In this post I ask the question, “what is quality and how do we know when we have it?” We all have experience as consumers. As consumers (customers) we have an intuition about quality, we believe we know what it is. When asked to define quality we might even say we know it when we see it, and in each individual instance that might be true but let’s take a moment now and try to define it.
A quick Google search might tell us that quality is” the standard of something as measured against other things of a similar kind; the degree of excellence of something.”
ISO 8402-1986 defines quality in manufacturing as “the totality of features and characteristics of a product or service that bears its ability to satisfy stated or implied needs.”
Lexus defines quality (according to a 1992 TV Commercial) defines quality as the ability to role a ball bearing along the edges of a car’s body panels.
The ASQ is a global organization that certifies Quality Engineers (http://asq.org.)
They define quality as: “A subjective term for which each person or sector has its own definition.” They go on to Quote the ISO definition, and note that according to Philip Crosby it means “conformance to requirements” and Joseph Juran says that it is “fitness for Use.”
As the owner of a “Job Shop,” focused in the B2B sector and supplying machined parts and machining services, I say quality is the only thing the customers are willing to pay for. Sometimes it means giving customers exactly what they asked for, and sometimes it means giving the customer what they need, regardless of what they asked for. The trick is to know what it means for each customer at each moment.
For manufacturing companies who are making products that they “own” and then sell to retail customers, creating a quality product means understanding the customer needs, figuring out how to fulfill those needs, and then manufacturing widgets that fulfill the needs.
For manufacturing engineers, and production managers at those companies, and at job shops it means making parts that conform to previously agreed to specifications provided by the customer (or design department.)
These design specifications, affectionately referred to as “The Art” in this book, combined with an accepted “Purchase Order” often constitute a legal contract.
Failure to deliver on time and “in spec” could mean that your company did all that work and bought all of those materials and won’t get paid, even if the customer took delivery and used the parts.
Quality means delivering what your customers need when they need it. To do this the manufacturer needs to understand what the customer needs, how to meet those needs, and how to know they did.
Profit may be King but Quality Rules the Realm
All businesses must make a profit to succeed overtime, this makes profit king, but if you simply go for short term profits with fancy marketing and sales tricks or worse by cutting costs in a way that reduces the quality of your parts you eventually alienate your customers because they stop receiving the value they expect.
Years ago there was a candy manufacturer, call them Acme Chocolates. They experienced this lesson first hand, or so the story goes.
Acme chocolates had the best selling candy bar on the market, sales were growing, but profits were flat, and so the board of directors brought in a new CEO to turn things around. The new CEO, let’s call him George, had an MBA from one of the best business schools, he was a SixSigma Black belt, and was a master of Lean Thinking. Not only that, George knew something about making candy too. There was no doubt at all that he could get the profits to line up with the sales, do doubt at all.
The first thing that George did was call a meeting of the production engineers and management team. They showed him Spaghetti Diagrams, charts, and spreadsheets. They talked about incremental improves they could make generating incremental savings and marginal profit gains. George said it wasn’t enough. He
thought for a second and told the production manager to go back to the slide with the list of ingredients in the Acme Bar sorted by annual costs.
“There” he said triumphantly. “Ingredient number 17 on that list. If we just stopped buying that one ingredient would save almost 8 million dollars a year! If you can figure out how to make me an Acme Bar that tastes just as good without that ingredient I’ll distribute half of that 8 million to your team as a one time bonus next year.
Well you can bet with an incentive like that, they went to work. Not only that, they did it. It took process tweaks, changes in the quantity of other ingredients, and an enormous amount of customer taste tests to ensure that the quality of the new Acme Bars was just as good as the old (Some of the tests even showed that the new one was better.) It turned out that the net savings was just over 6 million and George made good on his promise. Everybody was happy, profits were up sales were up, and the stock value was up.
With a success like that in his first year, George now had a reputation to maintain. He met again with the team who were also fired up from success. They studied the processes and ingredients list, picked another ingredient and went to work to see if they could repeat the success of the previous year. Sure enough they made it work, again relying on focus groups and side by side taste tests to ensure quality.
Spurred by success they doubled down and kept at it. Month after month and year after year George and his management team kept at their cost cutting pouring much of the savings into marketing campaigns, billboards and even a super bowl commercial. With all of this though sales began to fall, and profits were off.
George tried to rationalize it to the board, claiming the influx of low cost competitive products flooding the market after free trade laws were enacted. Even blaming the customers for having changing and evolving tastes, and the board bought it. All products have a life cycle they thought and the Acme Bar may have run through its peak. They urged George scale back production, take the savings from laying off the staff and start investing in new product development.
