Abstract

An industrial ethylene propylene rubber compounding process is used to illustrate some of the issues that arise in the monitoring of the manufacturing performance of a process comprised of both batch and continuous unit operations. The key issues relate primarily to the different formats of the data that are routinely collected. For example, on batch type processes such as mixers, measurements are collected for a fixed quantity of processed mass. In contrast, for continuous processes such as extruders, data is recorded at fixed time points. Finally, in order to monitor the end product quality this data has to be related to the final product quality measurements that are recorded after a fixed quantity of mass has been produced. This quantity differs from that used for the batch units. Thus, for the development of a manufacturing process performance monitoring scheme, it is necessary to provide a common sampling base. The approach proposed is to re-sample the measurements for a fixed sampling mass. Benefits arising from the proposed methodology include the more traditional advantages of multivariate statistical process control: that is, an enhanced understanding of the process and the early warning of changes in operation. In addition, the proposed approach removes the need to time-align the data.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call