In This Chapter
► Understanding what process optimization tools do ^ Knowing about statistical analysis tools ^ Seeing how these tools are used by the Six Sigma practitioner Finding out about available software tools and technologies
Ou don’t have to be a programming whiz or a Ph. D. statistician to apply Six Sigma in even the most rigorous of situations. All the necessary tools are well defined and readily available, and they enable the Six Sigma practitioner to directly connect Six Sigma theory to practice. Each of the strategies
And methods discussed in Part II are implemented through these tools.
The Six Sigma tools marketplace has many products available. But fear not — they all sort into just a few categories. Chapter 12 addresses the tools designed
For management. This chapter discusses the tools created and honed specifically for Six Sigma practitioners. When applied to Six Sigma projects, these
Tools help you make the outcomes accurate, presentable, and reusable.
Most Six Sigma tools are implemented in software programs, most of which run directly on a PC. These programs perform the many process and analytical functions for you. Unleashing your Six Sigma genius is just a few clicks away.
As a bonus, this chapter also provides advice on the platform and technology issues involved in tool selection and application.
V
The Practitioner’s Toolkit
To be a successful Six Sigma practitioner, you must be accomplished in the application of the Six Sigma analytical and statistical concepts and formulas defined in this book. In the early days of Six Sigma — that was way back in the 1980s — such analyses and statistical processing were largely a manual and complex effort, confined to the world of the statistical geeks. Unless you
Were both an accomplished statistician and a computer programmer, the
Practice of Six Sigma was off-limits.
Today, this has all changed. The methods and tools are now well defined.
Andthe power of desktop computing, combined with several generations of accelerated development in application software, has made all the analysis,
Including advanced calculations and data display, a relatively easy and simple thing to do. With the wave of a mouse, you can easily execute the most complex functions, run advanced simulations, conduct a Design of
Experiments and create impressive charts and plots.
In short, as a Six Sigma practitioner, you are now enabled by a fully capable
Set of practical application tools. As with all tools, you have to know how
To use them properly and interpret what they’re telling you, but after you understand the theory and strategy of Six Sigma, you can use these tools to directly apply your new-found knowledge — quickly, comprehensively, and
Accurately.
Practitioner tools come in many colors and flavors, but they all fall into one of two primary types: process optimization tools and statistical analysis tools. Each plays a critical role in the successful application of Six Sigma.
F Process optimization Tools enable you to design, simulate, and optimize
Work processes. These include tools for creating process and work flow diagrams, building cause-and-effect matrices, constructing fishbone diagrams, developing SIPOC (Suppliers-Inputs-Process-Outputs-Customers) diagrams, assessing process capabilities, and more. The goal of these
Tools is help you see how work is performed and identify where the source of problems is.
F Statistical analysis Tools enable you to analyze data collected either
From the real-world performance of a product or process, or as the output of a simulation or experiment. These include basic statistics tools, and tools for analyzing variance, conducting regressions, performing Design of Experiments (DOE), and building control charts, plots, tables, and graphs. The goal of these tools is to help you turn data into knowledge such that you can make informed decisions.
You have choices in selecting and applying these tools. Because each tool is based on fundamental principles or mathematical formulas, you could work
Them out longhand with pencil and paper. You could use a slide rule, or even
A calculator. But in this modern world of personal information technology, we have software programs that implement every tool — quickly, cheaply, and easily. These programs perform every function for you. They also combine
Multiple tools into kits and present them in a logical order.
Most of these tools run only on a desktop PC under the Microsoft Windows operating system. If you’re a Mac or Linux user, or if you wish to deploy these
Tools via an intranet or through the Internet as Web-based applications, it’s improving, but it has been slim pickin’s. The last section of this chapter
Explores this further, but the simple truth is that the Six Sigma toolkit is primarily a Windows environment.
Process Optimization Tools
You practice Six Sigma for one reason, and one reason only: to improve your
Business processes. Therefore, those tools, directly facilitating efforts to
Optimize the many types of work processes in a business, are your primary weapons in your battle against ineffectiveness, inefficiency, variation, and
Waste. All the other tools — be them managerial or analytical — are in a supporting role. It’s all about improving the process.
We use the term "process optimization" here as a catch-all to describe both
The subject area — processes — and the purpose — optimization. Within this
Broad category are many supporting tasks, for which sub-categories of tools exist. These are summarized in Table 11-1.
Table 11-1_Process Optimization Tools
Process Tool Role
The SIPOC Suppliers-Inputs-Process-Outputs -
Customers. Create a high-level process map with a few key details about each of the key contributing elements.
CT (critical to) tree Critical to. . . tree. Identify, organize, and
Display parts of the process according to areas of critical importance.
Modeling Define and design processes, including
The flow of work or material, the timing of activities, resources consumed, and points of decision, inspection, and delivery.
(continued)
Table 11-1 (continued)
Process Tool Role
Simulation Simulate the flow of work and material
Through a process based on the model, and analyze the results of the simulation for overall effectiveness and efficiency. Find defects, errors, bottlenecks, variation, and non value-added elements.
C&E (cause-and-effect) matrix For the outcomes of any process, define
All the contributors, weight their effects, and determine the significant contributors to the outputs.
Fishbone diagram Create a high-level C&E in the form of a
Tree structure, with categories for each major type of contributor. A method for capturing potential causes and inputs to a process.
FMEA (failure mode effects analysis) For any activity or item, define the potential failure modes, including the likelihood of occurrence, and the ability to detect and characterize the effects of those
Failures.
Capability and complexity analysis Analyze the tradeoffs between product
Complexity and process capability, and define the proper configuration of each
To achieve desired outcomes.
Plans Use the outputs of simulation and analy -
Sis to define how data will be collected and how the processes will be controlled
And audited.
The S1POC
SIPOC, pronounced sy-pok, is an acronym that stands for Suppliers-Inputs-Process-Outputs-Controls. The SIPOC is one of the most fundamental building blocks in the Six Sigma process. With this tool, you build your first controlled and organized view of your work process and set the foundation for applying
The breakthrough DMAIC strategy.
SIPOC is one of those handy reminder acronyms that contains the terms in their proper order, helping you remember not only the five high-level elements of a process map, but the order in which they occur.
|
Table 11-2 The SIPOC
|
|
S: Suppliers
|
Suppliers are systems, people, organizations, or other
|
|
Sources of the materials, information, or other resources
|
|
That are consumed or transformed in the process.
|
|
I: Inputs
|
Inputs are materials, information, and other resources
|
|
Provided by the suppliers that are consumed or trans -
|
|
Formed in the process.
|
|
P: Process
|
The process is the set of actions and activities that trans -
|
|
Form the inputs into the outputs.
|
|
O: Outputs
|
Outputs are the products or services produced by the
|
|
Process and used by the customer.
|
|
C: Customer
|
Customers are persons, groups of people, companies,
|
|
Systems, and downstream processes that receive the
|
|
Output of the process.
|
Developing a SlPOC
You build a SIPOC from the inside-out, beginning at the center, with the process — of course! It’s a six-step approach:
1. Identify the process you wish to map and define its scope and boundary points.
Using action verbs, describe what the process is supposed to do, and in
How much time. Define its starting and ending points.
2. Identify the outputs.
What are the products and the services that will be produced by the process?
3. Define by name, title, or organizational entity the recipients (the customers) of the outputs.
4. Define the customer requirements; what do the customers expect?
What will they demand? What will they be entitled to in their fair
Exchange of value?
5. Define the inputs to the process.
Identify the human, capital, information, materials, and natural resources required by the process to produce the identified outputs.
6. Identify the sources (suppliers) of the inputs.
With this information in hand, you now have a fully-contained high-level view of any process. This alone is one of the most powerful tools you can use, because it sets the conditions for the DMAIC of Six Sigma. With the SIPOC,
You now have the basis for defining and characterizing the process itself, thecontext for measurement, and the basis for analysis, identifying areas of improvement, and homing in on your targets of control. SIPOC software tools,
Like iGrafx, SigmaFlow, and Process Model, help you capture, organize, and display this information.
1 hear Voices
And who’s talking? The loudest voice you hear should be the Voice of the Customer. Known as VOC, Voice of the Customer is a practice within Six
Sigma process optimization for ensuring that the customer’s requirements, expectations, and entitlements are flowed into the process. But that’s not the
Only "voice" in Six Sigma. Competing with VOC in your mind are two additional voices you need to consider, the voice of the process, VOP, and the
Voice of the business, VOB.
V Voice of the customer (VOC): This is the voice calling back at your
Process from beyond the output that offers you compensation in return for satisfaction of its needs and wants. These voices are the needs,
Wants, and desires of the customer, generally spoken as the customer
Requirements.
V Voice of the process (VOP): The process must meet the requirements of the customer, and the ability of the process to meet these requirements is called the VOP. This is a construct for examining what the process is
Telling you about its inputs and outputs and the resources required to complete the functional transformation.
T-" Voice of the business (VOB): This is the voice of profit and return on investment. At the end of the day, every endeavor has to enable the
Business to survive, grow, and meet the needs of its employees, investors, and the community.
What’s critical} Look in the CT tree
In Six Sigma, you always look for the causes. You want to know what’s behind something, what’s causing the outcome — find those "critical *s." In optimizing a process, you have to understand what’s critical to the successful outcome of
Each step, so you can focus on optimizing the right things. This is what a CT
Tree is for.
CT stands for "critical to. . . ." Critical to What, You may ask? The answer is, simply: critical to whatever matters. Depending on what you are analyzing and optimizing, this could mean anything from the satisfaction of the customer, to the quality and reliability of the product, to the cycle time of manufacture or
The cost of the delivered product or service.
The Six Sigma practitioner will often refer to the general CT case as "CTXs," in reference to the many variables that influence a desired outcome. But there are specific cases in process optimization, and the CT tree is a tool that helps you identify and characterize the influencers on specific outcomes.
Most CT trees begin with the output of the SIPOC, customer satisfaction, at
The top, and the others are subordinate. And, while the CTXs in Table 11-3 are the most commonly used, you are free to invent and apply any X that fits your need. We’ve seen everything from CTQ (quality) to CTD (delivery), and even CTC (cost). It all depends on your application.
Table 11-3_Applications of the CT Tree
Critical To… Title Definition
Satisfaction CTS What contributes to customer success?
Quality CTQ What contributes to process or product quality?
CostCTCWhat contributes to the cost or final price?
DeliveryCTDWhat contributes to the cycle time to deliver?
