Working with Ideas

1
Working with Ideas

• Brainstorming
• Cause Effect Diagram
• Force-Field Analysis
• Affinity Diagram

RD120101
2
Learning Objectives

• In this module, we will become familiar with
  tools that help us
• Generate ideas to help improve our process.
• Organize our ideas so we can understand them.
• Prioritize our ideas so that we can get the most
  leverage from them.

3
Generating, Organizing, and Assessing Ideas

• We may need to generate ideas to help us
  determine where to focus improvement efforts
  related to a broader goal.
• Helps us collect and prioritize ideas for
  improvement.
• Helps us identify the right ideas that impact
  our key measures.
• We may also need to generate ideas to help us
  identify causes after a specific improvement
  effort is underway.
• Allows us to take a broad goal or scope and
  narrow it down to a manageable collection of
  highly-leveraged ideas for improvement.

4
Application Examples

• Manufacturing The manufacturing manager of a
  hose extrusion operation wants a list of possible
  causes for a recent batch of product that didnt
  meet adhesion specs.
• Transactional Due to a successful new marketing
  initiative, a motor company must quickly think of
  ways to meet the increased demand for quotations.
• Product Development / Revision A team needs to
  identify product design criteria to meet target
  cost and ease manufacturability.
• Manufacturing A team is trying to find ways to
  reduce scrap in a light machining-and-assembly
  process.

5
Basic Tools
Idea Generation and Assessment
6
What are the Basic Analytical Tools?

• Idea Generation / Prioritization
• Brainstorming
• Cause Effect (CE / Fishbone) Diagrams
• Nominal Group Technique (NGT)
• Multi-voting
• Affinity Diagrams

• Idea Analysis / Assessment
• Force-Field Analysis
• Pareto Analysis
• Check Sheets
• Cause Effect Matrix
• FMEA (Failure Modes Effects Analysis)

Not presented in this module.
7
Brainstorming

• What
• A structured method of generating unconstrained
  ideas and solutions, and gaining individual
  engagement or involvement in the improvement
  process.

8
Brainstorming (con't)

• Why
• Produces many ideas / solutions in a short time.
• Facilitates the creative thinking process.
• Separates idea generation from the organizing /
  assessment of the ideas.

Its important to augment and harness the
creativity of all participants in order to find
the best solution to a problem. In many
instances, this idea only comes about through
solid, group-based brainstorming.
9
Brainstorming (con't)

• How
• Review the problem definition.
• Clarify the goal / question and provide relevant
  information.
• Offer background, not possible solutions that may
  bias the group!
• Encourage fun and creativity.
• Start with a 5-minute practice session on
  something fun.
• Give everyone a few minutes of silence to think
  about the question and individually write down
  some ideas.
• Gather the ideas, round-robin, one at a time.
  Write these on a flip-chart and post them.
• (Important Allow no discussion of any ideas
  until the documentation session is complete!)
• Write down every idea (even if there are
  repeats!) without editing their words or ideas.

10
Brainstorming (con't)

• How (cont)
• Encourage participants to continue to write down
  additional ideas as they think of them.
• Continue round-robin-style until everyone is out
  of ideas.
• Consolidate similar ideas and discuss the
  complete list.
• Clarify the ideas with questions and ask for more
  specific information where necessary.
• Use other basic tools to assist in
  prioritization of ideas.

Maximize idea generation! Quantity (not quality,
not yet) is key. Create first, analyze
later. Never allow anyone to shoot down another
persons ideas!
11
Learn by Doing! Brainstorming Exercise

• How could we improve Buncombe or Madison County?
• Well spend the next 3 minutes and brainstorm to
  answer

Remember There are NO wrong answers in
brainstorming!
12
Brainstorming Conclusions

• A structured method of generating unconstrained
  ideas and solutions and gaining engagement /
  involvement in the improvement process.
• Produces many ideas / solutions in a short time.
• Facilitates the creative thinking process.
• Separates idea generation from the organizing /
  assessment of the ideas.

13
Cause-and-Effect Diagram(Fishbone or Ishikawa)

• What
• Represents the relationship between an effect
  (problem) and its potential causes. Categorizes
  causes.

