Predictive Modeling Example: Process Stability and Predictability For Non-Normal Response Data

By in ,
95

This PDF article provides a predictive modeling example for assessing the stability of a process and then provide a prediction statement for non-normal response data.

The high level goal of business should be the creation of More Customers and Cash. An organization’s Existence and Excellence depend upon it! Similar to Albert Einstein’s famous equation we can express this business need as E=MC**2. Management by hope does not lead to MC**2. To achieve MC**2, organizations need to orchestrate activities so that everybody is doing the right thing at the right time. If meaningful measures are not identified and monitored appropriately, management will not know if it is doing the right things correctly.

Behavior is driven not only by what is measured but also by how the measurements are tracked and reported. Organizations need measurements that lead to the orchestration of activities that align with achieving MC**2.

MC**2 can be achieved through:

Predictive Modeling Example: Process Stability and Predictability For Non-Normal Response Data

A predictive modeling illustration of a 30,000-foot-level report for non-normal response data is:

 

predictive modeling of lognormal data

 

In a 30,000-foot-level report:

  1. Process stability is assessed through evaluation of the individuals chart (left graph).  Since no point is outside upper or lower control limit, the decision is made that the process is stable. When processes are stable we conclude that they are predictable.
  2. The probability plot (right graph) is used to determine process capability against the specifications for the example of 0.5-7.0 for this example.
  3. The determined non-conformance rate from the probability plot analysis is reported out at the bottom of the 30,000-foot-level report-out.
  4. If the rate of non-conformance is not satisfactory, a process improvement undertaking is need.
  5. A 30,000-foot-level report that has an individuals chart which transitions to an enhanced level of performance is an indication that the process response changed because of improvement efforts.  CLICK HERE to see an example of a 30,000-foot-level report-out transition.

For reference, the data for creating the above 30,000-foot-level report above was generated randomly. This random generation of data resulted in the following data distribution from which this 30,000-foot-level report-out was created:

 

predictive modeling example data set

 

Applications and Additional Information about 30,000-foot-level Reporting

  • The Integrated Enterprise Excellence (IEE) Business Management System provides the framework for reporting the predictive analytics of 30,000-foot-level metrics throughout an organization, where there is a structured linkage of these futuristic measurement from the processes that created the metrics.
  • An IEE 5-book set provides the details of implementing the IEE system and its 30,000-foot-level reporting
  • Enterprise Performance Reporting System (EPRS) software provides the vehicle for providing predictive analytics for process response outputs that have automatic updates.
  • A free Minitab add-in is available for the creation of 30,000-foot-level report-outs is available through the EPRS Client link.

The IEE system provides a predictive modeling approach for addressing the business scorecard and improvement issues described in a one-minute video:

 

predictive modeling video

 

The link below provides a PDF that contains additional information on predictive modeling, the IEE system, and 30,000-foot-level reporting.

 

Contact Us to set up a time to discuss with Forrest Breyfogle how your organization might gain much from an Integrated Enterprise Excellence (IEE) Business Process Management System and its predictive modeling techniques.

Download