You are on page 1of 6

P R O C E S S I M P R O V E M E N T

How To Compare Six Sigma,


Lean and the Theory of Constraints
A framework for choosing what’s best for your organization
by
Dave Nave

W
ITHIN THE AMERICAN business Six Sigma
community a multitude of process
Six Sigma claims that focusing on reduction of
improvement champions are
variation will solve process and business problems.
vying for leadership attention.
By using a set of statistical tools to understand the
Each champion advocates the
fluctuation of a process, management can begin to
adoption of his or her improve-
predict the expected outcome of that process. If the
ment methodology in your organization. Almost all
plead that if you adopt their specific tools or follow outcome is not satisfactory, associated tools can be
a specific way of thinking, all your business prob- used to further understand the elements influenc-
lems will be solved. ing that process.
After listening to multiple champions advocate Through a rigid and structured investigation
their special methodology, how do you choose methodology, the process elements are more com-
what will be best for your situation? What method- pletely understood. The assumption is the outcome
ology fits the culture of your organization? of the entire process will be improved by reducing
Many process improvement methodologies the variation of multiple elements.
appear to conflict with each other or at least down- Six Sigma includes five steps: define, measure,
play the contribution of other
methodologies. This montage of
TABLE 1 Improvement Programs
tools and philosophies creates the
illusion of conflicting strategies. Program Six Sigma Lean thinking Theory of constraints
In this article, I will discuss the
Theory Reduce variation Remove waste Manage constraints
basics of the three improvement
Application 1. Define. 1. Identify value. 1. Identify constraint.
methodologies and present a model guidelines 2. Measure. 2. Identify value stream. 2. Exploit constraint.
to help you understand their con- 3. Analyze. 3. Flow. 3. Subordinate processes.
4. Improve. 4. Pull. 4. Elevate constraint.
cepts and effects and similarities 5. Control. 5. Perfection. 5. Repeat cycle.
and differences. Table 1 describes
Focus Problem focused Flow focused Systems constraints
the essence of each methodology.

QU A L I T Y P R O G R E S S I M A R C H 2 0 0 2 I 73
H O W T O C O M PA R E S I X S I G M A , L E A N A N D T H E T H E O RY O F C O N S T R A I N T S

analyze, improve and control zation can respond to cus-


(commonly known as DMAIC): tomer needs more quickly.
• Define. Practitioners begin by Six Sigma is founded on
defining the process. They ask two main assumptions.
who the customers are and First, people in an organi-
what their problems are. They
identify the key characteristics
Six Sigma
The assumption is the outcome
zation understand and
appreciate the fact that
important to the customer numbers can represent
along with the processes that features and characteris-
support those key characteris- tics of a process. They
tics. They then identify exist- of the entire process will be appreciate that a deeper
ing output conditions along understanding of data and
with the process elements. data analysis can be used
• Measure. Next the focus is on
measuring the process. Key
improved by reducing the to produce improvements,
and graphical representa-
characteristics are categorized, tions of data can provide
new and different per-
measurement systems are veri-
fied and data are collected. variation of multiple elements. spectives of the process.
• Analyze. Once data are col- Analytical types, such as
lected, it is analyzed. The engineers and scientists,
intent is to convert the raw generally respect this
data into information that pro- approach.
vides insights into the process. Another assumption is
These insights include identi- that through the reduction
fying the fundamental and most important causes of variation of all the processes, the overall perfor-
of the defects or problems. mance of the organization will be improved. But while
• Improve. The fourth step is to improve the process. it is hard to argue against improvement, the economic
Solutions to the problem are developed, and reality of business is we want the most improvement
changes are made to the process. Results of process for the least investment. Improving all of an organiza-
changes are seen in the measurements. In this step, tion’s individual processes could actually have a detri-
the company can judge whether the changes are mental effect on the company’s ability to satisfy the
beneficial, or if another set of changes is necessary. customer’s needs and provide product and services at
• Control. If the process is performing at a desired the right time at the lowest cost. The realized savings
and predictable level, it is put under control. This to the system might be less than the cost of all the
last step is the sustaining portion of the Six Sigma improvements.
methodology. The process is monitored to assure no So, an organization that improves things just
unexpected changes occur. because it can may be improving the wrong things for
Focusing on the primary area of variation reduction the business.
produces other secondary effects, too. Quality is
improved. Process investigation produces the re-eval-
Lean thinking
uation of the value added status of many elements. Lean thinking is sometimes called lean manufactur-
Some elements are modified, while others are discon- ing, the Toyota production system or other names.
tinued. Elements are refined and improved. Mistakes Lean focuses on the removal of waste, which is
and opportunities for mistakes are reduced. defined as anything not necessary to produce the
Some elements discovered during the Six Sigma product or service.
investigation constrain the flow of products or ser- One common measure is touch time—the amount
vices through the system. Flow is defined as the time of time the product is actually being worked on, or
from the input of raw material to the output of a sal- touched, by the worker. Frequently, lean’s focus is
able item. Improvement of a process that was restrict- manifested in an emphasis on flow.
ing flow results in reduced variation, better quality There are five essential steps in lean:
and improvement in the volume of the process out- 1. Identify which features create value.
put. Thus the organization has less money tied up in 2. Identify the sequence of activities called the value
in-process inventory. The time from paying for input stream.
material to seeing a profit is reduced, and the organi- 3. Make the activities flow.

