The text presented here is not precisely as published by OUP, but modifications are minor. Illustrations are another matter. Where images used in the original book were not my copyright, I have in most cases been able to substitute links to coloured images on the web. The sources are listed under Image Acknowledgements.
When this text was submitted as part of a PhD thesis in 1996, the Notes were greatly extended. Most readers may prefer to ignore them. They have been collected at the end of each chapter, with internal links leading to them and back to the text. They are a mixture of: simple page references; additional examples or quotations to justify generalisations; and some afterthoughts.
In the Oxford English Dictionary, the first meaning attributed to the word 'art' is "skill, acquired as the result of knowledge and practice". Design of structures as taught in undergraduate courses tends to consist of guessing the size of members required in a given structure and analyzing them in order to check the resulting stresses and deflections against limits set out in codes of practice. Where then is the 'art' in structural design in the sense of a skill which goes beyond the limits of precise logic?
Structural design can be seen as the process of disposing material in three-dimensional space so as to satisfy some defined purpose in the most efficient manner possible. To do this we must have philosophies of 'purpose' and 'efficiency'.
An engineer might see the purpose of the structure as being to carry the imposed loads to the foundation in the most direct manner possible. The word 'possible' recognizes that in doing so it should not interfere with the function for which the structure is intended. The structure of a building must be accommodated to the needs for clear space within the envelope, that of a bridge must allow for passage of water, or forms of transport beneath it. When these needs conflict with directness of structural action one or both must be compromised in some way. Already the designer finds himself attempting to balance the relative certainties of structural logic against the qualitative properties of interior space.
What is structural 'efficiency'? An engineer might define it as the ratio of benefit (output) to cost (input). It is possible to measure cost quite precisely in terms of the quantity of material used in a structure. Unfortunately this is of no practical use because it ignores the relative cost of different materials and most importantly the costs of fabrication and construction. A truss with intricate web tracery may do the same job as a solid plate girder while requiring far less material, but the high cost of labour involved in cutting and assembling its many parts may far outweigh the saving in material cost.
Again the designer must balance precise savings in material against his qualitative knowledge of the problems of fabrication and construction and the uncertainties of cost prediction in these areas.
The designer must use skills and imagination in envisaging the types and combinations of loads and internal forces that may arise in the life of the structure and in estimating their likely magnitude. Guided by the committees who write the codes of practice and the building regulations and in discussion with the owner, he must make many subjective decisions about the performance which is to be demanded of the structure. Should the building be designed at enormous cost to withstand the most severe earthquake it is possible to imagine, or the impact of crashing aircraft? If not, should it at least afford some protection, ending up bent but more-or-less whole?
Looking at more common types of loading the designer must decide what intensity of wind and occupancy loading he must allow for. His decisions must be based on the life of the structure, its importance to the community and on possible changes of ownership and usage, as well as on meteorological data and surveys of typical loadings.
All structures are designed with a factor of safety which may be defined briefly as the ratio of strength to reasonably expected load. This may be as low as 1.1 for slip-circle analysis of earth dams, but is typically of the order of 1.7 to 2.0 for structures. In arriving at such figures a large number of qualitative and probabilistic factors have been taken into consideration including quality of fabrication, variability of materials and loads, and accuracy of the theories employed in analysis. These figures also involve an assessment of the amount of risk which is acceptable to the owner, the users, and the community.
When the time comes for analysis, skill is required in the application of theoretical concepts and analytical techniques to the mathematical modelling of structures. All theories are approximations and the designer must choose with care the theory most closely approximating the reality of the structure, and encode its geometry and material properties into his mathematical procedures and computer programs in such a way as to achieve a valid analysis.
In order to do this he should be aware of the background to the development of the theories and the code rules which he applies. What are the approximations and assumptions inherent in the theory and especially in the computer program that he is using? What shortcomings should he expect in the picture they give of stress and strain in the structure, and how may he compensate for these?
The rules which are found in codes of practice for the analysis of reinforced concrete floor systems and the stability of beams and columns are gross simplifications of very complex phenomena. They incorporate ecisions about which parameters may be ignored, what are the likely dimensions and form of the structural elements under consideration, and in the case of stability what are the likely magnitudes of geometrical imperfections and of residual stresses in steel. Code rules governing fatigue, brittle fracture, and corrosion and other problems of weathering again give only patchy guidance on very complex and qualitative issues.
