American Printer's mission is to be the most reliable and authoritative source of information on integrating tomorrow's technology with today's management.

The shocking truth about scheduling

Oct 1, 2003 12:00 AM


         Subscribe in NewsGator Online   Subscribe in Bloglines

Few people are as passionate about scheduling as Udi Arieli, Printcafe's (Pittsburgh) director of product management for Web and CIM products. For 20 years, Arieli has argued that scheduling is not merely an activity that ensures on-time delivery, but a scientific tool that ultimately impacts a company's profitability, customer satisfaction and competitiveness. Indeed, the exec predicts that printing companies won't survive without dynamic scheduling systems that optimize the entire manufacturing process.

Inspired by Eliyahu Goldratt's “The Goal,” a business novel set in a manufacturing plant, in the mid-1980s Arieli developed the Theory of Global Optimization (TGO) specifically for the printing industry. TGO defines printing as a manufacturing operation comprising interdependent links. “Only a few constraints control the throughput, on-time delivery and cost of the entire printing operation,” explains Arieli. “If you manage the constraints, you can optimize your company's performance.”

What's a constraint?

In TGO parlance, a constraint occurs when a cost center (resource) has so many jobs lined up for it that it can't finish all the work on time — some jobs will eventually be late. For example, if the press is producing so many jobs that there will not be enough time for them all to go through the bindery and still be delivered on time, the bindery is the biggest constraint. To relieve this constraint, the press should produce fewer jobs, or produce jobs that do not need the bindery, or the bindery's capacity should be increased.

Poor management decisions also can result in constraints. “Suppose you cut overtime to cut costs,” says Arieli. “If a cost center is already constrained, you're only making it worse — the constraint becomes even greater while throughput is even less. People often complain about too much or too little capacity, but they forget that management can create a lot of problems.”

Initially, Arieli was either ridiculed or ignored. Unlike competitors, he didn't demonstrate the scheduling capabilities of a management information system (MIS) package when he spoke at industry conferences. “I was the only one talking about TGO,” recalls the exec. “Everyone would look at me like I was crazy — they had come to hear about scheduling — why was I talking about theories?”

By 2000, Arieli was no longer a voice in the wilderness. Arieli's company, AHP, merged with Printcafe, giving him the resources to put his theories into practice. With a team of developers, he helped create PrintFlow, a scheduling module that organizes the sequence of steps required to efficiently produce a job. Currently, it can be integrated with Printcafe's Hagen, PSI and Logic MISs. Built on the principles of TGO, PrintFlow combines weak-link analysis with dynamic scheduling. “It identifies ever-changing weak links and lets you resolve problems before they occur,” explains Arieli. “It shows you how a single change will affect the whole plant. PrintFlow chooses the optimum machine or cost center for a job, taking into consideration all other jobs, and sequences the jobs on every cost center to optimize the entire manufacturing process.”

Change or die

While acknowledging that the Printcafe merger enabled him to reach a broader audience, the exec also says changing times have made printers more receptive to TGO.

“The time is right,” relates Arieli. “For years, I talked about negative industry trends such as shorter print runs and lead times, lower prices and higher costs, the need to reduce waste levels and makeready times, and the cost of continuous improvement. You can't keep running your business the same old way — your biggest competitor is not the print shop next door. It's home computers, the Internet and other media. Unless the printing industry improves itself — by cutting costs, charging less and still making a profit — it won't exist for long. We have to be faster, better and cheaper.”

Although computer-integrated manufacturing (CIM) and other equipment-automation advances can help printers operate more efficiently, Arieli contends that TGO is the missing profitability link, since it provides a comprehensive look at the entire production from start to finish. Adapting this global view, however, may require printers to re-examine some long-standing beliefs and practices, including the following.

Bottlenecks aren't always bad

“A queue of jobs is not such a tragedy,” says Arieli. “Because all printing jobs are sold before they are manufactured, [it's not a problem] to have a few jobs waiting in a queue in front of a cost center — internal inventory is not an issue. The bigger problem is makeready — preparing a cost center between jobs… A bottleneck can be good for a company because it allows you to group similar jobs together, reducing makereadies and waste.”

The challenge, says the exec, is finding the correct balance between growing a bottleneck for optimum job-grouping purposes, but without holding jobs in a queue so long that they, or any other job, can't be delivered on time. “You have to find a compromise between the size of a bottleneck and job due dates,” explains Arieli. “But everything in PrintFlow is about compromises. Using what-if scenarios, you can give a job a higher priority and see the impact. If you need a job tomorrow, PrintFlow [might indicate] it can be done, but will take three hours of additional setup. Now it's up to you — do you really need the job tomorrow?”

Discounts should be based on lack of constraint rather than job quantity

Many printers' discounts are based on quantity or “best customer” status. Arieli argues discounts should be awarded based on time/constraint impact. “If you give me a job that goes to my busiest cost center, I don't want to give you any discount, even if you order a million,” submits Arieli. “But if you give me a job that goes to another cost center that's not a constraint, I'd be happy to give you a discount, because my throughput and profitability are higher.”

The exec cautions that merely increasing volume won't necessarily translate into increased throughput and profitability. “Even if you say to your salespeople, ‘Let's lower prices so we get more jobs,’ you already have constraints in your organization,” he says. “Everyone does. By just giving discounts across the board, you'll get a lot of the wrong jobs — jobs that are difficult to manufacture, that might require the same press or binding line. Getting all of this work won't help if you have to send it out.”

Similarly, Arieli claims that the customer may always be right, but shouldn't necessarily come first. No single job, cost center or customer should prevail over the needs of the plant at large.

Throughput is more important than job costing

Typically, most printers wouldn't switch a job from a two- to five-color press because the five-color press is more expensive to operate. “Even though the job cost is going to look bad, provided the five-color press isn't a constraint, you should move it,” asserts Arieli. “Say the job should take an hour — it's $200 on the two-color press and $500 on the five-color. That doesn't look good [on paper], but at the time, the five-color press wasn't a constraint. The capacity was available and you should have used it. On the other hand, the two-color press was a constraint. You had a lot of jobs waiting in line.”

The exec also cites this example to explain why the whole company — not just management — must understand TGO. Most employees have been taught not to make this kind of move, so you'll have to explain the big picture. “The more employees know, the better the company will run,” says Arieli.

A movie, not a slide show

Printers using PrintFlow range in size from 10 employees to more than 2,000. “We have one company that specializes in forms and business cards,” notes the exec. “Even a small company needs TGO to optimize its operation. Without it, they're leaving money on the table… And larger printers have the same amount of unpredictability as smaller ones. They might know they are going to print certain jobs, but there are always some surprises. Even if you're printing Time, every week is different. You can have different materals, page sizes and equipment problems.”

Ditch the scheduling board

“A scheduling board is a huge negative, because everything is constantly changing,” declares Arieli. “You need a real-time tool. A scheduling board is a snapshot of time — once in the morning and once in the afternoon, if you're lucky. You need a movie, not a slide show.”