Continuous Process Improvement
Continuous Process Improvement
Introduction
Improvement is made by
Process
Inputs may be materials, money, information, data, etc. Outputs may be information,
data, products, service, etc. The output of one process also can be the input to another
Page 1 of 26
process. Outputs usually require performance measures. They are designed to achieve
certain desirable outcomes such as customer satisfaction. Feedback is provided in
order to improve the process.
Process definition begins with defining the internal and/or external customers. The
customer defines the purpose of the organization and every process within it. Because
the organization exists to serve the customer, process improvements must be defined in
terms of increased customer satisfaction as a result of higher quality products and
services.
All processes have at least one owner. In some cases, the owner is obvious, because
here is only one person performing the activity. However, frequently the process will
cross multiple organizational boundaries, and supporting sub-processes will be owned
by individuals within each of the organizations. Thus, ownership should be part of the
process improvement initiatives.
At this point it is important to define an improvement. There are five basic ways to
improve: (1) reduce resources, (2) reduce errors, (3) meet or exceed expectations of
downstream customers, (4) make the process safer, and (5) make the process more
satisfying to the person doing it.
FEED BACK
INPUT
PROCESS
People OUTPUT
Materials
Equipment Information
Money
Methods Data
Information
Procedures Produce OUTCOMES
Data, et
Environment Services,
Procedures etc.
CONDITIONS
Page 2 of 26
Figure 5-1 Input/output Process Model
First, a process that uses more resources than necessary is wasteful. Reports that are
distributed to more people than necessary wastes copying and distribution time,
material, users and read time, and, eventually, file space.
Second, for the most part, errors are a sign of poor workmanship and require rework.
Typing errors that are detected after the computer printout require opening the file,
making the correction, and printing the revised document.
The fourth way to improve a process is to increase the satisfaction of the individual
performing the process. Sometimes, a little change, such as an ergonomically correct
chair, can make a substantial change in a person’s attitude toward their work.
Process improvement involves planning. One of the best approaches is the one
developed by Dr. Joseph Juran. It has three components: planning, control, and
improvement, and is referred to as the Juran Trilogy. It is based loosely on financial
processes such as budgeting (planning), expense measurement (control), and cost
reduction (improvement).
Page 3 of 26
Planning
The planning component begins with external customers. Once quality gods are
established, marketing determines the external customers, and all organizational
personnel (managers, members of multifunctional teams, or work groups) determine the
internal customers. External customers may be quite numerous, as is the case of a
bank supply organization, where they include tellers, financial planners, loan officers,
auditors, managers, and the bank's customers. Where there are numerous customers, a
Pareto diagram (see Chapter 18) might be useful to determine the vital few.
Once the customers are determined, their needs are discovered. This activity requires
the customers to state needs in their own words and from their own viewpoint; however,
real needs may differ from stated needs. For example, a stated need may be an
automobile, whereas the real need is transportation or a status symbol. In addition,
internal customers may not wish to voice real needs out of fear of the consequences.
One might discover these needs by (1) being a user of the product or service, (2)
communicating with customers through product or service satisfaction and
dissatisfaction information, or (3) simulation in the laboratory. Because customer needs
are stated from their viewpoint, they should be translated to requirements that are
understandable to the organization and its suppliers.
The next step in the planning process is to develop product and/or service features that
respond to customer needs, meet the needs of the organization and its suppliers, are
competitive, and optimize the costs of all stakeholders. This step typically is performed
by a multifunctional team. Quality function deployment (Chapter 12), Taguchi's quality
engineering (Chapter 20), and quality by design (Chapter 13) arc some of the
approaches that can he used. It is important that the design team, rather than a single
department, approve the final design and that the team be composed of all functional
areas within an organization as well as customers and suppliers.
The fourth step is to develop the processes able to produce the product and/or service
features. Some of this planning would have occurred during the previous step. This step
is also performed by a multifunctional team with a liaison to the design team. Activities
include determining the necessary facilities, training, and operation, control, and
maintenance of the facilities. Of particular concern will be the "scaling up" from the
laboratory or prototype environment to the real process environment. Additional
activities include process capability evaluation and process control typo and location.
Transferring plans to operations is the final step of the planning process. Once again, a
multifunctional team with a liaison to the other teams is used. When training is
necessary, it should be performed by members of the process planning team. Process
validation is necessary to ensure, with a high degree of assurance, that a process will
Page 4 of 26
consistently, produce a product or service meeting requirements. Positrol and process
certification, discussed later in the chapter, are excellent techniques to use to help
validate the process.