In order to apply these rules with any intelligence the engineer should be aware of their origin and the procedures through which they have been developed. Once he has analyzed the structure the designer must with the aid of codes make further decisions concerning performance requirements. All structures deflect under load. How much deflection is 'too much'? How much sway may be permitted in a multi-storey building under wind load? At what levels of vibration does the average person become uncomfortable then alarmed? Should the designer cater for the most sensitive person or the average?
It is of course possible for the designer to be lazy and depend for most of these decisions on the unthinking application of codes of practice and compliance with the letter of building regulations most if not all of the time. Such an approach is un-professional and at times dangerous. Designers of office buildings have for some years been allowing for much higher floor loads than required by law in order to cater for the weight of proprietary types of compact filing systems. A few years ago it was realized that the current dance craze was imposing dynamic loads with a static equivalent equal to several times the regulation design load.
It is obvious from this brief survey that sufficient skill, cunning and imagination are required in the 'design' of structures, even in its most narrow technological sense, for it to be properly described as an 'art'. The greatest part of this art lies of course in envisaging a suitable form for a structure: the disposition of material in space to perform the required function most efficiently. Our theoretical tools can provide us only with analyses of structures whose basic geometry has already been defined. At present our commercially available computer programs can do little more.
Space limitations prevent further detailed consideration of the qualitative and subjective aspects of the engineer's input into structural design. It is hoped that sufficient has been said to caution the engineer against a too arrogant assertion of the objectivity and precision of technology in comparison with other disciplines. On the other hand there is no need to adopt the defensive stance taken by some that engineering is all 'rule of thumb and fudge-factors'! Coping with the variable and the ill-defined is a challenging and a perfectly respectable occupation.
The structural engineer does not of course work in isolation. Politicians, executives in private and public organizations, and architects often play the major role in the definition of 'purpose'. In addition the structural engineer is usually involved in projects which are so large and complex that they must be conceived and designed by a team rather than an individual. To work effectively the structural engineer should know something of the concerns of these other participants in the project. It can thus be said that the art of structural engineering includes the art of communication, of relating successfully to the other members of the team, and of sharing in or leading the administration of the design process.
This is rarely an easy task because of certain conflicts inherent in the aims and limited viewpoints of the various participants. As a simple example, if the structural engineer increases the depth of floor in a multi-storey building in order to provide greater leverage between the flanges of his I-beams, he may add considerably to the area, and hence the cost of the external cladding. If he designs his floor too tightly with regard to depth or stress levels he will add to the problems of the services engineer who may wish to cut holes in the webs of the beams for water pipes.
In the wider context there is liable to be conflict between the structural engineer's desire to minimize the cost of the structure by providing regular and reasonably closely spaced supports and the architect's desire for appropriate clear spaces to suit the function of the building. In minimizing the direct cost of construction the engineer might choose a slower form of construction thus involving the investor in higher interest charges which could easily outweigh the saving in capital cost.
The moral of this is that all parties involved in the project should attempt to take into consideration all relevant factors. Obviously this ideal is unattainable. Given that no-one can 'know everything' it follows that many propositions must be put forward and many binding decisions made in the early stages of the design process in ignorance of the relevant facts. As we shall see shortly a classic example was the action of the architect of the Sydney Opera House in sketching the shell-like forms of its roof without the benefit of an engineer's understanding of the mechanics of thin shells.
As the implications of the original decisions become clearer the question arises of whether to attempt to overcome the difficulties imposed by the earlier decisions or whether to take advantage of other possibilities which were envisaged or have been revealed by the investigation. To what extent should the earlier decisions be modified? Should they be abandonned entirely? What relative weight should be given to the conflicting objectives and values of the various parties?
At this stage the personalities and social skills of the participants become as important as whatever objective facts may be to hand. As the architect who designed the Munich Olympic Stadiums wrote "often the louder, the stronger, the quicker will win" (Behnisch, 1980). The roots of what may be quite heated conflict lie in the varying objectives and value systems of the participants in the project. To simplify the situation to the point of caricature, the goal of a private client, say a property developer, is to make a profit. The politician wishes to achieve credit so that he will be returned at the next election. The architect might want fame for an outstanding design, leading to further commissions. All of these people are concered with different aspects of the project. Their goals are different and possibly contradictory, and conflicts of values are inevitable. [Note 1.]