In creating a CT tree, begin by defining your specific area of application, such as customer satisfaction (CTS). This is your trunk. Then, define the branches, by category, of the key contributors to customer satisfaction, which may include availability, price, selection, accuracy, presentation, performance, andso on. These are your subordinate CTXs. Lastly, define the leaf nodes: thecauses or influences on those categories of customer satisfaction.
An example CT tree is shown in Figure 11-1. In the main window, you can see that the root node of the tree is labeled as Critical to Satisfaction. The branch nodes for this CTS tree are identified in blue, as selection, price, performance, and availability. The leaf nodes are then identified and defined for each branch. In this example, one leaf node, configuration, has contributing sub-leaves as well, deemed worthy of inclusion as being critical to the configuration.
Also note in Figure 11-1 that the upper left window contains the same information, assembled in a different manner. That’s fine; it’s just another way of
Showing the same thing.
Modeling a process
A Six Sigma process is defined precisely — very precisely — down to the last detail of activity, resource, decision, dependency, and value. Only in this way can a process be sufficiently measured and analyzed, leading to breakthrough
Improvements and, ultimately, effective controls. The process model is our representation of this precise process definition, and the practice of process
Modeling is therefore at the very heart of Six Sigma.
Process modeling has been practiced for decades, and the fundamental concepts of process modeling are nothing new. The Six Sigma style of process modeling has a few different wrinkles, however, and as a result, the Six Sigma process model bears only a superficial resemblance to its ancestors. In the world of Six Sigma, the process model is characterized in mathematical terms, permitting us to perform a plethora of statistical analyses on its various parts and pieces. Each node, each function, and each activity is backed by numerical descriptions and quantifiable attributes, enabling us to see the
Process in this mathematical light.
^BC» In Six Sigma process modeling, you are characterizing a practical situation Y~f\ In ways that permit it to be described in statistical terms, allowing you to ( liW ) develop statistical solutions, which you then apply back into your practical environment. On the surface, a Six Sigma process model looks like a flow
Chart, but underneath, it’s a raging mathematical beast.
The 1, 2, 3s of process modeling
Process modeling is rigorous. It requires to you understand the whole of
Things as well as their detailed intricacies. A process model takes time to
Build. It crosses boundaries and borders, and sometimes just in the process
Of creation, you’re likely to uncover issues and even step on a few toes. But don’t despair — your process model speaks the truth! It’s the basis for
Understanding and breakthrough improvement. The time you spend building it will reap its rewards in performance and satisfaction.
Six Sigma process modeling begins with the building of process maps.
Thepaths, encounters, decisions, and destinations on these maps are then
Annotated and defined in quantitative terms, including such measures as
Value, time, resources, yields, and the statistical distributions around each. The outcome of this process is the statistical basis for simulating the process andanalyzing the results.
Drawing a process map
A process map looks like a flow chart, and, at the top level, that’s exactly
What it is. A process map is a picture of the activities and events in a process.
Figure 11-2 is an example process map.
You can draw a process with a pencil and paper, or with a drawing tool like Microsoft Visio. However, only more advanced tools like iGrafX Process, SigmaFlow, and Traxion permit sufficient definition and attributing to enable Six Sigma-class simulation and analysis. It’s important to follow a consistent set of conventions when using shapes, connectors, and other drawing elements.
Figure 11-3 is an example set of some drawing conventions used in process mapping. While these icons are typical, the exact shape may vary slightly
From one tool to another.
Use sufficient process mapping shapes across the organization. With the emergence of the Business Process Modeling Notation (BPMN) standard,
Process modeling is poised to become more routine and effective. In the
Meantime, your best bet is to adopt the conventions used in one of the software applications, and standardize their set of icons for use across your team or organization. Choose something, and then stick with it.
At this stage, you’re not worried about the details of what happens inside each of these boxes. Your goal is to capture each of the steps, identify their basic function, and connect them in the manner that represents the process.
( ) Start / End
Process
O
Decision
Document
Ј7
Data
Figure 11-3:
Mapping icons.
Storage
Transport
Inspection
Defining the process points
Once you’ve drawn a process map, the next step is to define explicitly each of the map’s objects. You must be precise — and quantitative; the accuracy of your process model depends on it. If you are using a process modeling tool,
Your tool will include prompts for the attributes at each node in the model.
These attributes are numerous. The categories of process element definitions include:
Operation cycle time Of the process element, including its average time to complete, the variation in time called the standard deviation, and perhaps a distribution curve to represent all the possible completion times as well.
F Resources Used in the process element, including human, capital, and natural resources. The better tools will permit you to identify resources by name and type, and then later track their utilization during simulation.
V Value added By the process step, in the units of measure that mean the most to your organization. At a minimum, you must be able to define whether the process step is value-added (VA) or non-value added (NVA).
F Costs Of the resources consumed. These include the costs of personnel,
Facilities, direct material, and can even include indirect costs.
The closer you can come to defining costs in the same terms as your accounting system, the better. Ultimately, you will be reconciling cost and claiming value that will be verified by your accounting department. Get the bean counters involved up front and make them your partner by counting your beans the same way they count theirs.
Swimming in fanes
A recent development in process modeling is a visualization technique called Swim Lanes. Remember how we told you that processes cross boundaries and borders? Imagine you are the customer: You’re in Lane 1. Sales is in Lane 2. Invoicing is in Lane 3. And so on. Time flows from left to right. The process crosses lanes as it traverses departments on its journey from start to finish. See Figure 11-4.
The swim lane is an effective visualization technique that enables each functional contributor to a process to understand their role, while at the same
Time giving everyone a chance to see just how complicated the process may
Be within your organization. Remember, each time we cross a lane we have in essence created a supplier — a customer interaction that implies needs,
Wants, and desires that must be met.
To be, or not to be
Process modeling is typically an exercise in defining how you envision your
Process can work sometime in the future, after implementing the changes
That would enable your new concepts. It’s the To-be State of affairs. Modeling the future in this way is powerful, because it provides you the opportunity to
Examine your plans in detail and consider the options before you implement
The changes. Coupled with available simulation capabilities, you wouldn’t
Dream of making changes without first modeling them.
The other application for process modeling is to create a model of today’s reality: the so-called As-is State. Few organizations do this. They are so eager to dismiss with today’s problematic world that they leap-frog straight to the dream of tomorrow’s possibilities. Big mistake! The only excuse for not modeling the As-Is process is if you are implementing something brand new and there is no existing process. Otherwise, if a process exists today, model it first. There are compelling reasons for this.
Set the baseline. Before you can measure the effects of your sweeping changes, you must first characterize the present conditions. By using
The same process modeling techniques to characterize today’s as-is state as well as the future to-be state, you have the basis for measuring the effectiveness of your process optimization effort.
See the process. There are three conventional views of a process:
• What you think is going on
• What is really going on and
• What should be going on
These are three distinctly different states of a process, and it is precisely what we are trying to do with Six Sigma. The only way to achieve this is by doing Step 2, mapping what is really going on. Then and only then can we move to Step 3.
F Stimulate closed-loop behavior. Your investment in modeling primes the
Pump for breakthrough performance improvement. To continue the cycle of improvement, your model should be a dynamic, living entity, where at
Any point in time, your model and reality are in sync. Modeling the as-is
Condition from the beginning stimulates this closed-loop behavior. Whither the S1P0C7
A SIPOC is not a process model, and vice-versa. But even if you are rigorously
Process-modeling your business, there is a role for the SIPOC. The SIPOC is applicable as an early-stage tool, and for high-level views of processes.
Simulating a process
Simulation is useful, but it’s static: It makes a process model come alive. A process map by itself just sits there. The process map is the first half of
Process modeling. While there is great benefit to the process map, its benefit
Is further enhanced when we can extend this static view or picture of our work activity into a dynamic view or movie of our work activity.
Simulation is the other half. Once you have mapped a process, simulation isthe practice of stimulating your model into action. Simulation tools are advanced computer-based programs that ingest all the parameters of your model and run dozens, hundreds, and even thousands of trials in your computer. By doing this, you generate simulated real-life outcomes without having
Made a single physical change to the process, and you generate vast amounts of detailed results for statistical analysis.
The more advanced simulation tools animate your process map as the simulations run, tracing for you where your process will operate smoothly and where your bottlenecks are — as they occur. From the act of watching your
Model in action, and then in analyzing the results of the simulations, you can pursue the goal of process optimization. It is an iterative process.
The simulation environment
Simulation requires both a computer and considerable expertise. The programs are sophisticated. You must set up the simulator precisely, with data
Generated specifically for the simulator by the modeler. Fortunately for us,
The advanced software tools on the market perform this task seamlessly. Simulators are demanding computational programs that can require advanced
Understanding as well as capable computers to run swiftly and properly. You
Will want to check the specifications of your computer against the requirements of your choice of simulator. To accommodate less-capable machines, several of the more advanced simulators also contain switch settings that permit you to operate simulations with selected real-time features disabled. All of the same results are generated, but the simulation will run longer and feature less animation.
Configuring simulations
To run a simulation of your process model, you must set a number of configuration parameters. These will typically include the following:
Is Number or duration of runs: You can specify the number of passes through the process that the simulator will perform, or specify the overall elapsed simulation time. The higher the number of runs or the longer the simulation executes, the more statistically representative the results.
Randomization: Many simulators permit the specification of random inputs, which better mimics the variability that occurs in real life. For
This to be effective, you need to understand the nature of the variability in your process inputs.
Patterns: Your process may encounter predictable patterns of variation,
Such as work shifts, days of the week, month-end effects, batch inputs, time zones, distributions, and the like. These can be specified in the
Simulation.
Data storage: Specify the nature of the data to be stored from the simulation. Also, specify the collection of data snapshots, either from elapsed
Time or after specified numbers of runs.
V Interactivity: Some simulators permit you to interact with the simulation as it runs. In this way, you can modify certain parameters and observe the
Resulting behavior. Simulation results
The results of simulations can be startling and invigorating! They are almost always full of surprises. Rarely is the outcome just what you expected. More
Often, simulation results reveal unexpected connections and dependencies that cause you to rethink, redesign, and re-plan your process.
The results of process simulations are usually available in the form of standardized reports generated by the simulator. In rare cases, the simulator
Stores the simulation data in a relational database, but in most cases, the
Data are presented as a list output or a set of canned reports. The more advanced tools permit you to specify custom statistics for additional viewing. The report statistics categories are summarized in Table 11-4.
Most simulators produce only rudimentary reports. Sufficient analysis will require you to export the simulator data to an analytical tool, like Minitab.
Table 11-4_Simulation Report Categories
Category_Purpose_
Time Show overall transaction times, and times per department,
Process, or activity.
Cost Report statistics on costs for all resources, transactions, and
Activities.
