Design for Six Sigma—Failure Mode and Effects Analysis


1. Introduction

The previous post covered Quality Function Deployment as one of the methodologies for designing in quality, with the House of Quality being a tool of that methodology. Another way of designing quality into a product from its conception is to do what is called a Failure Mode and Effects Analysis, or FMEA for short.

There are two types of FMEA:

a) those related to the design, which calls for a breakdown of the components of the product, and then an analysis of the possible defects or failures that can occur with each. This is Design FMEA or DFMEA.

b) those related to the manufacturing process, which calls for a breakdown of the processes involved in manufacturing the product, and then an analysis of the possible defects or failures that can occur with each. This is Process FMEA or PFMEA.

2. How does FMEA work?

The failure mode in every component (for DFMEA) or process (for PFMEA) is analyzed to see its effect on the other components or processes and for the required function of the product.

The effects of each failure mode are considered regarding their a) probability of occurring (OCCUR), b) their severity when they do occur (SEV), and c) ability to be detected if they do occur (DETEC). Usually each of these are expressed in terms of a 0 to 10 scale, with 0 meaning the best possible outcome, and 10 the worst.

Then these three factors are multiplied as the following diagram shows to create an or Risk Priority Number or RPN.

All failure modes are then ranked according to their RPN, and this gives you an idea in which priority to tackle either the components or processes that are contributing to the overall risk involving the product.

This is a very summary treatment of the subject, but that is all that is required for the Overview section of the Six Sigma Green Belt. More detail is gone into under the Define section, which will come in a later post.

Design for Six Sigma–Quality Function Deployment and the House of Quality


1. Quality Function Deployment

Quality control tries to reduce the defects in a manufactured product; quality assurance tries to put processes in place that will assure quality during the manufacturing process. What if quality can be designed into a product in the first place and not just assured and/or controlled during the manufacturing process?

Design for Six Sigma, then, is the third major topic in the Overview section of the Six Sigma Green Belt Certification body of knowledge, after Six Sigma and Lean Principles are introduced. One method for designing quality into a product is Quality Function Deployment, developed by Dr. Yoji Akao in Japan back in 1966. The idea behind Quality Function Development or QFD is to translate customer demands into specific technical design requirements that will, when deployed, achieve that quality demanded by the customer.

2.  The House of Quality

One tool that was developed to illustrate this method of QFD was the House of Quality, which first appeared in 1972. The reason why graphical tool is called the House of Quality is because it vaguely resembles a house. Here’s an example, and below it, I will explain the various features of the House of Quality or HOQ.

House of Quality Template

1. The left wall–VOC

This contains the customer requirements, which are sometimes called the “voice of the customer” or VOC. These are obtained through focus groups, surveys, or other methods to understand what potential customers would desire in a product or service. “Quality” in this sense simply means “what the customer wants.”    The relative importance of these customer requirements is given on the right-hand side of the left wall, usually with a scale from 1 to 5 with 1 meaning “not very important” to 5 meaning “very important”.   This becomes a factor in the “relative importance weight” of the design features which are shown as part of the foundation or basement of the house.

2. The right wall—Comparison with Competition

This contains the customer’s assessment of the competition. With relationship to the customer requirements listed in the left wall, how does the organization’s product or service stack up against the competition? Is it better, worse, or the same with regards to those requirements as compared to the competition?

3. The upper story—Technical Requirements

Here’s where the customer requirements get translated into details of the design. What are the technical requirements of the product or service? These are listed in the upper story above the main floor of the house (see paragraph 5 below).

4. The roof—Co-relationships between Technical Requirements

These technical requirements may reinforce each other, or they may conflict with each other. The price of components may be inversely related to their durability, for example. In the “roof” of the house, there are squares that relate to the intersection of various technical requirements in the upper story. Here is where you indicate whether the relationship is negative or positive, or whether it is strong, moderate, or weak.

5. Main floor—Relationship Matrix of the Technical Requirements

Here is where the Technical Requirements in the upper story of the house are ranked according to how well they actually achieve the customer requirements or elements of the VOC that are listed against the left wall.   They are usually given a weighting from 0 to 5, with 0 meaning “this design feature does not at all meet the customer’s requirements” to 5 meaning “this design feature totally meets the customer’s requirements.” It basically shows how strong the relationship is between the elements of the proposed design and the customer requirements in the VOC.   This relationship is another factor that is used in figuring out the “relative importance weight” of the design feature in the foundation or basement of the house.

