Articles by CBAssociates

Published, Newspapers and Technology, December 2001

Data + Analysis = Information

We must reduce waste, is a demand heard often in a manufacturing operation. Many times this demand is made after the publisher, GM or the CFO attends a meeting and hears that another newspaper produces a lower percentage of waste than his or her newspaper.

I cannot count the number of times I have been asked what percentage of waste a newspaper should run. Right up there in the question hall of fame is asking why one newspaper's waste is so much higher than another's.

I could have retired years ago if I had a $1,000 for every time someone asked why USA TODAY produced such high waste levels.

When we started USA TODAY, our goal was to make the first copy to the reader as good as the last. We did not know how much waste would be required to meet that criterion, but we knew what our quality goal needed to be.

It was not too long after start-up that we began to understand how to minimize waste and to know what waste levels were necessary to ensure our quality goal. As our knowledge developed, our replies to those who asked why we had high waste changed.

The replies changed because we knew why the waste was necessary. We understood the operation. And more important, we were in control. With this knowledge we became confident that few, if any, organizations could print the same circulation, pages, and quality night in and night out with the same number of waste copies.

So when people ask why waste is high, the answer is: "It is not high; it is appropriate for the volume of color, number of pages, circulation and the fact that we don't send out marginal copies."

That said, it's important to keep in mind that whatever level of waste a newspaper runs, there is always opportunity for improvement. The improvement might be gained through better procedures, training or additional equipment.

Most newspapers, not all, but most, are quite willing to send out substandard copies. Pasters are not pulled and they don't want the start-up copies to go to the advertisers. So when we discuss quality goals and start-up copies with organizations, we suggest that one option is to produce the first copy is as good as the last. Invariably one of the executives will ask the anticipated question: "How much would that cost?"

That question cannot be answered unless operational data and an understanding of the equipment exist. Most of the time there is no data, let alone the analysis to gain an understanding of the operation that would enable us to determine if the waste volume is appropriate to the pages, circulation and quality expectations.

Typically, useful manufacturing data is lacking. Operations may collect numbers, but often, the information they collect is too broad and doesn't go deep enough to produce an accurate picture of the factors that affect the results.

For example, The Total Waste Chart does a good job of showing what is happening with total waste and in this case it is pretty flatÑnot improving, but certainly not getting worse. No problems, no worries. It is typically what the money folks and unfortunately, too many operations folks watch. But it doesn't give an accurate picture.

For a better picture, the Waste by Categories chart is the one to watch. If this chart were the manufacturing managers' main tool, they would spot pretty quickly that big problems occurred with cold start-up waste. The total waste is only stable because the replates stopped.

This detail is necessary to keep the waste in check and to make improvements. The old math expression, that the total is the sum of the parts, is appropriate. Operations improvement is all about working on the parts and seeing the effect on the total.

An easy way to get in trouble is to look only at an operation as the CFO would look at it. The two charts are a good example of why different data is needed for different purposes.

And it's important to keep in mind that not all data must be analyzed every day. In fact some information would yield very little benefit for the effort. An example of data that should not be analyzed daily is web breaks by unit and reel. This should be analyzed weekly if the reels are old and analyzed monthly if the reels are in good shape. Of course, the maintenance folks should check the log daily to see which reels had trouble or breaks.

Operations improvement depends as much on understanding the current operation and its trends as on any other factor. The analysis that is used will influence how quickly managers realize that a problem exists.

The type of charts that are used will control the focus. Of course, over time, the focus should change to emphasize the areas that should be improved. To understand the components of cold start-up waste requires going to the next level of detail. Drilling down into the problem will create additional opportunities to create data and to let the data talk by using appropriate and good charting analysis techniques.

The start of any good analysis is to pose an appropriate and useful question. The better the questions, the better the analysis, and the more likely a manager will be in control of the operation.

[HOME]