A programming “knowledge library” is a collection of the tool, parameter and
toolpath choices associated with workpiece features that occur repeatedly in
parts throughout the shop. For example, a particular hole may always be machined
with the same series of spot drill, drill and tap. Or, a particular programmer
may prefer to always mill certain flats using the same series of rough and
finish tools. In cases such as these, CAM software can offer the ability to
store the preferences in a library so the programmer can easily call up the
relevant machining routines whenever the recurring features appear.
Of course, storing programming knowledge is nothing new. Programmers have
done it all along. Using a “crib book” or some similar notebook (often covered
in Post-it notes), a programmer can record what tools and strategies worked once
with certain materials and conditions to avoid having to rethink the process
when similar circumstances arise. In fact, according to John Callen, frequent
use of crib books is a clue that a particular shop ought to convert to a
CAM-based library.
Mr. Callen is marketing vice president for
Gibbs and Associates,
the Moorpark, California, creator of GibbsCAM. Beyond programming efficiency, he
says the far greater value of a knowledge library is the chance to standardize a
shop’s programming “knowledge” by having programmers draw upon a single library
together. Referencing standardized processes ensures that programmers machine
the same features in the same way, and in particular, that they use the same
tooling choices. The shop can thus make machining cycles more consistent and
predictable, and it can reduce the number of different tools it has to order and
track.
Knowledge libraries also address a potentially greater source of uncertainty
that weighs on many shops. Namely, what happens when an experienced programmer
resigns or retires? A CAM-based library allows a shop to define and maintain its
preferences independent of particular employees, and keep those preferences in
play as personnel change. The library also lets new programmers ramp up into an
effective role for the shop more quickly.
 |
|
The countersunk holes here involve multiple holemaking steps. Grouping the
sequence together saves the programmer from having to remember multiple
processes and tools per hole and having to call up each one
separately. |
Knowledge Politics
Mr. Callen says GibbsCAM
includes various features that relate to knowledge libraries. A “Save Process”
feature allows the library to be populated through the programmer’s regular
work. If the programmer makes a choice that deserves to be repeated for similar
features, then Save Process can save this choice as a library entry.
Complementing this is the “Multiple Process Programming” feature, which expands
the definition of a process by allowing a sequence of tools and steps to be
grouped as a unit and applied together. One advantage of this latter feature
relates to design updates, because rough and finish passes grouped in this way
are revised together if the underlying geometry changes. However, another
advantage relates to the knowledge library, because a series of tools and
steps—the aforementioned spot/drill/tap, for example—can be saved and applied as
a single library entry.
The knowledge library itself is based on a file structure. The user controls
the organization of the library by creating directories. A subtle advantage of
such a structure relates to politics, Mr. Callen says. If a shop sees the value
of standardizing programming, but has multiple long-time programmers with
separate opinions about the best ways to machine, then does the shop have to
choose among the different programmers’ methods?
Maybe not. The file structure permits a library that is not entirely
standardized. Specifically, for long-time programmers named Bob and Joe, the
file structure can allow for directories named “Bob’s Way” and “Joe’s Way.” The
shop can then standardize within the area where Bob and Joe agree—a level of
standardization that still stands to be significantly better than no
standardization at all. Then, when Bob or Joe retires, the shop can look to that
programmer’s preferences to decide which processes ought to be kept as part of
the shop-wide practices.
Intellectual Capital
One darker aspect of
knowledge libraries can’t be ignored—the fact that the library can be copied.
All knowledge tools have this weakness. If the purpose of the shop’s library is
simply to standardize programming, then this possibility may not be a concern,
but what about the shop whose library includes proprietary best practices? A CD
or memory stick might let someone carry this library away.
This danger is
still no reason to avoid using a knowledge library. After all, the alternative
of relying on individual programming knowledge simply impedes the spread of best
practices—and it doesn’t leave the shop any more secure.
Mr. Callen predicts that as more shops become specialists in particular
markets, and as more of these shops standardize their programming, the challenge
of securing programming knowledge will become more apparent. A technological
solution is likely to present itself, he says. If it doesn’t, then it is likely
that innovative shops will find some procedural solution that helps make
critical house knowledge more difficult to transfer to other settings.
MMS Online is a trademark of Gardner
Publications, Inc, copyright 1997-2008.
MMS Online and all contents are
properties of Gardner Publications,
Inc.
All Rights Reserved.
Del.icio.us
Digg
Google
Y! MyWeb
Slashot
Technorati
reddit
