This document describes a team-based variability reduction methodology used by a printing company to improve production processes. The methodology uses basic statistical tools within a 12-week roadmap to help teams reduce variability in key metrics. It was applied across 600+ teams, reducing variability by 2-15% on average. Lessons included the need for supervisor training, consistent auditing, and integrating the methodology into other improvement efforts. The benefits included improved conformance, efficiency gains, and sustained 4-10% increases in manufacturing throughput over four years.
2. Objectives
1. Show how to put data and basic problem-
solving tools in the hands of the
workforce to generate improved and
sustainable results
2. Provide a practical example of statistical
engineering.
4. Statistical Engineering
• Hoerl & Snee* define statistical engineering
as “the study of how best to use statistical
concepts, methods and tools and integrate
them with information technology and other
relevant sciences to generate improved
results.
* Hoerl, R. W., Snee, R. D. (2010). Closing the Gap: Statistical Engineering
Links Statistical Thinking, Methods, Tools. Quality Progress, 43(5): 52-53.
5. 5 Aspects Underlying Statistical Engineering*
1. A system or strategy to guide the use of
statistical tools is needed to effectively use the
tools.
2. The impact of statistical thinking and methods
can be increased by integrating several
statistical tools, enabling practitioners to deal
with highly complex issues that cannot be
addressed with any one method.
3. Linking and sequencing the use of statistical
tools speeds the learning of the approach,
thereby increasing the impact of the method.
* Hoerl, R. W., Snee, R. D. (2010). Tried and True: Organizations Put
Statistical Engineering to the Test. Quality Progress, 43(6): 58-60.
6. 5 Aspects Underlying Statistical Engineering*
4. Embedding statistical thinking and tools into
daily work institutionalizes their application.
5. Viewing statistical thinking and methods from an
engineering context provides a clear focus on
problem solving to the benefit of humankind.
* Hoerl, R. W., Snee, R. D. (2010). Tried and True: Organizations Put
Statistical Engineering to the Test. Quality Progress, 43(6): 58-60.
9. Which Process Would You Rather Have?
Process A
• Wide swings in process
performance
– Not predictable
– High costs for labor &
materials
– Constant fire-fighting
– Difficulty maintaining
improvements
Process B
• Swings in process
performance narrowed:
– Predictable
– Lower costs for labor and
materials
– More consistent process
output
– Effective corrective action
– Increased flexibility
10. Background
• Printing company with 20+ facilities in
North America.
• Began continuous improvement journey in
late 1990’s, including ISO 9000, workplace
organization (5S), problem-solving,
employee engagement and Six Sigma.
– Efforts were originally decentralized and
dependent on local leadership.
– Lack of standardization and stabilization
limited impact.
11. VR Methodology
• Design team chartered to develop a
common, team-based variability reduction
(VR) methodology October, 1999.
• Methodology defined as: A structured set
of tools, metrics, processes, and practices
to drive results in one or more parts of a
business.
12. VR Intent
• Squeeze the variability out of the process by
instituting a standard process, identifying and
eliminating sources of variability within the team’s
control, and establishing the daily disciplines
necessary to sustain the improvements over time.
13. Desired Benefits of VR
• Operations teams engaged in improving
and sustaining process performance.
• Standardize and stabilize the process to
yield fast and efficient results.
• Minimize capital expense and risk.
• Lower the “noise” in the processes so more
complex methods of understanding
sources of variation (such as Six Sigma)
can be used to create breakthrough
improvements.
14. VR Roadmap
• Twelve week structured roadmap using
data-based problem solving to reduce
variability in a primary process metric.
• Aligned with Six Sigma DMAIC.
17. Elements of VR Methodology
• Linkage to business strategy
• Roles & responsibilities
• Performance tracking
• Documentation
• Training
• Coaching/Audits
• Recognition
18. Linkage to Business Strategy
• Opportunities for VR were prioritized based
on savings potential, customer demand
versus current capability, availability of
resources, and other initiatives impacting
the process.
• Led by CI Director for each business unit.
20. Roles & Responsibilities
• VR Sponsor: Manufacturing VP
for business unit
• VR Champion: Department
Manager
• VR Team Leader: Process Supervisor
• VR Team: Cross functional process
members
• VR Analyst: Local Six Sigma Green Belt
21. Performance Tracking
• Key Metric: Downtime or Cycle Time
• Visual displays in the process created and updated
daily.
