This paper presents the development of a method to optimize a solar-assisted cooling system with limited budget constraints. Regression analysis is used to identify the relationship between the solar fraction and the system factors according to the data provided by experiments. In order to obtain an accurate model to estimate the problem using small number of experimental trials, the method of central composite design (CCD) from design of experiment (DE) is used as a key technique. The experimental trials are conducted in the transient energy system simulation (TRNSYS) tool. Finally, the optimization problem is formulated and solved by including the model as the objective function, the physical constraints of the system factors, and the budget limit. A case study was conducted to apply this optimization method to the design of a solar-assisted double-effect absorption cooling system installed in a small-sized office building in West Lafayette, IN, USA. The results show the developed optimal model strongly agrees with the physical system model in TRNSYS. This optimization method can be generally applied to different types of solar cooling systems, and other renewable energy systems. Published by Elsevier B.V.