Control
Control is used by operating forces to help meet the product, process, and service
requirements. It uses the feedback loop and consists of the following steps:
Statistical process control (see Chapter 18) is the primary technique for achieving
control. The basic statistical process control (SPC) tools are Pareto diagrams, cause-
and-effect diagrams, check sheets, histograms, control charts, and sea diagrams. In
addition, process capability information such as CP and CPK are used to determine if the
process is capable and is centered.
Improvement
The third part of the trilogy aims to attain levels of performance that are significantly
higher than current levels. Process improvements begin with the establishment of
effective infrastructure such as the quality council (see Chapter 2). Two of the cut the
council are to identify the improvement projects and establish the project with a project
owner. In addition, the quality council needs to provide the team the resources to
determine the causes, create solutions, and establish controls to the gains (see Chapter
4). The problem-solving method described in a later section may be applied to improve
the process, while the quality council is the driver that ensures that improvement is
continuous and never ending. Process improvement can be incremental or
breakthrough.
Page 5 of 26
CHAPTE
ER 5
Quality planning
p
40-i
C
Figure 5-2 The
T Juran Triloggy Diagram Adap
pted, with permission, from J. M.
M Juran. ed., Qu
uality Control Ha
andbook, 4th ed
d. (New
York: McGraaw-Hill, 1988).
Figure 5-2
5 provide es an exam mple of how
w the three continuous improvemment proce esses
interrela
ate.; In the figure, Ju
uran provides a distin
nction betw
ween sporaadic waste and
chronic waste. The sporadicc waste ca an he identified and corrected through qu uality
control. The chron nic waste reequires an improvemment processs. As a so
olution is fo
ound
through the impro ovement prrocess, lesssons learned are broought backk to the qu uality
planningg process so
s that new goals for th
he organiza
ation may be
b establish
hed.
Improve
ement Stra
ategies
There are
a four prrimary imprrovements strategic---repair, reffinement, renovation,
r and
reinventtion. Choossing the rigght strategyy for the rig
ght situation
n is criticall. It is also true
that prop
per integrattion of the strategies
s w produce
will e never-end ding improvvement.
Repair
This stra
ategy is simple-anyth hing broken n must be fixed so thhat it functions as de
esign'
There arre two levels to this sttrategy. If a customer receives a damaged product,
p a quick
q
Page 6 of 26
fix is required. This level is a temporary or short-term measure. Although short to
measures shore tip the problem, they should not become permanent. The second le
occurs when an individual or team identities and eliminates the root cause(s) of the pr
lent and effects a permanent solution. It is important to note that the repair strategy d
not make the process better than the original design.
Refinement
This strategy involves activities that continually improve a process that is not bra
improvements to processes, products, and services are accomplished on an
incremental basis. Refinement improves efficiency and effectiveness. It should become
an integral part of every employee's job. Both individuals and teams can use this
strategy. Typically it relies on doing things just a bit quicker, better, easier, or with less
waste. This is the concept behind Kaizen to be discussed later in the chapter. The
change may so gradual that there is no appearance of change. The primary benefit of
gradual change is that it produces little resistance from employees. However, because
the change is so gradual, management may not recognize and reward the affected
employees. Also, minor changes may not he documented or properly communicated.
Renovation
Reinvention
Page 7 of 26
activity is developed using teams based on a complete understanding of the customer's
requirements and expectations. Reinvention of reengineering begins by imagining that
the previous condition does not exist---in other words, a clean sheet of paper. Then the
team uses in-depth knowledge of the customer's requirements and expectations and
invents a new product, service, process, or activity. For example, the process of drilling
holes using lasers or water jets was a reinvention.
Additional Comments
The repair and refinement strategies require that all employees have the freedom to
solve problems and make incremental improvements in their jobs. Repair and
refinement improvements are almost immediate with very little cost.
Types Problems
There are five types of problems: compliance, unstructured, efficiency, process design,
and product design. The first three are performance problems where an existing system
is not performing satisfactorily, and the last two are design problems that require a new
or improved design.
