FabTime Newsletter Community Announcements Page

The announcements below are from the FabTime Cycle Time Newsletter, Volume 26, No. 3, published on June 17, 2025. We have highlights about recent and upcoming FOA meetings as well as new releases of our FabTime reporting software and Factory Digital Twin. 

Past issues of the newsletter are available for subscribers to download in PDF format. Existing subscribers can find the archive link in your most recent email newsletter. New subscribers will see the link upon registering. 

• Register for the newsletter here. 
• Main article for Volume 26, No. 3: What do we mean when we refer to tool availability?
• Subscriber discussion forum for Volume 26, No. 3
• Reach out to newsletter editor Jennifer Robinson here. 

The Q2 SEMI Fab Owners Alliance meeting was held at the Honeywell fab in Plymouth, MN. Jennifer was honored to moderate a panel discussion of device makers. The goal of the panel was to give the supplier community a picture of the concerns and priorities of the local device makers. Jennifer interviewed Dan Baseman of Honeywell, Alvin Kolkind of SkyWater and Craig Mattson of Polar Semiconductor (shown left to right below).

INFICON’s John Behnke also presented to the FOA device makers on Building Tomorrow’s Smart Factories today. INFICON sponsored the Thursday evening social event, which was held at WhirlyBall Twin Cities.

The next FOA meeting will be held July 29-30 at the Analog Devices fab in Beaverton, OR. That meeting will include a Women of the FOA event, which Jennifer plans to attend.

In other conference news, the Advanced Process Control - Smart Manufacturing conference (APCSM) will be held September 22-25 at Cliff Lodge at Snowbird in Snowbird, Utah. And, of course, SEMICON West will be on October 7-9 and will be (for the first time) held in Phoenix, AZ.

The FabTime team is delighted to announce the newest release of the standalone (SQL Server) version of FabTime. This version includes two major enhancements:

1. Boxplot support for stacked charts. This expands the range of boxplots in FabTime to answer questions like “Do our 5x Steppers have similar qualifications and performance?”  (shown below) and “Is our WIP distributed equally across our line segments?”
2. Natural language querying of past FabTime newsletters via the AskJen™ chat engine. Questions can be asked in a variety of languages, including English, Japanese, and German. The new AskJen control appears in the FabTime navigation bar (after running a few simple lines of code at the site level to enable it). A short example of a question and response is shown below.

This version also includes several bug fixes related to the introduction of our NextGen charting engine. While no action is immediately required, we do expect to switch the default to the NextGen engine in the next FabTime release (2509).

One other bug fix that is of special relevance to the newsletter community: the Operating Curve chart in FabTime now more accurately supports tool groups with multiple tools. The underlying calculations match the calculations in the Operating Curve Spreadsheet used in our Cycle Time Management Course.

Finally, patch installation is simplified. This release can be applied to any FabTime Standalone server running Patch 115 Build 16 or later. If you are interested in upgrading to this new version, please contact your FabTime site support engineer (or reach out to Jennifer, who can pass along your request).

One of the key pieces of predictive information built into the INFICON Factory Digital Twin is process time. When you consider a problem like scheduling, where algorithmic approaches often build millions of schedules before choosing the “best” one, you must be able to always and quickly answer the question: “If I put this lot on this tool, how long will it take to process?”

The INFICON Data Science team will be releasing a significantly upgraded machine learning model for process time prediction in June as part of Factory Digital Twin version 2509. The new model represents a ~30% improvement over the previous model in prediction accuracy as measured by reduction of mean absolute error. Better process time estimation means better scheduling, and ultimately, better factory efficiency. This is one of many examples of AI enhancements running live inside our Factory Digital Twin. Contact us for more information.

For more reading from the FabTime newsletter, see:

• Main article for Volume 26, No. 3: What do we mean when we refer to tool availability?
• Subscriber discussion forum for Volume 26, No. 3