Measurement

The normal path for a six sigma project is through DMAIC.  An interesting thing frequently occurs during the Measure step in the project.  It is not unusual; in fact it is quite common, for the performance metrics for the project to improve without any known action by the project team other than to following the process and start to collect data.

We have seen this happen even when the measurement area is not part of a six sigma project.  Quite often the fact that something starts to be measured is enough for a positive change in the performance. While not always the case this is often just the Hawthorne Effect in action.  When management pays attention to a specific area or to workers, in general, the response is to try to improve.  Measurement in secret or unknown to the workers involved seldom has the improvement effect.

Of course you can add some inducement to the effort.  As an example: In a unit with a history of spending beyond budget every month, a graph was posted showing budgeted spending on a straight line basis for the month.  Each day the actual cumulative total spending for the month was added to the graph.  The manager of the unit let everyone know that as long as the cumulative spending was less than the budget line no action by management would be taken.  If however the spending ever crossed the budget line there would be personal “management help” to keep costs under control.  In a unit that had struggled for years to maintain effective cost control, 36 consecutive months of within budget spending were achieved with no other management action.  Setting the standard and making sure it is know and then adding rewards – or in this case a lack of punishment “management help” – may be all that is needed to make substantial improvement in performance.

Not always will the simple act of measurement cause sufficient improvement and process improvements and control that arise from the remaining six sigma steps are required.  

Measurement remains a key step in any six sigma project. Without measurements you will not have effective demonstration of the performance at the beginning of the project.  Any improvements with then fall into the “I think, I feel” category rather than being data supported facts.

Measurement is much more than the mere collecting of data.  Issues of bias, linearity, repeatability, variation, etc. all may have to be addressed.  Measurement remains a key and is appropriate as soon as a project is defined – in fact measurement of some type  may be required before the define step is started. 

There is the old saying:

If you can’t measure it

You can’t manage it

And if you can’t manage it

You can’t control it

Adams Associates
10A Bayou Rd
Lake Jackson, TX 77566 
1- 979-297-5198

cadams@adamssixsigma.com