Maintains that since the evolution of primary-secondary pumping arrangement over 20 year ago, multiple chillers have been controlled by measuring flow in the secondary system and operating the appropriate chiller(s) to handle the flow. Argues that this control method is flawed in that a variable flow chilled water system is rarely a constant Delta-T system. Measuring only the flow does not provide an accurate indication of the load. As a result, most multiple chiller systems controlled in this manner do not operate at optimum efficiency and suffer from low Delta-T. Presents a case study of an actual system that was not optimised to peak efficiency and discusses the changes made to improve its performance. States these changes can be applied to other chilled water systems with multiple chillers.