Is Zero Rework Possible?

REWORK. Man, that’s one word I really hate.

Now, I love working in the printing business just as you do; if you didn’t, you’d be reading Car and Driver or O, The Oprah Magazine instead of In-Plant Graphics. But we both know that printing can be a tough way to make a living. Fussy customers picking you to death about their project and your prices; files that won’t RIP correctly to your devices; folds that crack even after they’ve been scored…you know what I’m talking about. It can be like hanging off the edge of a cliff by your fingertips. And, then, when you think the pain can’t get any worse, it happens: rework. And it’s like someone just ground their boot heels squarely on the tops of your aching fingers.

Of course a rework costs your in-plant time and money, but that’s just the beginning. Your carefully scheduled workflow has been disrupted. Will one or more of your customers’ jobs now be late? And if so, which customer wins and which one loses? What kind of hit will this have on your in-plant’s goodwill?

And if you’re like me, it’s still not over. When I cause a rework, I feel a dull ache all over my body, and it doesn’t go away when I walk out the door. My wife, the lovely Maureen, will ask how my day was, and I’ll mumble a few things, but I just can’t tell her about the rework. Its pain is still too fresh in my mind. She’ll hear about it the next day, when the ache has subsided enough that I can talk about it.

ANY REWORK IS TOO MUCH

I’ll wager that your reaction to rework is similar to mine, but whether it is or not, I think we can agree that the only acceptable level of rework for our in-plants is zero. As long as one has rework—any rework—there’s still work to be done.

But is zero rework really possible? Can we frail and imperfect humans pull off some system or methodology that can make that happen? Regrettably, in my in-plant that answer so far has to be “no.” Also, I’ve corresponded with many in-plant managers about rework; I know of one exceptional manager who has wrung so much rework out of his in-plant that it is now measured in parts per million, but even in his case, the answer is ultimately no.

Bottom line: I don’t know if zero rework is possible, but to accept anything less is to accept defeat. I’m not ready to make that concession, and if you’re not either, let me share with you a useful tool that has helped me reduce my in-plant’s rework rate and may help you reduce yours: Pareto Analysis.

WHAT IS PARETO ANALYSIS?

Let’s get the obvious questions out of the way: it’s pronounced PREE-toe, and while it’s named after Vilfredo Pareto, a 19th century Italian economist, it was actually invented by Dr. Joseph Juran, an American quality expert who is still with us today.

Pareto discovered that 80 percent of the wealth was held by only 20 percent of the people, and examples of his 80/20 rule abound today. Juran took that rule and applied it to quality: 20 percent of the causes of rework are responsible for 80 percent of the cost of rework. This 20 percent is the vital few—the problems that, if solved, will make the greatest impact. It’s really like organized common sense—easy to understand, implement and benefit from. Here’s how it’s done.

THE PARETO DIAGRAM

1| Identify and record the cause of every rework as it happens. (If you’re not already doing this, you should begin even if you have no intention of doing Pareto Analysis. Anything that is measured improves; no ifs, ands or buts).

2|Estimate the cost of each rework.

3| Note other factors of urgency that may give a cause of rework more weight than its dollar amount alone would indicate. One example of this would be a cause of rework that is usually caught by the customer only when the project is in his/her hands vs. a cause of rework that is usually caught and corrected by in-plant personnel before the project leaves the shop.

4| Estimate the cost of each factor of urgency. Most people, including me, find a proper estimation of this value too tough to find, but in the case of a rework discovered by my customer instead of by my people, I assign a triple value.

5| Rank these items in order from most costly to least costly using graphs or bar charts. This results in the Pareto Diagram.

Since we’re all printers here, the advantage of a using a Pareto Diagram instead of a ranked list should be obvious to us: a picture is worth a thousand words. Take a look at Exhibit A (right), which has both a ranked list and a diagram, and compare for yourself.

Exhibit A is last year’s Pareto Diagram from my in-plant showing $30,718 worth of rework spread out over some pretty typical rework causes. You probably see many of the same ones in your in-plant. All rework must be shown in your Pareto Diagram, as well as a cumulative percentage that quickly shows how big a problem it really is.

One category, “Shipped short,” is one of those that had a special factor of urgency because it is a rework that is always caught by the customer. Its actual cost was $593, but because of its urgency it was assigned a cost of $1,779, and you’ll see that it is listed as “Shipped short x 3” in recognition of its special place in my in-plant’s diagram.

But special place or not, in this case “Shipped short x 3” was not the problem to be immediately attacked, even if the reason it’s not higher is only because very few customers count their jobs. Here’s why: we only have so much time, money and resources. It’s not feasible to take on all causes of rework simultaneously and do the task justice. So if we can only tackle one or maybe two problems at a time, doesn’t it make sense to look at what costs our in-plant $12,687 a year instead of what costs it $1,779 per year?

SHARPEN THE FOCUS

Even though we concluded that we need to work on “not following job ticket instructions” instead of other rework causes, we still needed a little more information before proceeding. “Not following job ticket instructions” encompasses a broad range of rework causes, so the next step was determining which causes were costing us big bucks. The answer was, of course, another Pareto Diagram, this one a sub-diagram of the main diagram showing specific examples of failure to follow job ticket instructions. It’s shown in Exhibit B.

If you decide to do a Pareto Analysis in your in-plant, I promise that you’ll be surprised. For example, our analysis revealed that much of our rework was coming from one particular DocuTech operator—a surprise because she had been a reliable associate for many years. But when we sat down to talk with her about it, it was discovered she couldn’t read the job ticket well because her vision wasn’t what it used to be. One new pair of glasses later, her performance is back up to its previous level.

THE TRIVIAL MANY

The 80 percent of rework causes that only account for 20 percent of the rework costs are often called “the trivial many” for self-evident reasons. But that points out what I don’t like about Pareto Analysis: it’s like an acknowledgement that rework can never be eliminated, only reduced, and I’m not ready to accept that yet. But Pareto Analysis is a remarkably powerful tool in rework reduction, and we will continue to use it in our in-plant until we hit zero rework.

Make it a grand day. IPG

Richard Griffin has been a flyboy, a press helper, a pressman, a shift supervisor, a pressroom manager and a plant manager. He’s been a press mechanic and a press erector. He also sold printing, offset presses and Xerox machines, and founded Press Statistics, a consultancy for offset press acquisition and printing operations. He is currently the director of Campus Printing at Central Piedmont Community College. Contact him at (704) 330-6606 or richard.griffin@cpcc.edu