Resources Report statistics on resource utilization, time, activities, and
Costs.
Queues Report on bottlenecked processes or transactions waiting in
Queues due to resource, inputs, or other constraints.
Cause-and-effect (C&E) matrix
In Six Sigma, Y=f(x). All outcomes (Fs) are the result of some inputs s) and
The transformations that acted upon them: cause and effect. An effective tool
For the process analyst is the cause-and-effect matrix, known as the C&E
Matrix.
The C&E matrix Is an extension of the C&E diagram or fishbone chart, the brainchild of Kaoru Ishikawa, who pioneered quality management processes in the Kawasaki shipyards, and in the process became one of the founding
Fathers of modern business management. Cause and effect helps the Six Sigma
Practitioner to identify and prioritize the relationships between several inputs and the resulting outcomes. With the C&E matrix, you can identify, explore, and graphically display all of the possible causes related to a problem or condition and search for the root cause. An example of the C&E matrix is shown in Figure 11-5.
You can use a C&E matrix to examine a top-level complex process or system, and you can use it for mid-level, less complex processes and systems. For
Top-level applications, a C&E is used to relate process outputs to the customer
Requirements; it focuses your improvement efforts and identifies projects.
For mid-level applications, the C&E relates process inputs to process outputs and can be used to prioritize tasks and projects.
The system-level software tools that implement C&E automatically database
This information and carry it forward into further activities, including failure
Mode effects analysis (FMEA), control plans, and data collection plans.
Dem’ fishbones
A variation on the C&E matrix is the fishbone diagram, a brainstorming tool used to explore and display sources of variation or influence on a process. With the fishbone diagram, you can quickly create the inputs to a C&E matrix, identifying the key sources that contribute most significantly to the problem
Or process being addressed. The fishbone diagram also serves as an affinity mechanism for relating and categorizing inputs.
A fishbone diagram is so simple that it can be done on a whiteboard, notepad, or even a lunch napkin. However, the software-based tools will also capture,
Categorize, and promote the data for you.
To create a fishbone diagram, you identify the major categories of influence
On an outcome. Within those categories, you list the causes, as shown in
Figure 11-6.
Method
Man
Times – Process Time-Queue Time-Comp Downtime-Freq of Updates -
Figure 11-6:
The aptly-named fishbone diagram.
Holidays – Time of the day-Call Volume -
Carrier Updates Ticket Types
Postal Service – Call Routings -
Computers
PO Damage-Traveler Profiles -
Company Profiles
Training
Self
Internet—
Systems
Resources/Shift— Experience Level —
T1 – Lines-.
Dial up Line Phone Service Maintenance Servers Terminals
Material
Delivery Defects
Machine
FMEA: Failure mode effects analysis
Failure is nasty business. Product failures can mean everything from unhappy customers to harmful outcomes. Process failures result in poor products, lost profits, or both. The failure mode effects analysis is key to reducing or eliminating the risk of failures. The concept was first developed in the aerospace industry. And being from the aerospace industry, it is therefore universally
Referred to in its acronymic form, Fmea.
The FMEA provides you with a structured approach to identifying the potential ways a product or process can fail, how readily you can detect the failure
And the effects of those failures, so you can reduce the risk of either their
Occurrence, or impact, or both. Using the FMEA, you can further prioritize
The actions to be taken to reduce failure risk, and you can evaluate your design and control plans for their robustness to failure.
The FMEA is invaluable in applications where processes or products have safety or security implications, but it is equally applicable to any process or
Product where failures have a material impact on customer satisfaction or measurable business success.
The FMEA is a structured yet simple way of simulating the risk associated
With a particular event occurring. It helps us to find and focus our efforts on
The more significant contributors to our success or failure. It’s an excellent tool to funnel down the most likely contributors or *s to process optimization efforts.
Applications of FMEA include
Is Design FMEA, For analyzing product designs prior to release into production. The DFMEA is conducted early — well in advance of first builds — with a focus on product functionality.
Process FMEA, For analyzing design, manufacturing, assembly, distribution, services, support, and other processes. The process FMEA is
Directed towards process inputs.
F Product FMEA, Addresses failure modes possible in products or projects. t-" Software FMEA, For analyzing failure modes in software applications.
You can build the FMEA from your Process Map, the C&E matrix, or even the fishbone diagram. In any case, the approach is the same: You add three new
Categories of information to the identified failure outcomes or effects, as
Listed in Table 11-5.
Table 11-5_Primary Elements of the FMEA
FMEA Element Definition
Severity of impact Assign a normalized score to the severity of the
Impact in the event of a failure.
Probability of occurrence Evaluate and assign a probability score to the
Likelihood that the failure will occur.
Likelihood of detection Assign a probability score to the likelihood that
The current controls will detect the causes and therefore prevent either the failure itself or its effects from having impact.
Armed with these data, you are positioned to critically analyze the failure
Modes in your system, process, and products. The analysis phase of the FMEA
Is the process of determining probabilities and ranking the results. One of the primary indicators is the risk priority number (RPN). The RPN is simply the
Product of the three elements:
RPN = severity rating X Probability rating X Detection rating
Idiot-proofing
A Japanese manufacturing engineer at Toyota
Named Shigeo Shingo is credited with creating
And formalizing an approach to quality management called Poka-Yoke (pronounced "PO-kah YO-kay), which loosely means Mistake-proofing
(the literal translation is to avoid inadvertent
Errors). Poka-Yoke is used to prevent the inadvertent causes that result in defects, mostly
Using simple, low-cost methods for prevention.
A Poka-Yoke device is any mechanism or procedure that either intercedes to prevent a mistake from being made, or makes the mistake so
Obvious as to eliminate it. Poka-Yoke efforts or
Devices make it nearly impossible to make a mistake. They are especially relevant where
Humans are part of the process effort, because humans sometimes inadvertently forget or are likely to do things differently on occasion.
Examples include such everyday tools as lockout mechanisms, electrical connectors that are specially shaped to prevent reversed plug-in, and overflow prevention systems. The fact that your car’s gas tank has an inlet smaller than the
Size of the filling nozzle, preventing you from
Putting leaded gas into your car, which requires unleaded gas, is an example of Poka-Yoke.
For each failure outcome, you can plot the RPN in a Pareto chart, like the one in Figure 11-7. With this, you can see instantly where you should focus your
Attention.
The RPN is a primary indicator, but it is not the only important output of an
FMEA. It’s vital for you to consider the relationships between all the elements, including severity, probability, and detection, as well as cause, actions, conditions, and other circumstances. Because failure, by definition, is unwanted, you turn over every stone in your FMEA.
The outcome of the FMEA leads you directly to examine your fundamental
Designs, funnel reports, control plans, and data collection plans. Once you
Have updated these, you run the FMEA again — and again — until your failure
Risk profile is within acceptable tolerances.
KISS and tell: Capability-complexity analysis
Remember KISS? No, not that 1970s rock band, but that elegant acronym of simplicity: Keep it simple, stupid. Why do we say that? Because the simpler
You keep things, the fewer chances there are for something to go wrong. Everyone knows that.
At the same time, what about those special people who seem to be extra
Capable and can seemingly handle anything, no matter how much you heap on them? They don’t have as much trouble with the extra load. It’s obvious
That the more capable you are, the more you can handle.
When you put those two concepts together in the context of complex products and processes, it leads you to examine the precise relationship between
Complexity and capability. In Six Sigma terms, we can define, measure, and control both the complexity of our products and services as well as the capability of our processes. So, where’s the point of optimization?
F How capable must our processes be to handle the complexity in our
Products and services?
V How complex can our products and services be in order to handle them with the capability of our processes?
Is If we’re going to introduce new complexity into our products, how much must we ratchet up the capability of our processes to handle it?
Is If we introduce new complexity into our products and services and don’t ratchet up the capability of our processes, what’s the increase in our
Defect rate?
These are vitally important questions, and, to answer them, the Six Sigma practitioner applies capability-complexity analysis (CCA) in the pursuit of
Process optimization. Your process mapping and modeling depend on the balance of settings in a CCA, as do the C&E analysis, FMEA, and other tools.
Because the calculations required to compute the quantitative values of a
CCA involve the manipulation of multiple variables simultaneously, it is ideally suited for computer-based application software. A CCA program solicits your input for complexity parameters about your product or service as well as capability and control parameters about your sigma capability, static
Mean offset, and dynamic variation expansion factors. It will then compute your short – and long-term defect rates and yields per-element and per-unit.
An example CCA display is shown in Figure 11-8.
The more advanced CCA tools will permit what-if analysis where you can set outcome metrics and determine what changes in process capability or product complexity are required to achieve them.
Figure 11-8:
Capability-complexity analysis.
Funnel reports
^BE» Six Sigma is all about finding those Critical few Influencers out of the Trivial
4y^\ Many Candidates that affect the outcome. The number of possible contribu -
( iMl ) Tors can be extremely large at the beginning of a process optimization or
\UJ|/ Problem solving effort. In fact, it sometimes seems so overwhelming that we
Are de-motivated to solve the problem.
Six Sigma to the rescue. Inherent in the Six Sigma methodology is a process called variable reduction. Six Sigma is almost automatically reducing the number of contributors or funneling the *s to find the so-called "critical *s" or "vital causes." When you have these, you have the basis for obtaining
Breakthrough performance improvement.
Funnel Reports help you filter through the trivial many, extract the critical few, and manage them. The sources of information for the funnel report come
Primarily from the CT trees, C&E matrices, and FMEAs. The more advanced
Software tools will import these automatically. In the Funnel process, each candidate is subjected to a set of analytical and statistical considerations,
Which serve as tests to qualify if the cause is vital.
The outputs of the funnel report are considered to be the most likely causes
Of our problem or process deficiency. The next step is to further funnel down this list into the actual root causes using statistical inference and other experimentation tools.
Plans
The Six Sigma practitioner produces and manages a set of plans that affect the
MAIC elements of the breakthrough strategy. The data collection plan ensures
The measurements. The control plan ensures management of the critical *s,
And the audit plan addresses the ongoing monitoring of the vital causes.
Data collection plan
The data collection plan provides a concise and focused set of directives and
Actions required to collect all necessary data associated with a process or within a Six Sigma project. The data collection plan can be voluminous, in that it addresses not only the content, but the reliability, availability, and
Presentation or formatting of the data.
Identifying the information at a high level can come from a number of sources,
Including the process model, the CT tree, the C&E matrix, the FMEA, and the
Funnel report. The more sophisticated Six Sigma software applications will populate a data collection plan template directly from these other tools. The data collected feed both the control plan and the audit plan.
