Systems Thinking and Breakthrough Performance

1
Systems Thinking and Breakthrough Performance
2
Definition of a System

• A system is defined as a group of
  interdependent people, items, processes,
  products, and services that have a common purpose
  or aim.
• - Deming

3
Understanding Run Charts

• Run Chart - a line graph of data plotted over
  time. By collecting and charting data over time
  we find trends or patterns in the process, or
  process performance over time
• Run charts cannot tell you if a process is
  stable. However, they can tell you if the
  process is running.

4
Lesson Learned from Applying a Systems Philosophy

• All processes have inherent random variability
  common cause variability
• Unexpected events can result in special cause
  variation
• A process is in control if it exhibits only
  common cause variation Stable Process
• A process is out of control if it exhibits
  special cause variation Unstable Process

5
Elements of a Control Chart

• Upper Control Limit
• Lower Control Limit
• Average
• Performance over time

6
Remember a Control Chart

• Is a type of chart with statistically determined
  upper and lower controls limits
• Is used to monitor and control processes by
  showing variation over a period of time
• Allows you to analyze the data to determine if
  variation is due to common cause or to a special
  cause

7
Special Note

• Control limits should not be interpreted as
  desired goals or standards
• Control limits are merely arbitrary calculations
  based on the data from the process

8
Steps for Creating Your Team Control Chart

• Find the average of the four data points totals.
  This will be referred to as the AS (Average
  Score)
• Total
• Total Average Score Centerline
• Total
• Total

9
Use the BLUE marker and draw your
centerline/average on your control chart
10
Steps for Creating your Team Control Chart,
continued

• Calculate the range between each pair of scores.
  There will be one less range than there are
  number of data points. Dont worry about negative
  numbers.
• Score
• Range
• Score
• Range
• Score
• Range
• Score

11
Steps for Creating your Team Control Chart,
continued

• Find the average of the ranges. This will be
  known as the Mean Moving Range (MMR)
• Range Range Range X ? 3 MMR

12
Steps for Creating your Team Control Chart,
continued

• Multiply the MMR by 3.14. This determines the
  distance of the control limits from the center
  line. This will be known as R (result)
• MMR x 3.14 R

13
Steps for Creating your Team Control Chart,
continued

• Calculate the Upper Control Limit (UCL)
• Add R (result) to AS (Average Score) you
  calculated in Step 1 to get the UCL
• R AS UCL
• Use the GREEN marker to draw the UCL

14
Steps for Creating your Team Control Chart,
continued

• Calculate the Lower Control Limit (LCL)
• Subtract R (result) from AS (Average Score) to
  get the LCL
• AS - R LCL
• Use the RED marker to draw the LCL

15
Applying What Weve Learned
16
Traditional Improvement

• - Higher performance at a later time period than
  at an earlier time period

Lessons Data points that fall randomly within
the control range indicate that most variation is
coming from a common cause. The only way to make
improvements in a system that only has common
cause variation is to fundamentally change some
aspect of the system (e.g., processes, materials,
procedures, etc.)
17
Continual Improvement

• A positive trend in performance

Lessons Data points that fall randomly within
the control range indicate that most variation is
coming from a common cause. The only way to make
improvements in a system that only has common
cause variation is to fundamentally change some
aspect of the system (e.g., processes, materials,
procedures, etc.)
18
Traditional and Continual Improvement are

• RANDOM

19
Breakthrough Improvement

• A significant performance improvement of a
  systems indicator value over time.

Lessons breakthrough improvement means
significantly better performance over time. This
pattern of variation indicates the presence of a
special cause that deserves further investigation.
20
Statistical Process Control for Analyzing Data
If non-stable in a beneficial way, how do we stay
that way? If non-stable in a detrimental way,
what can we do better?
Was the process in control?
Build Control Charts
Collect data
No
Yes
Is the process acceptable?
Set up monitoring Control Charts
Identify and improve process
Yes
No