74 I M A R C H 2 0 0 2 I W W W . A S Q . O R G
4. Let the customer pull product or service through stant attempt to remove nonvalue activity, improve
the process. flow and satisfy customer delivery needs.
5. Perfect the process. While lean focuses on removing waste and improv-
Identify value. The determination of which features ing flow, it too has some secondary effects. Quality is
create value in the product is made from the internal improved. The product spends less time in process,
and external customer standpoints. Value is expressed reducing the chances of damage or obsolescence.
in terms of how the specific product meets the cus- Simplification of processes results in reduction of vari-
tomer’s needs, at a specific price, at a specific time. ation. As the company looks at all the activities in the
Specific products or services are evaluated on which value stream, the system constraint is removed, and
features add value. The value determination can be performance is improved.
from the perspective of the ultimate customer or a The lean methodology also makes some assump-
subsequent process. tions:
Identify the value stream. Once value is identified, • People value the visual effect of flow.
activities that contribute value are identified. The • Waste is the main restriction to profitability.
entire sequence of activities is called the value stream. • Many small improvements in rapid succession are
Then a determination is made as to whether activities more beneficial than analytical study.
that do not contribute value to the product or service • Process interaction effects will be resolved through
are necessary. Necessary operations are defined as value stream refinement.
being a prerequisite to other value added activities or People in operations appreciate this approach.
being an essential part of the business. An example of Lean involves many people in the value stream.
a nonvalue added but necessary process is payroll. Transitioning to flow thinking causes vast changes in
After all, people need to be paid. Finally the impact how people perceive their roles in the organization
necessary, nonvalue added activities have on the and their relationships to the product.
process is reduced to a minimum. All other nonvalue
added activities are transitioned out of the process. Theory of constraints (TOC)
Improve flow. Once value added activities and nec- TOC focuses on system improvement. A system is
essary nonvalue activities are identified, improvement defined as a series of interdependent processes. An
efforts are directed toward mak- analogy for a system is the
ing the activities flow. Flow is chain: a group of interdepen-
the uninterrupted movement of dent links working together
product or service through the toward the overall goal. The
system to the customer. constraint is a weak link.
Major inhibitors of flow are
work in queue, batch processing
and transportation. These
Lean
Lean focuses on the
The performance of the
entire chain is limited by the
strength of the weakest link. In
buffers slow the time from prod- manufacturing processes, TOC
uct or service initiation to deliv-
ery. Buffers also tie up money
removal of waste, which concentrates on the process
that slows the speed of prod-
that can be used elsewhere in uct through the system.
the organization and cover up TOC consists of five steps:
the effects of system restraints is defined as anything 1. Identify the constraint.
and other wasted activities. 2. Exploit the constraint.
Allow customer pull. After 3. Subordinate other processes
waste is removed and flow
established, efforts turn to let-
not necessary to produce to the constraint.
4. Elevate the constraint.
ting the customer pull product 5. Repeat the cycle.
or service through the process.
The company must make the
the product or service. Identify. The constraint is
identified through various
process responsive to providing methods. The amount of work
the product or service only in queue ahead of a process
when the customer needs it— operation is a classic indicator.
not before, not after. Another example is where
Work toward perfection. This products are processed in
effort is the repeated and con- batches.