The more elusive goal is in the manner by which the data themselves are collected. You will almost certainly need to work with members of your Information Technology team to determine how best to gather the data
Properly. The issues you need to address here include:
F Data sources: You’re best off if you can get the data directly from the point of origin, as generated — and verified — by the originator.
V Data timing: Transactional data from operations changes regularly. The timing of when you pull the data is as critical as the data itself. Getting the right data at the wrong time leads to bad data.
F Data stability: People change the basis of operational measures regularly and without notice. This is a configuration management problem — assuring that the definition of the data you’re depending on doesn’t
Change out from under you.
Data format: The physical formatting of data is critical. Be certain to
Identify how you want the data to be formatted in your collection plan.
F Data transfer: Specify how the data is to be shipped to you. By far the best way is for an automatic extract and transfer to occur on a scheduled basis. By far the most common way is for someone to periodically e-mail you some sort of extract. Press hard for the former.
Your plan will include all of the above and should be agreed to formally by all
Involved parties.
You’d be well served to apply some of these Six Sigma process tools to the process of data collection. The validity of your process optimization effort is only as good as the data upon which your decisions are founded. Data is a slippery beast; don’t underestimate the effort required to do it right.
Control plan
Based on the fundamental concept of Y=f(X), If you can control the Xs that dominate the outputs of interest (Y), you will have an improvement that
Lasts. The control plan directs your focus on the vital cause critical XS and ensures all participants understand the activities, items, and specification
Limits required for your process to be in control. The control plan is a proactive
Effort to assure long-term performance and also a call to action if a triggering event occurs, indicating the process performance is deteriorating.
Your control plan (see an example in Figure 11-9) is a key Six Sigma management tool. It’s a one-stop reference view of all the vital contributors to the
Success of your process, and it contains sufficient detail to exercise sharply-focused management controls.
Manage your control plan closely. Solicit broad support from management and affected contributors, including approvals and signoffs. Manage configuration changes to the plan closely as well, coordinating changes officially. If
Everyone operates according to this plan, you will be successful. Make it
Happen! Audit plan
The audit plan acts as the measurement tool for the control plan. When your control plan is in place, the audit plan is your means for regular measuring and monitoring of the outcomes.
Statistical Analysis Tools
At the heart of Six Sigma are the statistical tools (see Table 11-6). These enable the Six Sigma practitioner to first analyze practical problems statistically, and then to craft statistical answers that enable breakthrough practical
Solutions. The statisticians who pioneered Six Sigma forged the developmental application of these tools through grit and determination. Today, with the
Benefit of powerful desktop software applications, we merrily point and click
Our way through.
This section is an overview of the suite of Six Sigma statistical analysis tools and will show you which tools are applied in practice. These are the tools used traditionally by the Six Sigma Green Belts and Black Belts. Refer to Chapters 5 and 6 for the theory behind the application of these tools.
This section is not a tutorial on the statistical analysis tools of Six Sigma.
That’s a whole textbook in itself, and you have to invest in Belt training to master the applications! This general overview of the tools will show you where you should use them.
Table 11-6_Six Sigma Statistical Analysis Tools
Stats Tool Role
Basic stats The basic and descriptive statistics, such
As averages, ranges, variance, and so on, used routinely in Six Sigma analysis
Plots and charts Histograms, Pareto charts, control charts
Time series Specific tools for analyzing results of data
Collected over time — trends, decompositions, moving averages
ANOVA (analysis of variance) Analyze variances, test for equality of vari -
Ances, and determine whether there is a valid relationship between variables.
Tolerance analysis The analysis of margins and tolerances to
Determine optimal design specifications
DOE (Design of Experiments) Systematically investigate the process or
Product variables that affect product quality
Process capability analysis Determination of the capability of a process
To perform to expectations. The output is a numerically defined index of capability.
(continued)
Table 11-6 (continued)
Stats Tool Role
Regression Determining the strength of the relationship
Between a response variable (Y) And one or more predictors (Xs).
Multivariate analysis The analysis of data from multiple measure -
Ments on various items or subjects. The output is a graphical picture of the various
Relationships.
Exploratory analysis Methods used to explore data before apply -
Ing more traditional statistical analysis tools
Measurement Systems Analysis The analysis of the measurement system to
Determine the accuracy and precision of
The data obtained from the measurement.
Reliability and survivability Accelerated life testing, lifetime characteris -
Tics analysis, growth curves
The basics
At the root of Six Sigma is a set of statistical tools that drive most of the analytical activity, underlie the higher-level practices, and dominate the walk and
Talk of the Six Sigma practitioner.
In the practice of the statistical analysis side of Six Sigma, these tools are
Required fundamentals. You must understand them and be comfortable with
What they mean and how to use them. The good news is that you needn’t
Actually perform any of the calculations manually; they’re all done for you
By application software programs on your computer. Refer to the "Platforms and Protocols" section in this chapter for the overview on the applications
Software.
A picture’s Worth a thousand… dollars
Time is money. Plots and charts are a fast and powerful way to help you interpret and communicate the data. To get the message, use pictures — and lots
Of ‘em! Plots and charts can turn masses of unintelligible data into coherent information that leaps off the page and smacks you with the message. The
Most commonly used plots and charts are summarized in Table 11-7.
Table 11-7_Plots and Charts_
Plot or Chart_Description_Example
Histogram A bar chart that plots the spread of data
Into bins according to frequency of occurrence, immediately suggesting the
Distribution function.
Dot plot A type of histogram where data are
Displayed in a single-point format; used to assess a distribution or compare distributions.
Pareto chart A bar chart in which the bars are ordered
From highest to lowest, showing the critical
Contributors.
Scatter plot Shows the relationship between two
Variables, immediately conveying the nature of correlation.
Matrix plot A matrix of scatter-plots, showing the
Relationships between many pairs of variables at the same time.
3D scatter plot A three-dimensional scatter plot, useful for evaluating the relationships between three different variables at the same time.
Interval plot A two-dimensional plot of data values with
Added confidence intervals or error bars; useful for showing both the central tendency and the variability.
Box plot A side-by-side comparison of sample
Distributions. By convention, the central line is the mean, the boxes are ±25%, and the lines are the limits.
CDF (cumulative A stepped cumulative histogram (without distribution bars), overlaid with a best-fit normal
Function) plot cumulative distribution function. Used to
Fit a distribution to your data.
(continued)
Table 11-7 (continued)
Plot or Chart_Description_Example
Probability plot A scatter plot, overlaid with a CDF cumulative probability line. Used to determine how closely a particular distribution fits your data.
Time series plot A plot of data spread over time. Used to evaluate patterns in activity across time. By convention, time is plotted on
The x-axis.
Marginal plot A scatter plot with an added histogram
(or sometimes a box plot), used to assess the relationship between two variables and their distributions.
As powerful as the software applications are in crunching the statistical data in the first place, they really shine in creating these plots and charts for you. All the application programs on the market today will generate these types of plots and charts from the data automatically, with simple menu selections. They further will provide numerous plotting and charting options, including everything from curve and data fits to labels and legends and even colors
And fonts.
The time machine
Most human activity is measured, reported, and valued over time; hence, Time-Series Analysis Is closely correlated to the management and measures of performance improvement. Numerous Six Sigma statistical analysis tools are dedicated to time-series examination of every phase of a process. These include the following:
Is Trending: Fit a general model to past data and observe the trends. t-" Forecasting: Simple forecasting and smoothing methods help you
Decompose data into its component parts, and then extend the estimates into the future to predict ongoing performance.
Is Decomposition: Separate seasonal or cyclical trends into groups and
Profile repetitive performance.
V Moving average: Average consecutive observations and observe the trend over time. A pattern recognition tool called ARIMA (AutoRegressive Integrated Moving Average) can help you find patterns that may not be
Visible in plotted data.
Exponential smoothing: Smooth the time-series data using ARIMA and calculate the average level and, optionally, in a Double Exponential
Smoothing, both the average level and trend.
Autocorrelation: Discover repeating patterns in time-series data. Cross-correlation: Compute, plot, and discover the relationship
Between two separate time series.
Analysis of Variance: ANOVA
Because variance is one of the fundamental principles of Six Sigma (see Chapter 5), the analysis of variance is a major field of Six Sigma application. Analysis of Variance is so significant in both Six Sigma and in general statistics that it warrants its very own acronym: ANOVA.
ANOVA tools include such analytical marvels as: one-way and two-way analyses (variance testing with classification by one or two variables); Analysis of Means (test the equality of population means); balanced ANOVA (accounting for data collected by different designs or procedures), also sometimes referred to as the General Linear Model; fully-nested ANOVA (estimating the variance component for each response variable); MANOVA (multi-variate analysis of variance, for simultaneously testing the equality of means from different responses);
And the test for equal variances (determines the variance difference between
Samples from populations of different means).
All that’s a mouthful, but don’t despair; once again, software to the rescue!
Allof these tools are defined and executed in each of the major statistics applications programs on the market. These packages walk you through
These tools, holding your hand every step of the way. No sweat.
If the shoe fits…
One of the great challenges in this world is getting the right fit. The pen cap
Doesn’t stay on the pen; the lid doesn’t close on the jar; the door leaks air; the
Paint runs across the line. Things are too tight, or too loose, or off the mark.
How does this happen? It’s not because they were intentionally designed that way. It’s because the design didn’t take into account the combination of variations in manufacturing the different components.
Tolerance analysis Is the statistical analysis tool that helps you determine the
Right specifications and limits on individual parts and components to ensure
That they fit together properly as a system once manufactured. It’s treated as an advanced topic, as part of the field of Design for Manufacturability (DFM), and is usually taught in the advanced Design for Six Sigma (DFSS) courses.
Apply Tolerance Analysis in cases where parts or components must come together precisely for the system to function properly in satisfaction of the
Customer’s expectations.
Design of Experiments
Most people know DOE as the Department of Education. Well, in the Six Sigma case, that’s just about right, because in Six Sigma, DOE stands for Design of Experiments, A highly educational activity. Use DOEs to statistically
Investigate the variables that influence a process and the resulting quality of
Products and services in an experimental setting. You are then in position to effectively interpret the results and direct improvement efforts to enhance
The process in the production environment.
A DOE also allows the practitioner to simultaneously understand the effects
Of changing the settings of multiple variables. Without DOE you’re reduced to performing what we call OFAT experiments, which stands for one factor at a time. OFATs cannot detect the interactions that occur between variables. Besides, watching one-factor-at-a-time experiments takes forever.
Experiments are vitally important tools. They permit us to prototype, evaluate,
And test our hypotheses in controlled settings before unleashing them in the real world. Experiments are critical risk-reducers and confidence-builders.
