COMPUTER-INTEGRATED MANUFACTURING -- Strategies for a "Smart Fa

BY LARRY WARTER

In order to survive in an ever-changing industry, today's printer will have to become the "smart factory" of tomorrow. The "smart factory" concept integrates all of the internal printing plant processes with information flows linked to the outside.

Many of the criteria for being a true "smart factory" are related to standards (including accredited standards, as well as industry specifications), which address aspects of the process that everyone agrees should be standardized and, at the same time, leave opportunities for individual companies to differentiate themselves in other parts of the process.

One of the most crucial steps in becoming a "smart factory" is to make the press a connectable process. The printing process must have set limits and tolerances in order to ensure that the press becomes a preset reproduction machine. Image evaluation and adjustment should be relegated to the creative process upstream, allowing the press to be treated as a known output. This is not a new concept.

Our industry has been talking about this for quite some time; we've just never been able to achieve it. A lot has had to happen in order to make this dream work, but the pieces are finally in place for our industry to begin talking about becoming true "smart factories." Three steps are necessary in order to bring about this change: the press must be controlled, calibrated and connected.

The printing process needs to be controlled in three different ways. First, the press products must be standardized. There are ISO (International Standards Organization) standards for the size, thickness and squareness of plates. There is also an ISO standard that was the first industry-wide consensus on how to test for all performance properties of blankets. And, there is a whole series of standards that cover the rheology (printability factors) of inks and how to test them. The most basic step, then, in becoming a "smart factory" is to use products that have been tested and certified.

Secondly, there is an entire series of measurement standards that apply to printing. These were among the first standards developed by CGATS (Committee on Graphic Arts Technical Standards), and they can serve as tutorials for how to use densitometers to control the printing process and spectrophotometers to match inks and, ultimately, images. Much of this early theory has evolved into practical, closed-loop color control for sheetfed and, now, webfed presses.

The machines are sophisticated, but they are measuring the same basic properties documented in the standards as the best way to control a press. Measuring the process is essential to a "smart factory."

Thirdly, there is a series of standards that govern the process itself. The proper way to measure ink color and transparency "in the can" is defined in ISO 2846. There is also an ISO standard for platemaking, which helps define the reproduction argument (linear vs. bowed) that we are having with computer-to-plate. The "smart factory" of the future will properly define processes and establish shop standards based on good theory.

In an ideal world, each printed job should, at the very least, meet the customer's expectation. In the real world, however, the printing variables noted above prevent printers and their customers from accurately and reliably predicting how the image will appear on-press.

Once the printing process is controlled, it can then be calibrated, and the output can be documented and related to the customer's needs using the CGATS IT8 7/3 target and the ISO SCID images. And it is in this area of the process where most of the industry's groups are involved.

There are already a multitude of standards, specifications and general guidelines that cover all of the various printing markets and products. On the theoretical side, ISO has a family of standards that define the measurement targets for offset (commercial and publication), newsprint, flexo, screen printing, gravure packaging and digital printing. These standards give representative aim points and process tolerance limits for each of these types of printing.

The process tolerance limits are worth noting since this is one of the few times that the graphic arts industry has attempted to define process variation. A "smart factory" will need to know this type of information in order to set realistic expectations for its customers.

ISO has also recently revised the viewing standard for the printing industry. The previous standard allowed too much variation and failed to define conditions of use. In a subjective industry like printing, the "smart factory" will have to be able to relate process control conditions to customer needs. If this is not done, printers will see their profits decrease as time and money will be wasted in the makeready stage.

On the practical side, most industry markets have evolved user groups to represent them. Led by SWOP for the publication industry more than 20 years ago, these groups have established specifications, guidelines and best practices that define their different industries.

More recently, they have been working with the accredited standards groups to establish "reference printing conditions" that will better define their process in a color management world and set the stage for linking the "smart factory" of the future directly to its customers.

The theory is simple: Printing quality is largely dependent on the gamut or visual contrast of the image (which usually correlates with the grade of the paper, whether it's coated or uncoated, etc.). If the total spectrum of printing conditions could be grouped into a limited number of image quality reproduction options (based on their gamuts), then each commercial printing job could be identified as belonging to one of these groups in order to predict what it would look like on-press, greatly minimizing the communication problems between the printer and customer.

Simply Satisfying

If the gamuts are defined properly, a limited number of these reference conditions (5-6) should satisfy all market needs with all papers and printing conditions.

Once these processes are complete, printers, print buyers and creatives will be able to match the established characterization and, therefore, also accurately match one another, as well as the final outcome on-press. While this may seem somewhat daunting, it is something that the "smart factory" of the future will need to make a reality.

SWOP printed the first referenced printing condition as documented in the revised version of its industry specification (2001, the Digital SWOP). It has recently begun to certify vendor proofing systems as having the potential to match SWOP if they pass a rigorous visual test. SNAP also republished its Specification for Newsprint Advertising Printing in 2001.

The GAA gravure association has printed a reference printing condition for packaging gravure and submitted that to CGATS. For publication gravure, it uses SWOP and was the first example of how color management could be used to match two very different printing processes.

GRACoL is responsible for commercial offset. Besides revising its process control guidelines each year, it has printed referenced printing conditions for paper grades 1, 2 and 3. Grades 1 and 2 are being submitted to CGATS, although they will probably result in color-managed permutations of the same printing condition because they have the same density aim points and, therefore, the same gamuts. Grade 3 appears to be a color-managed variation of SWOP, but more testing is needed before it can be determined that it is representative of the industry and submitted to CGATS.

For its part, CGATS has documented SWOP Publication proofing as a technical report: TR 100, and this characterization should be used as the basis for all color management by any "smart factory" in the publication industry. TR 002 for SNAP, TR 003 for gravure and TR 004 for GRACoL Commercial should follow shortly.

These referenced printing conditions complete the calibration step, and all that is left is to connect the press by transmitting this information. As mentioned, the ICC (International Color Consortium) is responsible for the color management portion of the connection process.

ICC has been a "work in progress" for too long and, at the end of 2000, some major changes were made to the standard, which should have greatly improved interoperability between systems by now. However, there is still a need for ICC to address a "graphic arts workflow," which will allow the sender to indicate how the image reproduction should be done.

While this remains difficult with the current ICC system, there is a major effort under way within the ICC to understand the need and develop a solution.

As part of this connection process, color management will now be handled to the chosen file format. At present, ISO is working with Adobe and a number of vendors and users to adapt and limit the PDF standard for use in the printing industry.

There are two optional standards for "blind transfer," which means that the file is complete and unambiguous, and that the receiver can open the file and print the desired result without having to consult with the sender. PDF/X1/1a is for blind transfer of CMYK data, while PDF/X3 is for blind transfer of color-managed data. To accomplish this, the ISO committee has added output profiles to the appropriate PDF data dictionary. This gets around the ICC prohibition against sending two profiles within the color management data. Both PDF/X1a and X3 are out for ballot as ISO standards and should be ready in 2002.

Once the color management process and file format are set, there are just three additional questions that need to be answered before the press is connected. First, we need a color measurement definition of a match. This would include a limited number (5-10) of color patches that are sensitive to variation, along with some sort of weighting factor.

When two images are compared, the patches would be measured with a spectrophotometer and the weighting factors applied. If the total difference was less than a prescribed amount, the images would be declared a match. This assumes that this process would reflect visual experience. Second, we need to standardize the use of a "monitor as a proof," which will define what monitors can and cannot do.

Finally, we need to better address how this connected press fits with the new Job Definition Format from CIP4. CGATS is addressing the first two issues, and the third should evolve out of the PDF/X work, as many of the same companies are also involved with CIP4.

Once these processes have been accomplished, the press will be controlled, calibrated and connected—and the "smart factory" will be well on its way toward a bright future.

About the Author

Larry Warter is the director of new business development for Fujifilm Graphic Systems, a division of Fuji Photo Film U.S.A. During his career in the graphic arts industry, he has been an active volunteer in several industry associations. Warter and several co-authors completed new SWOP and GRACoL specifications for printers in 2001, and he is continually addressing other topics that concern the graphic arts industry, such as trends in digital proofing and the potential for industry-wide manufacturing standards.