Compliance
Unstructured
Unstructured problems resemble compliance problems except that they are not
specified by standards. The absence of standards may be due to system immaturity or
to the need for flexibility in system performance. For example, an expert woodworker
Page 8 of 26
adjusts her activities to the grain and moisture content of the wood, and customer
service workers adapt their behavior to individual customers. Identification of
unstructured problems is usually brought about by negative customer feedback. The
major challenges are to determine customer needs and to diagnose the causes of poor
performance. Because of customer variability, it is difficult to determine why a product or
service was unacceptable. Organizations need to treat each customer as an individual
and maintain a database on acceptable and unacceptable behavior.
Efficiency
Efficiency problems occur when the system is performing unacceptably from the view-
point of its owners or operators. In other words, the end user is satisfied; however, the
process is more costly than desired or working conditions are not acceptable. Problem
solving is directed towards reducing cost and providing safe working conditions.
Identification of such problems occurs from benchmarking and operator suggestions.
Process Design
Product Design
Product-design problems involve the development of new product, and the improvement
of existing products. A major focus is to prevent process and end user problems by
relying on customer needs. Although design work can be initiated as a result of poor
product performance, problem solving usually occurs as a natural part of a competitive
environment. A major challenge is translating, in a timely manner, user needs and
constraints into product attributes and specifications, usually using quality function
deployment (QFD).
The basic Plan-Do-Study-Act (PDSA) cycle was first developed by Shewhart and then
modified by Deming. It is an effective improvement technique. Figure 5-3 illustrates the
cycle.
Page 9 of 26
Act Plan
Study Do
The four steps in the cycle are exactly as stated. First, plan carefully what is to he done.
Next, carry out the plan (do it). Third, study the results-did the plan work as intended, or
were the results different? Finally, act on the results by identifying what worked as
planned and what didn't. Using the knowledge learned, develop an improved plan and
repeat the cycle. The PDSA cycle is it simple adaptation of the more elaborate problem-
solving method discussed in the next section.
Process improvement achieves the greatest results when it operates within the
framework of the problem-solving method. In the initial stages of a program, quick
results are frequently obtained, because the solutions are obvious or an individual has a
brilliant idea. However, in the long term, a systematic approach will yield the greatest
benefits. The problem-solving method (also called the scientific method) has many
variations depending, to some extent, on the use; however, they are all similar. There
are seven phases, as shown in Figure 5-4. Also shown is the relationship to the PDSA
cycle. The phases are integrated because each phase is dependent upon the previous
phase. Continuous process improvement is the objective, and these phases are the
framework to achieve that objective.
The objective of this phase is to identify and prioritize opportunities for improvement. It
consists of three parts: identify the problem, form the team (if one is not in existence)
and define the scope. Problem identification answers the question, "What are the
problems'?" The answer leads to those problems that have the greatest potential for
improvement and have the greatest need for solution. Problems can be identified from a
variety of inputs, such as the following:
Page 10 of 26
Phase 1: Identify the
Opportunity
Act Plan
• Pareto analysis of repetitive external alarm signals, such as field failures, complaints,
returns, and others (see Chapter 18).
• Pareto analysis of repetitive internal alarm signals (for example, scrap, network,
sorting, and the 100% test).
• Customer surveys.
• Employee surveys.
Page 11 of 26
Problems identified provide, opportunities for improvement. For a condition to qualify as
a problem, it must meet the following three criteria:
Identifying problems for improvement is not difficult, as there are many more problems
than can be analyzed. The quality council or work group must prioritize problems using
the following selection criteria:
In selecting its initial improvement opportunity, a work group should find a problem that,
if solved, gives the maximum benefit for the minim mi amount of effort. The second part
of Phase 1 is to form a team. If the team is .a natural work group or one where
members already work together, then this part is complete. If the problem is of a
multifunctional nature, as most are, then the team should be selected and tasked by the
quality council to address the improvement of a specific process. Goals and mile-stones
are established. More information on tennis is given in Chapter 4.
The third part of Phase 1 is to define the scope. Failure in problem solving is frequently
caused by poor definition of the problem. A problem well stated is half solved. Criteria
for good problem statement are as follows:
• It states the effect-what is wrong, when it happens, and where it is occurring, not why it
is wrong or who is responsible.
Page 12 of 26
As a result of a customer satisfaction survey, a sample of 150 billing invoices showed,
18 had errors that required one hour to correct.
The above statement describes the current state. We might also wish to describe the
sired state, such as "Reduce billing errors by 75%."
In addition to the problem statement, this phase requires a comprehensive charter for
the team. The charter specifies the followings:
2. Objective and Scope. What are the expected outputs and specific areas to be
improved?