They are a footbridge between models and reality.
Because of this keystone role, experiments must be done right. Time and resources for experiments are always limited, because people are impatient and see an experiment as a hurdle. Therefore, if you’re conducting an experiment, it’s very important that you get the most out of it. Well-designed experiments
Will yield much more useful results than tests that are casually thrown together. In fact, poorly defined experiments may yield the wrong results!
Experiments are mini-projects unto themselves. They consume resources, including personnel, equipment, and materials. They cost money and time. And because so much is riding on the results, they deserve the care and attention that any project or program would receive.
1. Define the problem.
Strictly define — in quantitative terms — the nature of the problem that you intend the experiment to clarify or solve.
2. Define the objectives.
Be certain that your experiment is focused on yielding specific, practical, and useful information.
3. Design the experiment.
Using the many available DOE tools, design a robust experiment that will satisfy the objectives.
4. Develop the plans.
Thoroughly analyze the environment, the background, and the conditions that will guide and constrain the experiment, and develop a plan that will
Meet the objectives with the time and resources allotted. Develop a Data
Collection Plan that ensures you have the measurement systems in place to capture all the required information, and a Data Analysis Plan that ensures you have accounted for the work required to properly interpret
The results.
Well-honed through years of experience, Six Sigma practitioners have defined a suite of tools to aid you in developing your DOE.
V Factorial designs: Factorial designs help you study simultaneously the effects that several different factors may exert on your process or product. This improves experimental efficiency, by enabling you to vary the levels or settings of parameters simultaneously during the experiment.
F Response surface designs: Response surface designs help you examine
The relationship between one or more response variables and a set of
Experimental variables. This approach is particularly useful after you’ve determined which parameters constitute the "vital few," and you want to
Find the settings that optimize the output.
Taguchi designs: Named after Dr. Genichi Taguchi, who is widely regarded as the foremost authority in robust parameter design, Taguchi experiment
Designs help you find the settings that permit your product or service to
Operate consistently over a variety of conditions.
Ho© capable is your process?
Process capability analysis is the next of kin to statistical process control
(SPC), and is how you determine if your process, once in control, is also
Meeting specifications. Process Capability Analysis is a critical component of the Six Sigma methodology, and Six Sigma practitioners calculate a variety of indices and measures and draw numerous plots and charts to assess and
Optimize process capability.
In summary, capability analysis takes the voice of the process (VOP) and compares it to the voice of the customer (VOC) to see if it is capable of meeting the requirements.
We cover the definition of Process Capability at some length in Chapter 6. The tools for process capability analysis are extensive. We’ve listed the most commonly applied tools in Table 11-8.
Table 11-8_Process Capability Analysis Tools
Tool_Application_
Normal analysis Analyze process capability when the data are
From a normal distribution.
Non-normal analysis Analyze process capability when the data are
From a non-normal distribution.
Between/within analysis Analyze process capability for between -
Subgroup and within-subgroup variation.
Multi-variable analysis Analyze the capability of an in-control process
When each of multiple continuous variables follow a normal distribution.
Binomial analysis Analyze the process when the data are from a
Binomial distribution — when examining the number of defective items out of the total number of items sampled.
Poisson analysis Analyze the number of defects observed, where
The item occupies a specified time or space.
Capability six-pack An set of six charts, which collectively contain
Key process capability metrics. An example Six-Pack is shown in Figure 11-10.
Figure 11-10:
Process capability six-pack.
|
1=-i-m-.- -.-v I .liiLiliilily Tfxp,!. K 1
|
In
|
|
-ham
|
-
|
|
I U I n U N
Hti^NjI T*J*n
|
!* i> :■■
|
|
J:
|
|
J J
<■
|
1
|
|
K
|
Trail
|
|
;» >»
|
Ht IM
|
Regression
Regression analysis Is used to discover and characterize the relationship
Between a response and one or more predictors. In regression analysis, you fit models or distribution functions to observed data. And depending on the
Data, this can lead you to a variety of functions.
The goal of regression analysis is to fit a line and create an equation to explain
Or predict the way your process output is behaving. As an example, imagine plotting your car’s gas mileage for different driving speeds. We all know intuitively that the faster we drive the lower the gas mileage, but could you come up with an equation to predict your car’s mileage as a function of its speed?
Yes you can. Regression analysis helps you do it.
Fitted line: For when the best fit to the data is linear or logarithmic.
Derivatives include second-order (quadratic) and third-order (cubic) fits.
F Least squares: When the response variable is continuous.
F Partial least squares: When the predictors are highly correlated or if they outnumber our observations.
V Logistic: Used with categorical response variables. There are three types: binary (two levels), ordinal (three or more levels), and nominal
(no natural ordering of the levels).
F Stepwise: A technique of removing or adding variables to the regression model in order to identify a useful subset of the predictors.
V Best fit: Examining all the subsets, identify the best-fitting models that
Can be constructed with the specified predictor variables.
Multivariate analysis
Quite often, you will have multiple measurements on a given item or subject. Multivariate Analysis helps you understand the structure in this mix of data.
It helps you assign different observations to statistically-significant groups
And visually explore the relationships among the grouped variables. Multivariate analysis Begins with applying tools to understanding the covari -
Ance structure in the data. Principal component analysis and factor analysis are two methods for helping you determine structure, alignment, and dimensions of the variables within the data. Grouping tools then help you aggregate data. These tools include data clustering from similar observations, clustering of variables, grouping by known similar averages (called K-means), And grouping by comparison to a sample group (known as discriminant analysis).
When you have statistically segmented your variables, multivariate analysis
Explores the relationships among them.
F Simple correspondence analysis Explores variation between two variables.
F Multiple correspondence analysis Extends the process of simple correspondence analysis to the case of three or more variables.
Exploratoru analysis
Sometimes, you’re not quite sure where to begin or which statistical tools you should apply to a given situation. Sometimes, you’re not sure which tools apply — or if any of the traditional tools apply at all. That’s okay! A variety of Exploratory analysis Tools let you examine data in nontraditional ways, giving you the ability to work outside traditional boundaries and see your data in a different light. A sampling of these tools are summarized in Table 11-9.
Table 11-9_Exploratory Data Analysis Tools
Tool_Application_
Stem-and-leaf A quick way to examine the spread and shape of your
Data.
Box plots Assess and compare sample distributions (see exam -
Ple in Table 11-7).
Letter values Assign data into broad buckets.
Median polish Analyze variance relative to the median instead of the
Mean.
Resistant line Fit a straight line to your data while ignoring the
Outliers.
Resistant smoothing Smooth your data, removing random fluctuations,
Before examining trends.
Rootogram Plot your data up as a histogram, fit a normal distribu -
Tion to it, and examine how closely the data fit or deviate from the normal distribution.
Measurement systems analysis
All the emphasis on the collection and analysis of data begs the obvious
Question: How good is your measurement system? If your measurement
System is faulty, your data is faulty, in which case your analysis is no good, and
You may as well put your plans to better use on the bottom of the bird cage.
Measurement systems analysis Is the practice of determining the extent to which observed process variation is due to variation in your measurement systems. Any time you take measurements, you will encounter variation. Thesource of this variation is two-fold:
Real variation in the actual process Imperfections in your measurement system
Measurement system errors are classified into two broad categories: accuracy and precision. With most measurement systems, both errors are present.
Accuracy
Accuracy is the difference between your observed measurement and the true value. Three sources contribute to accuracy error:
Is Linearity Is a measure of how the observed error is somehow related tothe size of the measurement. If your measurement is accurate in the middle of the measurement range, but not for very large or very small measurements, you have linearity error.
V Bias Is the condition when your measurement system is skewed — like when you dial the zero value on your bathroom scale down a few pounds before stepping on to check your weight!
V* Stability Is the tendency of your measurement system to vary over time
Or some other condition, such as temperature or humidity.
Gauge Linearity, Gauge Bias, and Gauge Stability studies help you analyze measurement systems accuracy issues.
Precision
Precision erroris the condition of observing variation from measurement to measurement, or from part to part. The two components of precision error are as follows:
F Repeatability: Variation in the Measuring device. All other conditions
Being equal, there is variation in the measuring device itself.
F Reproducibility: Variation in the Measurement system. The device performs properly, but the system of measurement — including the procedures, human error, and support systems — introduce variation.
Gauge R&R (repeatability and reproducibility) studies help you determine
The extent to which device and process variation contribute to your overall measurement system variation.
Back to the future
Reliability and survivability analysis Helps you use all the measurements and data from the past to predict what is most likely to occur in the future.
F Develop tests that demonstrate compliance with reliability specifications to specified confidence levels.
V Determine the number of tests needed to develop precision estimates of percentiles and reliabilities.
Is Define criteria for accelerated life tests to determine the relationship between failure time and key predictors.
Platforms and Protocols
Throughout this chapter, we’ve been telling you that the many tools and methods of Six Sigma are nicely encapsulated in application software packages,
Available for your immediate use. It’s true. While the statisticians, analysts,
And progenitors of Six Sigma were busy refining the methods and tools of Six Sigma, so, too, were the software programmers. The Six Sigma software marketplace is now brimming with well-tuned packages. Every tool we address in this chapter is programmed into a number of nicely designed software products. And what’s more, they’re relatively inexpensive, and easy to use.
By and large, the software programs follow the tools: process optimization
And statistical analysis. As the software matures, the overlap is increasing, but the areas of expertise are well-established.
Software products
It’s a crowded field. By our last count, over 120 commercial software programs supporting Six Sigma process and statistical analysis were available on the market. Many of these come from small shops with niche products for specialty
Purposes. As Pareto would have it, however, the market is led and dominated
By the critical few larger and very professional organizations, who bring solid and fully capable commercial-grade application software products to a demanding market.
Statistical analysis software
The leaders in Statistical Analysis software include the following:
Is Minitab: The undisputed leader. Minitab is taught extensively in colleges and universities, used extensively in major corporations, and fostered universally by Six Sigma consultancies. In its 14th release as of this writing, Minitab is packed full of features and is completely capable of stellar performance for you on every statistics tool discussed earlier in this chapter. It runs only on a PC.
JMP: From the prestigious SAS Institute, JMP (pronounced Jump) Is a
Professional statistics package that rivals Minitab in its features and capabilities. Its greatest advantage is in its multi-platform support: JMP runs on Windows, Macintosh, and Linux.
Is Excel: Yes, as in Microsoft Excel. That ubiquitous spreadsheet program that comes on nearly every PC is also a powerful computational and display program, and it’s used extensively in Six Sigma statistical analysis
Applications. Unlike Minitab and JMP, you have to program Excel to perform the calculations, but there are also a plethora of software companies that sell Excel add-ons and extensions for statistics.
