Method for controlling and/or monitoring a chemical production plant

A computer program, specifically an application, for controlling and/or monitoring a chemical production plant, wherein the chemical production plant comprises at least one process chain comprising at least one batch process, wherein the computer program comprises instructions which, when the program is executed by a computer or computer network, cause the computer or computer network to carry out the following steps:

• i) determining of input data, wherein the input data comprises at least one quality criterion and production plant layout data, wherein the step comprises retrieving the production plant layout data and receiving information relating to the quality criterion via at least one communication interface;
• ii) determining operating conditions for operating the chemical production plant by applying at least one trained model on the input data, wherein the trained model is at least partially data-driven by being trained on sensor data from historical production runs;
• iii) providing the operating conditions, and determining a production plan for runs of the chemical production plant based on the operating conditions and scheduling the production plan using a scheduler,
• wherein the trained model comprises a combination of a white-box-model and a black-box-model, wherein the white-box-model is based on physico-chemical laws, wherein the black-box-model is based on the sensor data from historical production runs, and
• controlling the chemical production plant based on the production plan.

A controlling system for controlling and/or monitoring a chemical production plant, wherein the chemical production plant comprises at least one process chain comprising at least one batch process, wherein the controlling system comprises at least one communication interface configured for determining of input data, wherein the input data comprises at least one quality criterion and production plant layout data, wherein the communication interface is configured for retrieving the production plant layout data and for receiving information relating to the quality criterion, wherein the controlling system comprises at least one prediction unit configured for determining operating conditions for operating the chemical production plant by applying at least one trained model on the input data, wherein the trained model is at least partially data-driven by being trained on sensor data from historical production runs, wherein the controlling system comprises at least one output device configured for providing the operating conditions, and further comprising a scheduler configured to determine a production plan for runs of the chemical production plant based on the operating conditions and to schedule the production plan, and wherein the trained model comprises a combination of a white-box-model and a black-box-model, wherein the white-box-model is based on physico-chemical laws, wherein the black-box-model is based on the sensor data from historical production runs, and

• controlling the chemical production plant based on the production plan.