Estimated reading time: 10 minutes
Solving DAM Problems
By Gray McQuarrie
< Back To Columns
Solving Your DAM Problems: Those DAM Laminate Suppliers
Editor's Note: This article originally appeared in the May 2012 issue of The PCB Magazine.
Escaping the commodity mindset and becoming more competitive and valuable starts with “us” and how we choose to treat our suppliers. Because of today’s market turbulence, we need suppliers who can perform. If it is all about price, we are going to have a supplier who doesn’t care about performance. If our suppliers do perform, then we must pay them fairly, based on their performance; we have to show them we care so that they are motivated to continue to perform. We need to keep it simple. Less is more.
We are being paid very well for higher technology, shorter lead times, and responding to customer emergencies. The costs are very high to us if our supplier fails. The incentive is great to change our supplier attitude. A perfect place to start with this new attitude of breaking the price game and moving into the performance game is with our laminate supplier. They need to be our true collaborative partner if we are all going to win.
When I entered the laminates industry in 1984, nobody was being rewarded for performance. It was all about making a few products as cheaply as possible. I remember watching operators stop a treater in the middle of a run, cut our three large pieces of material, weigh it, crunch it up, collect the resin dust, melt it on a hot plate, record the resin and cure percent, and then start up the treater almost an hour later. Several questions went through my mind: Is this data useful!? What was the detrimental impact caused by the test? How did the large holes screw up the tension across the web?
As I observed the insanity of it I asked, “Who wants this test?” I expected to hear that it was the operators who needed it, but it wasn’t. The operators were smart enough to know this data was useless. The answer I got back was, “It was what the customer wanted.”
The data was more or less a way to placate the customer, who only cared about price. But they wanted quality, too, and they didn’t trust the supplier’s ability to produce quality. So they forced this laminator to take process tests, like one I observed. The resulting problems and negative interactions wasted both time and energy.
It got to the point where this laminator hated its customers, and for good reason. It was like they had a gun pointed to their head and at any time the trigger could be pulled. If a customer’s quality manager demanded a SCAR, they would write one up, but there was no incentive to actually fix anything. If they were late with a delivery, they would explain it would never happen again, but it would. And the customers we were dealing with continued to tell us we had no reason to live. The business should just be shut down and we should collectively jump off a cliff. There was absolutely no motivation to become a better supplier at all.
When I returned to the laminates industry in 1996, I pushed to get a Measurex system into a facility. This was a machine that could measure the resin percent in real time across every inch of material. In turn, this would enable a process capable of providing better than +/- 1% resin, which was unheard of back then. As great as this technology was, it was almost impossible to justify. When our engineering team pushed for it, the answer we received was, “Our customers don’t pay for higher quality. They don’t pay us for performing better. They only care about price.” This attitude pervaded our supply chain at that time and fueled our demise.
Figure 1: Measurex data.
Figure 1 represents an example of what is achievable using a Measurex system for measuring resin % or cure %. Notice the trough on the right that persists through the rapid changeover of going from one product type to another. This happens when the warp is above 40. This data suggests a mechanical issue that can be fixed that otherwise would go unnoticed using the traditional method of stopping the treater and cutting out three samples.
Too many of us today, while trying to perform and get the revenue and profits we want, treat our laminate supplier as if we were living in George Orwell’s 1984. If we are going to truly take advantage of what is happening in our industry and win BIG, we have to change how we treat our laminate supplier now. For example, we have to replace our “supplier commodity manager” with a “supplier performance manager.” We need to remove the word “commodity” from our mindset and from our conversations and replace it with “performance,” and have performance-mindset conversations. You don’t want to choose a laminate supplier based solely on price ever again. It must be based on cost and performance.
Once you have chosen a laminate supplier, you want them to stay motivated. The motivation has to be simple and real. If the laminate supplier can’t perform they need to understand that it costs you money. When they do well they will be rewarded. If they don’t do well there will be consequences. One of the greatest mistakes we make with our laminate supplier is that we put them in a box: No matter what they do they aren’t good enough. If your supplier can’t win then eventually they are just going to give up. One of the ways we make it impossible for them to win is by requiring them to supply a boat load of data with hundreds of metrics. This is confusing, and often the metrics are in competition or in conflict with one another. This type of data is expensive and not useful to our strategic need to have a supplier that will perform when we need them to.
Simple is best. Less is more. What you need is a laminate supplier who can meet on-time delivery all of the time. Don’t even bother with a quality metric. Why? Because when you separate quality from on-time delivery you allow for these two metrics to be at odds with one another. The supplier thinks, “Do we slow everything down to improve the quality, or do we cut corners knowing we will have quality escapes in order to meet on-time deliveries?”