Process optimization software
The Process Optimization space is much broader — and younger — than the
Statistical analysis market, and no one software program does it all. You have
To be willing to piece together your solution from several different software vendors. Following is a short list of leading software companies whose products support the core process tools. They all run on the PC platform, with a
Few exceptions.
Process tools are categorized in two classes:
V Business Process Analysis (BPA) Tools enable you to model and map processes, simulate how they’ll work, and analyze the results.
Is Business Process Management (BPM) Tools enable you to connect with
The information systems in your business and provide process measurement and control functions.
The leading tools in the process management arena are:
Traxion: From CommerceQuest. Traxion is a complete BPM tool, with modeling and simulation capability, similar to iGrafx and SigmaFlow. But Traxion has the unique additional capability to collect information out
Ofyour operational environment and give you feedback on your performance in real time. This so-called "closed loop" solution integrates your Six Sigma design and simulation with the measurement and control systems in your business. More on this in the "Technology architectures"
Section.
V IGrafx: The iGrafx mapping and process simulation software is by far the most widely used full-featured process analysis tool in the marketplace. As a company, iGrafx has bounced around, and as of this writing, it’s a division of Corel, Inc. But the iGrafx pedigree goes back nearly 20 years, and the product set is well-regarded as the BPA leader.
SigmaFlow: Although a relative newcomer in the market, SigmaFlow is
Earning fans through its more business-focused approach to modeling
And simulation. It’s fully integrated internally and automatically populates one tool with the information and output from another.
Visio: Surprisingly, a lower-level tool from Microsoft is a major player in the Six Sigma marketplace. Visio has long been used for drawing process maps. No detailed model attributing or simulation, though.
Is Varyx: From Savvi International, Varyx is the only Tolerance Analysis software on the market. A niche player, but with a powerful and important product. In the interest of full disclosure, you should know that two
Of your authors work for Savvi. But we still recommend these tools!
3-Cs Explorer: Also from Savvi International. Capability-Complexity Analysis is made simple with the 3-Cs Explorer analysis and display.
Technology architectures
Software is complicated stuff. Beneath the covers and underneath all the windows, dialog boxes, drop-down menus, and snazzy reports of commercial
Application software swims an ocean of program code. Software programs are
Developed according to an architectural design, and a program’s architecture dictates how the code will perform, what types of computing environments it will work in, how it can interact with other programs, and how accessible its
Functions and data will be to the outside world.
You must be aware of the implications of the architectures in the software
Products you consider for implementing the tools of Six Sigma. These products have been built to a range of architectural models, and while many may have similar features, underneath they’re different and may not suit your
Application. This section addresses these differences and gives you some guidance to help you choose the right solutions.
PC-Windo©s
With a few notable exceptions, most of the programs for Six Sigma process optimization and statistical analysis have been developed exclusively for the PC-Windows operating system. They run on PC-Windows computers only. If you have Macs or Linux systems in your environment, these PC programs will
Not run on them. But that may be okay. If you’re operating in a PC-Windows
World, these tools may be sufficient.
Few company environments are this monolithic. For this reason, most of the statistical analysis programs have import and export capability, so they can
Move data into flat files or Excel files. This permits users in other environments
To generate or view the results. The Excel file type is used as a universal translator in this manner.
The gotcha with PCs is that many PC programs operate as standalone systems, and provide little by way of connecting or operating with other people or systems. E-mailing data files back and forth gets really old, and file-sharing is
Cumbersome, error-prone, and difficult to manage. In short, the PC desktop
Environment by nature often works against your Six Sigma philosophy and
Goals.
This is what’s known as Functional sub-optimization — the tool may be great for
You, but it doesn’t help you work well with others. The solution to the sharing problem is in what’s known as enterprise technology, addressed later in this section and in Chapter 12.
Mac and Linux platforms
These two main platform alternatives to PC-Windows have been classically
Under-served by the Six Sigma tools market. Very few providers have offerings for these platforms, with the exceptions of Microsoft Excel and the very capable JMP statistical analysis product. JMP is what’s known as a cross-platform
Tool; it runs on all three desktop platforms — PCs, Macs, and Linux. If you
Arein a mixed-platform environment, JMP is a viable solution you should strongly consider.
Using a mix of platforms can also present problems when you want to share
Information and work in a team environment. Typically, the architects of Mac
And Linux solutions include more integration and connectivity, but many programs offer you little more than the chance to e-mail your data files around. Remember that you can always run PC programs through emulation software on the Mac. It’s not pretty, but it works.
Enterprise platforms
The solution to the challenge of working together with computers is provided
By what’s known as Enterprise technology. Enterprise systems are designed for interoperability — permitting people to fulfill their business and functional
Roles while interacting with a coherent system of information management.
Large software systems have been built this way for decades.
The Six Sigma technology and tools industry grew up in the PC era, and as a
Result, precious few of the process or analysis tools have been crafted to an
Enterprise architecture. (Note that this is Not The case for the management
Tools, which have mostly been developed to an enterprise architecture. This is discussed further in Chapter 12.) All the integration and interoperability in enterprise systems comes at a price, however. It usually also means that each user makes some personal sacrifice, in terms of performance or independence,
For the greater good. But if your business environment requires extensive
Sharing of analytical or process information in a controlled manner, look for
Enterprise solutions.
The best way to recognize an enterprise software solution is the user interface. Most enterprise systems are accessed via a Web browser like Internet Explorer or Mozilla Firefox. The programming logic and the database of information
Reside across the network somewhere, rather than on your local machine.
Chapter 12
Mastering Six Sigma Manner Tools
In This Chapter
^ Assessing your Six Sigma management tool requirement
^ Exploring the different types of Six Sigma Management tools
^ Understanding where and how Six Sigma Management tools are used
Finding out about available management tools and technologies you can use
The only thing more important than practicing Six Sigma is managing Six Sigma. You think it’s important to understand all the technical practices
And analytical tools? It is, but it’s even more important to manage resources (people), schedules, and budgets — and to be accountable for bottom-line
Results.
Managing Six Sigma projects and programs requires that you understand your
Area of application as well as the use of process methods and tools. But it also
Requires you to apply your methods and tools of the management process. These tools support your need to manage the many interactions between multiple contributors, who occupy different roles in the organization at many different levels. The tools must also support the complex technical interactions between information systems. We suppose that’s why the managers get
The big bucks.
In this chapter, you discover the methods and tools of Six Sigma management.
These begin with program leadership tools and include project definition and tracking, as well as business reporting. In addition, this chapter covers the tools you need for accessing reference information. The chapter concludes with a look at knowledge transfer systems and the emerging world of online learning.
The Manager’s Toolkit
To be a successful Six Sigma manager, you must not only understand Six
Sigma but also be skilled in the methods and tools of project management.
A Six Sigma initiative is an endless series of projects — of various sizes and shapes — cascading together in a programmatic fashion, creating an unending stream of breakthrough improvements in business performance. These improvements are made one project at a time, and each project is an encapsulated universe of Six Sigma activity unto itself.
The Six Sigma manager’s toolkit is, therefore, a set of project and people management tools in a portfolio. These tools are inspired by leadership, enabled by infrastructure, tailored to the Six Sigma methodology, and implemented through technology. Some of these tools are relatively straightforward, such
As tools to help you manage project deliverables and tools to help you remember how to do Six Sigma. Others are more involved, like tools for tracking and
Communicating critical business indicators, and those for helping you manage cost and schedule.
Unlike the tools used by practitioners, where most are used in an individual, standalone fashion, most management tools have a focus on integration and communication. Management systems must robustly link the daily work products of many individual contributors. As such, Six Sigma management tools are typically built to an enterprise information architecture, with core data repositories and shared access to the application logic, thus minimizing operational error. To be sure, plenty of management tools are integrated via Sneakernet—the practice of running files back and forth (usually with e-mail) — a practice that begs for error. In any case, management tools provide specialized information access and control to each of the constituents with a stake in the success of the initiative.
The management process is a little different for Six Sigma than it is for most other activities. Managing scope, schedule, and budget is still the manager’s job, but there are specific methods and tools in the case of Six Sigma. You
Want to integrate existing management tools and the knowledge you currently possess with specific management tools you need to make Six Sigma
Easier and more beneficial.
The qatteru
At the management level, everybody’s watching and listening. And asking why.
Six Sigma management tools support each of these constituents, including both participants and spectators:
Executive management: Because most Six Sigma initiatives are strongly and directly endorsed by executive management, tools must support the
Executives and provide them the information and interaction needed to continue their endorsement.
F Six Sigma champions and Deployment Leaders: Champions are steering the overall Six Sigma initiative and are accountable to the executives
And the operational business units for the results. They’re in the cat-bird seat, must have a direct line of sight on people and projects, and must
Constantly know the pulse of the Six Sigma activity.
F Financial executives: Because Six Sigma projects contribute directly to the bottom line, their performance is of great interest to financial managers, who want to maintain close touch with their progress and results.
F Process Owners: These managers own the profit and loss (P&L) or the budget and productivity of the processes that either support or deliver
The value proposition of the company. Process Owners must have immediate access to the information and rationale behind any changes in
Order to support and implement those changes.
C Black Belts: Black Belts are the team leaders of major Six Sigma projects.
These are the projects with the most complexity, difficulty, and the greatest impact and return to the business. As team leaders, Black Belts use project tracking and management tools as well as the process optimization and statistical analysis tools discussed in Chapter 11.
V Green Belts: Green Belts have traditionally acted as support staff but they also act as project managers within their areas of responsibility. When acting as Six Sigma project managers, Green Belts use similar tools
As Black Belts with a lower level of analytical prowess.
Is Yellow Belts: Six Sigma initiatives affect everyone in the business, and
The successes and performance of these initiatives are communicated to
Everyone in the company. Training and reporting tools are used by Six Sigma managers to involve and inform the staff, and by the staff to initiate new projects and participate in existing ones.
Suppliers: As suppliers become more integrated into business processes
And enabled by the enterprise architecture in many of the management tools, Six Sigma tools securely enable the management of vertical value
Chains.
F Customers: The external customer is the customer who pays for the product or service that we deliver. Management is ultimately accountable to the customer and uses the tools and techniques of Six Sigma to
Direct improvements on behalf of the customer.
Types of management toots
Because the constituencies served in the Six Sigma process are such a broad
Set, the tools of management are a diverse lot. In total, these are the tools of communication and leadership, project management, reporting, knowledge management, and learning.
