Implementation of diffraction in a ray-tracing model for the prediction of noise in open-plan offices

Sound prediction in open-plan offices is a real challenge because of the complexity of the layout of such offices, and therefore because of the multitude of acoustic phenomena involved. One such phenomenon, of primary importance, and not the least challenging of them, is the diffraction by screens and low dividers that usually partition the workspace. This paper describes implementing the equations of the Uniform Theory of Diffraction [McNamara et al. (1990). Introduction to the Uniform Theory of Diffraction (Artech House, Boston)] in an existing ray-tracing model initially dedicated to sound prediction in industrial premises. For the purposes of validation, a series of measurements was conducted in a semi-anechoic chamber in the same manner as Wang and Bradley [(2002). Appl. Acoust. 63, 849-866] but including real desktops instead of single screens. A first phase was dedicated to controlling the quality of the installation by making comparisons with McNamara's solution for a single screen on a rigid floor. Then, the validation itself was conducted with measurements on real desktops, first without a ceiling, and then with a rigid ceiling suspended above the double desk. The results of the comparisons between calculations and measurements in this configuration have demonstrated that the model is an effective tool for predicting sound levels in an open-plan office.