QU A L I T Y P R O G R E S S I M A R C H 2 0 0 2 I 75
H O W T O C O M PA R E S I X S I G M A , L E A N A N D T H E T H E O RY O F C O N S T R A I N T S

Exploit. Once the constraint is identified, the improvement programs: that many programs use a
process is improved or otherwise supported to mass, one size fits all approach to improvement. With
achieve its utmost capacity without major expensive the mass approach, a company hopes that by refining
upgrades or changes. In other words, the constraint is and improving each process individually and inde-
exploited. pendently to maximum output,
Subordinate. When the con- the entire system output will
straining process is working at improve.
maximum capacity, the speeds TOC methodology operates

TOC
of other subordinate processes on several assumptions:
are paced to the speed or • As in the case of lean, the
capacity of the constraint. Some
processes will sacrifice individ-
The performance of the organization places a value
on the speed at which its
ual productivity for the benefit product or service travels
of the entire system. through the system. Speed
Subordinate processes are entire chain is limited by and volume are the main
usually found ahead of the con- determinants of success.
straint in the value stream. • Current processes are essen-
Processes after the constraint tial to produce the desired
are not a major concern—they the strength of the output.
are probably already producing • The product or service design
under capacity because they is stable.
have to wait on the constrain-
ing process.
weakest link. Value added workers do not
need to have an in-depth
Elevate. If the output of the understanding of this improve-
overall system is not satisfac- ment methodology. Suggestions
tory, further improvement by the workforce are not con-
is required. The company sidered vital for successful
may now contemplate major implementation of the theory of
changes to the constraint. constraints. Organizations with
Changes can involve capital hierarchical structure and cen-
improvement, reorganization tralized knowledge value this
or other major expenditures of approach.
time or money. This is called elevating the constraint
or taking whatever action is necessary to eliminate it. Comparing the three methods
Repeat. Once the first constraint is broken, another There are some commonalities and general criti-
part of the system or process chain becomes the new cisms of all improvement models. In addition, all
constraint. Now is the time to repeat the cycle of process improvement theories and methodologies
improvement. The performance of the entire system make a few of the same assumptions. The main points
is re-evaluated by searching for the new constraint of each methodology are summarized in Table 2.
process, exploiting the process, subordinating and Improvement methodologies begin by taking the
elevating. product or service configuration at face value and
By focusing on constraints, this methodology pro- improving the processes or system. They assume the
duces positive effects on the flow time of the product following:
or service through the system. Reduction of waste in • The design of product or service is essentially cor-
the constraint increases throughput and improves rect.
throughput time. When the constraint is improved, • The design of the product or service is the most eco-
variation is reduced, and quality is improved. nomical.
Constraint focus does not require intimate knowl- • Customer needs are satisfied with that design.
edge of data analysis or that a large number of people • The current product configuration fulfills the func-
understand the elements of the system. Understanding tional requirements of the market and customer.
by a few people with the power to change things is all • The management structure supports and nourishes
that is necessary. The effort can be localized with mini- change.
mum involvement of the workforce. These assumptions may not be valid and require
TOC overcomes one criticism of most process exploration.

