AECEF NEWSLETTER 1/1999

Information Technology in the Modern Design Office

by Ye ZHOU, Ph.D. Candidate
and S. F. STIEMER, Dr.-Ing. ( Ph.D.), Professor

Department of Civil Engineering
University of British Columbia
Vancouver, B.C., Canada

Introduction

With the end of the millennium approaching, it is useful to look at the development of information technology over the past few decades. The evolution of information technology can be shown as an exponential growth curve. It was only 50 years ago when the first commercially available computers became a reality and just 20 years ago when the personal desktop computer began to pervade daily life. The advancement of information technology in the last 20 years has paralleled the growth of computers. Twenty years ago the best personal computer an engineer could obtain at a reasonable cost was a 2 MHz machine with 180 kB storage space. This was only good for simple word processing. Serious engineering analysis tasks, such as Finite Element Analysis, had to be resolved by renting computing time on large, expensive commercial main frame computers. The computer technology of today shows a brighter picture. Engineers can now easily own personal desktop computers with 450 MHz speed and up to 10 GB of storage space with extensive adaptable software. The evolution of the Internet and its effect on life is well known. Finite Element Methods (FEM), the most time consuming (and primitive) process in engineering computing, can be carried out anywhere, anytime, with a trivial cost compared to that of 20 years ago. The rapid development in information technology has had a significant effect on the engineering design process.

However, the advancement of information technology in engineering is progressing only in a linear fashion while the information technology is advancing on an exponential scale. In other words, there is much potential for improved application of information technology in engineering design processes. Engineers are good technology adopters; quickly accepting and using popular modern computer tools, such as word processors, spreadsheets, computer aided drafting (CAD) and databases. Yet, to fully utilise the strength of the latest information technology, one has to have more than just average computing skills. The recent advancement of information technology requires rather specialised knowledge to use its full potential. Today's engineers are trained with only a few skills in information technology, and computer professionals, on the other hand, possess only little engineering knowledge. General-purpose programs like word processing or spreadsheets were no challenge to computer professionals. However, many of the engineering processes, although possessing the potential for large-scale automation, are beyond the knowledge and skills of most computer specialists. The diversity of required knowledge, together with the exponential advancing speed of information technology, creates an ever-widening gap that needs to be filled.

The Needs

Any engineering design starts with two elements: the conceptual stage, and the detailed design stage. Although there is great demand for advanced tools for the conceptual design, as explained in the following, very few are currently available or well developed. Nowadays detailing design tools are abundantly available, such as FEM and CAD tools for structural engineering. However, there is ample room for improvements. Information flow in an engineering office is not only required in computation and drafting. The application of information technology needs to be expanded to the entire spectrum of tasks. Many innovative tools for knowledge elicitation and collection such as groupware, data mining, and automated brain storming, which have been well developed in areas such as business administration, are rarely seen in the field of engineering design. This scenario contradicts the fact that engineering work requires the generation of innovative ideas and the collection of valuable past experience, which is probably at least as important here as in the business world. Fortunately, these needs, i.e. conceptual design tools, and knowledge generation/collection tools, could be accommodated by state-of-the-art development in information technology.

Figure 1. Project cost at different design stages [Davies]

While most computation-intensive tasks in engineering have been well handled by fast computers and specialised algorithms, few computer tools have yet been employed in the conceptual phase of engineering designs, which, as a matter of fact, affects the majority of engineering costs. Figure 1 illustrates how the costs relate to the progressing design stages of an engineering project. It is noteworthy to recognise that the curve slope is greater in the early stage of a design, i.e. the conceptual stage. The steeper slope reflects the fact that the conceptual design affects the project costs much more than the later phases. In the field of structural engineering, the detail design consists of mostly FEM analysis and CAD drafting, for which many matured software tools are available, while the conceptual design is still being carried out by experienced engineers using a pencil and a scratch pad, as has been done for many years, decades, centuries.

In engineering, conceptual design is often a process of rapidly turning ideas into virtual models, evaluating them and selecting the most suitable one. Up to now, converting ideas into models is done mostly by mental imagination. The evaluation of certain parameters (suitability, price, feasibility, etc.) is mostly performed using experience, intuition and a few simple calculations. Conceptual designs can be well conducted in this fashion provided that the system is simple and the designer is experienced. For more complex systems, visualisation of the models becomes difficult, and it is even harder to evaluate them against other design variations. For example, to perform a fast comparison among different structural schemes that might involve various materials such as steel, timber and concrete, is virtually impossible with currently available commercial design tools. To resolve these problems, the required software tool should be able to aid the designer by quickly visualising ideas. A qualitative evaluation would be highly desirable, too. This tool should be simple, quick and intuitive, then design creativity will be fostered and innovative and efficient designs can be found.

