This paper illustrates how actual chiller plant performance data collected from an energy management system (EMS) can be used to adapt chiller plant control to optimize energy savings. Adaptive algorithms that can be programmed in to a direct digital controller (DDC) are applied to three basic optimization control strategies: (1) optimizing the staging on or off multiple chillers, (2) optimizing condenser water temperature, and (3) optimizing chilled water temperature when variable-speed chilled water pumps exist. The staging of chillers is determined by the chiller combination that can meet the refrigeration load with the least power requirement. The condenser water temperature is obtained by minimizing the sum of chiller plus cooling tower fan power consumption. The chilled water temperature is obtained by minimizing the sum of chiller plus variable speed pump power consumption For each control strategy, a DDC control algorithm is illustrated by means of a flow chart.Units: Dual