To predict the quality of final products and to enable staff to adjust the relevant parameters to improve the actual quality we offer our module inPROC. The module warns your staff about the deviation from the planned quality and recommends adjustments to production parameters to reach the specified target quality.

The module inPROC of our IPN IoT/ML platform is used to predict product- & process quality. Complementary, inPROC recommends adjustments to production parameters to reach a specified target quality. In contrast to the module IPN inCARE, this module uses process & status data for its predictions which are collected only once / a few times (representative spot samples) during a production run or a batch. In that case the data has to be transformed according to the requirements and assigned to the event which shall be predicted. The transformation and the assignment is performed by our module IPN inPOLL.

inPROC covers the phases “Modeling”, “Evaluation” and “Deployment” of CRISP-DM as well as the continuous application of the models to your real-time data.

Same as for the module IPN inCARE the prediction models can be developed and deployed both by IPN or directly by the user himself. In case you do not plan to develop the predictive models yourself IPN is offering “model development” as a service. The model development is based on the IPN methodology which is the basis of all our consulting services.

IPN inPROC is used in IPN projects to help reduce scrap production in a batch production, to optimize the rigging phase of a production machine and for to optimise process control of a chemical reactor.