6. Foundation—Target Values

The three elements of the foundation of the HOQ are distilled from the other elements. For each Technical Requirement, the following three values are derived, which are listed in the stories of the “foundation” or “basement” of the house.

1)  Relative Weight—the ranking of a) how well the Design Feature meets the needs of the customer requirement (on a scale from 0 to 5) multiplied times b) how important the Design Feature is from the customer’s standpoint.  The result is the “relative importance weight” of each design feature or “technical requirement”.  

2)  Benchmark Value—a measure of the specific value of the design element of the competitor’s version of the design, for example, “weighs < 5 lbs”.

3)  Target Value—a measure of the target value of the design element of the organization’s version of the design, for example, “must weigh < 4 lbs”. How much it should improve upon the benchmark value of the competition will depend on how much relative weight each element receives.

This gives a general idea of the House of Quality is laid out. It is a tool which shows visually how the customer’s requirements are mapped onto technical requirements which meet those requirements to a level exceeding that of the competition.

The next post in the Design for Six Sigma will discuss failure mode analysis.

Theory of Constraints: Drum-Buffer-Rope Methodology


Theory of Constraints: Drum-Buffer-Rope

The 5 Focusing Steps of the Theory of Constraints are summarized as follows (see earlier post for details):

Step Explanation
1. Identify system constraint Find the part of the process that limits throughput of system or rate at which goal is achieved.
2. Exploit system constraint Use incremental improvements or kaizen to the throughput by getting the most out of the system with existing resources.
3. Subordinate everything else Adjust rate of other activities in the process so that they are aligned with constraint; align organization to support decision.
4. Elevate system constraint Consider further actions to eliminate constraint, including additional investment in capital equipment and/or technology if necessary
5. Repeat process Once constraint has been eliminated, avoid inertia and search for next constraint to be removed.

One manufacturing execution methodology that utilizes the first three focusing steps is called “Drum-Buffer-Rope” after its three components:

Component Question Purpose
1. Drum What is the physical constraint or “drum” of plant? Throughput is maximized when constraint operates or “drum” beats at maximum capacity.
2. Buffer What work flows into the “drum”? Throughput is maximized when inventory feeds into the drum on a steady basis, and this system is called the “buffer”.
3. Rope When are new resources required for the “buffer”? Throughput is maximized when the signal requests new inventory to be fed into the drum on a timely basis, and this signal is called the “rope”.

Here’s another schematic comparison between the three components.

This system maximizes the throughput while minimizing both the inventory and the work flow that feeds the inventory into the constraint or “drum”.

There are actually two buffers required, one before the constraint called the “constraint buffer” (no surprise there), and then one right before the product ships out to the customer, called the “customer buffer.” Even if your factory is very efficient with regards to throughput with very little inventory feeding into the system, if your finished products stack up to the ceiling before they get shipped out to the customer, then that is another waste in the system that must be reduced.

This concludes the material on the Theory of Constraints that is expected to be known as part of Lean Tools & Techniques (chapter 2 of the Six Sigma Green Belt Body of Knowledge). The next posts will discuss the D of DMAIC, or “Define” phase of the Six Sigma process.

Theory of Constraints—3 Thinking Processes and 5 “Trees”


1. Three Thinking Processes

To remove constraints and increase throughput, the current state of the system needs to be analyzed, and the constraints identified. Then the undesirable effects or UDEs need to pinpointed, which is the answer to the question “what needs to be changed?”

Then a cause-and-effect analysis is done to find out those actions which, if taken, will result in the elimination of those UDEs. This is the answer to the question “what actions will cause the change?” Finally, the future state of the system without the constraints must be visualized, which is the answer to the question “what should it be changed into?”

2. Tools or “Trees” in Theory of Constraints

NOTE: In the terminology of the Theory of Constraints, UDEs stand for UnDesirable Effects. These three focusing questions are explored with the use of the following tools or “trees”.

Tool or “Tree”

Explanation

1. Current Reality Tree  

Shows the current state, identifies UDEs and traces them back to root cause.

