Lack of Interoperability Continues to Vex Manufacturing Industry
The manufacturing industry is abuzz with all things product lifecycle management. PLM has become a catchall term to describe the vast array of product creation strategies in place at organizations across the globe. In its simplest form, PLM allows manufacturers to create, manage, share, store, and maintain product data in an integrated IT environment. PLM systems provide a backbone or infrastructure to support product design, manufacturing, maintenance and, ultimately, product retirement. It's no secret that manufacturing companies struggle with the same competitive pressures, regardless of the consumers they serve: the need to innovate more quickly, create more product variants than ever before, reduce cost to the absolute minimum, and create IT infrastructures that foster collaboration with partners, suppliers, and customers. PLM technology, it would seem, has matured just in the nick of time. Or maybe not.
True, PLM has created never-before-realized
gains in manufacturing productivity: fewer physical prototypes are
built now than ever before (in the automotive industry, building a
single physical prototype can cost upwards of $500,000). Collaboration
with suppliers has reduced time-to-market in many verticals, and
advanced data management capabilities mean that product variation or
made-to-order design and manufacture strategies are no longer just
futuristic notions. Advances in digital technology have made it
possible to carry the mockup of an entire aircraft or automobile on a
laptop to a design review with clients. Yet, with all of these
technological advances, one roadblock perpetually slows progress and
diminishes the spoils of PLM's technological achievement:
Interoperability, or the lack thereof, between applications means that
much PLM's advantage is left on the table.Inadequate Interoperability
Indeed, in 1999 the National Institute of Standards and Technology
estimated that inadequate interoperability cost the automotive supply
chain $1 billion per year. Move that forward to 2005, include the
proliferation of Internet-based and other technologies not available in
1999, and then extend this to include many other industrial sectors,
and the costs of insufficient interoperability are likely in the tens
of billions of dollars per year.
Why does a problem identified as significant more than five years
ago continue to bedevil the manufacturing verticals? How are software
suppliers addressing this challenge? What are manufacturers doing to
cope? There are no easy answers to these questions-nor is there an easy
answer to any one interoperability problem. This problem is so
widespread that many products are not even backward compatible with
themselves, let alone other systems. According to Daratech, Inc.
(Cambridge, MA)-an information technology market research and
technology assessment firm-original equipment manufacturers and
suppliers from a bevy of industries such as automotive, aerospace,
defense, consumer products, electronics, and off-highway say that
better integration and interoperability are required because
multisystem deployments are the reality in manufacturing industry.
Manufacturers want newly purchased PLM solutions to integrate with
their existing infrastructure and be interoperable with the product
development tools already in place; and they want their legacy systems
to work with what they're going to buy.
Technology buyers lay much of the blame for the lack of
interoperability on the IT suppliers. Many providers have made laudable
strides in recent years by opening up their technology, while others
merely pay it lip service. In a panel discussion from a recent Daratech
conference, a representative from one vendor said that in order for
interoperability to become a reality, it was up to the customers to
demand that the vendors work more closely together. This unified voice,
he reasoned, would ultimately lead to interoperability. After the
simultaneous groan from the audience subsided, a manufacturer quipped,
"I've been asking you to do that for the last five years."
The rapid evolution of technologies used by manufacturers also leads to
whole new problem areas. One is the urgent need for better
interoperation between CAD and CAE environments-a critical requirement
if manufacturers' burgeoning initiatives to deploy digital prototyping
and simulation more strategically and effectually are to fully succeed.
Another pressing need is for better connectivity and interoperation
between disparate product data management (PDM) systems. Growing
numbers of manufacturers say the accelerating pace of mergers and
acquisitions-where the company being acquired often has a different PDM
system than the new parent-has made the goal of converging on a single
corporate PDM system unrealistic. Likewise, companies that rapidly
partner with numerous suppliers and customers face the problem of
somehow making their PDM system support the goal of collaborating with
companies using rival solutions.Developing Standards
To address interoperability concerns, many suppliers have raced to
develop standards. Standard data exchange formats, however, are no
longer seen as the panacea they once were. Indeed, many, including
Daratech, feel the proliferation of standards threatens to become a
bane in its own right. Why? Because the proliferation of
standards-intended to help manufacturers exchange engineering data-is,
paradoxically, diminishing the prospects for practicable
standards-based interoperability.
Many of the new standards boast of their openness, but most of these
claims have yet to be backed by the manufacturing community.
Manufacturers see a need for a new approach to developing and adopting
industrywide standards as the gap widens between what needs to be
standardized and what standards organizations and competing suppliers
have accomplished. Consider the long and still-growing list of widely
used data formats-IGES, STEP, DWG, DWF, DXF, XT, SAT, STL, JT, PLM XML,
3D XML, and other standards. This list does as much to cloud the
picture as to clarify the best route to universal interoperability.
But to focus solely on the supplier community is a bit unfair.
Suppliers, after all, are for-profit entities whose goals are often not
aligned with those of the open software movement. Someone must pay for
adherence to standards, for availability of tool sets on multiple
hardware platforms and operating systems, etc.-and it is to the
suppliers' advantage to keep their customer bases captive and thereby
ensure a healthy revenue stream. It is also true that much of the
interoperability problem is due to legacy systems: who could have
envisioned that the company enterprise resource planning (ERP) system
put in place five years ago must now interoperate with the PLM system?
Or the CAD system of an acquired company with the acquirer's PDM
system? The technological challenges of making different systems
communicate with one another are far from trivial, and would require
substantial development resources to be added or diverted from other
work. Too, while users have long complained about the situation, few
appear eager to pay extra for what many feel should somehow be a
standard capability of the product.
The lack of interoperability is partly by design, as PLM companies seek
to keep their user bases captive, but also partly due to changes in the
manufacturing marketplace and the rapid proliferation of IT tools of
many kinds. Given that these last two are unlikely to ever diminish,
interoperability will continue to be a challenge, and smart
manufacturers are doing the best they can to come up with their own
strategies for dealing with the problem-regardless of support from the
IT solution providers.
How are companies overcoming interoperability challenges? They do so in
a number of ways, though none of them optimal. They perform
translations into recognized standards such as IGES or STEP, which
strip the model of intelligence. This can be time-consuming and
error-prone. Sometimes they wholly recreate models in another system.
Or they employ third-party software translation tools, of which 100
percent reliability is still a question. Some manufacturers overcome
interoperability difficulties by shear force: dictating that suppliers
use a particular system. This approach works terrifically for the
manufacturer, but is expensive and labor-intensive for the supplier who
must now maintain at least one seat and one trained operator for each
system. Suppliers in this situation do their work in whatever system is
most applicable to their needs and then redo much of this work in the
system mandated by the OEM-the ultimate in wasted productivity. But far
from uncommon; many manufacturers still enter by hand data generated by
a bill of material system into their corporate ERP system or type in
part specification data from a printed catalog.
Clearly, how interoperability will ultimately play out remains in
doubt, but Daratech has long believed that locking in data is
ultimately a losing proposition for software vendors. In the meantime,
manufactures are coming up with their own strategies and processes to
overcome inadequate interoperability, though coping mechanisms are
hardly ideal. Most software developers are making commendable strides
in this arena, but forgive the manufacturing community for being
skeptical.
For more information contact Tim Hickey of Daratech, Inc. (Cambridge, MA). Go to wwww.daratech.com