Communication and leadership: Communication and leadership tools
Are both formal and informal: company Intranet sites, video messages,
Letters and memos, reports, and other messages. And don’t forget the most important leadership tool: face-to-face contact.
F ProJEct management: Management tools include everything from the
Capture of ideas into project assignments, staffing, budgets, and performance. The more advanced tools include multi-project and cross-project portfolio management in a shared enterprise architecture.
F Reporting tools: These are tools that query data and create reports
Provide standard and repeatable ways to communicate detailed information. These reports include tables, plots, and charts of analytical and
Process performance data. These are combined with budgets, schedules, resources, and business-impact information to create comprehensive
Pictures of project and program status, progress, and trends. When
Aggregated together, these tools are typically called dashboards. Is Knowledge management tools: These tools are extensive collabo -
Rationtools, granting individuals and teams access to information
Repositories. By having access to the right knowledge at the right time,
Managers and practitioners can expedite their return on improvement investment.
Learning tools: Beyond traditional training, learning tools provide direct, just-in-time, and lower-cost training to individuals, teams, and companies. These tools are critical enablers for the job of training large
Numbers of people in the concepts, ways, and methods of Six Sigma.
Because these tools integrate people, functions, and systems, utilities known as Application Integration tools help tie together and share the information
They generate and use. A class of these utilities, called Enterprise Application Integration (EAI), or Middleware, Helps you move data between and among
Not only these management tools, but also between and among transaction systems, including customer management, accounting, design, and shop-floor systems.
Through the Looking Glass
After you get the basic concepts of Six Sigma, you’re changed forever. You’ll have insights and vision that dramatically enhance your abilities. The knowledge and tools of Six Sigma well up in you an emboldened sense of personal empowerment. With Six Sigma, you command the power and have the ability to foster significant positive change in the world. As a result — regardless of your title or official duties — you become a leader!
Your leadership role compels the use of the single most important tool for
Any leader: communication. Your Six Sigma knowledge and capabilities grant you significant influence, and you apply that influence through all manner of
Communication. The tools of communication you must use are the broadest
Set of communication tools possible and they’re summarized in Table 12-1. We can’t overemphasize the power of leadership within every Six Sigma
Practitioner. Whether you’re an analyst, executive, manager, engineer, or
Administrative assistant — it doesn’t matter. When you know how to apply the methods and tools of Six Sigma, there’s no turning back. You have this special insight, a new ability, and you’re going to use it. With that ability comes the essence and responsibility of leadership.
|
Table 12-1 The Tools of Leadership Communications
|
|
Communication Tool Role
|
|
Face-to-face
|
The most powerful leadership tool is your personal
|
|
Communication communication. Direct interaction is the best way to
|
|
Listen and influence.
|
|
Formal presentations
|
Using a presentation tool like PowerPoint, a formal
|
|
Presentation is a common, effective, and repeatable
|
|
Leadership and communication tool.
|
|
Impromptu
|
White boards and flip charts make ideal platforms for
|
|
Presentations conveying important ideas and information, conducting
|
|
Brainstorming sessions, developing early designs, and
|
|
Troubleshooting.
|
E-mail Messages, directives, requests, and reports can all
Be communicated via e-mail, which communicates directly and by passing along through different audiences. Using attached files, e-mail is a powerful communication conduit. It’s poor and inappropriate
_For resolving issues, however._
(continued)
Table 12-1 (continued)
Communication Tool Role
Shared repositories Systems like intranets, file servers, groupware, and enterprise application systems help communicate broadly and consistently.
Phone calls Particularly when there’s an issue or problem, there’s
Nothing quite as effective as just picking up the phone
And calling them.
Memos and letters Formal memoranda and letters are most useful for
Communicating in an official manner, such as a policy directive or formal announcement.
Bullhorn Hey, whatever it takes! Just make sure you get the
Message out.
In addition to communication tools, you must use other leadership tools,
Including motivational tools and the tools of influence.
Project Management
Six Sigma benefits are derived from a series of projects. Lots of projects.
Bigprojects, little projects. Long projects and short projects. Projects within a single department, and projects that cross departments. Projects inside
Companies, and projects that cross company boundaries. Dozens of projects — and, in big companies, hundreds of projects. Dr. Seuss could write a book on all the Six Sigma projects!
At the business level, Six Sigma projects are the players in the overall game
Plan of a breakthrough performance improvement initiative. The business perspective is that a Six Sigma project is the agent of action that executes
The business strategy and returns the results. Selecting the right projects is,
Therefore, critical, as is executing them properly. This means that the effective management of projects is core to the success of a Six Sigma initiative.
The skills and tools required to manage a Six Sigma project are similar to those
Required to manage other types of projects. It’s rigorous, but you don’t need
A certification from the Project Management Institute to do it.
First, you define the fundamental problem or need you intend to solve or
Address with your project effort. Then, you define the objectives and results you seek to achieve. From this, you define the project plan, which includes
Scope, schedule, and benefit. Upon gaining approvals, you’re off and running. When you’re underway, you must track and manage the project to plan, and
Deliver the results to the Champion or Deployment Leader.
The application domain of the project may be unique to Six Sigma, but the
Management of the project follows many standard project management rules
And guidelines. In Table 12-2 are listed the major categories of project management tools used in the management of Six Sigma projects. Application
Software packages of various types are available on the market to assist in the execution of these tools.
Table 12-2_Six Sigma Project Management Tools
Project Tool Role_
Ideation Capture ideas for potential Six Sigma projects.
Definition Establish the project scope, write a problem and objective
Statement, set a schedule, and assign initial team members.
Selection Establish priorities for projects, manage the queue of projects,
And launch projects.
Tracking Track and manage project progress. Identify and manage vari -
Ance to plan. Ensure deliverables to the established objectives
And schedule.
Reporting Communicate the status and results of the project — to the pro -
Ject team members, business owners, Six Sigma Champions, executives, and other constituents.
Eureka!
Projects begin with a problem or a need to improve. Someone, somewhere,
Realizes that it can be done better. Improve a process. Reduce defects. Eliminate waste. Projects can be motivated in countless ways. (Refer to Chapter 4 for more on sourcing and defining Six Sigma projects.)
The process of discovering the opportunity to perform a Six Sigma project is known as Project ideation. This funny word, Ideation — short for idea creation — refers to the process of creativity and insight within the formality
Of the controls that permit the idea for the project to flourish. Tools for project ideation enable you to capture the essence of the idea, along with supporting information, in a central database to evaluate and consider. An
Example of a Web-based project ideation portal is shown in Figure 12-1.
Figure 12-1:
Project ideation portal.
Ideation tools like these are powerful ways to enroll everyone in the
Organization — including customers and suppliers — in the process of
Identifying potential improvement projects.
Pick a ©inner
Project selection is a delicate act of evaluation, alignment, and prioritization.
Your Six Sigma projects must be of proper value and contribution in their own right, but they must also be set in the context of the improvement of the business and in meeting its stated goals. Rogue Six Sigma projects can
Solve the wrong problems.
As part of the selection process, a project must first be defined in rough
Terms — but sufficiently quantified in scope, schedule, difficulty, and expected
Impacts on the business. Then you can determine if it’s worth doing. Evaluation
First, evaluate any proposed project for its direct contribution to its spe -
Cificarea of business and its alignment to the overall business strategy or objectives. These contributions should include quantifiable measures, such
As significant percentage of defect reduction or measured customer satisfaction improvements, as well as the financial contributions to profitability.
V* Quantifiable improvements: 70 percent or greater improvement over
Baseline performance on key metrics
Quantifiable returns: Return on investment is less than 1 year
Alignment
Next, evaluate the project in terms of its alignment to the goals and strategies
Of the business, and for its context relative to core or enabling business
Processes. The Six Sigma Champion or Deployment Leader should evaluate how the project will contribute to the overall business needs.
Categorize the project in terms of hard dollar value or soft contribution.
No more than 25 percent of Six Sigma projects should be soft-savings
Projects.
Align the project profile to the overall business to ensure its efforts and contributions are placed strategically.
Consider the learning value and the contributions toward generation of momentum as part of the total Six Sigma improvement initiative.
Priority
As a result of your evaluation and alignment exercise, assign a numerical priority to the project. Typically, use a range of 0 (project disapproved) to 10
(assigned top resources and budget).
Use the priority scheme to identify those projects that have the largest potential impact on the organization, either strategically or financially, and that have the highest probability of success with the lowest level of required
Resources. A priority matrix is a useful tool in comparing parameters and prioritizing projects.
Project definition
Project definition is the critically important process of transforming a practical business problem into a Six Sigma project. The output is a well-defined
Problem statement and a well-scoped set of objectives, including approvals
From those who are either involved in the project or affected by its results. The Six Sigma management community believes that 50 percent of the success of a project is in the quality of its definition.
The project definition worksheet breaks down the many elements of defining
A project into easy-to-handle pieces. Refer to Chapter 4 for more on sourcing
And defining Six Sigma projects.
Management — in the form of the Champion, Deployment Leader, and/or Process Owner — is responsible for defining projects. Management must decide which projects will achieve business (VOB) goals and meet customer (VOC) requirements. Six Sigma practitioners are closely involved, assisting
Management in this effort, contributing input to the evaluation and alignment of candidate projects.
At the highest level, that’s really all there is to the essence of project definition. But while it may sound easy, a lot of information and work goes into defining
Projects, which means you have a lot to track and manage.
You have difficult hurdles in front of you. Resources— budget, people, and equipment—are always in short supply. Schedules are always tight. You have to account for the constraints and risk factors, too. However, if you follow the
Project management process, you will produce a well-defined project plan and enable your project team to be successful.
Behind the magic trio of a problem statement, an objective statement, and approvals are a number of supporting elements that make up a sound project definition. These are summarized in Table 12-3. Every project definition should contain a concise and accurate description for each of these elements.
Table 12-3_Basic Elements of Project Definition
Element Definition
Purpose The reason and motivation for doing the project. This includes a precise statement of the problem, and its impacts.
Objectives The core set of objectives that must be met if the project is to be
Judged a success. Be quantitative in identifying the anticipated
Levels of improvement.
Benefits How everyone will gain from successfully meeting the project objectives. For Six Sigma projects, this specifically includes the
Bottom-line benefits.
Team Identify the team of individuals and skills needed to complete
Members project. The team should be small to remain agile, yet have the
Sufficient expertise and representation. Typically, a core team of six or fewer are required, with additional help on a part-time basis.
Schedule The schedule includes the total duration of the project as well as the individual duration of each project phase
Risk and Scope, schedule, and objectives hang in a delicate balance. A controls change in any one affects at least one of the other two. Estimate the risk and impact of possible and probable changes, and identify the controls you apply to prevent them from occurring and to respond if they do.