76 I M A R C H 2 0 0 2 I W W W . A S Q . O R G
After extensive refinement of the exist-
ing processes or systems, many improve- TABLE 2 Comparison of Improvement Programs
ment methodologies begin to look at the
product or service design. However, Program Six Sigma Lean thinking Theory of constraints
each views the design through its theory Theory Reduce variation Remove waste Manage constraints
and tools.
Application 1. Define. 1. Identify value. 1. Identify constraint.
Quality function deployment and guidelines 2. Measure. 2. Identify value stream. 2. Exploit constraint.
value management are two techniques 3. Analyze. 3. Flow. 3. Subordinate processes.
used to help connect the product or ser- 4. Improve. 4. Pull. 4. Elevate constraint.
vice design to customer needs. Both 5. Control. 5. Perfection. 5. Repeat cycle.
bring marketing, finance, operations, Focus Problem focused Flow focused System constraints
design, customer and suppliers together Assumptions A problem exists. Waste removal will Emphasis on speed
to systematically explore how the prod- Figures and numbers improve business and volume.
uct performs the function the customer are valued. performance. Uses existing systems.
System output improves Many small improvements Process interdependence.
needs. if variation in all are better than
An interesting part of this investiga- processes is reduced. systems analysis.
tion is that cost can be associated with Primary effect Uniform process output Reduced flow time Fast throughput
function. When marketing and cus-
Secondary Less waste. Less variation. Less inventory/waste.
tomers know the cost of specific features, effects Fast throughput. Uniform output. Throughput cost
they make informed choices about the Less inventory. Less inventory. accounting.
configuration of the product or services. Fluctuation—performance New accounting system. Throughput—performance
measures for managers. Flow—performance measurement system.
Major obstacles to improvement Improved quality. measure for managers. Improved quality.
Improved quality.
There are major obstacles to the
Criticisms System interaction Statistical or system Minimal worker input.
improvement methodologies: not considered. analysis not valued. Data analysis not valued.
• They address management theory as a Processes improved
secondary or tertiary issue. independently.
• They don’t address policies, either for-
mal or informal.
• They don’t address how managers are measured Beneath a theory of management is a system of
and rewarded for process improvements. organizational values. Is the purpose of the organiza-
• They don’t address the general theory of manage- tion solely to increase the wealth of the stockholders?
ment used by the organization. Or is the existence of the organization to benefit soci-
• They don’t address the organization’s values. ety, the nation or some other group? Do not get caught
In any organization many activities are driven by in the trap of thinking an organization’s only purpose
policies whose purpose has been lost in time. All is to make money. Money, or profit, is the result of
change programs challenge the existing ways of doing good management toward satisfying a societal need.
things. This necessitates asking what purpose a specif- Also, consider that some organizations are created not
ic policy serves and whether that purpose is still valid to make a profit.
in today’s environment. The issues of theory of management and organiza-
One technique for assessing an organization’s man- tional ethics and values are beyond the scope of this
agement theory is to search for the underlying article but have been raised to point to other areas
assumptions supporting each policy. A challenge of needing consideration when looking at process or sys-
assumptions provides a starting point for determining tem improvement programs.
whether the current policy is still supporting some- Champions of each of these methods say they can
thing of value today. overcome these drawbacks because implementation of
The management theories of W. Edwards Deming their particular methodology and focus on their tools,
may help organizations challenge current manage- methods and theories will allow an improved theory
ment practices and assumptions—not by suggesting of management and business strategy to emerge.
incremental improvements but by pointing to a new
way of managing. Through this line of study, leaders How to choose
achieve a greater understanding of the way they can To help work through the apparent conflicts of dif-
influence the social and economic well-being of their ferent improvement programs, use a model that iden-
organization. tifies a hierarchy of cause and effect relationships.