• Automated scorecards developed with assistance
of IT to inform senior leadership of status.
22. Documentation
• VR Implementation Guide
– VR Roadmap
– Tools
– Roles & Responsibilities
– Audit Process
– Certification
– Examples
• Intranet site for sharing best practices and
success stories.
23. Training
• Champion Training
– ½ day focused on basics of VR and Champion role
• VR Team Training
– 2 consecutive days focused on VR roadmap and tools
– ½ day focused on team chartering and project
management
• Provided by plant Certified VR Trainer.
24. • Audits integrated into the VR Roadmap at
4-week intervals.
• Used to assess team progress and provide
firm, honest feedback to the team and
champion.
• Used to coach team and champion on next
steps.
• Conducted by a second party appointed by
the CI Director of the business unit.
Coaching
25. Recognition
• Team recognition at two levels:
–Silver Certification – Granted to teams
that demonstrated effective application of
VR at the completion of 12 weeks.
–Gold Certification – Granted to teams that
demonstrated sustained improvement of
at least 15% in their primary process
metric for a minimum of 90 days.
26. Example: Bindery Line
• VR initiated on bindery line in mid-2000.
• VR Team created flowcharts of line startup and
operation, analyzed for non-value-added
activities and different practices between shifts
and individuals.
• VR Team agreed to standard process for startup
and operation of the line; documented in work
instructions and used to train all operators.
• Defined categories of downtime that negatively
impacted line performance; incorporated into
recently developed downtime reports and
used to create a “delay Pareto.”
27. Example: Initial Bindery Line Delay Pareto
High Book Jam: VR
Team Initial Focus
Missing Signature: Six
Sigma Black Belt Project
29. Example: High Book Jam Root Cause Analysis
Effect
Multi-vote
Count
% of Total
Infeed Trough 90 19.48
Not Jogging
Signature 72 15.58
Infeed Pin 68 14.72
Transfer Setting 60 12.99
Curled Paper 56 12.12
Packer Set-up 48 10.39
Feeders 36 7.79
Book Size 32 6.93
The trough found to be defective due to design and
wear. Stainless steel trough eliminated future wear
concern.
C&E MATRIX
30. Example: Bindery Line Missing Sig
• VR Team worked with
Black Belt to identify
critical input: vacuum drop
at the sucker cup.
• New gauging
methodology perfected to
measure and control.
• Control methods for
vacuum drop and other
critical inputs were
transferred to other lines
and plants.
31. Example: Bindery Line Delay Pareto Results
Missing Sig: Two Missing Signatures within Top 10
High Book Jam: Moved From #1 Delay to #3, With
Significant Reduction in Hours & Occurrences
32. Implementation Across the Organization
• Applied to over 600 process teams
– Manufacturing
– Customer Service
– Finance
– Logistics
• Variability reduced 2-15%.
• Average performance improved 4-15%.
33. One Function’s Experience
• 40% reduction in cycle time.
• Error rates reduced 70-89%.
• Rework reduced 25%.
• Over $300,000 annual savings in reduced
spoilages.
• New revenue: Process controls put in place
by VR Team enabled $700,000 of work for
a new customer.
34. Lessons Learned
• The structured roadmap and tool set
allowed most teams to successfully
implement VR in 12-20 weeks.
• Although not originally developed for
nonmanufacturing processes, VR was
found to work in customer service, finance,
and logistics.
– Language and examples in the training
materials and Implementation Guide were
modified to be more “user-friendly” for
both audiences.
35. Lessons Learned
• Process supervisors have not often had
training in teambuilding, conflict resolution or
team dynamics necessary to lead such a team
effort and require additional training and
coaching.
• VR Trainers need the equivalent of Green Belt
training to train and coach teams on the data-
based tools.
• Teams need the discipline to perform root
cause analysis versus jumping to solution.
36. Lessons Learned
• Consistent audit protocols are needed to
ensure consistency from audit to audit,
auditor to auditor. Audit checklists were
developed and used to train auditors. Also,
coaching skills.
• A compliance management system that
provides mechanisms for document
management & control and training & audit
records frees the VR team from these
types of administrative tasks.
