CAD is Dead!
Page Created By: Kaushik Sahu on 06/26/2000 at 09:45 AM
Category: CAPPD Lab
INTELLIGENT DESIGN TOOLS: THE NEED OF THE HOUR
(by Dr. Kaushik Sahu)
"CAD is DEAD!!". This is what we were told way back in 1989 by Prof. John R Dixon, who was heading the Mechanical Design Automation Laboratory of the University of Massachusetts, Amherst, USA. I was perplexed. Then, as I sat through his lectures, gradually the realization dawned on me. What he was trying to say was that CAD tools lacked the ability to "think intelligently" and hence were no better than mere drafting tools. The letter 'D' in the acronym CAD was indeed very misleading. CAD really meant Computer Aided Drafting and not the widely used term Computer Aided Design. The design potential was grossly misused and CAD continued to remain isolated even in a computer-integrated environment.
The first step towards Computer Integrated Manufacturing (CIM) was in bridging the gap between design and manufacturing. Since human beings have a fascination for automation, the result was Computer Aided Design and Manufacturing (CAD/CAM), or the larger wheel, Computer Integrated Manufacturing (or maybe Computer Integrated Management). My focus is on a smaller domain of CIM, that is, CAD/CAM, where we will try to understand the need for intelligent design tools.
There is no doubt that a computerized drafting tool makes life easier. CAD, as in Computer Aided Drafting, helps in improving productivity and quality while introducing a higher degree of standardization. In no way does it help in innovative design, other than just by representing a "skeletal" geometric shape in the form of either points, lines or curves on one end or a "lifeless" solid model on the other. Be it a wire frame model or a solid model, the fact of the matter is, it needs further processing downstream for either performing engineering analysis or computer aided manufacturing. This does not help in bridging the rift between design and manufacturing, thereby, making Computer Integrated Manufacturing (CIM) a difficult strategy to adopt. The need to overcome this problem by addressing all downstream concerns was partly a motivating factor in the development of intelligent design tools.
Here is a brief sketch on how intelligent design tools were developed. Early efforts were made at bridging the gap between design and manufacturing by trying to translate the design into some manufactured reality. Systems were developed to interpret the drawings (of designs) into process plans for manufacturing. Typically these systems would access the geometric database and identify features relevant from the manufacturing viewpoint. Once the features were identified the corresponding process plans were automatically generated. Several such systems were developed in the 80s and these were known as computer aided process planners (CAPP). However, the trouble with these systems was that they were repeating the task of the designer by trying to understand what had been already designed in the first place, and then trying to develop the means of manufacturing that design. The barrier (gap) between the designer and the manufacturing process planner still remained, only now replaced by a computerized database.
Towards the late 80s, researchers started thinking differently. They realized that the drawings stored by the designers are nothing but low level data having limited or almost no knowledge. The CAD tools had advanced from wire frame modelers to solid modelers, and yet they lacked knowledge. Automated process planners still had to abstract relevant information from low level data. Higher level representations were badly needed to store the designers' intent and assist the manufacturing planner thereby providing intelligent support right up front.
As mentioned before, relevant features were being identified from low level geometric data to develop detailed process plans. This word, FEATURE, gradually started picking up momentum. There were researchers talking about design-by-features and some believing in design-with-features and yet some others researching just on features (feature mapping, feature extraction, feature recognition, etc.). Their efforts gave birth to the feature based design tools. Designers could now depend on a library of features to assist them with their designs. There were several different viewpoint specific features. A feature, relevant from a manufacturing viewpoint, was defined as an entity representing manufacturing knowledge (or, for that matter, any other viewpoint specific knowledge). Thus, using a feature from the library, the designer would be forced to design-for-manufacturability. Though these features imposed constraints on the design space, they helped the designer to do his work right the first time, an essential principle of Total Quality Management (TQM). This reduced rework and enhanced concurrency thereby saving engineering and design cycle times that ultimately helped improve the lead time in new product development.
As we see, research has progressed in enhancing the intelligence of the design tools to a large extent. However, we in India have not yet been able to realize these benefits. Our country has been the dumping ground of obsolete technology from developed nations. With them now realizing the enormous market potential of Asian countries like India, the rate of dumping has also increased. This has resulted in our people being faced with products having really short useful life cycles. Before one gets acquainted with a certain product a new one is introduced in the market. Hence our people are forced to move from older models to newer models without realizing the maximum benefits of any one of them. For instance, we have offices acquiring pentium processors without having their people realize the full potential of the 286s, 386s and the 486s. This fast rate of change, coupled with attitudinal barriers, bring in a certain cultural resistance. This is because of the fact that there is a natural reluctance and inability to learn about and use new products. This is true of any emerging hard or soft technology.
The same is also true of the class of software products used by our designers. Even though the design software have evolved over the years, our industry has not been able to take the full advantage of the promises of CAD tools. We are still awed by the ancient drafting (not design) tools. Even if there is a slightly improved design tool available we under-utilize it in terms of its capabilities. Nor do we find our industry aggressively adopting product innovation as a major competitive strategy. Naturally therefore we are not facing any pressures to research and develop real design tools for new product development.
The point I am trying to make here is that we should be focusing on the development and use of intelligent design tools and not mere drafting tools. A good and intelligent design tool fosters concurreny. Concurreny helps in reducing the lead time thereby making the manufacturing organization competitive. Manufacturing is the heart of any country -- and India is no different. With globalization and liberalization the Indian Industry is faced with the problems of being competitive in the global arena. Unless we gear up for this challenge now, we lose. Our industry needs to be innovative. Product innovation is the need of the hour. For this, we need to exploit the full potential of the available design tools and, if necessary, we have to indigenously develop our own design tools. The fastest and effective way to indigenously develop is to customize the available design tools to think intelligently.
I am not saying that high-tech is the solution to all our miseries. What is appropriate for our socioeconomic conditions is a mix of both high-tech and manual systems -- Human Oriented CIM (more on this later) is perhaps one answer. Most importantly we have to learn to realize the maximum benefits of the available technology and be ready to catch up fast with the next to next generation products.
Original: January 1996