Auto stop-start (or Engine stop-start) has become a widely used feature to reduce fuel consumption and CO2 emissions particularly in congested cities. Vehicles that are equipped with such system typically include two DC power sources that are coupled in parallel: a primary and a secondary one. The primary power source supplies energy to the starter to crank the engine, while the secondary power source supplies current to the rest of the vehicle electric loads. During an auto-stop event, a controllable switch decouples the two power sources. Moreover, operating current and other parameters form are monitored to ensure enough battery SOC (State of Charge) for the next auto-start event. When any of these operating parameters exceeds a threshold, the controllable switch opens to isolate the two batteries and the engine is automatically started. This paper introduces a strategy to control the power usage of the Electric Power Assist Steering (EPAS) system for enhanced stop-start availability, optimal steering/EPAS assist, maximum fuel economy saving and minimum CO2 emissions. The algorithm predicts the energy demand by the customer and determines the next engine state based on the above mentioned criteria. The power system architecture uses a Lead Acid battery (Lead Acid, AGM and/or Enhanced) in parallel to a Lithium-ion (Li-ion) battery. Such architecture combines the advantages of the two technologies and supplies the required vehicle power over a wider operating range of ambient condition. Moreover, the Li-Ion has voltage characteristics that are similar to the lead acid battery, and hence there is no need for any adjustments between the two power supplies. Measured data and results assure the ability of the introduced strategy to optimize between engine auto-stop/start and the EPAS assist.