Operationalizing Maintenance of Calibration Models Based on Near-Infrared Spectroscopy by Knowledge Integration

PurposeIn compliance with good manufacturing practice, the use of process analytical technology based on near-infrared spectroscopy requires that a calibration model be maintained to assure the continuous estimation accuracy. The present research proposes knowledge integration-based operationalization of calibration model maintenance.MethodsA model maintenance framework was developed with reference to the regulatory guidelines and the United States Pharmacopeia. To utilize the model maintenance effectively, individual knowledge about the detailed processes was integrated using business process modeling. This was based on the type zero method of the integrated definition language (IDEF0) and a role assignment matrix, RACI.ResultsKnowledge integration clarified the detailed processes, as well as the necessary and sufficient roles for the operation. This clarification improved the operational efficiency with proper procedures. Moreover, knowledge integration enhanced knowledge management by acknowledging the need to exploit empirical knowledge from the continued operation. Based on the knowledge integration results, model maintenance was established for a drug product in a commercial manufacturing plant. The operation was successfully validated through model maintenance practice three times for four calibration models in process validation.ConclusionKnowledge integration based on business process modeling is beneficial for operationalizing model maintenance intelligently.