Come to Jan. 16 , 2020 MAJUG Meeting - 1:30 - 4:00pm following New in JMP and JMP Pro session (Arlington, VA)
Dec 11, 2019 6:54 AM
| Last Modified: Dec 11, 2019 7:17 AM(1053 views)
Please join us for an afternoon session with presentations about practical applications of JMP. It will be held after lunch following the New in JMP 15 and New in JMP Pro 15 morning presentation by JMP staff who will demonstrate and answer questions about new capabilities.
SAS Arlington Training Center
1530 Wilson Blvd Suite 1000 Arlington, VA 22209
1:30 - 4:00 pm
The afternoon MAJUG meeting agenda and registration form are being finalized. Presentations will include:
Extracting Valuable Practical Information From Experimental Models Created for Quality by Design Presented by Rob Lievense, Senior Systems Engineer, JMP
Extracting Valuable Practical Information From Experimental Models Created for Quality by Design JMP DOE models can be used dynamically to provide stakeholders with reliable estimates of the quality performance for new products. The design and execution of structured, multivariate experiments allow scientists and engineers to efficiently define a robust design space for pharmaceutical and medical device manufacturing. Models created through the DOE platform in JMP are used to determine settings of the critical process parameters (CPPs) that ensure a robust process to make products that meet the requirements for Critical Quality Attributes (CQAs). Regulatory submissions that include such QbD elements demonstrate that risks have been mitigated; however, a great deal of practical information can be extracted with simulations of the model. This presentation utilizes the experimental model with historical information and subject matter expertise to project the likely operational performance of a product. The DOE model Prediction Profiler is used with the simulator to dynamically predict a population of future results with patterns of real-world variation included in the inputs. This dynamic modeling is an excellent tool for setting manufacturing card limits determining a manufacturing control space defined with estimates of the percentage defects. The analyses allow for the inclusion of the measurement uncertainty as an added noise factor for the response.