A hybrid simulation approach to predicting noise in restaurants

High levels of noise are a well-known source of occupant discomfort in dining environments, and noise ranks highly among the most common customer complaints in restaurant reviews. Prior work has focused on measuring existing restaurant soundscapes and attempting to predict acoustic performance; however, current techniques are still limited in accuracy for restaurants with time-varying occupancy and complex layouts. This presentation demonstrates a novel approach to predicting ambient noise in restaurants via a hybrid simulation framework. Acoustical computations for a given restaurant space are performed in a room acoustics model, while a discrete event simulation (DES) model is used to simulate the arrival patterns of patrons and their conversational behaviors. Predictions from the room acoustics model inform the DES model, allowing for noise level predictions at each seat position. Results from the prediction model are validated against measurements in a case study restaurant. Further development of this framework may support the practical assessment of design interventions in any restaurant while also accounting for time-varying patterns of occupancy.