Thursday, July 30, 2015

Not All of Big-Data Is Hype -- Self-Service BI with Data Analytics Can Yield Quick Results for Manufacturing

Rapid Insights and Operational Intelligence for the shop-floor

By Gopal GopalKrishnan, P.E., Solution Architect, Partners and Strategic Alliances, OSIsoft, LLC. Member, MESA International Technical and Education Committee 

Not all big-data is hype. Self-service BI with its slice and dice user experience came along just as big-data was gaining mindshare.  And, when targeted to specific use cases, self-service BI with data analytics can yield rapid insights – big data or not.

This is Part 2 of a blog post series describing the use of self-service BI and data analytics from real-life scenarios in MOM (manufacturing operations management).  This use case comes from one of Georgia Pacific’s (GP) paper machine operation.  

To read Part 1 click here:

If you are not familiar with how paper is made, please see this interactive video tour. The data and analysis described below is for the “Papermaking” step in the video tour.  Here, several chemicals are added to the feed stock to influence its properties such as whiteness, strength, and others.  

Variability in Chemicals Consumed

The quantity of chemicals consumed for each product grade run varied considerably.  Business was keen to get a quantitative understanding of this variability – sliced and shown by product SKU, by grade run, by feed-stock, by day, by month, by shift, and so on.

A data infrastructure for MOM helps capture time-series sensor data such as flows, temperatures, pressures, machine speed, etc. Then you can shape them with the correct context with relational data such as product SKUs, recipes (feed-stock blend percent for virgin pulp, brown stock, etc.), and quality targets (basis weight aka paper density, whiteness, strength, etc.).  And then you are ready for the slice and dice user experience in the self-service BI tools.

The drill-down helps to hone in on the answers for questions such as:

       Did I follow the recipe? Show me by grade…
       How often, how much, and when do I deviate from recipe?
       How does brown stock in the feed affect whiteness and strength?
       How does refiner energy consumed affect the strength?
       When and where did I use the least amount of chemicals?
       What and when is my best production run?


Rapid Insight

The visuals with aggregated values offer a bird’s eye view of all the production data for multiple years and across multiple production lines and SKUs, and across a variety of feeds and quality targets – and you can methodically drill down to specifics with an Expedia-like user experience mentioned in Part 1.

The GP video is in this rapid insights using data analytics talk.  The chemical usage variability will be of interest to every paper mill manager.  If you are in a different industry, the take-away from this use case is how the data infrastructure and the self-service analytics tools enable rapid insights for operational intelligence.

I’d love to hear about your experiences with data analytics and MOM.


Gopal GopalKrishnan, P.E. has been involved in several roles at OSIsoft and has been working with the PI System since the mid-1990s, in software development, technical and sales support and field services. Attached to the Philadelphia office, he is currently a Solution Architect in the Partners & Strategic Alliances Group.  Previously, he was a Product Manager with a focus on Enterprise and Asset Integration and PI System data access.

Gopal has a Master's in engineering, continuing education in business administration and is a registered professional engineer in Pennsylvania. He is also active in the MESA Technical Committee and the Education Committee and the MESA Continuous Process Industry Special Interest Group and active in topics such as data mining, energy efficiency, manufacturing intelligence, sustainability, including green and sustainability initiatives in facilities and data centers.

No comments: