The effects of moisture on sensible and latent conduction loads are shown by using a heat and mass transfer model with variable material properties, under varying boundary conditions. This model was then simplified to reduce calculation time and used to predict conduction peak load (CPL) and yearly integrated wall conduction heat flux (YHF) in three different cities: Singapore (hot/humid), Seattle (cold/humid) and Phoenix (hot/dry). The room air temperature and relative humidity were calculated with the building energy simulation program DOE-2.1E. The materials studied were aerated cellular concrete (ACC), brick (BRK), lime mortar (LMT) and wood. It is shown that the effects of moisture can be very significant and that simplified mathematical models can reduce the calculation time with varying effects on accuracy. (C) 2002 Elsevier Science B.V. All rights reserved.