3. Composition. Who are the team members and process and sub-process owners?
4. Direction and Control. What are the guidelines for the internal operation of team?
5. General. What are the methods to be used, the resources, and the specific
milestones?
The objective of this phase is to understand the process and how it is currently formed.
Key activities are to define process boundaries, outputs and customers, it and suppliers,
and process flow; determine levels of customer satisfaction and measurements needed;
gather data; and identify root causes.
The first step is for the team to develop it process flow diagram (see Chapter 18). A flow
diagram translates complex work into an easily understood graphic description. This
activity is an "eye-opening" experience for the team, because it is rare that all members
of the team understand the entire process.
Next, the target performance measures (see Chapter 7) are defined. Measurement is
fundamental to meaningful process improvements. If something cannot be measured, it
can not be improved. There is an old saying that what gets measured gets done. The
team will determine if the measurements needed to understand and improve the
process are presently being used; if new ones are needed, the team will:
Page 13 of 26
Once the target performance measures are established, the team can collect all able
data and information. If these data are not enough, then additional new inform is
obtained. Gathering data (1) helps confirm that it problem exists, (2) enables the team
to work with facts, (3) makes it possible to establish measurement criteria for baseline,
and (4) enables the team to measure the effectiveness of an implemented solutions. It
is important to collect only; needed data and to get the right data for the problem. The
team should develop a plan gat includes input from internal and external customers and
ensures the plan answers the following questions:
Data can be collected by a number of different methods, such as check sheets (see
Chapter 18), computers with application software, data-collection devices like hand-held
gauges, or an online system.
The team will identify the customers and their requirements and expectations as well as
their inputs, outputs, and interfaces of the process. Also, they will systematically re-view
the procedures currently, being used.
Page 14 of 26
• Supplier information, such as process variation, on-time delivery, and technical
competency.
It is important to identify the root cause. This activity can sometimes be determined by
voting (see Chapter 18). It is a good idea to verify the most likely cause, because a
mistake here can lead to the unnecessary waste of time and money by investigating
possible solutions to the wrong cause. Some verification techniques are the following:
3. Check the process when it is performing satisfactorily against when it is not by using
the who, where, when, how, what, and why approach.
4. Utilize an outside authority who plays "devils advocate" with the data information,
and reasoning.
6. Save a portion of the data used in the analysis to confirm during verification.
Once the root cause is determined, the next phase can begin.
This phase has the objective of establishing potential and feasible solutions and
recommending the best solution to improve the process. Once all the information is
available, the project team begins its search for possible solutions. More than one
solution is frequently required to remedy a situation. Sometimes the solutions are quite
evident from a cursory analysis of the data.
In this phase, creativity plays the major role, and brainstorming is the principal
technique. Brainstorming on possible solutions requires not only a knowledge of the
problem but also innovation and creativity.
Page 15 of 26
There are three types of creativity: (1) create new processes, (2) combine different
processes, or (3) modify the existing process. Thu first type is innovation is its highest
form, such as the invention of the transistor. Combining two or more processes is a
synthesis activity to create a better process. It is a unique combination of what already
exists. This type of creativity relies heavily on benchmarking. Modification involves
altering it process that already exists so that it does a beta. It succeeds when managers
utilize the experience, education, and energy of empowered work groups or project
teams. There is not a distinct line between three types-they overlap?
Creativity is the unique quality that separates mankind from the rest of the kingdom.
Most of the problems that cause inefficiency and ineffectiveness in organizations are
simple problems. There is a vast pool of creative potential available these problems.
Quality is greatly improved because of the finding and fixing o number of problems, and
morale is greatly increased because it is enormously satisfying to be allowed to create.
Areas for possible change are the number and length of delays, bottleneck, equipment,
timing and number of inspections, rework, cycle time, and materials. Consideration
should be given to combining, eliminating, rearranging, and executing simultaneously
the process steps. in particular, reducing cycle times, lowering I levels, and searching
for non-value-added activities are excellent sources for change, as these typically have
many hidden costs that, if minimized or eliminated, effect a member of processes in the
organization. For example lowering inventory level allows there to he less WIP to be
transported, frees floor space and lessens the management and accounting of the WIP,
particularly if the inventory is time-dated material.
Once possible solutions have been determined, evaluation or testing of the solutions
comes next. As mentioned, more than one solution can contribute to the situation.
