1.  World Class Quality by Keki Bhote

For the Analyze and Improve phases of DMAIC.  In my opinion, this is the most powerful, hands-on problem solving book available for manufacturing professionals.  Keki Bhote’s passion for quality and highly opinionated discussions make this book an entertaining and highly educational read.  This is the book that propelled my career at a time when I was a quality engineer in an assembly plant, swamped with quality problems and a very unhappy sales team.  The tools taught in this book enabled me to solve a number of complex problems that others struggled to get their arms around.  The book focuses heavily on the Shainin Methods, which emphasize simple and very effective approaches for isolating root causes on the production floor.  Keki’s mantra is, “Talk the the parts, not the engineers!”  He focuses on fact-based problem solving using physical evidence instead of collective opinion.

2.  Goal-QPC’s Memory Jogger Series

For all phases of DMAIC.  Goal-QPC has a series of quick reference books that are very well written and work very well as training materials.  The book shown is the second edition of th original Memory Jogger, which I have used countless times for training shop-floor associates.  Some of the tools covered include brainstorming, fishbone diagrams, control charts, process flowcharts, histograms, and pareto charts.  A total of 27 tools are covered, and the graphics and examples in the book are outstanding.  Goal-QPC also has a number of other process improvement books, including several quick reference guides on Six Sigma.

3.  Mistake Proofing for Operators

For the Control phase of DMAIC.  I am amazed at the number of businesses that have never introduced their teams to mistake proofing.  This book provides an excellent grasp of the concept, and is loaded with a number of specific examples that show how part and process design come together to produce defect free products.  I have purchased multiple copies of this book for shop floor operators, mechanics, and technicians.  As long as I’ve bought a copy for each person in the class, I have no problem scanning the examples and adding them into my training material.  Mistake proofing is a powerful and lasting defect prevention tool, and is an excellent way to lower risks identified in PFMEA’s.

4.  Strategic Error Proofing:  Successful Processes and Smart FMEA’s

Also for the Control phase of DMAIC.  This is an outstanding book that covers error proofing and mistake proofing (you’ll learn the difference between these two terms when you read the book!), and lays out a framework for designing, implementing, and maintaining devices across large facilities.  The books also provides a framework for rating error proofing and mistake proofing devices (in terms of simplicity, maintainability, and several other factors), something I had not seen until this book was published.  Quality leaders for larger organizations will also appreciate the focus on error proofing and mistake proofing metrics that can be rolled up across plants and businesses.  This is a terrific book.

5.  AIAG’s FMEA Manual

For the Analyze and Control phases of DMAIC.  The Automotive Industry Action Group (AIAG) publishes a number of excellent how-to guides for quality planning, and this FMEA guide is the most comprehensive that I have seen on the subject.  It includes complete guidelines for facilitating and documenting FMEA’s, and covers every detail of the subject.  The AIAG also publishes a number of other excellent reference publications, including Advanced Quality Planning and Control Plan.  These books are must-have references for quality professionals.

6.  Statistics for Experimenters

Box Hunter & Hunter’s book is a classic DOE training guide that starts with the basics.  If you want to follow sound experimentation techniques (which make all the difference between true conclusions and false conclusions), then read this book.  Here are a few comments from the publisher:  ”Catalyzing innovation, problem solving, and discovery, the Second Edition provides experimenters with the scientific and statistical tools needed to maximize the knowledge gained from research data, illustrating how these tools may best be utilized during all stages of the investigative process. The authors’ practical approach starts with a problem that needs to be solved and then examines the appropriate statistical methods of design and analysis.”

Implementing an overarching quality system is essential for sustaining best practices across any organization, and the ISO:9001 framework the most commonly followed model for achieving this goal.  There are several elements inside the ISO:9001 framework, and one of the most critical is QMS (Quality Management System) audits.  When I was a young quality engineer, I viewed QMS audits as outdated practices that added no value.   As my responsibilities have grown over the years to include global, multi-plant responsibility, I have come to appreciate the tremendous value of these audits.

What Does a QMS Audit Accomplish?

QMS audits verify that the quality system is being followed, and without these audits, compliance to the quality system is guaranteed to deteriorate over time.  This is what happened in the plant I visited.  A particular process had been implemented several months prior to my visit that required a great deal of extra effort on the part of the equipment operators.  Unfortunately, there were no audit mechanisms in place to ensure that the equipment operators were carrying out their new responsibilities over the long run, and compliance with the new process gradually deteriorated, particularly on the night shift.  As a result, the plant produced a great deal of suspect product over time, which cost the business hundreds of thousands of dollars in warranty returns.  All of this could have been prevented with a simple routine audit, conducted by the quality team, to ensure that operators across the plant were complying with control plan requirements.

Choose Your Crisis

Businesses have two choices when it comes to the crises they respond to:  they can let their customers pick the crisis-of-the-day when poor quality product is shipped, or they can created their own internal crises by auditing compliance to key processes (business processes too) and implementing solid action plans when audit findings uncover major problems with compliance.  This latter approach provides far lower quality costs for obvious reasons.

