A refugee housing unit design approach in terms of comfort and environmental performance through a two-step decision-making method

The global increase in displaced populations and refugee camps driven by factors such as natural disasters, political conflicts, and climate change underscores a pressing challenge. Furthermore, it is evident that the inadequate design of the refugee housing unit (shelters)—which is often misaligned with the local climate conditions—has contributed to substandard living conditions, thereby undermining both the environmental and social well-being of refugees. In this context, this research aims to improve thermal comfort, reduce environmental impact, and enhance energy performance through shelter design proposals. Considering the global distribution of refugee camps across diverse climate zones, Zaatari Refugee Camp—located in a hot arid climate—was selected as a representative case for analyzing existing shelter conditions and informing design improvements. Six design variables were defined to optimize the base case: 14 plan layouts, three envelope materials, three insulation thickness options, five window-to-wall ratios, three shading device options for the WNW and ESE orientations, and 11 window opening rates for the West North-West, East South-East, and North North-East orientations—resulting in 188,669,250 possible design combinations. Among them, the Pareto front optimization of DesignBuilder generated 63,154 initial iterations, resulting in 243 Pareto-optimal scenarios. Then, applying the equal-weighted decision-making (EWDM) method, the ideal scenario is identified, outperforming all EWDM scenarios and the base case by improving thermal comfort by 44.48%, life-cycle assessment (LCA) by 69.46%, embodied CO ₂ by 84.55%, operational CO ² by 40.44%, and primary energy by 60.09%. Consequently, the sensitivity analysis indicates that material optimization alone is insufficient. The LCA results are primarily driven by the window-to-wall ratio, while thermal discomfort is largely influenced by window opening rates.