14
Cause-and-Effect Diagram

• Why
• To help ensure that a balanced list of ideas have
  been generated during brainstorming.
• To sort and relate the factors affecting a
  process when little quantifiable data is
  available.
• Helps serve as a discussion guide to assist in
  determining root causes.
• To determine the real cause of the problem,
  versus solely identifying a symptom.
• To refine brainstormed ideas into more detailed
  causes.
• To identify a team's level of understanding.

15
Cause and Effect Diagram (con't)

• How
• Name the problem or effect of interest.
• Effect Excess length of cut wooden frames
• Decide the major categories for causes.
• Major causes may include the 6 Ms or 4 Ps
• Manpower (or personnel) 1. Policies
• Machines 2. People
• Materials 3. Procedures
• Methods 4. Plant / Technology
• Measurements, and
• Mother Nature (or environment).
• Major causes may also include the process steps

16
Cause and Effect Diagram (con't)

• How (cont)
• Brainstorm for more detailed causes. Ask "why"
  each major cause happens at least 5 times (follow
  up each answer with a question about the answer).
• Effect Excess length of cut wooden frames.
• Cause The operator has a hard time setting the
  correct length.
• Why does the operator have a hard time setting
  the correct length?
• Because the measurement gauge is hard to
  read.
• Why is the gauge hard to read?
• Because its too dark in the work area.
• Why is it too dark in the work area?
• Because 2 of the 4 light sources are broken.
• Why?
• Eliminate causes that do not apply.

17
Cause and Effect Diagram (con't)

• How (cont)
• Discuss the causes and decide which are most
  important.
• Work on the most important root causes.
• Brainstorm for more ideas in categories
  containing fewer items. (This helps counter the
  theme effect common in brainstorming by forcing
  consideration of other topics.)
• Perform another iteration to determine root
  causes, if necessary.

18
ExampleCause and Effect Diagram
6 Ms Man Material Machines
Methods Measurements Mother Nature
4 Ps Policies People Procedures /
Process Plant / Technology
19
Refine Brainstormed Ideas to the Root CauseAsk
"Why?" 5 Times
Flow Rate Varies 1. Why? Tank pressure drops 2.
Why? Tank becomes empty 3. Why? Tank not
changed promptly 4. Why? Nothing prompts
operator to change tank 5. Why? "Because
we haven't installed a gauge"
________________ 1. Why? 2. Why? 3. Why? 4.
Why? 5. Why?
20
Cause-and-Effect Diagram Conclusions

• Represents the relationship between an effect
  (problem) and its potential causes. Categorizes
  causes.
• Helps ensure that a balanced list of ideas have
  been generated during brainstorming.
• Helps us overcome the theme effect (or piling
  on).
• Sorts and relates the factors affecting a process
  when little quantifiable data is available.
• Serves as a discussion guide to assist in
  determining root causes.
• Helps determine the real cause of the problem
  versus a symptom.
• Helps us refine brainstormed ideas into more
  detailed causes.
• Helps us identify a team's level of understanding.

21
Force-Field Analysis

• What
• A tool to assist in examining the driving and
  restraining forces of change.
• A tool to help a team understand the forces that
  keep things the way they are.

22
Force-Field Analysis (con't)

• Why
• To force creative thinking focused on the issues
  of change.
• To build organizational consensus concerning the
  fuel for and the barriers to change.
• To provide an entry point into process
  improvement initiatives.

23
Force-Field Analysis (con't)

• How
• List all of the driving forces and all the
  restraining forces to change. Brainstorming and /
  or nominal group technique can be used to assist
  in list development.
• It may be useful to assign weights to the drivers
  and restraints to indicate the relative strengths
  of each.
• Establish a plan to eliminate or reduce all
  restraining forces.
• Market and use the driving forces in your
  implementation planning.

24
Force-Field Analysis Example
Buying a Car
From The Complete Guide to Six Sigma, p. 304, by
Thomas Pyzdek
25
Force-Field Analysis Conclusions

• Assists in examining the driving and restraining
  forces of change.
• Helps a team understand the forces that keep
  things the way they are.
• Forces creative thinking focused on the issues of
  change.
• Builds organizational consensus concerning the
  fuel for and the barriers to change.
• Provides an entry point into process improvement
  initiatives.