QU A L I T Y P R O G R E S S I M A R C H 2 0 0 2 I 77
H O W T O C O M PA R E S I X S I G M A , L E A N A N D T H E T H E O RY O F C O N S T R A I N T S

First, identify the primary theory. What is the core use, the main issue left to explore is the speed at which
emphasis of the program or methodology? This core a method will be accepted into an organization:
emphasis is usually a few words or a short phrase: Six • If your organization values analytical studies and
Sigma’s is variation reduction, lean’s is waste reduc- the relationships of data, charts and analysis, Six
tion and TOC’s is constraint reduction. Sigma is a perfect program for you to start with.
Then identify the relationship between the primary • If your organization values visual change and right
theory and the primary focus of the tools and method- now time, then lean thinking might be the way to
ology. This relationship indicates how the primary the- go.
ory manifests itself in tangible results—what I call the • If your organization values a systems approach
primary effect. This is an if/then type of relationship: where total participation is not desired and if it val-
• For Six Sigma: “If we focus on reducing variation, ues the separation between worker and manage-
then we will have more uniform process output.” ment, then TOC might be a good way to start.
• For lean: “If we focus on waste removal, then flow More and more organizations are trying to deter-
time will improve.” mine what improvement method will work best and
• For TOC: “If we focus on constraints, then through- fit best with their culture. When you are working
put volume will improve.” through the apparent conflicting claims of perfor-
The next to last level of the model in Table 2 (p. 77) mance improvement programs, my advice is to con-
identifies secondary effects. Secondary effects can be centrate on the primary and secondary effects of their
described by using an if/theory and primary philosophies. Once the values of a specific improve-
effect/results type statement. While the primary theo- ment program are identified, the comparison of those
ry to primary effect relationship is usually one-to-one, values with the values of the organization can make
the secondary effects are several-to-many, including: the method of selection easier, if not obvious.
• For Six Sigma, focus on reducing variation and Never stop learning. Each improvement methodol-
achieving uniform process results in less waste, less ogy contributes valuable concepts, ideas and tech-
throughput time and less inventory. niques to your organization. Your challenge is to use
• For lean thinking, focus on waste and flow time whatever strengths the methodology possesses to help
results in less variation, uniform output and less your organization improve.
inventory.
• For TOC, focus on constraints and increased BIBLIOGRAPHY
throughput results in less inventory and a different Deming, W. Edwards, The New Economics for Industry,
accounting system. Government, Education, second edition (Cambridge, MA:
Each improvement methodology appears to be dri- Massachusetts Institute of Technology, 1994).
ving toward common tools and concepts. However, Dettmer, William H., Goldratt’s Theory of Constraints: A Systems
different methodologies begin the journey from differ- Approach to Continuous Improvement (Milwaukee: ASQ
ent perspectives. At the secondary effects level of the Quality Press, 1997).
model, the results from each methodology start to Goldratt, Eliyahu M., The Goal: A Process of Ongoing
look similar. Many of the secondary effects of one Improvement, second edition (Great Barrington, MA: North
methodology look similar to the primary effect or River Press, 1994).
focus of another methodology. Goldratt, Eliyahu M., It’s Not Luck (Great Barrington, MA:
Extending the fundamental philosophy through each North River Press, 1994).
methodology’s primary, secondary and tertiary effects, Womack, James P., and Daniel T. Jones, Lean Thinking: Banish
you might conclude each method strives to achieve Waste and Create Wealth in Your Corporation (New York:
similar results. Even along the journey, each methodol- Simon & Schuster, 1996).
ogy incorporates the primary effects of other improve-
ment programs. Can we infer that after extensive time
and effort implementing a single methodology, the end DAVE NAVE works on assignment to Boeing Commercial
result will be similar no matter which path we take? Airplanes for Comforce Technical Services. He holds a master’s
Where does that leave us? As a manager, how do degree in management systems from the Deming Scholars
you select an improvement methodology or program Program at Fordham University in New York. Nave is a member
to overcome your obstacles? of ASQ. QP
Selection of a process improvement methodology is
dependent on the culture of your organization. If many
popular programs appear to end up in the same place
addressing the same issues after a number of years of © 2002 American Society for Quality. Reprinted with permission.

78 I M A R C H 2 0 0 2 I W W W . A S Q . O R G

You might also like