|
Knowledge-based
Engineering complements traditional CAD/CAM systems to provide
a new strategic approach for improving effectiveness in eManufacturing.
DEEPAK SHIKARPUR explains what KBE is all about and opines
that KBE is becoming an essential part of strategy as manufacturing
companies struggle to respond to global competitive pressures
Todays
economy presents a new set of challenges for the manufacturing
industry. Global competition has empowered customers who now
constantly demand new and better quality products. Design-to-delivery
cycle time is rapidly collapsing along with prices and margins,
which is hurting RoI. As a result of all this, businesses
are forced to increase customer and market focus and leverage
seamless technology solutions to stay competitive. Organisations
engaged in discrete manufacturing have problems in managing
the design-to-delivery life cycle of the product. The cycle
starts with concept design of the product, which is where
abstract ideas are tried and tested.
Current CAD/CAM technologies do not address the requirements
of the integrated design engineering process completely. All
too often, the CAD model is merely a geometric interpretation
of the design intent. Design iteration is achieved by iterative
procedures that are cumbersome and inevitably stretch design
timescales. In todays dynamic marketplace, customer
needs and specifications change often and as a result Change
Management becomes an issue. Version control of changes
and mapping them across from design to manufacturing is where
IT can be used innovatively to gain competitive edge.
Knowledge-based Engineer-ing (KBE) offers a significant advantage
over current CAD/CAM technologies because the fundamental
element is not geometry but a rule. Rules are used to describe
the knowledge about the product, the technologies used to
design, analyse and manufacture it. Rules can be in the form
of physical equations, graphical and tabular relationships
and even anecdotal rules of thumb, which reflect
years of design experience but which defy a more rational
relationship. This is where an artificial intelligence approach
can be leveraged.
KBE and objects
The underlying basis of a KBE system is an object-oriented
structure. An object is an assembly of rules that can create
and store data, and relate data to other objects. The definition
of an object is self-contained and can be developed and tested
in isolation. The same object can be re-used in many different
contexts.
Unlike conventional computing environments, it is not necessary
to flow chart the order in the engineering process. KBE only
requires description of the individual objects required to
define the solution. The order of solution is demand driven
and hence is worked out by the KBE system in response to user
inputs. The rules provide a generic description, which is
used to drive the design engineering process.
What KBE achieves
The KBE system can model the complete design process, integrating
analysis and manufacturing considerations, with design geometry.
It can be defined as an engineering method in which knowledge
about the product and the techniques used to design, analyse
and manufacture a product are stored as a series of rules.
KBE creates a design that is, therefore, a single instance
of a generic definition and can adopt its geometric representation
to suit the requirements of analysis and manufacture and react
rapidly to fundamental design changes.
Knowledge-based Engineer-ing extends the product design process
beyond geometric modelling to capture the knowledge of a companys
most important assetthe experience of its engineers.
Unlike CAD systems that manipulate mainly geometric information,
KBE systems capture the engineering intent behind the productthe
how and why aspects of the design.
A KBE system stores knowledge about a product in a comprehensive
product model composed of engineering rules that describe
how products are designed, analysed and manufactured. These
rules can be design rules, standard engineering rules, or
experiential rules of thumb about attributes of the physical
product such as geometry, material type. The model created
by the rules is similar to a spreadsheet and allows engineers
to quickly evaluate many design alternatives or create new
designs by altering the input parameters or the rules, dramatically
improving the productivity and ensuring consistently high
product quality.
Another strategic benefit of KBE is that the product model
also includes various outputs such as reports of engineering
results, data of engineering analysis, 3D geometric models,
and manufacturing instructions. When the design changes, so
do the outputs. Information is always up-to-date and reflects
the current state of the design, automatically.
Strategic benefits
More companies are turning to KBE systems to help them respond
to worldwide competitive pressures.
With CAD systems, people define part geometry and document
the result of the design process with annotated drawings.
KBE systems complement CAD systems by adding the engineering
knowledge that drives the product design process. They also
produce other information besides drawings such as Bills of
Material, cost analysis, process plans, MRP inputs and user-defined
reports. Companies can use CAD systems as either the front-end
to supply geometric constraints or other user specifications
to a KBE system, or as a back-end for detailing, documentation,
analysis and NC tool path generation.
Who needs KBE?
Companies with complex design and manufacturing problems will
benefit the most from KBE. Many of the leading aerospace,
automotive, consumer products, defence and industrial equipment
manufacturers have installed KBE systems to help them reduce
time-to-market and capture valuable engineering knowledge
within the company. Complex designs that require much iteration
for optimisation and designs with a high degree of repetition
are perfect candidates for KBE.
|