Author(s): Smita Sabnam & Ulrike Passe
Dhaka, Bangladesh, is one of the most densely populated cities in the world. Due to increasing temperatures, reliance on air conditioning units is rising, yet most households cannot afford either the units or the electricity bills. Interior wall panels made of locally available materials like adobe offer a promising retrofit solution. However, retrofit decisions often prioritize steady-state insulation metrics, assuming lower U-values ensure lower operational energy. This short paper examines that assumption by analyzing the convergence and divergence between thermal transmittance (U-value) and annual Energy Use Intensity (EUI) for adobe panels as interior retrofits in Dhaka. Using Climate Studio simulations in Rhino and Grasshopper with a thermal comfort range of 24–32°C, a baseline brick wall is compared against 11 adobe panels with different thicknesses and insulating materials. Results show that although the rice husk adobe panel achieves the lowest U-value (0.24 W/m²K), it does not deliver the lowest EUI. In contrast, the wood-chip option (1.12 W/m²K) and rammed earth panel (1.51 W/m²K) reduce EUI by 22–23%. The findings demonstrate that annual energy performance can diverge from U-value rankings because thermal mass combined with ventilation opportunities can outweigh extreme insulation. The paper concludes that in hot-humid climates where natural ventilation supports thermal comfort, moderate insulation combined with high thermal mass can outperform super-insulating composites. Effective retrofit panels must store heat, not just block it across the diurnal cycle.
Volume Editors
Reed Kroloff & Francisco J. RodrÃguez-Suárez
ISBN
978-1-944214-52-4