Of course, people are far more complicated than this. They have a need for self-esteem and for public recognition of their significance. Competition and the ruthless pursuit of profit are limited by convention and personal beliefs. Some people seek public recognition directly, either from the public or their peers, while others are satisfied to earn a high salary and demonstrate their status by the way they spend it. It is important for some to establish bureaucratic 'empires' while others prefer the anonymity of 'back-room' jobs. It is necessary for all consultants to convince others of their competence in order to ensure a continuing livelihood.
However, even the less contentious motives may give rise to friction. The architect may see himself as an artist, bringing warmth and humanity to a staid and colourless environment. The engineer may pride himself on his practicality and his ability to conserve capital and resources. A local government engineer may gain satisfaction from his roles as a sort of policeman of the building industry. The potential for conflict is obvious. Furthermore a person's belief system is usually based on personality characteristics rather than rational analysis.
Fortunately, all the participants also have motives for banding together to form a team and this normally overcomes the divisive effects. Consultants would be lost without clients and vice versa.
Architects would, one hopes, be lost without engineers and regulatory bodies would be out of a job if there were no-one to regulate!
An interesting point concerning motivation is that salary earners on the whole have a different outlook from the self-employed, particularly regarding the importance of money vis-a-vis perfectionism in the search for solutions, in computation and detailing and in aesthetics. Contrary to popular belief, government engineers may be more innovative than privately employed ones because they run less personal risk in trying out new ideas.
As previously mentioned, these matters are beyond the scope of the book, except for the difference in attitude between engineers and architects which is of immediate importance to most structural designers. The manner in which people interact in groups as a result of personality factors has been researched by sociologists and psychologists under the headings of 'Group Dynamics' and Behavioural Psychology. Reference is made in the Notes to some sources dealing with motivation, value systems and group behaviour.
Although this book concentrates on the middle ground the influence of the politics of the general community must be given at least a passing mention. Politics may have a considerable influence on the work of the structural engineer, the ultimate being the cancellation of a project to which he has devoted several years of his life.
We shall see in the section on aesthetics the influence which pressure groups concerned with environmental protection may have on the design of structures in both the country and the city. [Note 2] Applications to the local authorities for planning permits may be opposed by local interest groups who are capable of obtaining major modifications to, or even complete abandonment of proposals. With publicly funded projects there may be similarly strong resistance to the basic concept and there is the added problem that taxpayers' money is being spent. It is always difficult if rises in cost occur (an apparent rise is impossible to avoid in times of inflation) or if completion does not occur on schedule.
The Snowy Mountains Hydro-Electric Scheme was a politically sensitive project in its early days and there was much debate about its economic viability. As a result great attention was paid to public relations. The design of above-ground power stations was considerably affected, and the Authority became a pioneer in the aesthetic design of industrial structures. [Note 3] In the Murray River stations, one side of the machine hall was fully glazed to allow visitors to see into the station, despite the resulting problems in structural design. Areas were provided for visitors to observe the machines at close range, and in the Murray 2 station this facility was incorporated in a two-storey building spanning a 15m gap - an expensive requirement. Such decisions are entirely political; but involving as they do considerable additional expense they tend to make the cost-conscious engineer wonder just why he struggles so hard to save a few millimetres on the diameter of his reinforcing bars or to cut a few centimetres off the thickness of a concrete slab.
National, state or city prestige often becomes tied up in structures, with consequences for those involved in their design and construction. In the following chapter we shall see how politics hastened the start of the Sydney Opera House project, compounding the difficulties of the design and how it entered into the final resolution of the conflicts within the design team.
The city of Montreal was host to Expo '67 in an attempt 'to put itself on the world map'. One of the major exhibits was 'Habitat', a high-density housing project whose designer, Moshe Safdie, wished to prove the practicability of mass-production methods in the building industry. This aim became entangled with the desire of the city to prove it could 'produce the goods' on time.
Fig. 1.1. Habitat, Expo '67, Montreal.
A prototype for an industrialized building concept which became entangled in politics. (Archt: Moshe Sadie. Engr: initially A. E. Komendant.) [Photo: see the Vieux Montreal website.]
Private industry is also affected by the desire for prestige and this may in turn affect the instructions given to the architect, particularly in the case of banks and head-offices.