 

2. Strategy and Tactics Tree Implementation plan for changes to achieve the future state (replaces old Prerequisite Tree).

 

3. Evaporating Cloud Tree Identifies changes (aka “injections”) to system in order to eliminate UDEs; resolves conflicts between alternative approaches.
4. Future Reality Tree  

Shows result of changes designed to eliminate UDEs.

 

5. Negative Branch Reservation Identifies possible new UDEs caused by changes from current reality to future reality that eliminate the old UDEs.

The result should be the elimination of the constraints or UDEs from the system.

The last post covers the way that the work flow is controlled in the Theory of Constraints, with the controlled pull system called Drum-Buffer-Rope for short.

Theory of Constraints: Five Focusing Steps


 

This post is part of a series that explains the basic components of the Theory of Constraints. So far I have dealt with the types of constraints (internal/external) and the measures of constraints used as part of Throughtput Accounting (throughput, investment, and operating expenses).

This post will explain about the five steps used in focusing an organization on dealing with the constraints or bottlenecks in the system that limit its efficiency and thus its profitability.    It will then compare the five steps used in the Theory of Constraints and the five stages of the Six Sigma process.

1. Five Focusing Steps

Step Explanation
1. Identify system constraint Find the part of the process that limits throughput of system or rate at which goal is achieved.
2. Exploit system constraint Use incremental improvements or kaizen to the throughput by getting the most out of the system with existing resources.
3. Subordinate everything else Adjust rate of other activities in the process so that they are aligned with constraint; align organization to support decision.
4. Elevate system constraint Consider further actions to eliminate constraint, including additional investment in capital equipment and/or technology if necessary
5. Repeat process Once constraint has been eliminated, avoid inertia and search for next constraint to be removed.

Fig. 1 The Five Focusing Step Cycle

2. Relationship between Theory of Constraints’ Five-Step cycle and DMAIC process in Six Sigma

The relationship between this five-step cycle to removing constraints in the system and the DMAIC approach of Six Sigma is somewhat complicated. The Define, Measure, and Analyze would be analogous to Identify, whereas the Improve would correspond to all of Exploit (improving throughput with existing resources), Subordinate (reducing inventory and operating expenses to open up new resources), and Elevate (eliminate constraint with new and existing resources). The Control part of Six Sigma, the extension of improvements from the short term to the long term, would be analogous to the Repeat process in the Theory of Constraints, where improvement to throughput are extended by focusing on another constraint.

The next post will deal with some analytical tools used in the Theory of Constraints.

The Theory of Constraints: Throughput Accounting vs. Traditional Accounting


The Theory of Constraints tries to make an operating system more efficient, i.e., more profitable, by analyzing its components into inputs, processes, and outputs, and then measuring them.    Then constraints are removed to improve the system.   This post talks about Throughput Accounting, which is a way of measuring the efficiency of a system.  After this explanation, there will be a brief explanation of how this kind of accounting differs from the traditional type of accounting (particularly with regards to how it treats inventory).

1. What are the three basic measures of an operating system?

The way you measure the efficiency of an operating system is through what is called Throughput Accounting. The theory of constraints says that you try to maximize your output or throughput and minimize your input, of which there are two types, money or operating expenses and capital or investment.

Sales are an example of throughput; inventory, equipment, and real estate are the investments needed to run an operation, and the money it takes to actually create throughput would be the operating expenses.

In general, the goal of an organization will be first to increase throughput, and next to decrease both operating expenses and investment.

2. What are the four derived measures of an operating system?

Given the basic statement in paragraph 1, that the goal is to first increase throughput and next decrease operating expenses and investment, let’s turn to the four derived measures of an operating system which are derived from the three basic measures.

Derived Measure Formula
1. Net Profit Throughput – Operating Expenses

 

2. Productivity Throughput / Operating Expenses

 

3. Return on Investment (ROI)

 

Net Profit / Investment

 

4. Investment Turns Throughput / Investment

1. Net Profit

The net profit can be increased by either increasing the throughput or decreasing the operating expenses.

2. Productivity

The same type of measure as “net profit”, but this time expressed as a ratio rather than the difference between throughput and operating expenses.

