Artificial intelligence based prediction models for rubber compounds

In the rubber industry, rheometric properties are critical in defining processing times and temperatures. These parameters of rubber compounds are determined by time-consuming and expensive laboratory studies performed in a rheometer. Artificial intelligence approaches, on the other hand, may be used to estimate rheometric properties in seconds without the need for any samples or laboratory experiments. In this research, artificial neural network, Gaussian process regression, and support vector regression techniques are used to predict minimum and maximum torque, 30% and 60% cure time of a rubber compound using both process parameters and raw material composition as input. The dataset comprises 1128 batches of the selected rubber compound. A detailed sensitivity analysis is performed to determine the best performing hyperparameters and the prediction performances are expressed as mean absolute percentage error (MAPE). Minimum, maximum, and average MAPE values are presented for each artificial intelligence technique. Besides this research contributes to fill the gap in rubber industry literature, the results obtained also strongly improve the existing literature results.