Many in print consider themselves to have an automated workflow. Few actually have one.
Perhaps the incoming files are submitted directly into a prepress workflow, the PDF checked and preflighted automatically, passing into a production queue and signalling this to the MIS. The MIS will sequence the creation of plates and which press receives which job next, and information about makeready and press performance statistics are again gathered and submitted to the MIS. Repeat for the finishing area and into ordering delivery times.
In the fully automated workflow, there will be no need for a paper job ticket, no need for any operator to enter the parameters of a job. Heidelberg UK calls its automated workflow concept Push to Stop because the start up and the entire process to start printing can take place without intervention. Others will take the job set up to the pass sheet, leaving the operator to press to go.
There are not enough of these presses in the field to reflect the percentage of printers claiming to be fully automated. And this is without robotics to move plates from the platesetter or paper from press to folder, without the data standards that communicate seamlessly from customer to printer and back to customer, and without the data being sent to equipment manufacturers for remote monitoring of performance and the need for maintenance. All this comes under the broad heading of Industry 4.0.
Unquestionably, some printers have implemented highly automated systems, particularly in prepress, where a job can progress from file submission to exposed plate with no intervention required, except when the system flags up a problem, as is the case with, perhaps, 5% of jobs.
Some printers have extended a full JDF implementation from a web portal for job submission through to connection to couriers or transport companies. Likewise some have built, or used integration companies to build workflows around Enfocus Switch. Some have implemented Prinect workflows to connect their Heidelberg supplied equipment, or Connex for Muller Martini. And so on.
These are islands of automation, paving the way for automation of some elements of production, but needing bridges to be built to link the islands. And it has not been easy to construct these bridges. JDF was supposed to make this easier, but in practice each implementation has needed to be bespoke. And JDF has been process focused rather than product focused, which constrains the flexibility of any artificial intelligence of machine learning that comes along. This year CIP3, the curator of the JDF standard, released JDF 1.7 and XJDF 2.1. The latter version is the slimmed down version of JDF intended to make implementations quicker and easier, but with little feedback as yet on user experiences.
Then comes artificial intelligence, emerging from cloud hosted applications that can share anonymised data about hundreds of thousands of jobs and uncover optimised ways of handling similar jobs. In an ideal world it will be like drawing on the experience of the world’s most experienced press operator who recalls the settings needed to print on that awkward stock or with those challenging inks, to cut out the waste of materials and time.
However, in order to capture the promise of AI, a common data structure is needed so that jobs can be compared and settings shared. JDF may not be up to the task, focused as it is on the process rather than the product. The spin off XJDF version holds more promise, but currently demand is low.
Enter companies from outside the industry. The need to describe a job in a way that automates printing in a process agnostic way is the promise of the formats that Zaikio has described and also how Color Concepts has been able to create a data structure for display graphics substrates.
A generation ago a number of suppliers proposed JDF, quickly finding many more companies joining in. However, like any committee, the more members that are involved the slower it can be to achieve a consensus.
There is no question that automation is needed. Already short print runs, sometimes printed digitally, at other times more efficient on a litho press, are going to become shorter. This means that printers need to attract an ever increasing number of jobs and must process these without increasing their costs. And they only way to do this is to replace human touch points with machine made decisions.
The speed of turnaround that is necessary is another factor that will force printers to adopt automation. There is simply not the time for human decision making. Online trade printers, which have frequently built their own automated information and production systems, can already spin a job overnight, making money by batching small format jobs on the same large format sheet. In order to survive commercial printers will need to match or better this speed of response across a wider range of products than web to print can currently cope with.
HP‘s PrintOS has gained the support of many add on apps, but remans a closed environment for users of HP presses only.