3. Return on Investment

This can be increased by an increase in net profit or a decrease in investment required to create that amount of profit. The net profit, in turn can be increased according to the two factors described above in paragraph 1.

4. Investment Turns

How much throughput is created per unit of investment? This is what is measured here.

3. Throughput vs. Traditional Accounting

The above measures are part of what is called Throughput Accounting. How does it differ from traditional accounting? The difference is most notable with inventory, which is part of investment. Traditional investment considers unused inventory as an asset, because it could theoretically be sold and become part of the throughput and thus contribute to the net profit. In reality, however, it just sits there and that is why inventory in Throughput Accounting is considered a liability, and not an asset. As a liability, the emphasis is on reducing it as much as possible.

Now that we have covered the way the Theory of Constraints measures the efficiency of an operating system, let us discuss in the next post the five focusing steps on how to increase the efficiency.

Theory of Constraints—Types of Constraints (internal/external)


The Theory of Constraints is part of the material on Lean Principles that is covered in the Certified Six Sigma Green Belt material (chapter 3 of the primer put out by the Quality Council of Indiana).   The next few posts including this one will discuss this theory by explaining its components.  All of these were presented by our instructor in our Green Belt Six Sigma class; for more details, one online resource I found that is reliable is www.leanproduction.com.

  1. What is a constraint? (internal/external distinction)
  2. Three measures of a system (throughput, inventory, operating expenses)
  3. Five focusing steps (Identify, Decide, Subordinate, Elevate, Repeat)
  4. Three questions of Thinking Processes (identifying and analyzing constraints)
  5. Tools or “Trees” (formalizing Thinking Processes) (Current Reality, Evaporating Cloud, Future Reality, Strategy and Tactics)
  6. Drum-Buffer-Rope (relationship of production to constraints)

1. Theory of Constraints—Introduction

The Theory of Constraints is a theory developed by Dr. Eliyahu Goldratt, who popularized it in his novel The Goal, first published in 1984. The basic principle behind the Theory of Constraints could be summed up in the saying that “a chain is at strong as its weakest link.” In a manufacturing process, the link that is weakest is the constraint or bottleneck.

2. Types of Constraints (Internal/External)

Let’s start with the explanation of a constraint. A constraint is that which keeps a system from achieving more of its goal. If you are a manufacturer creating a product, then a constraint would be that which prevents you from selling more of your product.

What’s the reason for your not being able to sell more of your product? There are two general answers to this question:

  1. There is not sufficient demand in the market to match the supply of your product.
  2. There is not sufficient supply of your product to match the demand in the market.

In the first case, the source of the constraint is in the market, which is external to your organization, so it is called an external constraint.

In the second case, the source of the constraint is in your organization, so it is called an internal constraint.

Another way to look at this is with the diagram below. In comparing supply and demand, one of them is going to be greater than the other (except if they exactly balance each other and there is therefore no constraint). Whichever one of them is smaller is the constraint. If the constraint is the demand, that comes from the market, so it is external, and if the constraint is the supply, that comes from the organization, so it is internal.

3. Types of Internal Constraints

There are three types of internal constraints. The way the equipment is used may be limiting output, the training and/or mindset of people could be preventing the system from producing more, or it could be the policy
of a company that is the real source of the system not being able to achieve its goals.

How do you identify which of the constraints is the one that is preventing your organization from achieving its goals? There are measures of a system which may help provide clues, and that is the subject of the next post.

Six Sigma Green Belt—5 FAQs on Certification Exam


The American Society for Quality sponsors many certifications relating to quality, one of which is the Six Sigma Green Belt (SSGB). This post talks about the certification test, so you understand what is involved in preparing for it.

1. When is the exam offered?

The date to take the next SSGB certification exam being offered at a local ASQ chapter is December 1, 2012, but the deadline for applying for this exam has unfortunately already passed. For those on the West Coast, you can take the exam at the 2013 Lean and Six Sigma conference which will take place in Phoenix, AZ on March 3, 2013. The application deadline for that exam is coming up on January 20, 2013.  Other exam dates and locations are to be found on the ASQ website.

2. What is the cost of the exam?

It costs at the present time $359 to apply and take the exam, but only $209 if you are a member of ASQ. Since the ASQ membership costs $139, and you get a $150 discount on the exam if you are an ASQ member, I strongly recommend that you become an ASQ member if you plan to take the SSGB exam.   In addition, if you become a member of ASQ, you will be able to connect to other members in the quality community locally, and networking is an essential part of your career.