Software tools for use during detailed design still have ample room for improvement. In the field of structural engineering, most analysis can be easily and accurately handled by FEM, which have been matured in the last few decades and are now equipped with efficient algorithms. Fast computers free the designer from the concern about the time needed for running the computation. However, building up the FEM models and to process the results, FEM programmes have received better pre/post-processing tools, but they are still not sufficiently connected with other phases of the engineering design processes. The information technologies required to smoothen the interaction between FEA (Finite Elements Applications) and other design processes should be available in principle, and many tools could be adopted and implemented to transform the time-consuming process into trivial efforts. The other major process in detail design is drafting. In the last few years, the advancement of CAD programmes have made 3D drafting possible and many designers have started using it. 3D representation of engineering structures brings drawings closer to reality. Although not yet popular in the field of structural engineering, 3D drawing will make it easier for the designer to verify the model, visualise the analysis and give clients better product presentation. Some useful utilisation of 3D drawings has emerged in the field of architecture, in which existing surroundings were constructed and the effect of a newly inserted building could be easily examined. The next step after 3D drawings would be animation.

Communication is essential in any complex engineering designs, which always involves several parties, and effective group co-operation is necessary for successful completion of projects. However, few tools have been designed with only engineering in mind. One critical part of engineering communication is the idea generation, which represents the basis of all engineering designs. The most conventional method for this is to hold a physical meeting and have a verbal discussion. This process has not changed much since the emerging of engineering itself. Sometimes this method is effective. But in many cases, conducting meetings does not always result in yielding good ideas. The recent development in computer network technology introduces many new possible ways for interaction among engineers and their ideas. The approach should encourage idea generation and should not in any way prevent ideas from flowing freely.

Collection and storage of ideas are also of great importance in engineering designs. Traditionally, databases are constructed to store information. Nowadays, a database can be adopted to collect knowledge. In engineering, often knowledge is lost or cannot be described by conventional ways of documentation and storing knowledge. Some schemes should be able to collect engineering knowledge and also retrieve them easily when needed.

The Solutions

The following paragraphs discuss some methods for solving the above-mentioned needs with the new possibilities created by the fast advancement of information technology.

Applying information technology to engineering applications needs to fully utilise the strength of computers and try to avoid their shortcomings. Compared to human activities, computers excel in their high speed and accuracy in data reduction, graphics, data processing, long-term memory and numerical processing. These merits often come with drawbacks such as the requirement of tedious data input and lack of flexibility, which sometimes constrains any creativity. Those problems can easily deter the engineer from a partial or complete implementation of information technology. It should be noted though, that these shortcomings could be reduced or eliminated by intelligent development of computer applications and their user interfaces. For example: The time-consuming process of setting up FEM models could be automated by generating models directly from CAD drawings. The lack of flexibility of a data analysis programme could be reduced by improved software design. The creativity in engineering design could be boosted by many information techniques such as mind mapping and data mining.

The popularisation of information technology has brought many powerful software and hardware tools into engineers' offices. Now, the widely adopted software applications include electronic spreadsheets, word processors, databases and project management packages. Telephones, facsimiles and local area networks (LAN) are necessities for most engineers' daily operation; the use of Internet is also infiltrating many areas of engineering. All these tools have raised the efficiency and quality of engineering design. However, information technology is far from being fully utilised in engineering designs. General-purpose tools have been widely adopted by the engineering profession. These tools were unquestionably of great help but in order to further the effectiveness of computers, specialised information tools for the engineering field need to be developed. The following paragraphs propose several such tools that are directed towards specialised usage in engineering.

Figure 2. A sample catalogue system for structural engineering design

Catalogue approach: As discussed, conceptual design plays a critical role in a complete design process while few tools are currently available for assistance. Appreciation of an engineering design concept needs quick visualisation and evaluation of various existing design schemes. A long process of such visualisation and evaluation may not only leave many good designs untouched but also cramp a designer's creativity. A useful tool will be a catalogue composed of many predefined but customisable design schemes. Once a scheme is selected it can be evaluated quickly through some analysis process, say an FEM programme. Using a catalogue is swift since all processes are automated, such as transferring a design scheme to an analysis process and processing the feedback from the analysis process. The catalogue approach deviates from conventional paper-based catalogues in that everything is digitised and automated, so that design schemes cannot only be quickly selected but also evaluated against each other. An example of this approach to structural engineering could be the evaluation of a steel portal frame against ones with different materials, say concrete or timber. Figure 2 illustrates a sample framework of such catalogue system, in which the FEM analysis engine can be any commercially available programme. The catalogue tools can quickly return the weight, cost and efficiency of these three designs and the designer will have the optimum scheme in a very short time. Conventional paper based catalogues confine creativity since they cannot be expanded or customised. Digitised catalogue tools may booster creativity by expanding or customising the pre-defined schemes, so that designers will always face increasing number of choices. The development environments for such digital catalogues are readily available from the software development industry. A few of such cataloguing tools have already emerged to the engineering world.

