Control Charts are the most commonly used tools of Statistical Quality control to assess if a process is in control or not. If one looks at the Stock Market in the last six-eight months, one can easily comment that it has been out of control! Let us try to understand what a control chart is and what its applications in manufacturing industries are.
Typically, a Control Chart is a graphical representation of a business process and explains whether the process is in a state of control or not. Let us consider a few hypothetical examples. A lot of consumers started complaining that they were being charged the value of two items when they bought one at a Target store. This was an unusual case and a deep analysis into the case revealed that one of the check-out machines was malfunctioning. Over thirty five consumers were refunded the amount that they were charged due to the error and given a complimentary meal at the restaurant in Target. The situation was out of control by all standards and the situation was shared with all the stores in the country. A pen manufacturing company reported that in the last few months the number of defectives had increased by over four times. This was clear as the Control Charts used to monitor the number of defectives each week clearly told their story. A deep analysis for the causes of this variation led to the discovery that the pen nibs were not of good quality and therefore, compromised the writing ability of the pen itself.
It is clear from the above two examples that it is impossible to maintain the stability of the process under all circumstances. Things do get out of hand, and necessitate action. The question is when. Control Charts help a user to observe how consistently his process is in control, and if not, what are the actions that he may take to correct the situation (this is not the primary goal of a Control Chart). The objective of constructing a Control Chart is to understand if there are chances of variability in any process, and how a diagram can solve this problem. To have realistic results, instead of comparing the process values to a single line (denoting the averages), it is better to have an upper control limit and a lower control limit. After the values of the process have been plotted on such a chart, it can be concluded whether the process is in a statistical state of control or not. It is suggested that 95% or 99% of data should fall within the control limits. Values outside the control limits mark statistically significant changes and may indicate a change in the underlying process. Many changes are controllable and can be attributed to change of material quality, change of craftsman or operator, slight change of production processes etc. There are other changes which are not so random and may be caused due to special reasons. They are called uncontrollable causes and should be dealt with care. The present scenario in the US Stock Market is a classic example of stocks being totally out of control for the past six to eight months.
A simple Control Chart has the following components:
• An X-axis showing time periods
• A Y-axis showing the observed values
• UCL line shows the Upper Control limit
• LCL line shows the lower control limit
Control Charts help a user to identify special causes of variation, shifts, trends and cycles. An ideal example in this context can be a seismograph. Earthquakes generate seismic waves which can be detected with an instrument called seismograph. Yet if you look at the drawings with a Statistician’s eye, it does look like a Control Chart and one does not wish to see it out of control!
Learn more information about 6Sigma.com’s Lean Six Sigma training coursework, available as classroom, onsite, or online options.
No responses / comments so far.