One of the essential requirements of injection molding is to ensure the stable quality of the parts produced. However, numerous processing conditions, which are often interrelated in quite a complex way, make this challenging. Machine learning (ML) algorithms can be the solution, as they work in multidimensional spaces by learning the structure of datasets. In this study, we used four ML algorithms (kNN, naïve Bayes, linear discriminant analysis, and decision tree) and compared their effectiveness in predicting the quality of multi-cavity injection molding. We used pressure-based quality indexes (features) as inputs for the classification algorithms. We proved that all the examined ML algorithms adequately predict quality in injection molding even with very little training data. We found that the decision tree algorithm was the most accurate one, with a computational time of only 8–10 s. The average performance of the decision tree algorithm exceeded 90%, even for very little training data. We also demonstrated that feature selection does not significantly affect the accuracy of the decision tree algorithm.

The simultaneous improvement of injection molding process efficiency and product quality, as required by Industry 4.0, is a complex, non-trivial task that requires a comprehensive approach, which involves a combination of sensoring and information techniques. In this study, we investigated the suitability of in-mold pressure sensors to control the injection molding process in multi-cavity molds. We have conducted several experiments to show how to optimize the clamping force, switchover, or holding time by measuring only pressure in a multi-cavity mold. The results show that the pressure curves and the pressure integral are suitable for determining optimal clamping force. We also proved that in-channel sensors could be effectively used for a pressure-controlled SWOP. In the volume-controlled method, only the sensors in the cavity were capable of correctly detecting the end of the filling. We proposed a method to optimize the holding phase. In this method, we first determined the integration time of the area under the pressure curve and then performed a model fit using the relationship between the pressure integral and product mass. The saturation curve fitted to the pressure data can easily determine the gate freeze-off time from pressure measurements.