Department

College of Science of Statistics

Department of
Statistics

Collaborating Around
Six Sigma Innovation & Design
… and …

Six Sigma Information Technology:

Averting Disaster in Washington, DC
Dr. Rick Edgeman, Professor & Chair and Six Sigma Black Belt
Tel. +1 208-885-4410 Fax. +1 208-885-7959 Email: [email protected]
 College of Science Department of
Statistics
Six Sigma is …
… a highly structured strategy for acquiring, assessing, and applying
customer, competitor, and enterprise intelligence for the purposes of
product, system or enterprise innovation and design.

Innovation Algorithm
DMAIC (Define-Measure-Analyze-Improve-Control)
Design for Six Sigma Algorithm
DMADV (Define-Measure-Analyze-Design-Verify)

Makes Integrative Use Of:

Various strategies and tools from Statistics, Quality, Business, Engineering
and … ???
 College of Science Department of
S S Statistics
O ix igma rganizations
• GE … All 300,000+ GE employees must Rapidly Increasing Areas of Application.
be Six Sigma certified. All new GE
products developed using the “Design for – Healthcare – GE Heathcare - SLC
Six Sigma” (DFSS) approach.
– Financial,
• 3M … New CEO (from GE) requires all
3M employees to become Six Sigma – Military – NSWC, Pentagon, etc.
certified.
• Dupont • Fueled by:
• AlliedSignal • Strategic Contexts.
• Sun Microsystems • Notorious bottom-line orientation & results.
• Raytheon • Adaptable to multiple bottom lines.
• Motorola • Process orientation: rigorous and systematic
• Boeing approaches to innovation and design.
• Lockheed-Martin • Focus on the customer.
• Bank-of-America • Successful track record elsewhere.
• American Express • “Industry Buzz”.
• HSBC
• SAS Institute

While Six Sigma is new at, for example, 3M – its benefits at others of
these organizations is measured in the multi-billions of US dollars.
 College of Science Department of
The Villain
S tatistics Cost of Poorly Performing
Processes
σ level DPMO CP3
2 308,537 Not
Applicable
3 66,807 25%-40% of sales
4 6,210 15%-25% of sales
5 233 5%-15% of sales
6 3.4 < 1% of sales
Sigma (σ ) is a measure of “perfection” relating to process
Each sigma shift
performance provides…
capability a 10%
the net income
“bigger improvement
the better.”
A process operating at a “Six Sigma” level produces only
3.4 defects per million opportunities (DPMO) for a defect.
Without dedication of significant and appropriate attention
to a process, most processes in leading U.S. companies
operate at a level between 3 and 4 sigma.

Cost of Poorly Performing

Processes (CP3)
Why is Six Sigma Important?
 College of Science Department of
Statistics
… and the Hero What Does Six Sigma Tell Us?
• We don’t know what we don’t know.
• We can’t do what we don’t know.
• We won’t know until we measure.
• We don’t measure what we don’t value.
• We don’t value what we don’t measure.
• Typical Results: companies that properly implement
Six Sigma have seen profit margins grow 20% year
after year for each sigma shift (up to about 4.8s to
5.0s. Since most companies start at about 3s, virtually
each employee trained in Six Sigma will return on
average $230,000 per project to the bottom line until
the company reaches 4.7s. After that, the cost savings
are not as dramatic.
• However, improved profit margins allow companies
to create products & services with added features and
functions that result in greater market share.
 College of Science Department of
Statistics
Six Sigma COPIS Model
Outputs Process Inputs

Customers Suppliers

Steps How does Six Sigma Work?

The Voice of the Customer (VOC) is aggressively sought and rigorously
evaluated and used to determine needed outputs and hence the optimal
process configuration needed to yield those outputs and their necessary
inputs for which the best suppliers are identified and allied with.

From Concept to Market: the Voice of the Customer

 College of Science Department of
D
S tatistics
Six Sigma: How Do We Innovate?
efine
Define the problem and customer
requirements.

Control Measure M easure defect rates and document

the process in its current incarnation.

Analyze process data and determine

the capability of the process.

Improve the process and remove

defect causes.

Control process performance and

ensure that defects do not recur.
Improve Analyze
“Common sense” doesn’t mean “commonly done” or when done, done well.

Six Sigma Innovation & the DMAIC Algorithm

 College of Science Department of
Define
S tatistics
Six Sigma: How Do We Design?
Define customer requirements and
goals for the process, product or service.

Verify Measure M easure and match performance

to customer requirements.

Analyze and assess the design for

the process, product or service.

Design and implement the array of

new processes required for the new
process, product or service.

Verify results and maintain

Design Analyze performance.
All new products at GE are designed using a DFSS algorithm.