26
Affinity Diagram

• What
• A tool for organizing facts, opinions and issues
  into natural grouping as an aid to diagnosing a
  complex problem. The inputs are listed on cards
  which are then rearranged until useful groups are
  identified.

27
Affinity Diagram (con't)

• Why
• When breakthrough is needed.
• To help organize.
• To help develop central themes.
• Facts on a problem are not well organized.
• Breakthrough beyond traditional thinking is
  needed.

28
Affinity Diagram (con't)

• How
• Assemble the right team.
• Clearly state the problem to be addressed.
• Brainstorm ideas and place them on index cards
  (or Post-It notes).
• Clearly display the cards on a wall as ideas are
  generated.
• Without talking, have the team sort the cards
  into related groups.
• In many cases, cards will be moved from group to
  group, because members will have different
  perspectives!
• When finally grouped, create header cards for
  each group.
• Draw the completed diagram.

29
Affinity Diagram Conclusions

• A tool for organizing facts, opinions and issues
  into natural grouping as an aid to diagnosing a
  complex problem.
• Helpful when breakthrough is needed.
• Helps organize ideas and / or facts.
• Allows the development of central themes.

30
Takeaways

• Proper brainstorming is the backbone of good idea
  generation.
• A cause effect diagram helps us balance our
  brainstorming among the many categories and helps
  us visualize relationships.
• Force-field analysis helps identify the driving
  and restraining forces of change so that we can
  better facilitate change.
• Affinity diagrams organize and group ideas to
  help us get a better grasp on complex processes
  or large numbers of ideas.

31
Takeaways (con't)

• Using the tools covered in this module can help
  us collect and prioritize ideas for improvement.
• These tools help us identify the right ideas
  that impact our key measures and results.
• These tools allow us to take a broad goal or
  scope and narrow it down to a manageable
  collection of highly leveraged ideas for
  improvement.
• It is not always necessary or even useful to use
  ALL of the basic tools in this module for any
  given problem.

32
Handling andAnalyzing Data

• 3 Steps to Analyze All Data
• Stratification and Segmentation
• Graphical Displays of Data

RD112801
33
Studying the Data

• LSS and breakthrough improvement rely on
  data-driven decision-making, not just gut
  instinct or guesswork
• Collecting and analyzing data are integral
  components of the improvement process. We need
  to analyze data to see how the data support or do
  not support our hypotheses.
• But, before moving to conclusions and
  implementing radical change, you need to verify
  that the data is
• Accurate and Precise (subject to little
  measurement error)
• Unbiased and Representative (across the spectrum
  of experiences)
• Reliable (properly collected and recorded)
• Informative (provides insight into inputs,
  processes or outputs)
• Understandable by others (to improve
  understanding and buy-in)

34
Analytical Overview

• 1. Scan the Data Visually
• 2. Show the Data Graphically
• 3. Describe the Data Analytically

35
Case Study

• Take a look at the following survey data

36
Overview Look at the Data Visually!

• 1. Scan the Data Visually
• What kinds of data have been collected or
  presented?
• Categorical Data in mutually exclusive
  categories (e.g., male female)
• Continuous Data with values across a continuum,
  from a minimum to a maximum value (e.g., 0 to
  10000, where any number within the range is
  possible, even something like 116.68 feet)
• Discrete Data with values across a continuum,
  but where only certain values are possible (e.g.,
  number of individuals you cannot have a ¼ of a
  person! survey responses using letters of the
  alphabet)
• How many samples were collected? (sample size)
• When and how frequently were the data collected?
  (time period)
• Who collected the data? (data recorders)
• How was the data collected? (data forms)
• Where and from whom was the data collected? (data
  sources)

37
Overview Look at the Data Visually!

• 1. Scan the Data Visually (cont)
• Are there any gaps or duplications in the data?
• Is the data recorded in the right place?
• Are there any obvious recording errors or
  patterns in the data?

Recv'd User Size Day 1 R1 L Day 1 R15 M Day
1 R12 L Day 1 R6 S Day 1 R7 L Day 2 R2 M Day
2 R5 S Day 2 R14 M Day 2 R11 L Day 2 R3 S Day
3 R4 S Day 3 R8 L Day 3 R9 S Day 3 R10 M Day
3 R3 S
Day 1 R6 S 3 2 3 3 -2
38
Overview Look at the Data Graphically!

