|
|
|
Control Chart
What is it?
A Control Chart is a tool you can use to monitor a process. It
graphically depicts the average value and the
upper and lower control limits (the highest and lowest
values) of a process.
Who uses it?
The management, the team.
Why use it?
All processes have some form of variation. A Control Chart helps you
distinguish between normal and unusual variation in a process. If you
want to reduce the amount of variation in a process, you need to
compare the results of the process with a standard.
Variation can exist for two reasons:
- Common causes
are flaws inherent in the design of the process.
- Special causes
are variations from standards caused by
employees or by unusual circumstances or events.
Most variations in processes are caused by flaws in the system or the process, not by the employees. Once you realize this, you can stop blaming the employees and start changing the systems and processes that cause the employees to make mistakes. (It is important to remember, however, that some variations are not "mistakes" introduced by employees, but, rather, they are innovations. Some variations are deliberately introduced to processes by employees specifically because these variations are found to be more practical.)
When to use it?
First, you need to define the standards of how
things should be. Then, you need to monitor (collect data) about processes in your organization. Then, you create a control graph using the monitoring data.
How to use it:
- Select the process to be charted and decide on the type of control
chart to use.
- Use a Percent Nonconforming Chart
(more information available from Health Tactics P Chart) if you
have data measured using two outcomes (for example, the billing can be correct or incorrect).
- Use an Average and Range Control Chart
(more information available from Health Tactics X-R Chart)
if you have data measured using a continuous scale (for example, waiting time in the health center).
- Determine your sampling method and plan:
- Choose the sample size (how many samples will you obtain?).
- Choose the frequency of sampling, depending on the process to be
evaluated (months, days, years?).
- Make sure you get samples at random (don't always get data from the
same person, on the same day of the week, etc.).
- Start data collection:
- Gather the sampled data.
- Record data on the appropriate control graph.
- Calculate the appropriate statistics (the control limits) depending on the type of graph.
- Observation:
The control graph is divided into zones:
______________________________ Upper Control Limit (UCL)
______________________________ Standard (average)
______________________________ Lower Control Limit (LCL)
- Interpret the graph:
- If the data fluctuates within the limits, it is the result of common
causes within the process (flaws inherent in the process) and can only be affected if the system is improved or changed.
- If the data falls outside of the limits, it is the result of special
causes (in human service organizations, special causes can include bad instruction, lack of training, ineffective processes, or inadequate support systems).
- These special causes must be eliminated before the control chart can be used as a monitoring tool. In a health setting, for example, staff may need better instruction or training, or processes may need to be improved, before the process is "under control." Once the process is "under control," samples can be taken at regular intervals to assure that the process does not fundamentally change.
- A process is said to be "out of control" if one or more points falls
outside the control limits.
Click here to learn more about statistical process control from Health Tactics.
|
Case Study
Welcome