Design for Six Sigma (DFSS)

 College of Science Department of
Statistics
A Specific Example

Six Sigma Innovation & Design for Information Technology:

Averting Disaster in Washington, DC

Dr. Rick L. Edgeman, Professor & Chair, Department of Statistics, University of Idaho
Dr. David Bigio, Associate Professor of Mechanical Engineering, University of Maryland
Thomas E. Ferleman, Information Technology Consultant,
Office of the Chief Technology Officer, Government of the District of Columbia
 College of Science Department of
Statistics
Imagine … able towaking up and being
Magdalena Forsberg – Swedish

simply do nothing.
Olympian - 2002

Not having to train.

Jumping in the car and

driving to the country.

Making it through a whole day

without feeling any pain.”
“Goldfinger”, January-February 2002, Scanorama, pp. 22

… Working toward an IT future at OCTO where …

 College of Science Department of
Statistics
Imagine what was once Unthinkable …
Imagine the potential implications of a well-timed and successful terrorist
strike initiated by crippling or otherwise compromising the integrity of the
information technology infrastructure of the government of Washington,
DC – tourist haven; within 100 kilometers of three major international
airports; home of key defense, biotechnology, and information technology
firms; host to diplomats; the military and political capitol of the free world.
Truly the potential impact on the world order is staggering and likely well
beyond what the majority of us are reasonably able to guess.

Goal: to immunize / protect the IT infrastructure of the government of the

District of Columbia.
Means: Six Sigma Innovation and Design applied to critical IT functions.
 College of Science Department of
Statistics
The Office of the Chief Technology Officer (OCTO) is the primary provider
of information technology (IT) services to all 68 agencies of the government
of the District of Columbia (Washington, DC).
Agencies of the District of Columbia provide services that
simply must be operational at some minimal level or –
having failed – must quickly recover.

This includes, for example, police, fire and other emergency response,
public education, fire, public utilities, and Department of Motor Vehicles.

Service availability was promised to client agencies was well in excess of

OCTO’s ability to deliver. Moreover, the harsh reality of September 11, 2001
underscored the critical role of OCTO. Much of OCTO’s funding comes from
the United States Congress with a large amount of funding from the
Department of Homeland Security.
 College of Science Department of
Statistics
Five Critical Areas of Information Technology Service to OCTO
Client Agencies were examined and methods of improvement,
design, and integration explored. These areas were:
• Service Level Management (SLM)
• Capacity Management (CaM)
• Availability Management (AM)
• IT Service Continuity, a.k.a., Disaster Recovery (SCM), and
• Financial Management (FM)

These are detailed on the following slide.

 College of Science Department of
Statistics
Service Level Management (SLM) ensures that SLAs are met and that adverse impacts on service quality are minimized, assessing the impact of
changes on service quality and SLAs, both when changes are proposed and after their implementation. Key targets set in SLAs relate to service
availability thus requiring incident resolution within agreed periods. SLM is the hinge of service support and delivery and relies on the effective
and efficient working of underpinning support processes, without which an SLA is useless, since these are foundational to content agreement.

Capacity Management (CM) ensures constant availability of adequate capacity to meet agency business requirements. CM involves incident
resolution and problem identification for those difficulties related to capacity issues and generates requests for change (RFCs) that ensure
sufficient capacity. RFCs are subject to a change management process and implementation often affects hardware, software and documentation
and requires effective release management.

Availability Management (AM) concerns design, implementation, measurement and management of IT services to ensure that stated
availability requirements are met and requires IT service FMEA and the understanding the time taken to resume service. Incident management
and problem management provide key inputs ensuring that appropriate corrective actions occur. Availability targets specified in SLAs are
monitored as part of the AM process that also supports the SLM process by providing measurements and reporting to support service reviews.

IT Service Continuity Management (SCM) or “Disaster Recovery” manages an organization’s ability to provide a pre-determined agreed
upon level of IT services to support minimum business requirements. Among the means used are resilient systems and recovery options such as
back-up facilities. Configuration management data is required to facilitate this prevention and planning. Infrastructure and business changes
need to be assessed for their potential impact on continuity plans, and IT and business plans are then subject to change management procedures.

Financial Management (FM) accounts for costs and returns of IT service investments and cost recovery from clients. FM requires interfaces
with CM, configuration management, and SLM to identify the true costs of service. FM works together with business relationship management
and the IT organization during the negotiation of IT budgets and client IT expenditures.

KEY: SLA = Service Level Agreement. FMEA = Failure Modes and Effects Analysis. RFC = Request for Change.

Definitions
 College of Science Department of
Approach
Charter
Statistics
Description or Example Use

resolution plan.
1

Purpose is to define the business case; project goals and limits; way of working together; and conflictAll
ITIL Areas

Brainstorming Uses included cause identification and solution generation. All

Affinity Diagram Uses included associations among OCTO needs (CTQs) All

Interrelationship Digraph Primarily used to explore causal relationships between enablers (“hows”) to capture correlations and formAM, CtM
the roof in the HOQ.

