Leveraging Statistical Process Control for continuous improvement of the manufacturing process

Abstract

Statistical Process Control (SPC) has been applied to manufacturing processes for several decades as a means of ensuring product quality and has become a primary tool for the application of continuous improvement efforts. Continued Process Verification (CPV) is a Food and Drug Administration requirement that requires biopharmaceutical companies, such as Amgen, Inc., to demonstrate control of commercial manufacturing processes. Furthermore, the Food and Drug Administration's guidance on CPV specifically calls for the use of SPC. This thesis suggests including the use of the Akaike information criteria (AIC), a recognized statistical model selection criterion, for objective model selection for the purpose of establishing the most representative control limits in the application of SPC. The most representative control limits are instrumental in eliminating unnecessary use of resources in the evaluation of manufacturing data. Thus, the use of AIC is one way to reduce waste in the entire process of monitoring the manufacturing process, evaluating data, and making improvements to the manufacturing process. In addition, this thesis forms several key concepts for effective use of SPC and continuous improvement efforts when working with contract manufacturing organizations (CMOs). Finally, this thesis will discuss the applicability of the work done related to SPC as the foundation for effectively monitoring, evaluating and improving the manufacturing process.