Distributed energy systems based on cogeneration offer significant potential to save energy since they effectively utilize waste heat from power generators. However, unless there is an appropriate combination of machinery and operations, the planned performance cannot be achieved. Thus, it is quite difficult to determine the optimal combination of machinery and operations. For this, an optimal design approach is needed. In this study, a new optimal design method for building energy systems is proposed. There are an enormous variety of combinations with regard to energy supply and demand. This method designs the most efficient energy system by optimizing the operation of available systems with consideration for the optimal capacity of machinery in the systems. Optimization algorithms known as "genetic algorithms" (GAS) with the capacity to deal with non-linear optimization problems have been adopted in this optimization analysis. In this study, a single-building energy system is evaluated. The result shows that the proposed method is sufficiently capable of optimizing the design, and has the potential to be applied to very complex energy systems with appropriate improvements. (C) 2010 Elsevier B.V. All rights reserved.