • Annual reduction in heating ≈5% and cooling ≈11% vs no-PCM baseline. This study investigates the integration of a bio-based phase change material (BioPCM®) into the external walls of a mid-rise residential structure in Tehran, Iran, to enhance thermal efficiency and reduce energy consumption. Sixty parametric simulations were performed using DesignBuilder (EnergyPlus) by modifying four design variables: the melting temperature of the PCM, the thickness of the PCM, and the setpoints for the heating and cooling thermostats. An artificial neural network (ANN) surrogate model was developed using the simulation dataset to effectively anticipate heating (gas) and cooling power loads (R 2 > 0.91), hence facilitating an efficient exploration of the design space with little effort and cost expenditure. A sensitivity analysis confirmed that thermostat setpoints are the principal factors influencing energy consumption, while PCM properties exhibit nonlinear but substantial effects. Subsequently, NSGA-II was employed for multi-objective optimization to concurrently minimize both heating and cooling loads. The findings indicated that both the weighted sum and ideal distance decision-making strategies produced comparable optimal layouts. The setups featured a 10 cm PCM layer, a melting temperature of 27 °C, and setpoints of 17 °C for heating and around 25 °C for cooling. Quantitatively, the weighted-sum optimum reduced annual heating demand by ≈5% and cooling electricity by ≈11 % relative to a no-PCM baseline. Against an alternative baseline configuration, heating decreased by ≈20 % with a modest cooling trade-off of ≈12 %. The findings indicated that optimal arrangements resulted in significant reductions in annual heating and cooling requirements while maintaining occupant comfort. The results demonstrate the significance of concurrently optimizing PCM integration and HVAC controls. This provides valuable insights for the construction of energy-efficient structures in semi-arid regions.