Nominal Group Technique: NGT Nominal Group Technique used as part of QFD / HOQ to prioritize OCTO needs. All

Matrix / Priority Matrix Diagrams Various uses including distribution of tasks to team members and relating OCTO needs (CTQs) to enablersAll
(“hows”) in QFD.

SMART Goals Specific, Measurable, Attainable, Relevant, and Time-Bound goals and problem statements. All

Process Maps Included high-level COPIS maps and detailed process maps. Both “before” and “after” improvementAll
versions were used.

VOC Tools Approaches included surveys, focus groups, customer complaints, and interviews. All

Drill Down Trees Process-Product Drill Down Tree. All

FMEA Failure Modes & Effects Analysis. AM, CtM

HOQ / QFD House of Quality / Quality Function Deployment. Integrated use of Matrix Diagrams and NGT to assessAM
internal and external customer needs and deploy solutions.

RESULTS Total Savings Estimated by Deputy Director = $2M to $3M from 2003-2007.
ITIL Areas A = Availability Management, CaM = Capacity Management , CtM = Continuity Management,
FM = Financial Management, SLM = Service Level Management

Table 1. Examples of Approaches and Their Use in the OCTO DC Project

 College of Science Department of
Approach

Pareto Chart
Statistics
Description or Example Use

Used to identify dominant issues / defect causes.

1

All
ITIL Areas

Fishbone Diagram Also called Cause-and-Effect Diagrams display “effects” representing a problem or an opportunityAll
with the “causes” being real or potential drivers of the effect.
SWOT All ITIL areas were assessed for strengths, weaknesses, opportunities and threats with motivationsAll
being improvement or leverage of strengths, diminution of weaknesses, welcoming opportunities, and
countermanding threats.
Benchmarking Benchmarking of South Dakota Bureau of Information Technology and private sector ISPs. CtM

Chi-Square Tests of homogeneity examined whether differing approaches yielded similar results and tests ofAM, CaM, CtM,
independence explored CTQ-Enabler relationships. SLM

Correlation & Regression Used to explore, assess, characterize and exploit CTQ-Enabler relationships. SLM

Design of Experiments DOE approaches included experiments with operating parameters, critical elements, or both.AM, CtM, SLM
(DOE) Operating parameters are enablers (X’s) that vary in amount while critical elements are enablers (X’s)
that differ in type or categorically. Screening and factorial designs were used.
SPC Charts Statistical Process Control Charts recommended or used included p charts, I-MR charts, and X-barAM, CaM, CtM,
and R charts to (directly) control the X’s, hence indirectly the CTQs. SLM

RESULTS Total Savings Estimated by Deputy Director = $2M to $3M from 2003-2007.
ITIL Areas A = Availability Management, CaM = Capacity Management , CtM = Continuity Management,
FM = Financial Management, SLM = Service Level Management

Table 1. Examples of Approaches and Their Use in the OCTO DC Project

(continued)
 College of Science Department of
Statistics
References on Six Sigma at OCTO:
Mission Critical: Six Sigma and Business Excellence for Information Technology.
Rick L. Edgeman, David I. Bigio, and Thomas A. Ferleman (2005 expected)
World Class Applications of Six Sigma: Case Studies from Manufacturing and Service Industries
Elsevier Science, Oxford, UK. Jiju Anthony & Mohammed Zairi, Editors. – BOOK CHAPTER

Six Sigma or Business Excellence: Strategic and Tactical Examination of IT Service Level Management at the
Office of the Chief Technology Officer of Washington, DC. (Invited Contribution)
Quality & Reliability Engineering International, Vol. 21, No. & pp. pending, 2005
Rick L. Edgeman, David Bigio and Thomas Ferleman

Six Sigma Availability Management of Information Technology in the Office of the Chief Technology Officer of
Washington, DC. Total Quality Management, Vol. 15, No. 5/6, 2004.
David Bigio, Rick L. Edgeman and Thomas Ferleman

General Six Sigma References:

Edgeman, R. and Bigio, D. (2004). “Six Sigma as Metaphor: Heresy or Holy Writ?” Quality Progress, Vol. 37, No. 1, 25-30.

Six Sigma in Communities of Care: Improved Care via Institutionalized Genius

Business Briefing: Global Healthcare 2002, Vol. 2, 46-49 (Invited Contribution)
World Medical Association – 53rd General Assembly. London, UK. – Rick L. Edgeman – BOOK CHAPTER

Klefsjö, B., Wiklund, H., and Edgeman, R. (2001). “Six Sigma Seen as a Methodology for Total Quality Management”,
Measuring Business Excellence, Vol. 5, No. 2, 31-35.

The Six Sigma Handbook Expanded and Revised. Thomas Pyzdek (2003). McGraw-Hill, New York.