Why Control Chart Your Processes?

Control charting is a tool used to monitor processes and to assure that they remain in control or stable. Proposed by Walter Shewart in 1924, control charts help distinguish process variation due to assignable causes from those due to unassignable causes.

Table 1: Types of Process Variation Found on a Control Chart
Types of Variation Variation Characteristics
Assignable cause, also known as special cause
  • Meaningful factors of process; not always present
  • Cause can be avoided and should be investigated
  • Not normal to process
Unassignable cause, also known as common cause or chance cause
  • Factor caused by chance; always present
  • Unavoidable and inherent in a process
  • Normal and expected within process

Elements of a Control Chart

A control chart consists of:

  1. A central line,
  2. An upper control limit,
  3. A lower control limit, and
  4. Process values plotted on the chart.

If all process values are plotted within the upper and lower control limits and no particular tendency is noted, the process is referred to as in control. If the process values are plotted outside the control limits or show a particular tendency, however, the process is referred to as out of control (see red-circled data points in Figure 2 below).

Figure 1: In Control Process Control Chart

Figure 1: In Control Process Control Chart

Figure 2: Out of Control Process Control Chart

Figure 2: Out of Control Process Control Chart

There are many types of control charts. In a future article, we will investigate the different types of control charts by process values and purpose.

Handpicked Content:   Leverage Six Sigma to Manage Operational Risk in Financial Services

In a control chart, control limits are calculated by the following formula:

(average process value) (3 x standard deviation)

where the standard deviation is due to unassigned process variation only.

Constructing a Control Chart

Here is the general process for making or constructing a control chart for your process:

  1. Select the process you would like to chart
  2. Determine your process sampling plan
  3. Collect data from your process
  4. Calculate the control chart specific statistics
  5. Calculate your control limits
  6. Construct your control chart

Comments 4

  1. Rosalie Dimen

    What is the difference of special cause variation between subgroups and special cause variation within subgroups?

  2. Anshuman

    special cause variation between subgroups happens when unexpected variation occurs due to some unforeseen circumstance in one group and the other group goes unaffected. The key is that both the groups have to have profiles similar in nature. their Key process indicators and goals have to be similar and their deliverables are exactly the same. For example Dell has two call centres in India and Philippines. due to heavy rains there was huge employee outage in the indian call centre for a day. The philippino call centre had no such issues for the very same day. Hence when you look at the Average call handle time, you would see there were significantly high call drops(customer calls not picked up within the stipulated timelines) in the Indian call centre due to not enough executives being present.
    For the second part of the question, lets have the same example and look at the indian call centre for average calls handled in a day. Employees living closer to the office(A) reached on time however people living away from the office(B) managed to reach 2 hours late. The productivity of A would be 25% higher than B(8 hour day, A having 2 hour advantage).
    We can include a control chart with UCL and LCL and this very example would show as a gap when comparing both the call centres on weekly performance. The variation is the variation caused by special cause with two subgroups.
    A daily productivity analysis would show B under the LCL due to the special cause variation.

  3. zulfiqar Parkar

    what is significance of rule of seven ?

  4. Name (Required)

    I have been discussed about control used in investigation in quality risk management

Leave a Reply