ABSTRACT In an automotive air-conditioning system, the passenger comfort in the vehicle cabin gained importance and designing a right HVAC (Heating Ventilating and Air conditioning system) unit plays a vital role during the upfront design stage. Predicting the performance of cabin cool down rate upfront in the initial design stage will help us to reduce the overall product development time. To meet the customer comfort it is necessary to validate the HVAC performance at vehicle level in the extreme hot or cold ambient conditions. The vehicle which is having higher seating capacity will have higher thermal load and providing the thermal comfort to the passenger is the challenging task for the automotive HVAC industry. The dual HVAC unit is generally used to provide uniform cooling to the larger cabin volume. In this paper, optimization studies are carried out to understand the effect of various parameters that influences the performance of dual HVAC system and arriving at the optimal values. 1D (One Dimensional) simulation is extensively used to predict the HVAC performance during the initial design stage of the program. The refrigerant loop is modeled with components such as compressor, condenser, TXV, evaporator and with Cabin. The complicated vehicle cabin including the glazing surfaces and enclosures is modeled as three row duct system along with cabin using 1D tool AMESim® which includes the type of material, density, specific heat capacity and thermal conductivity of the material. The actual vehicle driving conditions as per standard were simulated to validate the HVAC performance. The transient 1D simulation was carried out as per the vehicle boundary conditions. The heat gain values of the panel ducts were adjusted to reduce the deviation from the test. The simulated results for average cabin temperature and grill temperature were compared against the available surrogated vehicle test data. The detailed comparison of test data and simulation results were plotted and identified the simulation parameter which affects the correlation. This optimized model has been used for various future programs to predict the performance.