3. What material do I study to prepare for the exam?
The material covered in the exam is called the Six Sigma Green Belt Body of Knowledge, and that is contained in the Certified Six Sigma Green Belt Primer published by the Quality Council of Indiana (QCI). The primer contains all the Six Sigma and Lean topics covered in the exam. It also includes sample test questions for each of the chapters, which are outlined in the chart below under question 5. The cost of the primer is $70. There is a solution manual for the sample test questions that costs an additional $35, and an electronic exam CD gives additional practice questions from past SSGB exams, and costs an additional $70. So the entire “deluxe” set of preparatory materials from QCI costs $175 (plus shipping and handling).

The exam is an OPEN BOOK test, and you may bring the CSSGB primer with you to the exam. However, the sample test questions in the primer that come from previous exams must be removed; these are printed on green paper so you can easily locate these pages.

An important hint was given us by our instructors, namely, to put tabs on the various pages where we locate important concepts or formulas that come up on practice exam questions. Also take the index out of the primer and put it at the FRONT of the book so you can remove it during the exam and have it ready to be referred to. This is because you will need it to search for important terms, etc., and having it out in front of you will save some time as compared to having it at the back of the binder where you constantly have to flip back and forth between it and the material in the front of the binder.

4. What kind of calculator is allowed on the exam?
Since there will be some questions that require calculation, you can bring a calculator as long as it is not a programmable one. Smart phones are not allowed to be taken into the test.

What our ASQ instructor for our SSGB class recommended we use is the TI-36X Pro calculator which costs somewhere around $20-$25, depending on where you buy it. Added to the cost of the QCI primer materials, you are looking at a total cost of about $200 for the preparation materials plus your calculator. It is important to practice using your calculator before the exam so that you know where all the relevant buttons are in order to do your statistical or other calculations. Taking a “calculator guide” which outlines where the important statistical functions are located on your calculator is something highly recommended by our instructor, so you don’t waste time trying to hunt for the right buttons during the exam.

5. How is the exam scored?

There are 100 questions, and you have 4 hours to complete the exam, so that means you have 2.4 minutes per each question. This should be plenty of time to answer the questions. Here’s how they are broken down in terms of subject area of the SSGB Body of Knowledge or BoK. You will notice that after the overview, the categories correspond to the acronym DMAIC, the basic methodology of Six Sigma.

Category of BoK Number of Questions Chapters in CSSGB Primer
Overview 15 2, 3
Define 25 4, 5
Measure 30 6,7,8
Analyze 15 9
Improve 15 10
Control 11

The two largest areas are Measure, containing the statistical tools and techniques used in Six Sigma, and Define, where you go from defining the problem to proposing a solution. Then come the three areas of a) Analyze, where you design an experiment to test the solution, b ) Improve and Control, where you implement the solution and make sure it stays implemented, and c) Overview, which explains the basic principles behind Six Sigma and introduces some of its tools & techniques.

I hope that this post answers questions you may have on the exam and what you should do to prepare for it. If you do have additional questions, leave them in the comments to this post.

So far in my posts on Six Sigma and Lean I have been covering topics that are in the Overview section of the SSGB BoK as given in the CSSGB Primer. I will complete these in the next few so I can go on to the next section, that of Define.

20 Six Sigma Lean Tools: A Detailed List by Category


1. Six Sigma Lean Tools & Techniques–Categories (recap from previous post)

To implement Six Sigma and Lean tools and techniques, you need to break down the manufacturing process into individual processes which can then be labeled as a) non-value-added activities or NVAs, b) activities which are required by the manufacturing process which I will refer to as “work flow”, and c) value-added activities or VAs. In Six Sigma you try maximize the effectiveness of value-added activities (category c) by reducing variances in the outputs. Lean manufacturing consists of minimizing non-value-added activities (category b) and streamlining the work flow (category b). Then, there are systems of Organization that try to rationalize the manufacturing process as a whole, and systems of Improvement which apply to the individual processes AND the entire overall manufacturing process.

These different categories are related schematically in the diagram below.

