Print Page | Close Window

Six Sigma Process Improvement -- (D)MAIC

Printed From: One Stop Testing
Category: Quality Assurance @ OneStopTesting
Forum Name: Quality Methodologies / Streams @ OneStopTesting
Forum Discription: Any Good Testing Engineer must know about All the Quality Certifications & Methodologies like ISO, IEEE, CMM, PCCM, CMMMi, XP, Agile and many more.
URL: http://forum.onestoptesting.com/forum_posts.asp?TID=3156
Printed Date: 23Dec2024 at 3:42am


Topic: Six Sigma Process Improvement -- (D)MAIC
Posted By: tanushree
Subject: Six Sigma Process Improvement -- (D)MAIC
Date Posted: 23Oct2007 at 7:03am

Six Sigma Process Improvement -- (D)MAIC


A more quantitative version of Deming's PDCA (Plan-Do-Check-Act) Process Improvement methodology was developed to implement this statistical approach -- it is commonly referred to as MAIC.

    MAIC
  • Measure
  • Analyze
  • Improve
  • Control
Key product-process performance variables are measured, analyzed, improved, and controlled using statistical methods. The simple "statistical" quality tools that were popularized in the Total Quality era are reinforced with Design of Experiments (DOE) and more sophisticated Statistical Process Control techniques.

Process sigma is the primary unit of measure. It is determined from an analysis of the number of defects observed in a process. Performance is compared to the Best-In-Class sigma for that process to determine whether the process needs to be improved or the product / service needs to be re-designed. When improvement is necessary, Design of Experiments (DOE) are used to determine which product or process parameters are most important and specific parameter values that will give the best performance. SPC is used to continually monitor product and process performance.

DMAIC Similar to the problem-solving models where an initial step to define the problem was frequently added, some practitioners prefer to precede MAIC with a Define step. They feel that selecting and defining the right process is critical. Effort can easily be wasted working on poorly selected, ill-defined processes -- as illustrated by many TQM failures.


Print Page | Close Window