The DOE’s 2014 reports evaluated membrane heat pumps technology as one of the most promising alternatives to conventional vapour compression methods. The vapour compression methods maintain evaporator temperature lower than dew points to deal with the latent heat load which generally occupies 20-30% of total cooling load. In membrane heat pump systems, only the water vapour transfers and there is no phase change. This migration is caused by the difference in vapour pressure before and after the membrane. A vacuum pump or blower is used to make the pressure difference. However, there is no methodology for predicting dehumidification performance of membranes when it is used as a part of cooling system. In this study, with the assumption that there is a similarity between heat transfer and moisture pervaporation, the performance indexes of the membrane were derived using well-known heat exchanger method, ε-NTU models. The performance estimations were conducted for two representative system layouts – bypass layout and vacuum layout. Simple relations between design parameters were suggested, and these would give design guides for researchers