Qualitative Reasoning approach: In the planning and design of engineering projects, engineers are very often required to decide upon a course of action irrespective of the completeness and accuracy of available information, especially at an early stage of the design. Qualitative reasoning is one of the well-developed techniques in the field of Artificial Intelligence, which has the ability to draw conclusion from partially defined information. Qualitative reasoning is augmented with interval analysis methods and can handle qualitative and partial numeric input data. The unique feature of qualitative reasoning makes it an ideal candidate for the early stages of engineering designs, and can be well harmonised with the above mentioned catalogue tool.

Idea generating and collecting tools: Innovative ideas compose the essence of engineering designs, while few software tools are currently available to assist creativity in engineering. Besides thinking alone, the traditional method of idea crunching in a team of engineers is to have a meeting and hold discussions, in which the flow of thoughts may not be unhindered and innovation might not always be encouraged. Furthermore, many valuable ideas get lost because of the difficulty or inconvenience of documentation. Computer network technology is an effective way to bring people closer, and brainstorming techniques excel in creating ideas. If constructed properly, the recent rapid advancement in network technology combined with brain storming techniques, can boost engineers' creativity as well as preserving valuable engineering knowledge and experience.

Figure 3. An information system for engineering offices

A useful network technique, which is also the most popular way for networking in information technology, is to use Transfer Control Protocol and Internet Protocol (TCP/IP), the basis of Internet and Intranet. Through TCP/IP, any networked computers at any location can communicate with each other. This effective linkage of computers has brought many practical communication tools to the engineers, e.g. e-mail, newsgroups (or discussion groups), audio and video conferencing, etc. If brainstorming techniques, such as mind mapping, are embedded into to a well-constructed communication tool, the process of generating innovative ideas as well as preserving them will be easier and smoother than ever before. Figure 3 depicts a system framework for this purpose. The centre of the system is a database, which is capable of storing knowledge information, and is hierarchically structured, searchable and also "sortable". The contents of the database are supplied directly from engineering designers, or through a discussion group server or a networked application server for brainstorming. The operation of the discussion group server is initiated by a discussion topic, say the dynamic properties of some structure, raised by a designer. Upon seeing the topic, all designers in the team may participate by adding relevant information, in the forms of text, graphic, audio or even video, to the server at anytime, from anywhere. The contributed information is then sorted into the database for decision making and knowledge preservation. Various features, such as knowledge rating tools, are built into the system for effective knowledge management and retrieval. The brainstorming application server provides networked software tools, such as mind mapping, to the engineering design team. With the help of this set-up, innovative opinions can be generated, stored and processed by designers and are visible to other designers in real time. By using TCP/IP, many constraints on idea generation and collection, such as time and location, will no longer be existing, so that designers' creativity can be carried forward at maximum capacity.

Summary

The exponential-scaled advancement in information technology provides great opportunities for optimising the information process and management in design offices. At present, modern design offices lack effective tools to assist conceptual designs and encourage creativity. These tools, however, exist in other fields and it should be possible to implement them into engineering. State-of-the-art information technology and knowledge management methods, such as cataloguing software, qualitative reasoning and brainstorming techniques, might soon become commonplace in engineering. This will contribute to an effective knowledge flow among designers and to an enhanced engineering creativity.

References
Buzan, Tony; Use both sides of your brain: new mind-mapping techniques, 3rd ed. New York, N.Y., U.S.A.: Plume, 1991.
Davies, P.R.B. The Contractor's Role. In G. Somerville (Editor), "The Design Life of Structures, Blackie and Sons Ltd. 1992. Colloquium proceedings of British Group of IABSE at Pembroke College, Cambridge, UK, pp 229-232, 1990.
Gao,Wen; Knowledge discovery in database, China Computer World Daily, China, 28 Sept. 1998.
Kramer, Matt. A smart way to brainstorm, PC Week, USA, October 1998.
Hares, John S. Measuring the value of information technology, New York: J. Wiley, 1994.
Huitema, Christian; IPv6--the new Internet protocol, Prentice Hall PTR, c1996.
Thorne, Paul; Organizing genius: the pursuit of corporate creativity, Cambridge, Mass., US: Blackwell, 1992.

Return to the page "Newsletter 1/1999"