• 2. Show the Data Graphically
• A graph is a visual representation of a data set.
  Graphing helps us analyze data we have
  collected.
• A graph shows a relationship between 2 or more
  quantities.
• Graphs can demonstrate aspects of the data more
  clearly and more meaningfully than solely
  presenting the collected data in a table.
• In order to graph well, you must be able to
  choose appropriate graphs for particular data
  sets. The most commonly used graphs
• Line Graphs / Time-Series / Run Charts
• Bar Graphs / Histograms / Pareto Charts
• Scatter Plots

These graphs will be described in the following
pages.
39
Time Series / Run Charts

• Why Use It?
• To study observed data for trends, cycles,
  patterns, or shifts in patterns over a specified
  time. Useful to discover patterns that occur
  over time
• Useful in demonstrating before vs. after
  comparisons.
• How do I Interpret It?
• Look for obvious patterns or changes in results
  over time.
• If there are no obvious trends, calculate the
  mean and draw it into the graph. (Hint Do not
  redraw the mean every time you add data unless
  there is a significant change.) Is the average
  where it should be relative to customer needs or
  expectations?

Also known as Line Graphs or Time Plots
40
Time Series / Run Charts (cont)

• Tips in Usage
• Draw vertical lines across the x values to
  separate time intervals such as weeks
• Draw horizontal lines across the y values to show
  where trends in the process or operation occur or
  will occur
• Add trend lines to the graph to more easily
  demonstrate overall change and to forecast future
  results
• Add lines for baseline performance and/or
  averages
• Note peaks and valleys in the graph for further
  investigation

41
Time Series / Run Charts (cont)

• Where do we use Run Charts in our plant?

42
Histograms

• Why Use It?
• To summarize data and graphically present its
  frequency distribution.
• To evaluate process centering, spread or
  variation, and shape (accuracy and precision).
• What Questions the Histogram Answers
• What is the most common response or result?
• What distribution (center, variation, shape) does
  the data have?
• Does the data look symmetric or is it skewed to
  the left or right?
• Does the data contain outliers (extremes in the
  data points)?

A Bar Graph is similar to a histogram, except
that, instead of intervals for one response, each
group receives its own bar (e.g., machine 1,
machine 2, etc.)
43
Histograms (cont)

• How do I Interpret It? (cont)
• The intervals (or categories) are shown on the
  X-axis and the number of scores in each interval
  is represented by the area (or height) of a
  rectangle located above the interval.
• Look for significant differences in the area (or
  height) of the bars. These indicate categories
  or intervals that occur more frequently than
  others.
• Is the process capable of meeting my customers
  needs?
• Where is the distribution centered? Is the
  process running too high? Too low?
• What is the variation or spread of the data? Is
  it too variable?
• What is the shape? Are there multiple peaks
  (suggesting the existence of different sources
  and the need to stratify)?
• Be suspicious if the histogram suddenly stops at
  one point (such as a specification limit). Do
  you have all the data?

44
Histograms (cont)

• A Special Case Pareto Charts
• A Pareto Chart is a special form of a histogram.
  A Pareto Chart is used to graphically summarize
  and display the relative importance of the
  differences between groups of data.
• In a Pareto Chart, the categories are re-ordered,
  so that the most frequent categories are shown to
  the left side of the histogram.
• What Questions the Pareto Chart Answers (80/20
  Rule)
• What are the largest issues facing our team or
  business?
• What 20 of sources are causing 80 of the
  problems?
• Where should we focus our efforts to achieve the
  greatest improvements?

45
Histograms (cont)

• Where do we use Histograms and Pareto Charts in
  our plant?

46
Overview Look at the Data Graphically!

• 2. Show the Data Graphically (cont)
• Is there obvious stratification, segmentation or
  grouping?
• Are there any errors, patterns or correlations
  visible in the graphs?