2. Six Sigma Lean Tools & Techniques—Details

Here are the various tools & techniques used in Six Sigma Lean manufacturing based on the color-coded category listed above. Terms that come from Japanese are listed in italics.

Category

Tool or Technique

Explanation

NVA 1. Elimination of muda Minimization of waste in NVAs
2. SMED or Single-Minute Exchange of Dies Rapid changeovers of production machinery
3. Motion study and material handling Reducing unnecessary motion in the handling of materials from one station to another.
4. JIT or Just In Time Just-in-Time is a system of ordering parts so that they are delivered just when needed in order to reduce unnecessary inventory.
5. POUS (Point of Use Storage) Parts are stored near where they are to be used to reduce transfer time and storage costs.
6. Cellular flow Machines are grouped together according to the families of parts produced, reducing travel time and storage costs.
7. Kanban—”Pull” system Notifies operators to order parts or orders them automatically from suppliers when needed.
Work Flow 8. SUR or Set-Up Reduction Breakdown into internal/external set-up operations and preparation of parts & maintenance.
9. Visual Controls For purposes of visually displaying instructions to operators regarding proper work flow.
10. Poka-yoke Means “error-proofing”; visual and manual controls that help detect/prevent operator errors.
VA 11. Decreased cycle time Reduction of time required to process one batch.
12. Batch reduction Reduction of units per batch in order to avoid overproduction and inventory.
13. Quality at Source Quality inspection done by every operator at every stage rather than leaving quality up to “final inspection” alone.
14. Continuous or one-piece flow Goal is the culmination of the above three techniques: reduce batch to one piece at a time in which operator adds value at every step in the most time-efficient manner.
Organization 15. SOPs or WIs Documentation of Standard Operating Procedures (SOPs) or Work Instructions (WI)
16. TPM or Total Productive Maintenance Doing regular maintenance not just to correct malfunctioning equipment but to prevent such malfunctions from occurring.
17. 5S System System for workplace organization to make sure workers have safe, reliable access to all necessary tools.
18. Plant Layout Planning layout of production facilities in plant so as to minimize work flow.
Improvement 19. Kaizen Teams Continuous, incremental improvements on existing processes using teamwork.
20. Value Stream Mapping Breakdown and analysis of manufacturing processes to facilitate reduction of NVAs, and streamline work flow.

The Relationship between Project Management and Quality Control


Having just finished a summer-long course in project management, I am now studying quality management in a Six Sigma Green Belt (SSGB) class.  

I’ve been thinking about the intersection between project management and quality management since I started the started the SSGB class and wanted to sketch out a few of these ideas.

1.  Quality Management => Project Management

One of the relationships between quality management, particularly as done through the Six Sigma process, and project management, is that each quality improvement “experiment” is done as a project.  The Black Belt acts as the project manager of the project, and the Green Belt acts as a member of the project team. 

The project is chartered by the Black Belt with the approval of management or perhaps a Master Black Belt, and it is carried out by the Black Belt who has the Green Belts gather the data.   He or she then

  1. Defines the problem with the process,
  2. Measures the performance baseline data and the variances from it,
  3. Analyzes the root cause of the variances, proposes a solution, and designs an experiment to test it,
  4. Improves the process and verifies the hypothesis that the solution worked, and
  5. Controls to make sure that the solution stays in place.

This is the DMAIC methodology based for quality process improvement.  So each Six Sigma improvement process is run like a project.

2.  Project management => Quality Management

A project manager must manage the quality on his or her project, of course, but managing the cost, schedule, and even scope of the project also take on characteristics of quality management.  Instead of a performance baseline that defines quality, there is a cost performance baseline (i.e., the budget), and a schedule performance baseline (the schedule).   Variances from these performance baselines are monitored or measured at various points in the project, and they are controlled, meaning that any significant variances call for some sort of correction action.  

Since in a project you are often dealing with the output of people, not machines, analyzing and correcting the root cause of a variance may differ from the precisely laid-out DMAIC methodology of Six Sigma.   But the same problem-solving spirit is certainly there as it is in Six Sigma and this is one common point I find between project management and Six Sigma methodology:  they both require a lot of problem solving, a penchant for working with teams, and an ability to get all stakeholders to buy into any proposed solution.