An approach to optimizing the energy efficiency of a Heating, Ventilating, and Air Conditioning (HVAC) system is presented that utilizes computational predictions of the effect of heat load distribution on moist air temperature, density, and humidity variation. Lumped-HVAC (L-HVAC) is a new lumped parameter code that couples fluid transport, energy transport, thermodynamics, and psychrometrics in an HVAC system. This code contains a nonlinear implicit solution algorithm for steady-state and transient calculations for flow resistance, water mass balance, and energy conservation. L-HVAC has been validated using a simplified analytical model, the commercial lumped parameter code SINDA/FLUINT, and experimental measurements. Steady-state calculations for a single-room system suggest an order of magnitude greater energy savings using a variable chiller power control approach compared to control damper and variable-drive fan approaches. L-HVAC was also applied to predict that the fraction of latent to total heat load influences the steady-state system temperature by up to 0.4 degrees C for the example system in this study. (C) 2010 Elsevier B.V. All rights reserved.