Data collection is achieved through multiple automated collection programs, utilizing a class factory (plug-in) architecture and job manager program. The job manager and scheduling programs call and execute the data collection methods without having to know the specifics of the data collection program.  By using a plug-in architecture, we can deploy new collection methods without having to rewrite or edit existing programs or the job manager. These new collection methods can be tailored to specific machine tools or measurement systems, without affecting existing data collection methods. This approach also reduces errors to existing data collections.

LAI International recognizes the criticality of accurate and traceable data for the aerospace and medical industries. To ensure full traceability, the raw data is imported without any manipulation, and all changes to the data are tracked via audit and log tables at the database level.  Any analysis is done via separate reporting tables. When duplicate or mal-formatted data is detected, that file is not imported and is pushed to separate areas, then notifications are sent to address the errors.

We have implemented a series of cross-reference tables to normalize the incoming raw data. By normalizing, we can then identify the KPI’s across multiple data sources and report those KPI’s to online Statistical Process Control (SPC) systems.  This reduces the amount of data for the online SPC system and improves the responsiveness for the end users.  The cross-reference tables also provide data for reporting tables that are access via standard Open Database Connectivity (ODBC) type connections for ad-hoc analysis tools.  By combining the cross-reference tables, KPI’s and reporting tables, we can serve both the production and engineering communities in our company. The cross-reference tables are maintained by the responsible engineering staff utilizing an editing program including import and export to Excel format. All changes to the cross-reference table are logged at the database level.

We will continue to expand the data collection to machine data and environmental data with the eventual goal of providing direct feedback to the machine tools to correct and adjust parameters from the proven KPI’s.