Evaluation and/or testing determines which of the possible solutions have the greatest
potential for success and the advantages and disadvantages of these solutions. Criteria
for judging the possible solutions include such things as cost, feasibility, effect,
resistance to change, consequences, and training. Solutions also may he categorized
as short range and long range. At a minimum, the solution must prevent recurrence.
One of the features of control charts is the ability to evaluate possible solutions.
Whether the idea is good, poor, or has no effect is evident from the chart.
Once the best solution is selected, it can be implemented. This phase has the objective
of preparing the implementation plan, obtaining approval, and implementing the
process.
Page 16 of 26
Although the project usually has some authority to institute remedial action more often
than not the approval of the quality council or other appropriate authority is required. If
such approval is needed, a written and or oral report is given.
Answers to these questions will designate required actions, assign responsibility, and
establish implementation milestones. The length of the report is determined by the
complexity of the change. Simple changes may require only an oral report, whereas
other changes require a detailed, written report.
After approval by the quality council, it is desirable to obtain the advice and consent of
departments, functional areas, teams, and individuals that may be affected by the
change. A presentation to these groups will help gain support from those involved in the
process and provide an opportunity for feedback with improvement suggestions.
The final element of the implementation plan is the monitoring activity that answers the
following:
• What information will he monitored or observed, and what resources are required?
Measurement tools such as run charts, control charts, Pareto diagrams, histograms,
check sheets, and questionnaires are used to monitor and evaluate the process
change.
Pylipow provides a combination map to help formulate an action plan to help measure
the results of an improvement. The map, shown in Table 5-1 provides the dimensions
of: what is being inspected, the type of data, timing of data collection, by whom, how the
Page 17 of 26
results will be recorded, the necessary action that needs to be taken based on the
results, and who is to take the action.
This phase has the objective of monitoring and evaluating the change by tracking an,
studying the effectiveness of the improvement efforts through data collection and review
of progress. It is vital to institutionalize meaningful change and ensure ongoing
measurement and evaluation efforts to achieve continuous improvement.
Table
The team should meet periodically during this phase to evaluate the results to see that
the problem has been solved or if fine-tuning is required. In addition, the team will want
to see if any unforeseen problems have developed as a result of the changes. If the
team is not satisfied, then sonic of the phases will need to be repeated.
Once the team is satisfied with the change, it must be institutionalized by positive
control of the process, process certification, and operator certification. Positrol
(positive control) assures that important variables are kept under control. It specifies the
Page 18 of 26
what, who, how, where, and when of the process and is an updating of the monitoring
activity. Standardizing the solution prevents "backsliding." Table 5-2 gives an illustration
of a few variables of a wave soldering process.
Finally, operators must be certified to know what to do and how to do it for a particular
process. Also needed is cross- training in other jobs within the process to ensure next-
customer knowledge and job rotation. Total product knowledge is also desirable.
Operator certification is an ongoing process that must occur periodically.
This phase has the objective of achieving improved levels of process performance.
Regardless of how successful initial improvement efforts are, the improvement process
continuous.
Table 5-2
Belt Speed
Preheat
Temperature
Table 5-3
TABLE 5-3
Page 19 of 26
Visible, audible alarm Temp/humidity control Combe tasks
signals
Foolproof inspection Electrostatic discharge Encourage suggest
Neighbor and self- Storage/inventory control feedback of result
inspection
Continuous improvement means not only being satisfied with doing a good job or
process but also striving to improve that job or process. It is accomplished by
incorporating process measurement and team problem solving in all work activities.
TQM and techniques are used to improve quality, delivery, and cost. Organizations
must continuously strive for excellence by reducing complexity, variation, and out-of-
control processes.
Note that there are many similar approaches to problem solving that deviate slightly
from the one presented here; however, all such approaches provide similar features. A
similar problem-solving approach was used by the Federal Communications
Commission (FCC) when it began its quality effort in 1990. Their seven-step approach
to continuous improvement included identifying improvement opportunities, prioritizing
and selecting potential improvements, analyzing root causes, developing alternative and
selecting choice solutions, testing of solutions, implementing them, and tracking the
retentiveness of the solutions. When it began its quality effort in 1991, the FCC was
receiving roughly 900 license applications per year and had a nine-month backlog of
600 applications. During the first step, the quality improvement team brainstormed on
20 possible improvement ideas, which were then narrowed to five. Next, they prioritized
the opportunities based on an estimate of the cost of poor quality. Upon collection of
Page 20 of 26
data, it was found that over half of the licensing applications received contained errors
that doubled the processing time. Once the team identified this information, they
developed an "opportunity statement" to "increase the percentage of error-free license
applications from 40% to 70'0." To analyze the root causes, the team flow-charted the
licensing process and constructed a cause-and-effect diagram. Through it data
collection effort, it was found that 50% of the errors were due to unqualified persons
completing the license application, and that the license form and instructions were
unclear. As it result, the team redesigned the application process so that the instructions
were clearer and processing guidelines were understandable by the applicants. Before
implementing the final solution, the form was pilot tested to a sample group. A little ov er
it year, the team implemented the new form. Nine months after implementing their new
solution, the percentage of applications received without errors increased from 40% to
80%, the backlog was reduced to 16 applications without staff increases, the speed of
service improved 47%, and the cost of poor quality was reduced by 67%.
