«

»

Apr 08 2014

Print this Post

Establishing a Hierarchy Model for Manufacturing Analytics

W_E_Deming_manufacturingW. Edwards Deming was an American statistician, professor, author, lecturer, and consultant. He gained notoriety in the manufacturing world by his extensive promotion of the “Plan-Do-Check-Act” cycle, a strategy to implement continuous process improvement. One of his quotes was “The most important things cannot be measured.” His point was that strategic, long term issues can’t really be measured up front – you don’t know what will be important until time has passed.

In the world of manufacturing operations management, however, I would argue that nothing will improve unless it is measured.

If you are to successfully measure something for improvement, then three conditions must exist:

  1. We must have at least one consistently defined attribute
  2. We must have a way to consistently measure that attribute
  3. We must have a way to compare that measurement, either to a baseline or to a peer attribute

Let me explain with an example. If I have a machine I am responsible for managing, then I need to know the “health” of that machine. Here, an attribute might be “output.” I can quantitatively and consistently measure output quite easily – at the end of the day, my machine produces 20 widgets. But, the question is whether 20 is good?  Should I be producing 25? Or 30? This question can only be answered by comparing my performance to others.

Defining an Attribute

Manufacturers are quite clever in defining and identifying attributes that cast a favorable light on our performance. Nuances best known by those on the shop-floor can be capitalized to help skew performance figures in the most positive light. Regardless, try to be objective and pick a group of attributes that the group can “own” and that are most relevant and applicable. The only possible problem here is if you select too many to measure, thereby diluting your ability to effectively manage these business drivers.

Measuring an Attribute

Here is where manufacturing IT systems come into play. With today’s increasingly global manufacturing operations, no longer is it viable to perform manual measurements – systems must be programmed to perform the necessary calculations. More importantly, these measurements must be done consistently. Here is where a robust analytics platform can be a real benefit, and can make a critical impact on your Plan-Do-Check-Act cycle. Further, a consistent perspective and understanding of the IT systems running in a manufacturing environment is important, so as to not only select the right attribute to measure, but to achieve consistent measurements on a go forward basis.

The following IT systems hierarchy map was prepared by our friends at LNS Research. It is a good illustration of how these systems should be organized, and what role each plays as part of an overall framework of applications running across manufacturing operations.

Establishing a Hierarchy Model for Manufacturing Analytics

 

With a consistent application architecture or framework, you will now be in a position to best accomplish the last step necessary for continuous process improvement – comparing your performance to others.

Comparing an Attribute

Here is where attention to detail and consistency will either “make” you or “break” you!

If you have bad data as the basis for your measurements, then we all know how that will come out. Now it is necessary to figure out how well you are doing, which can only be accomplished with a comparison – either to your own prior performance, or to that of the industry. Ideally, you can do both. If you have consistently architected your manufacturing IT systems according to your industry’s best practices (such as what is illustrated above), then you are more likely to have results that can be effectively compared to that of your peers.

How do you know if your analytics measurements and strategies are consistent with that of your peers? And, what ways do you have to obtain market performance metrics such that you can effectively compare? One way is to purchase research reports. Another is to attend industry conferences and events, such as the Manufacturing Analytics Summit being held in Chicago on May 21 & 22. Manufacturing Transformation is a media partner for this event, so can offer you 20% (use this code: MFG20).

Regardless of how you collect your market data, be sure to figure out some way to do so. Then, you will have all the pieces for your own Plan-Do-Check-Act cycle, and the sky is the limit!

Permanent link to this article: http://www.apriso.com/blog/2014/04/establishing-a-hierarchy-model-for-manufacturing-analytics/

Leave a Reply

Your email address will not be published. Required fields are marked *


+ 3 = eleven

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>