Development of an Empirical Acoustic Model for Predicting Reverberation Time in Typical Industrial Workrooms Using Artificial Neural Networks

In noisy workrooms, acoustic treatments may be very costly and subject to different limitations. Reverberation time prediction can be employed for discovering acoustic treatments opportunities so that, more acceptable conditions are obtained. Using artificial neural networks, this study aims to develop an empirical model for predicting the reverberation time in industrial embroidery workrooms. Geometrical model was also employed and its results were compared with those of neural networks. The main acoustic features and reverberation times were determined in the studied workrooms. Different structures of networks were developed and finally, the networks with one hidden layer along with five neurons were found to be the best model for predicting reverberation time (RMSE = 0.17(s) and R ² = 0.77). The relative errors of the networks and the geometrical models for a new case of workroom were 6% and 4.7%, respectively. Although networks are empirical in nature, the results confirmed the potential of this approach for minimizing the uncertainties in acoustics' modeling. The developed model gives professionals the opportunity to have an optimum decision about the effectiveness of acoustic treatment scenarios in embroidery workrooms.