Kaizen
Kaizen is a Japanese word for the philosophy that defines management’s role in
continuously encouraging and implementing small improvements involving everyone. It
is the process of continuous improvement in small increments that make the process
more efficient, effective, under control, and adaptable. Improvements are usually
accomplished at little or no expense, without sophisticated techniques or expensive
equipment. It focuses on simplification by breaking down complex processes into their
sub-processes and then improving them.
Page 21 of 26
6. The five S's for workplace organization, which are five Japanese words mean proper
arrangement (seiko), orderliness (seiton), personal cleanliness (seiketso), cleanup
(seise); and discipline (shitsuke).
7. Visual management by means of visual displays that everyone in the plant can use
for better communications.
8. Just-in-time principles to produce only the units in the right quantities, at the right
time, and with the right resources.
10. Team dynamics, which include problem solving, communication skills, and conflict
resolution.
Reengineering
Page 22 of 26
only relied on kaizen but have developed policy management (hoshin tenkai) and policy
deployment (hoshin tenkai) in large part to produce the kind of large-scale
breakthroughs that Hammer and Champy promote. Nor is this concept uniquely
Japanese. Joseph Juran has had a long-standing emphasis on breakthrough efforts
aimed at achieving unprecedented levels of performance.
Six-Sigma
Statistical Aspects
According to Janes Harrington, "Six sigma was simply a TQM process that uses
process capability analysis (see Chapter 18) as a way of measuring progress. Sigma, δ,
is the Greek symbol for the statistical measurement of dispersion called standard
deviation. It is the best measurement of process variability, because the smaller the
deviation value, the less variability in the process.
Figure 5-5 shows a process that is normally distributed and centered with the upper and
lower specification limits (USL and LSL) established at ± 6 (r. For, this situation,
99.9999998% of the product or service will tween specifications, and the
nonconformance rate will be 0.002 parts per million, per billion. The situation
diagrammed represents a process capability index (C,,) A C P of 1.33 has been a
defacto standard. Table 5-4 shows the percent between specifications, the
nonconformance rate, and process capability for different specification limit locations.
According to the six-sigma philosophy, processes rarely stay centered-the center tends
to "shift" above and below the target, µ. Figure 5-6 shows a process normally
distributed, but has shifted within a range of 1.5δ above and 1.5δ below the target. For
the diagrammed situation, 99.9996600% of the product or service will be between
specifications and the nonconformance rate will-be 3.4 ppm. This off-center situation
gives it process capability index (Cpk) of 1.5 with 1.0 being the defacto standard. Note
Page 23 of 26
that the index is calculated differently and, therefore, has a different symbol (CP vs. Cpk).
See Chapter 18 for a detailed analysis of the differences. Table 5-5 shows the percent
between specifications, the nonconformance rate, and process capability for different
specification limit locations 'I ue magnitude and type of shift is
Table 5-4
Figur 5-5
Table 5-5
a matter of discovery and should not be assumed ahead of time. None of the case
studies in the literature have indicated a shift as great as 1.5δ. The automotive industry
recognized the concept in the mid-1980's, evaluated it and deemed it unacceptable. In
fact, the original work of six sigma was based on only a few empirical studies of a single
process.
The statistical aspects of six-sigma tell its that.we should reduce the process variability,
(jr, and try to keep the process centered on the target, p;. These concepts are not new.
They have been long advocated by Shewhart, Deming, and Taguchi and are covered in
Chapter 18.
