Invest in IT know-how
I am often surprised at the lack of IT-knowledge in some companies. While this may work as long as you can rely on vendors to each take care of their own solution which they implemented at your company, things get different when you start crossing the boundaries and implement systems that integrate all of those processes.
It may be wise to invest part of the savings that automation brings in the training of IT-people or in enhancing the IT-skills of some key prepress operators. Once automated processes are up and running, you need someone who understands how it all ties together and where to look if something goes wrong.
JDF is far from being a self-adapting technology. As soon as new equipment gets added or requirements change, it is often more cost-efficient to have the in-house skills to adapt the automated processes instead of having to rely on expensive consultants or others who need a lot of (chargeable) time to understand your set-up.
Is the flow what you expect it to be?
A company decides to automate prepress by sending job related data from the MIS system to prepress. While rolling out the system, they suddenly realize that over 30% of their jobs don’t even exist in the MIS system. For repeat orders such as magazines or newsletters, no job data are entered in the MIS system because prepress always initiated those jobs. The same was true for loyal customers or jobs done with a fixed tariff. Prepress would process pages and make plates without there ever being a job definition in the MIS system. Suddenly management realizes that they wanted to automate a single process but it turns out they run two or more processes alongside each other.
To avoid such situations, map the internal processes before trying to automate them. A surprising number of companies have never really thought about the way they do things. When things get changed reality suddenly catches up with them.
Standardization is the key
JDF works best for standardized processes. It cannot (yet?) cope with situations where each job is done in a completely different fashion. Many people think that the jobs they process or the customers they work for make it impossible to standardize processes. Sometimes it takes an outsider to take a helicopter view and see how a few changes are all it takes to get rid of a lot of production variables.
No better recipe for failure than a company that isn’t capable of clearly communicating their goals talking to a vendor who is all too eager to make some vague promises in order to get the deal.
It often isn’t easy to define the exact solution that you are looking for when you aren’t sure about technical limitations or a specific solution’s scope. It is however fairly easy to clearly define your starting point:
- Providing vendors with a copy of the paper job form that is currently used gives them a clear idea of the data that are currently exchanged between departments.
- A basic diagram of the company structure with the names of key people who will be involved in automation is a useful reference, even for internal use.
- Don’t talk about integrating with finishing equipment but talk about integrating with model XXX of vendor YYY with software revision ZZZ. A list of hard- and software that is currently in use or that will be used in the future is easy to assemble and can avoid misunderstandings about the exact scope of the project.
- A flow diagram that details how jobs get processed, which job data are transferred, who gets involved, and which systems store or process which data takes some time to create but it can also be a useful tool in negotiating with vendors.
Any vendors worth its salt will at some point in the process ask for the above data. If you are negotiating with multiple vendors regarding a JDF automation project and you have made your homework, it may not be a bad idea to withhold these data a little and see who asks the right questions.
One thought on “How to implement JDF”
how can i fine T3 fonts?