George walked back to his office munching on an Acme Bar glad to still have his job and thinking about how to break it to the management team that they would probably have to shut down part of the factory and let go of key people who had worked for Acme Candy for decades. He sat at his desk staring at the wall for what seemed like hours but was probably only 2 or 3 minutes. Chocolate had always cheered him up so he fished around his desk unwrapped a bar and took a bite.
“Man that’s a good chocolate bar” he said out loud. How could sales be off he wondered. “This is the best chocolate I’ve had in ages.” He continued talking to the empty room “Damned customers and their fickle tastes. They wouldn’t know quality chocolate if it kicked them in the ass.” Then he looked down at his desk and the wrapper from the bar in his hand. It was the old style wrapper. This was one of the original Acme Bars. He quickly walked back to the now empty board room and picked up one of the new bars from the table. He took a bite and it was like night and day, the original bar was so much better.
Instead of scaling back production and starting over from scratch, they re-introduced the original Acme Bar. It took a few years but they finally regained the number one spot in the market. And at the same time they figured out how to grow the profits by introducing new complimentary products.
It was an incremental degradation of product quality that eventually caused the problems at Acme. The root cause of the degradation was the fact that they were measuring the wrong thing when they decided that each new version of the bar was as good as the previous. They relied on anecdotal evidence based on comparing each new bar with the previous. It would have been if they tested each new version as compared to the original, or even better could have a metric that wasn’t based on anecdotal evidence from side by side tests and focus groups.
To deliver quality to our customers we have to know what they need.
Years ago an engineering colleague of mine came back from an international design conference hosted in Italy. As an engineer he has always been interested in the how part of problem solving and design. We were chatting about the conference over a martini one evening and he commented that “all the Italians wanted to talk about was the customer needs and they weren’t interested in the design part of designing.” I laughed, as a business owner, I know you don’t get paid for designing or knowing how to design unless you design something the customer needs.
It is up to the designers to understand what the Customer Needs, CNs, are. Once the designers understand the CNs it is possible for them to figure out how to meet those needs by establishing a set, or sets, of Functional Requirements FRs that define how the needs will be met. For example if the designer decides, the customer needs a way to travel to Mars, the highest level FRs might be:
- Accelerate the customer to escape velocity from the Earth’s surface.
- Guide the customer to the point in Mars’s orbit where Mars will be when the customer arrives.
If they then design and manufacture a system that can accomplish these two functions and try to sell it to the customer they may not make the sale. What if the customer has additional needs that the designer didn’t consider? Maybe the customer wants to live through the experience. Maybe they want to return home after the trip. Do they want to fly by Mars or land? If they land how long will they stay?
It is possible that I cannot stress enough that it is up to the designer to understand and interpret the customer’s needs. Once they have done this and have developed a comprehensive list of FRs The designer can then establish Design Parameters, DPs, that will satisfy the FRs. The DPs will include things like drawings, solid models, and lists of specifications and they need to convey enough information to the manufacturer so that they, the manufacturer, can make the part(s) described.
Let’s first consider a simple physical component with no moving parts, remembering that anything with moving parts can probably be assembled from a series of such components.
Before a part can be manufactured, he designer needs to, at least, tell the manufacture:
- the size and shape of the component,
- the material it will be made out of, and
- how much the manufacture can screw up and still make an acceptable part.
And, the purchaser, who may or may not be the designer, needs to tell them:
- how many they want and
- when they need them.
Geometric Dimensioning – the size and shape of the component
In the earliest days of manufacturing the designer was almost always the manufacturer and often had the exact same needs as the customer. Processes were developed by trial and error which eventually enabled the the production of parts which could be sold or traded. These manufacturers became skilled craftsmen. Generation after generation they would pass on their craft(s) as Masters to Apprentices and Journeymen who would in turn become Masters to continue the cycle.
As the industrial revolution took off and production lines became the norm the people performing technical crafts began to specialize more and more focusing in tighter and tighter niches. In one of my first engineering jobs, in what we might consider the dawn of the information age, I was a Design Engineer working in the R&D department of a medical device manufacturer. It was my job to envision the solutions to the customers needs (as defined by the marketing department.)
Once I thought I knew what to do I would sketch a part or parts I wanted and give the sketch to a Drafts Person. She would then take my sketch and create dimensioned drawings of the part or parts I wanted. I should review the drawings, she would make adjustments and then we would have some one in Purchasing send the drawings to vendors for pricing. once the best price was established Purchasing would order the parts.
