Data Mining with JMP – Product Performance and Customer Behavior
Peter Wiebe, Krista Birch, and Anthony Orzechowski, Abbott Diagnostics
JMP provides a powerful platform to explore large data sets. At Abbott Diagnostics, we have the capability to query data from our customer’s instruments from around the world. Using JMP we have applied methods to address questions about our products, our customers, and our instruments that were previously too challenging to address in a reasonable timeframe. This presentation will show how we can monitor product performance, customer compliance, and diagnostic prevalence geographically around the globe. We achieve this capability using an ODBC in combination with JMP Scripting Language (JSL). Together, this set up allows for data to be rapidly and flexibly queried an imported, allowing the user the capability to begin exploring the data with all the statistical power available in the JMP software. Since each exploration into the dataset may vary, yet may be required to be well documented, JSL offers a means for traceability as reproducibility. Simply by capturing the script as the analysis is performed, it can be reproduced and verified, but the analysis can remain flexible for innovation.