Life-cycle assessment and multi-objective optimization of natural-insulated envelopes across Iranian climates
Peyman Naghipour, Afshin Naghipour, Tarana Bakirova
Abstract
The building envelope plays a pivotal role in achieving sustainable energy performance, particularly in regions characterized by extreme climatic variations. This study investigates the holistic optimization of building envelopes—comprising walls, roofs, floors, and fenestrations—by integrating natural insulation materials with passive and smart design strategies. The research aims to enhance energy efficiency, reduce total CO₂ emissions, and improve occupants’ thermal comfort across four representative low-income climatic regions of Iran: Yazd (hot-arid), Tabriz (cold-dry), Rasht (temperate-humid), and Bandar Abbas (hot-humid). Addressing the existing research gap, the study extends beyond operational energy analysis by incorporating a full life cycle assessment (LCA), including embodied energy and life cycle carbon footprint. Multi-objective optimization (using NSGA-II) was performed to minimize annual energy demand, life-cycle cost (LCC), and environmental impact simultaneously. Building performance simulations were conducted using IES-VE and EnergyPlus, while LCA and economic analyses were executed via SimaPro and HOMER Pro. The results indicate that hybrid natural insulations-particularly straw–hemp composites-combined with passive strategies (dynamic shading, natural ventilation, and thermal mass enhancement) can reduce total CO₂ emissions by 42–58% compared with conventional materials. Also, the results demonstrate that the optimized design solutions can reduce annual energy consumption by approximately 25–35% compared to the baseline design, while achieving a 15–25% reduction in life-cycle costs over the building lifespan. Additionally, orientation-sensitive optimization improved thermal comfort indices (Predicted Mean Vote—PMV, Predicted Percentage of Dissatisfied—PPD) throughout the year. The developed predictive models based on machine learning (Random Forest) exhibited robust accuracy in estimating energy consumption. The findings provide an integrated framework for sustainable, low-cost, and climate-responsive envelope design, supporting the transition toward net-zero energy buildings in developing regions.