37. Benefits of VR
• Operations teams owned daily process
improvement and were more inclined to
sustain the improvements over time.
• Capital expenditures for the improvements
were minimal.
• Conformance to schedule improved with
the reduction in variability. Customer
Service was able to commit to new work
within a specified time period with more
confidence.
38. Benefits of VR
• ISO 9001 activities were integrated into
daily work in VR processes.
• Efficiency of Six Sigma resources
increased.
• The elements of the methodology apply to
other improvement approaches (Six Sigma,
5S, quick changeover, ISO 9001, etc.).
• Over 4 years of implementation, observed
4-10% increase in sustained throughput
of manufacturing assets.
39. How is this Different from Your
Improvement Efforts?
40. How Might This be Applied in Your
Workplace?
What Would be the Challenges?
41. More Information
• Article on VR appeared in recent special
issue of Quality Engineering titled:
Variability Reduction: A Statistical
Engineering Approach to Engage
Operations Teams in Process
Improvement.
• Contact Susan Schall at:
susan@soschall.com or 540-636-1418.
The key tools used throughout the 16 week structured roadmap and beyond may be split into three categories:
Those that help the team to understand and document the process and its performance.
Those tools that are used to analyze the process.
Those tools that are used by the team to aid in their team processes and communication with key stakeholders.
We’ll spend the rest of today looking at most, not all of these tools.
Roles & Responsibilities
VR Sponsor
Understand and commit to using VR within the function/facility.
Prioritize VR opportunities.
Review VR scorecard for each team weekly.
Review progress monthly with VR champions.
Help teams and champions overcome barriers.
Follow-up on VR audit corrective actions.
VR Champion
Understand and commit to using VR within the department.
Participate in VR Champion training.
Assist sponsor to prioritize VR opportunities.
Provide resources to begin and sustain VR within the department.
Review visual display and scorecard for each team at least weekly.
Ensure VR teams follow the roadmap and utilize the tools to make data-based decisions.
Help team(s) overcome barriers.
Conduct work instruction audits.
VR Team Leader
Understand and commit to leading VR within their operation.
Participate in VR team training.
Guide the VR team through the roadmap and utilization of the tools to make data-based decisions.
Engage other process team members in VR implementation.
Communicate team progress and concerns to leadership and key stakeholders.
Conduct work instruction audits.
VR Team Members
Understand and commit to leading VR within their operation.
Participate in VR team training.
Use the VR roadmap and tools to make data-based decisions to continuously improve process.
Record accurate and timely data.
Adhere to agreed work instructions.
Take positive action to address sources of variation.
Engage other process team members in VR implementation.
Six Sigma Green Belt
Assist team with data analysis and making data-based decisions.
Coach the VR team leader on use of data analysis tools.
The second major problem area was High Book Jams. The Green Belt led the crew through creation of a cause & effect diagram (fishbone) to brainstorm all the possible causes of high book jams. They then took these ideas into C&E matrix and prioritized all the possible causes into the top few they believed were likely root causes and began investigating them. The top cause – inadequate trough (in-feed track) was found to be the primary root cause. Due to poor design, the trough was worn unevenly and causing one or more sigs to be high in the track. The trough was replaced with a stainless steel trough to reduce wear.
The Black Belt and the Process Team brainstormed possible causes, collected data and determined that the critical factor to the sucker picking up and inserting the sig was the vacuum drop at the sucker cup. Now, you might think that this was obvious, but up until this time, the team had focused on the cups themselves – size, type of material, location, etc.
There was no gage on the vacuum drop, so the team worked with engineering & the shop to perfect a new gage to measure the vacuum drop at the point where the sucker cup picks up the sig. They then experimented with the vacuum pressure and found that maintaining 22.5 +2.5-inches of vacuum at the sucker orifice to be the critical range.
Now the crew can monitor the vacuum drop and prevent missing sigs.
The gage has been installed on Line 953 and is being installed on other lines across the plant and other plants besides Danville – it is considered a best practice.
With these improvements in place, only two missing sigs appear in the Top 10 delay Pareto and the High Book Jams went from being the number one delay to the number three. This improved the performance of Line 953 to achieve Certification.
Note that these improvements did not eliminate the two problems – the problems often have more than one source. We identify one and knock it out of the picture it is easier to identify the others.