The geometric dimensioning happened on the drawings prepared by the drafts person with my help. She prepared the drawings instead of me because she had spent two years in school learning how to create drawings in compliance with national and international standards. She knew what symbols to use to precisely indicate to any manufacturer exactly what we wanted. She knew how to represent complex shapes in simple views, and how to show dimensions and tolerances with no ambiguity. Deciding what the dimensions and tolerances were on the other-hand was my job.
Material Specifications – what it will be made of
It is almost always the job of the designer to specify the materials to be used. The choice of materials often depends on the required strength and durability of the part(s) to be made. It can be impacted by the need for conductivity or resistance of electricity, resistance to corrosion, and heat transfer coefficients. Sometimes it is the physical appearance or feel of a material that dictates the choice. Material specifications can be vague like “Low Carbon Steel” or they may be specific indicating to within parts per million of how much of a particular element can be present in the alloy.
Tolerancing – how much we can screw up and still make a good part.
I stated above that determining tolerances was the designers job, but didn’t explain what a tolerance was. Simply stated a tolerance is the amount of leeway the designer is willing to give the manufacturer for any particular dimension or specification. If the designer wants the part to be five inches long but is willing to accept four and three quarters or even five and a quarter the tolerance is half of an inch.
Why do we need tolerances, why not just make all of the parts five inches long? There are two reasons really:
- no manufacturing process i perfect and
- no measurement tool is perfect.
Tight tolerances are when the designer is only willing to accept small variations in a particular dimension or specification. Tolerances are considered to be loose when the designer is willing to accept relatively large variations in particular dimensions or specs. and in general you can assume parts with tighter tolerances will be more expensive to make than parts with loos tolerances. This is because the tighter the tolerance for a particular dimension the more expensive the equipment needs to be to make it and the more expensive the process of measuring it is.
If you cannot measure it, you cannot make it.
It may be possible to make things you cannot measure, but it is inadvisable to ship them to your customer. Customers don’t want to pay for parts that do-not meet your specifications and may even back charge you for the cost of receiving and disposing of parts that are out of spec. I once had a customer try to back charge me for a part they had received and accepted in writing because they later lost the part in their warehouse.
I state above that quality in manufacturing is delivering what your customer needs when they need it. Unless you are the inventor and designer and are selling directly to end users of your product it is up to the customer to tell you what they want, and to provide a detailed design and specifications that allow you to know what they want.
As the manufacturer you then need to make sure you don’t ship any bad parts to your customers. You can do this by hoping you don’t make any bad parts, or by establishing a quality plan that includes inspection.
The Quality Plan | How we make sure our customers receive quality parts. First we need to decide what it is the customer needs usually based on what they asked for. Then we need a plan for how we can make and ship the parts without defects. ASQ will tell us that a quality plan is an overriding document that defines how a company or organization intents to create quality products (http://asq.org/learn-about-quality/quality-plans/,) but that isn’t what I’m talking about. We need one of those too, but what we need at this step is the specific plan for this part that we are making.
This plan includes work instructions that step through the processes of transforming the raw materials into the finished part our customer wants. These instructions will often be based on standard practices outlined in our organization’s plan but may need to be customized based on specific customer needs. The plan will also include details for in process measurements to be made and final inspections to be made (if needed) before shipping the parts to the customer.
There are really only two ways we can ensure we don’t ship bad parts to our customers.
- Don’t make any bad parts.
- Find bad parts and don’t ship them.
If we think about value and Lean Manufacturing it seems that the first option is the best. Bad parts we make and don’t ship are by their existence wasteful. In fact measurement of any kind rarely adds value to the parts we make. The One notable exception is when the customer is willing to pay more for a part with an inspection report.
With that said let’s talk about the 2nd option first. Practices that keep us from shipping bad parts that we might make to our customers are usually governed by a Quality Control (QC) or Quality Assurance (QA) program or department at our company. It is often a Quality Control Engineering or QC Manager who is in charge of making sure we don’t ship any bad parts. The last bastion of quality at many companies is a final inspection step at or before the packaging step.
Inspection is any step that compares a part with the specifications that define it. It almost always includes measuring:
- The size of parts and features (length width and height,)
- Location of features / surfaces,
- Orientation of features / surfaces,
- Surface Texture / Roughness,
- Thickness of Coatings,
- Bulk Material Properties, and even
- Visual Appearance.
Inspection usually involves the use of measurement tools, and the act of measurement is often called metrology. In general you can measure anything you can define with a number and put a tolerance on. The trick in Quality Control whether you are using methods that keep you from making bad parts, or tools that simply keep you from shipping bad parts is knowing what to measure, how to measure it, and how often to make measurements.