It is hoped that by now the reader will be growing aware of the very human side of structural design and has begun to realize that while technical competence is very valuable it is only a part of the overall skill required for successful design of structures. In the hope of confirming this impression and giving some taste of the nature of the design process, a part of the story of the design of the Sydney Opera House is provided in the next section as a case study.
[Top.]
[Contents.] [References.]
[Preface.] [Next Chapter.]
A study was made of work published by psychologists and sociologists under the heading of 'motivation and value systems'. This was found to be very broad in character. There appeared to be little that dealt specifically with professional group problem solving. However, introductions to the broader concepts were found in books such as Collins and Guetzkow (1964), Argyle (1974), and Stein (1974-1975). An interesting 'non-specialist' account of group psychology by Misha Black was found in the final chapter of Middleton (1967) and there are some interesting notes on motivation in Lawler (1973), Herzberg (1968) and Dubin (1958). Statements concerning the importance of the human element in design were found in RIBA (1973), 131-3, and Armstrong and Jack (1969) amongst others. [Return.]
The following sources provided confirmation of the role of politics in the planning and design of built form.
Ruchelman (1977).
Moses (1970), pp.161-5 and 197-220. (Moses was the author of the scheme and his is a highly personal account. An alternative view was found in Caro 1974.)
Stubbins (1976), pp.158-63; Beranek (1962), pp.251-5; Siegel (1962), pp.288-9; Bauwelt, 49, No.1, 7-16 (1958); Architectural Forum, Jan. 1958, 117-21, 170, 172; Baukunst und Werkform, No.1, 1958, pp.13-24; NCE, 16 Oct. 1980, p.8; 29 May 1980, pp.4,5; 5 June 1980, pp. 22, 23; Beton und Stahlbetonbau, Dec. 1980, pp.281-4; Bautechnik, Aug. 1982, pp.253-60. (This building was a gift to the German people from the American people, with the result that the architect strove for a particularly symbolic form. This caused problems of compatibility between the symbolic and the technical aspects, leading to a lively debate in which both nations' prestige was involved. The roof failed in 1980.) [See also Holgate (1997) on Jörg Schlaich, 58-9.]
Suckle (1980), pp.107-21; RIBA J. Dec. 1978; Colquhoun (1981), p.110 ff; Architectural Design, Feb.1977; Proc. ICE, Nov.1979 and Aug. 1980; Architectural Review, May 1977; Architectural Record, Feb. 1978; NCE 27 Mar. 1975 and 3 Feb. 1977.
[See also Silver, Nathan. The making of Beaubourg: A building biography of the Centre Pompidou Paris, MIT Press, Cambridge, Mass and London, 1994.]
Hayward (1976), Civil Engineering - ASCE, June 1976, pp. 56-9 and Dec. 1976, pp.50-4; Xercavins and Ostenfeld (1976); Travaux, Aug.-Sept. 1976, pp.28-49; NCE, 3 June 1976, pp.29-35; 14 May 1981, p.5; 11 June 1981, pp.8,9; 2 July 1981, pp.20,21; 27 Aug. 1981, p.4; 22 Oct. 1981, p.18; NCEI, Nov. 1982, p.55; Apr. 1977, pp.25-36; Structural Engineer, Sept. 1976, pp.329-39.
Safdie (1970). [Return.]
This statement was a loyal reflection of the perception current within the SMHEA at the time I worked for that organisation. It must be qualified in the light of subsequent study. The Authority certainly stood out at that time as a leader in aesthetic design of industrial structures in Australia. However, as stated elsewhere in TAISD (p.126) and AOBF (p.199) the topic was a preoccupation of the Deutscher Werkbund early in the 20th Century. Engineers and architects in many countries were also concerned with the subject both before and after that time. It might be more correct to say that the SMHEA restored the subject to prominence in Australia in the 1950s and 1960s. See Stilgoe, J. R. Moulding the industrial zone aesthetic: 1880-1929. Journal of American Studies 16 (April 1982) 5-24. [Return.]
Vieux Montreal. Link.
[Top.]
The art in structural design navigation:
[Preface.]
[Chapter 2.]
[Contents]
[References]
Site navigation:
[Aesthetics of built form.]
[Papers.]
[Work of Jörg Schlaich.]
[John Monash's early engineering.]