In room acoustic refurbishment/renovation projects, it is common to create a digital room model for use in room acoustic prediction software such as ODEON. Before simulating changes it is desirable to match the model, as well as possible, to existing conditions so that measured room acoustics parameters are in fair agreement with the ones simulated in the digital model. The acoustic data of the surface materials may be imprecise or indeed unknown. Therefore calibration has to be done manually by the room acoustician, who changes the absorption coefficients of the different surfaces in the room model in order to match measured parameters such as EDT, T30, TS, SPL, C50 and C80 against the simulated ones in an iterative process. This process is time consuming; requiring many iterations, and even so it can be difficult to obtain a reasonable match. This paper presents an implementation of a calibration tool utilizing a genetic algorithm to search through the M-dimensional search space defined by the number M of the unknown surface materials.