Other Aspects
Harry and Schroeder use a methodology called DMAIC, which stands for define,
measure, analyze. improve, and control. This approach is somewhat similar but not as
comprehensive as the seven phases of the problem-solving method discussed in this
chapter. Their methodology features a breakthrough strategy using the project
approach, which is similar to the same strategy advocated by Juran. Projects are
identified by the amount of savings they can generate, which is simply good business
sense and is one of the duties of the quality council as given in Chapter 2. The body of
knowledge is not as comprehensive as that for a Certified Quality Engineer or the
information in this textbook. Harry and Schroeder and others emphasized the need for
management commitment and involvement, but these factors are necessary for any
successful program and have been emphasized in this textbook. Except for the
information in the next paragraph, six sigma has little to offer that isn't already available
through oth -r approaches.
Page 24 of 26
project leaders and have five days of classroom training. Black Belts help Green Belts
define their projects, attend training with their Green Belts, and assist them with their
projects after the training. Black Belts receive 160 hours of classroom instruction and
one-on-one project coaching from Master Belts. Master Black Belts provide the
technical leadership and much of the training for the program.
Problems
There are a number of problems associated with the six-sigma methodology. It would
be very difficult and not very cost effective for a small business to develop the required
infrastructure. Even a medium-sized business would have difficulty paying for the high
cost of the training. General Electric has spent over two billion dollars to develop their
infrastructure.
In large companies, there is a great danger that the infrastructure will become a
bureaucracy. At one flagship six-sigma company, a technical employee was
admonished by a Black Belt for fixing several processes rather than turn them into six-
sigma project. It is possible that operating personnel who know the most about the
process will be outside the improvement loop. Certainly the concept of Kaizen is not
compatible with having projects that average $175,000 in savings.
According to stamatis, “Six sigma presents absolutely nothing new to the quality field of
defect prevention. It's little more than an old appraisal methodology that focuses on
problems after they've already occurred."
Iomega, maker of Zip drives, began the implementation of its six-sigma program in
1998. The infrastructure of over 400 Master Black Belts, Black Belts, and Green Belts
out of 3400 employees made impressive improvements. I however, by May, 2001 the
company was losing money, the share price went from $4.00 to 51.75 and the CEO, a
former General Electric manager, was tired along with many employees.26 It should
also be pointed out that Motorola, where the six-sigma concept originated, has lost a
considerable amount of market share of its wireless business and did not show a profit
in 2001 for the first time in many years.
Located 20 miles north of New York City, the Pearl River School District (PRSD) has
five schools: three elementary schools (kindergarten through grade four), one middle
school (grades 5 through 7), and one high school (grades 8 through 12). The district has
about 330 employees and approximately 2.460 students. Ninety-four percent of the
students beginning their schooling in Pearl River complete their high school education in
Page 25 of 26
the district. The percentage of students graduating with a Regents diploma, a key
objective of Pearl River School District, has increased from 63rc in 1996 to 86% in
2001, while the percentage of students in schools with similar socio-economic profiles
has decreased from 61% in 1996 to 58% in 2000. In comparison, the highest reporting
district in New York had a 90% passing rate in 2001. In 2001, PRSD exceeded their
benchmark on three of the eight State Regents content exams and was within three
percentage points of the highest reporting district on the other five content exams.
'Helping students to prepare for college, PRSD has improved Advanced Placement
course performance from 34% of the students achieving a "3" or better in 1997 to 76%
in 2001, while dramatically increasing the percentage of students taking the AP courses.
PRSD has reduced per-pupil expenditures (PPE) 9% over the last 11 years (from
$14,563 in 1990/1991 to $13,180 in 20W200 1.
PRSD's overall staff and faculty satisfaction rate, as measured using a recognized
national survey, has increased over the past four years from 89% to 98% for staff and
860, to 96% for faculty. These results exceed the survey databank's highest reported
score for overall faculty and staff satisfaction combined (the composite faculty/staff
satisfaction benchmark is 89%).
PRSD uses curriculum maps, developed by teams of teachers and senior leaders, to
align its entire K- 12 curriculum to state and national standards and to align instruction
within and across all grade levels. These maps, which detail the content area covered
as well as the method of instruction and assessment techniques used, are adjusted
quarterly based on data analyses and through benchmarking the best practices of other
school districts.
PRSI) has a systematic nine-step process for strategy development at the district level,
executed by the Board of Education and the Administrative Council. PRSD's
educational design and delivery process uses a PDSA cycle to systematically plan,
prioritize, design, and pilot new educational programs before they are introduced widely.
Page 26 of 26