47
Stratification

• Technique used in combination with other data
  analysis tools to separate the data in order to
  better see patterns.
• When data come from several sources or
  conditions, such as from different shifts, days
  of the week, lines, suppliers or populations.
• Line 1 vs. Line 2 vs. Line 3 vs. Line 4 scrap
  rates, production yields, and per-hour
  productivity
• Day Shift vs. Evening Shift vs. Swing Shift
  efficiency
• Machine A vs. Machine B setup times, cycle times,
  and batch sizes
• Method 1 vs. Method 2 assembly processing times
• Supplier A vs. Supplier B vs. In-house parts
  quality

48
Segmentation

• Technique, similar to stratification, to separate
  the data from different groups in order to better
  see patterns or cause-and-effect relationships.
• Segmentation Examples
• Large Customers vs. Small Customers
• East Coast vs. West Coast vs. Midwest vs. Deep
  South pricing
• Region 1 vs. Region 2 vs. Region 3 product
  returns
• U.S. Supplier Quality vs. Canadian Supplier
  Quality
• Masters Degree Graduates vs. Bachelors Degree
  Graduates vs. Associates Degree Graduates vs.
  High School Graduates

49
Overview Look at the Data Analytically!

• 3. Describe the Data Analytically
• Transforming data into information starts with
  summary statistics. With these statistics, we
  can then compare and relate things to one
  another.
• Describe the data, overall and using
  stratification or segmentation, using Basic
  Statistics to explain and understand the
  variation in a process that is under study
• Mean (arithmetic average)
• Median (50th percentile)
• Mode (most frequent observation)
• Min (lowest point)
• Max (highest point)
• Range (Max Min)
• Standard Deviation

On average
Variation or Degree of spread
50
Overview Look at the Data Analytically!

• 3. Describe the Data Analytically (cont)
• Use 1, 7, 3, 5, 9, 7, 12, 15, 6, 8, 4, 7, 3, 5,
  6 (i.e., 15 data points)
• Mean (arithmetic average)
• Procedure Add up all of the numbers in the list
  or data set, and divide by the number of numbers
  in the list.
• Example (1, 7, 3, 5, 9, 7, 12, 15, 6, 8, 4, 7,
  3, 5, 6) / 15 6.5333
• Median (50th percentile) Reduces the impact of
  outliers (extreme data points) as compared to use
  of the mean.
• Procedure Sort the list of data points in
  ascending order. Take the middle data point in
  an odd number of numbers in even number of
  numbers, take the arithmetic average of the two
  middle numbers.
• Example 1, 3, 3, 4, 5, 5, 6, 6, 7, 7, 7, 8, 9,
  12, 15 6

51
Overview Look at the Data Analytically!

• 3. Describe the Data Analytically (cont)
• Use 1, 7, 3, 5, 9, 7, 12, 15, 6, 8, 4, 7, 3, 5,
  6 (i.e., 15 data points)
• Mode (most frequent observation)
• Procedure Sort the list of data points in
  ascending order. Find the data point which
  occurs most frequently.
• Example 7
• Min (lowest point)
• Procedure Sort the list of data points in
  ascending order. Take the smallest data point.
• Example 1
• Max (highest point)
• Procedure Sort the list of data points in
  ascending order. Take the highest data point.
• Example 15

52
Overview Look at the Data Analytically!

• 3. Describe the Data Analytically (cont)
• Use 1, 7, 3, 5, 9, 7, 12, 15, 6, 8, 4, 7, 3, 5,
  6 (i.e., 15 data points)
• Range (Max Min) (a description of spread or
  variation in the data, but impacted by outliers
  or extreme data points)
• Procedure Subtract the Min from the Max
• Example 15 1 14
• Standard Deviation (average distance of each data
  from the mean of the data set)
• Procedure Use Excel or a statistics package to
  calculate
• Example 3.5630
• Standard Deviation is a statistic used to measure
  the variation in a data sets distribution. The
  measure is presented as a measure of the spread
  in the data in relation to the mean. A process
  with a smaller standard deviation is more
  consistent in results than a process with a large
  standard deviation.

53
Overview Look at the Data Analytically!

• 3. Describe the Data Analytically (cont)
• Compare actual results against the expectations.
• Does the data agree with your suspicions or
  hunches?
• Why? Why not?
• What could be driving these results?
• How do the variables (inputs) relate to the
  process results (outputs)?

54
Takeaways

• Stratify and segment your data to help make
  patterns in the data more visible.
• Before moving on your data, take time to review
  the quality of the data and its implications
• Visually
• Graphically
• Analytically
• Use various graphical and statistical tools to
  summarize the process data and results.