Project planning and tracking
The preceding section describes the tools of project definition. In this section, you find out about the tools for Six Sigma project planning and tracking.
Planning the project
After the Six Sigma project definition phase is complete, the next step is to plan the project. Project planning is important, because a project plan is no better than all the effort and consideration that went into it. A project is a process, and the effort to plan a Six Sigma project is just as rigorous as the development of the process you’re setting out to improve. As General and two-term U. S. President Dwight D. Eisenhower once said, "Plans are nothing. Planning is everything."
To prepare a project plan, you must first collect and organize the information from your project definition effort.
Is Methodology: In a Six Sigma project, the project method follows the DMAIC process (see Chapter 3). Any of the project planning and tracking software tools for Six Sigma have built-in templates for DMAIC. The project milestones and subordinate deliverables will follow this approach.
V Roles: Choose and assign the people and skills you need to complete
Theproject. This includes Black Belts, Green Belts, and Process Owners. If the project is very complex, spans many organizations, or has been unsolvable in the past, you need a Black Belt to contribute in-depth technical analyses and leadership.
V Schedule: Six Sigma projects are short, usually three months or less.
Torealize the advertised returns, the project team should complete its
Work in a short timeframe. Project milestones normally coincide with the application methodology, such as DMAIC, with deliverables and
Checkpoints at each phase boundary.
V Reporting: The communication of project status is critical in Six Sigma
Projects, just as it is in other types of projects. Each stakeholder and
Participant must be regularly informed as to progress and results. This includes not only the project team members, but the finance group,
Process owners, and executive leaders. The tools of leadership communications (refer to Table 12-1) facilitate the project-reporting effort.
V Cost/benefit: Because the ultimate deliverables and returns on Six Sigma
Projects are measured in terms of bottom-line contributions, your most important project metrics are the returns you’re generating as a result ofthe project effort. Manage to the return on investment (ROI), and use methods of valuation that are consistent across your business.
The output of the project planning process — the Project plan — is a specific
And controlled set of information. The project plan includes text documents, supporting spreadsheets for financials, and a Gantt or similar type of project schedule, with milestones, resource information, and reporting mechanisms. Collectively, this information set is reviewed and, after it’s approved, is set
Down as what is known as the Baseline Project plan. This baseline plan is what everyone works to. Any changes to the baseline plan must be reviewed
With the Champion or Deployment Leader as well as other stakeholders to determine if or how the project should continue.
The supporting application software tools to complete the project plan include:
Word processor: A tool like Microsoft Word captures all textual information, and stores it in files. You want to print one or more copies for
Physical reference and include signature authority to allocate resources
To perform the project.
Is Financial calculator: Typically, a spreadsheet tool like Microsoft Excel supports financial planning information. However, in more sophisticated environments, this may be replaced by built-in capabilities in your company’s ERP system or managed from within an integrated project planning tool like Instantis or SixNet.
Scheduler: Project schedulers are like process mappers for projects.
They enable you to capture all planned work tasks, resources, costs, and risks and place them in a scheduling format according to a formal organization of work. Numerous project management software tools are on the market; tools specifically designed to support Six Sigma projects include Instantis and SixNet. These are powerful, and they perform project scheduling for Six Sigma-specific projects from an Internet-based architecture. There are many others; the most commonly used generic project scheduling tool is Microsoft Project. Refer to Figure 12-2 for an example of a project schedule.
F Reporting: Project status reports are typically generated directly by the project planning and scheduling tools. You may need to extract information from your project scheduling tool to generate the precise type of reports you need. The following section addresses this in more detail.
Document manager: Often overlooked but of critical importance are the tools for managing the plethora of project documentation. Document
Management is the practice of securing a set of data files in a repository with strict access and revision controls. These systems are invaluable for controlling updates to official or reference documentation.
Tracking the project
Project plans are really great — until the day the project begins. Tracking and managing the project to plan is critically facilitated by the tools, and this is where they really shine. These tools help you track and report project status;
Make changes to resources, budgets, and schedules; and redefine work and
Deliverables. In many cases, the tools will also manage changes to the plan, and compare the real results, called Actuals, To the original baseline to produce variance-to-plan information.
The saving grace for Six Sigma projects is that they’re short. While longer projects become increasingly complicated to manage, Six Sigma projects typically last only a few months. This way, projects — and their models — can’t get into too much trouble, because they don’t last long enough to be overly complicated.
Just the Facts, Ma’am
Reports communicate results. They are the trailing indicators, demonstrating
The outcomes of your initiatives. Reporting tools are communication tools
That tell everyone how well your initiatives are performing. These tools are vitally important, because they provide visibility into the bottom-line results
Of your projects and programs.
As key communicators, you must pay close attention to your project reports. Without them, your constituents have no way of knowing all the great things you’ve accomplished and, therefore, have no way of supporting you or your
Project team. After all the work and all the achievements, you want to ensure that everyone gets the good news.
Reporting tools for Six Sigma projects and programs are available in many
Flavors.
F Generic reporting tools: The information management marketplace has a category for what is known as Business intelligence tools. These tools
Are useful for culling information from databases and presenting nicely formatted reports on a repeating basis. This class of tools is not specific
To Six Sigma, but because they are generic, they are adaptable to the
Application. Among this group are tools like Cognos, BusinessObjects, and even Microsoft Office.
V Integrated tools: The enterprise integrated tools used for project planning and tracking also include a suite of reporting tools. If you’ve used these tools on the planning and tracking side of your management effort,
They are efficient and effective as reporting tools. See Figure 12-3 for an
Example.
V Balanced scorecard: A specialized system of reporting specific business measures and performance in a proscribed manner, known as the Balanced scorecard, Has emerged in recent years. The balanced scorecard is an entire field of study unto itself, and its methodology is specified by the
Members of the Balanced Scorecard Institute, which include both corporations and governments. In some cases, Six Sigma results must be presented in this manner. Figure 12-4 shows an example of a balanced scorecard.
|
—
|
M
|
55
|
3
|
|
L-J
|
‘-’
|
O
|
T
|
|
O
|
|
■:JH
|
Ln
|
|
_-p
|
|
W
|
%
|
5
|
En
|
|
1
|
|
Ol
|
|
>-
|
W
Figure 12-4:
Balanced scorecard.
1
35
Knowledge Management
Knowledge Management (KM) tools are extensive collaboration vehicles that help you transfer and share knowledge and information across your organization. These are by no means specific to Six Sigma, but the savvy Six Sigma manager employs these tools to ensure a widespread effect of the behaviors and results of a Six Sigma initiative.
KM tools are almost exclusively enterprise-class applications that operate across networks. They work together with learning tools to provide effective
And efficient mechanisms for sharing intellectual capital and creating an environment of responsiveness and furthering innovation. KM tools are an
Informal, bottom-up way to bring together and share information.
The marketplace for Knowledge Management tools is quite broad, and in
Thebroadest sense it includes all information, reporting, and content management technologies. More specifically, KM tools enable people to collect,
Access, manage, and share all relevant information on a variety of topics.
Thisincludes
F Six Sigma methodologies
Statistical analysis methods and tools
F Process optimization methods and tools
F Results of past statistical analyses
V* Past process management activities and results
V Project definitions and charters
F Project plans
Project results
The field of Knowledge Management considers this list a treasure trove of
Intellectual value within a business. All manner of tools and technologies, from Internet and intranet access tools to structured access and control tools to library management systems are fair game for your consideration relative to how you choose to integrate this information into the regular daily lives of
Your staff and co-workers.
Knowledge Management is not an esoteric field of academic ballyhoo about life in the electronically-enabled information age. Six Sigma initiatives produce a wealth of vital information that contribute significantly to the intellectual capital of your business. KM is the technology that releases this capital into your organization to generate sustained growth and performance improvement.
An Apple (or \lour Apple
A key tool of Six Sigma management is training. Corporate training in the principles and practices of Six Sigma is traditionally delivered by consultants or internal Master Black Belts. This training follows a prescriptive training regimen, whereby executives, Champions, Deployment Leaders, "Belt" staff, and functional support staff receive the training necessary to conduct themselves in their respective Six Sigma roles.
More and more, instruction and learning are taking place through the computer. This is called online learning, or E-leaRNing. The field of online learning has been growing and maturing dramatically in recent years, and represents an increasingly larger piece of formalized instruction and corporate training. Chances are you’ve experienced some form of e-learning.
Online learning for Six Sigma is an attractive management tool for several reasons:
F On a per-student basis, e-learning is far less expensive than conventional
Stand-up training.
V E-learning eliminates most of the time and expense of travel — either for
The student or for the instructor.
Is With e-learning, students are often able to take classes during off-times, reducing work interruptions.
F E-learning reduces instructor time, permitting companies to deploy
More Master Black Belt trainers back into the workforce.
F Because e-learning is conducted within information technology environments, it directly facilitates understanding of technology tools.
E-learning is a natural companion to knowledge management, and the
Materials for e-learning quickly translate to reference knowledge.
You can see how attractive e-learning for Six Sigma can be. E-learning technologies and platforms do not replace conventional Six Sigma training in one
Key area, however: Students must still complete workplace projects and be
Mentored closely through the project process. The Master Black Belt is an
Important mentoring component of the student’s knowledge development
Process. Automation has not fully replaced the instructor — at least not yet! Several methods of e-learning for Six Sigma are available as discussed in the
Following sections. Computer-based training (CBT)
This is the most famous and common of all e-learning techniques. In this approach, the training is delivered either via a compact disk (CD) or through
The Internet on the Web. Students receive the training by clicking through pages of material and, once completed, they usually verify their knowledge by
Taking online examinations.
This approach is common in lower-level types of training, but has fallen short of delivering value with the higher-level constructs and learning required in a Six Sigma environment. The timing and interaction of the mentor is also difficult to schedule through this approach.
Synchronous mentored learning
This approach, often referred to as "the instructor in a box," is more intimate
Than CBT. With synchronous learning, the student attends class online at a fixed time of day, during which the instructor presents the materials. The
Student usually has a real-time link and can interact with the instructor and
Other students with questions and discussions.
This approach is favored by several universities, and we have observed it in
Several corporate training environments. The benefit includes a certain spontaneity in the instructor’s lecture materials, but the intimacy of the real-time
Online interaction comes at the price of high network bandwidth. Asynchronous mentored learning (ALN)
The Asynchronous Learning Network approach is a hybrid of the preceding two. With ALNs, students follow the lectures independently, and yet interact with the classes and their classmates on a fixed routine. ALNs are increasing in popularity in corporate and government training applications because of
This duality.
