Time Flies: Querying and Exploring Large Time-Dependent Data Sets (2026-US-30MP-2969)

Defectivity reduction efforts drive large-scale investigations of historical process data in photoresist manufacturing. For production engineers at Inpria, this data is stored as time series data on an SQL database. Data is recorded from equipment sensors every two seconds. As a result, large-scale investigations of new data metrics are time-consuming and present serious data integrity challenges. So, what happens if you need to investigate a time-dependent parameter across historical production records?

For the HVM and Pilot engineering teams, this question presents very different challenges. The HVM systems have access to years of high-quality batch records. The Pilot systems are R&D equipment. Batch records exist in various forms: lab notebooks, Excel spreadsheets, and SQL data tables.

JMP makes it easy to import batch record data, regardless of data source. The SQL query parameters are generated using this batch record data. Once a subset is created, querying the time series data is simple using JSL SQL integration. From there, the time series data is cleaned and explored using JMP visualization tools. Lastly, a summary table is generated that can then be joined with defectivity data sets.