You need just one metric: On-time delivery. If your laminate supplier ships bad product then it doesn’t count as an on-time delivery. They are late. They failed to perform. They are subject to the consequences for failure. One thought, one result. Simplicity increases the probability for success. For example, a golfer has only one swing thought to connect all of the pieces required for a flowing swing. You want your laminate supplier to have only one thought in their head in order to align their people, process, methods, and controls to what you need. If you make your needs confusing, your supplier won’t be able to meet those needs.
Since missing on-time delivery can be financially devastating, you need to replace the quality mindset with a reliability mindset. A reliability mindset means you focus on the mean time for failure and the mean time it takes to recover from the failure. For example, if every quarter you have an average of three orders that are late from your laminate supplier, and it takes five days, on average, to recover and get the part shipped, then the meantime for failure is three orders late per quarter, or 90 days, and five days meantime for recovery or repair. You can now model this fact and estimate the costs of this kind of performance rather easily.
Stop the SCARs immediately. The word itself has a negative connotation and is extremely demotivating. The scars created when Captain Bligh tried to whip his crew into submission didn’t work for him and it will not work for you. You will just get information that is meant to temporarily placate you. Nothing real will improve. If you don’t believe me, try SCARs on your kids and watch what happens.
Stop with the requests for seeing or knowing average yield data. Averages are close to meaningless. What you need to know is what the performance is of every lot produced for the job they are having the most difficult time building for you. If you go back to my October Article, Don’t shoot the Messenger, Make Your Quality Meetings Fun, you want data that looks like Figure 2. The job you want to focus on is a low erratic yielding part, like Job E in this figure. Focusing on the problematic job will test your suppliers resolve to improve and be a great measure if sustainable improvement has occurred that deserves to be recognized and rewarded.
Stop forcing your laminate supplier to collect SPC information. Instead, like with their treaters, you want real-time measurement for resin percent and cure percent, such as what is possible using a Measurex system or equivalent. With so much data it is silly to think of trying to use SPC for a host of technical reasons, one of which is the problem of autocorrelation and the fact that averaging data hides the real information that operators and engineers truly need to improve the process.
Extremely large data sets that come out of real-time measurement equipment are like a picture from a high-megapixel camera. The higher the megapixel is the more resolution and detail you see. Could you imagine using SPC to process the photographic information from your camera and determine if a fine-line detail is really there? As you take more high-resolution pictures you can produce a movie. You can play back this movie and use it like a football coach would study a game. Technology is changing the future of how we analyze data! We need to embrace this fact and use it to our advantage.
By dumping SPC, using real-time data sets that provide a picture and movie playback that can be studied, and focusing on the problem jobs with a reliability mindset, your laminate supplier has a good chance of performing the way you need him to. In order to amplify this, you need to show you care. You need to visit your laminate supplier with the intent of wanting to celebrate his success, reward him with price for performance, and express your real concerns on his ability to perform based on data.
If you want to collect supplier data, have it centered on detecting if they are cutting corners with the process, whether the process is valid, and how they are using data to get better. This data needs to be collected on the most problematic jobs they are building for you. In this way, real problems will come to the surface quickly, and be exposed. Once exposed, you will know how committed your supplier will be in the future, and you can decide to keep them or find somebody else. Simple is best. Less is more.
On top of all this, recognize nothing is going to improve unless you care. Don’t visit your supplier or call them with the intent to criticize, but rather with the intent to encourage them to be better and to have honest open discussions in the spirit of being collaborative partners. Moreover, get to know the operation and the people directly and let them know you care about the work they do and that you notice what they do well. Nothing will produce a more engaged supplier than showing the people making your product that you care about how they do their job.
Gray McQuarrie is president of Grayrock & Associates, a team of experts dedicated to building collaborative team environments that make companies maximally effective. McQuarrie is the primary inventor of the patent, Compensation Model and Registration Simulation Apparatus for Manufacturing PCBs. For more information, e-mail McQuarrie at firstname.lastname@example.org or click here to learn more about how your company can win big through collaboration.
More Columns from Solving DAM ProblemsSolving DAM Problems: Solving a True DAM Problem
Solving Dam Problems: Why Removing Your Bottleneck is a Bad Idea
Change Your DAM Thinking: Do We Need to be More Innovative?
Solving DAM Problems: Can Scrap be Beaten? A Strategy for 2014 and Beyond
Solving Dam Problems: How to Compute Your Plant's DAM Capacity
Solving DAM Problems: Standards are DAM Important!
Solving DAM Problems: What's the DAM Problem with Scheduling?
Solving Your DAM Problems: Examine Your DAM WIP