The Keys to Successful QMS Audits

1. Have the discipline to create and follow an internal audit calendar – this is the Quality team’s responsibility.
2. Make the audits specific and meaningful.  Target the critical areas in the business and focus on specific process requirements.  General audits, where the auditor randomly selects where to focus, provide very little value.
3. This is very important – the management team must react strongly to internal audit results, as if a customer had reported a major quality issue.  This type of behavior will ensure that the entire team understands the importance of adhering to the quality system, and that systemic causes behind noncompliance are corrected.

Six Sigma certification is about building the skill set to understand and control processes to achieve substantial, permanent gains. Have you ever found yourself in a project where the team, despite a great deal of effort, is struggling to bring a process to an acceptable level of performance? Sometimes, despite an effective use of the Six Sigma tools, teams cannot “crack the nut,” – they cannot identify and control the key inputs that are driving unwanted variation. The temptation in these situations is to stay on course, applying the DMAIC process within the confines of the existing process, when a faster and more effective approach might be a process or product redesign.

Let’s look at an example that some can relate to – let’s say that John, a recently certified Six Sigma Black Belt , decides to use the DMAIC approach to find a girlfriend. Having spent too much time in front of the television in his early years, John believes that perfecting his “pick-up lines” is the only way to realistically meet a girl. So John sets up a simple experiment where he tries four different pick-up lines at various locations, from bars to bakeries, also varying a few factors like eye contact, the strength of his smile, and the color of his shirt. John is a hard worker, and ultimately introduces himself to 60 women in three months. Here are the results

- 52 of the 60 women had no interest in John whatsoever
- 4 women took John’s phone number but did not call him
- Another 4 took his number and called him, and two of these resulted in dinners out with John
- Of the two women that went to dinner with John, one did not show interest in seeing him again, and the other had a boyfriend who was traveling overseas at the time.

So where did John go wrong? He was working inside a lousy equation, trying to optimize a fundamentally lousy process. How often do we find ourselves in this situation – we work within the confines of our existing process, lured by the fact that we sometimes get good results, believing that we have failed if we do not find the “root cause” and obtain zero-defect performance. This is a common flaw in Six Sigma certification training: too much emphasis is placed on understanding and controlling existing processes, with little focus on building cross-functional teams to generate “out of the box” solutions that may result in faster, more effective fixes.

Let’s look at John’s situation. There were some big factors that John will never be able to control, like the fact that many women are immediately put off by superficial pick-up lines. Also, John has no way of knowing which women are already in relationships. These two factors alone provide with a John a best-case success rate under 50% for a first date, even if he is the best looking guy on the planet! Clearly John’s issues go beyond using the DMAIC approach to optimize his “process” for finding a girlfriend. His process is fundamentally broken, and no amount of optimization will provide meaningful results.

John is persistent and hard working, but also smart enough to know when he is “barking up the wrong tree,” so he meets with a relationship consultant who writes a whole new equation for John, one that has been proven time and time again. Two major changes are in order: (1) use a reputable dating service to find women looking for a relationship with a guy like John, and (2) actually engage in genuine conversations with these women during his introductory discussions with them – no superficial pick-up lines.

John’s life changed. First, his success rate in landing a first date went from 3% to 67% (two out of three) – a rate that the best pick-up line in the world couldn’t have provided. Second, he was able to stop working his equation altogether, because the second of the two women who agreed to go out with John turned out to be the love of his life.

It’s important to recognize when a team is struggling with a lousy equation and needs a game changer – a significant process or product redesign that results in a new equation with tremendous potential for success. This is where having a cross-functional team is crucial, because Six Sigma Black Belts are usually not process and product design experts. So the message is, seek the most productive path to a lasting solution, even if it means abandoning the DMAIC approach within the confines of a current process or product design.

Here are a couple of examples –

Magnet Adhesion Project

A motor manufacturer bonds magnets to motor housings using epoxy glue that is heat-cured. The Cpk for bond strength is 0.4, not good! Despite a number of studies, the Six Sigma Black Belt and her team cannot find and control the variables that are driving poor bond strength. Finally she holds a brainstorming session, seeking “out of the box” solutions. A team member proposes a secondary grit-blasting operation on the magnets to improve adhesion. This additional process ends up solving the problem, and the short term Cpk improves to 2.5! A solid control plan is put in place to monitor the grit blasting operation, and the process produces zero loose-magnet defects in the years that follow. Had the team stayed on the path of finding and controlling root causes within the confines of the existing process, they might still be working on the project.

Cracked Pump Housings

A pool pump manufacturer is receiving field complaints due to cracked pump housings. A Six Sigma team is put on the project, and after an extensive DMAIC study cannot find which factors are causing cracked housings. A number of studies are done, including chemical analysis, melt viscosity on the (plastic) housing material, dimensional analysis on good versus failed housings, and a geographical analysis of the failures to understand possible environmental factors. After several weeks with no root cause identification, a design engineer looks closely some of the field failures and proposes a redesign to the housing, adding more material and more generous radii in the areas where the cracks are forming. The approach ultimately solves the problem completely, and the “fix” is robust, meaning that it does not require extensive monitoring in the manufacturing process.