Eric Hunting on Adaptive Architecture (2): Types of Adaptive Architecture

We continue our serialization of Eric Hunting’s essay on the topic, which we started yesterday.

Eric Hunting:

There are basically three ‘schools’ of adaptive architecture; adaptive reuse, functionally generic architecture, and adaptive systems. These further break-down into more specific building systems and design approaches. Adaptive reuse is based on the repurposing of a ‘found’ structure -often a pre-existing piece of architecture that has become obsolete in its original purpose. This is most common in the context of commercial, municipal, and industrial structures with large span interiors that allow for easy retrofit or sometimes the erection of whole light independent structures within the shelter of the larger structure. Adaptive reuse also applies to vehicles and other industrial artifacts like ISO shipping containers and has been explored with everything from culvert pipe to the external fuel tanks of the Space Shuttle. Adaptive reuse also can apply to entirely new prefabricated structures and building systems which are simply employed to a purpose they were not originally designed for. This is common with light industrial and farm structures and their sometimes highly modular building systems.

The chief limitation of adaptive reuse as a strategy for adaptive architecture is that one is limited to the very providential adaptive potential of a found structure one has no control over the form of.

Potential adaptability thus varies greatly, usually being greater the simpler the form of found structure and the larger its structural spans. In general, this approach is most often limited to discrete dwellings and does not suit community development unless the found structures are truly vast -like very large industrial, commercial, transportation, and municipal buildings. Early era industrial buildings, of course, are the basis of most ‘loft’ apartment conversion based on their large size, large spans, easy compartmentalization using built-up partition walls, and numerous windows. Few other types of buildings have been so comprehensively reusable. Modern industrial buildings, which are predominately based on steel frame and panel structures rather than masonry, are nowhere near as versatile in this respect and require radically different approaches to reuse.

Functionally generic architecture is based on structures intentionally designed for perpetual adaptive reuse -a level of design foresight that’s rare today and typically limited to large scale commercial and industrial buildings. These are structures with no pre-determined purpose for any of their interior space -except, perhaps, in a very generalized sense relative to the environmental character of large zones of the structure. Instead, they are designed to accommodate as many uses as possible anywhere within them as necessary over time and with the aid of non-permanent retrofit that conforms to the dimensional limits of the larger structure. This concept is closely related to the notion of ‘skybreak’ architecture where a large independent roof or enclosure structure like a dome is used as a simple weather shelter for light independent modular, and freely adaptable structures built inside it, the elimination of the burden of weatherproofing allowing these structures that lightness and adaptability. This is how Buckminster Fuller actually intended the geodesic dome to be used for housing -as opposed to the rather tricky and unreliable wooden frame dome houses we commonly see today- and it has seen more recent interpretations in such designs as Shigeru Ban’s Naked House where a greenhouse like structure provided skybreak shelter for a compound of traditional Japanese rooms put on boxes on casters like pieces of furniture.

This functionally generic design concept was once almost the exclusive province of commercial office building design, until in the wake of the Lofting movement a few residential developers realized the advantage of designing new buildings as ready-made loft apartment structures. We tend to think of buildings as being whole structures when, in practice, they actually tend to be organized into several primary and largely independent elements; superstructure (which does the work of holding the building up), foundation, roof, floor/deck, ceiling, outer enclosure, partition walls, and furnishings. In some vernacular architectures, superstructure, foundation, floor, and roof were the only substantial or ‘permanent’ elements of a structure. Everything else was temporary, moveable, and light. This strategy was adopted in modern times for the design of commercial office buildings where it was necessary to lease space on a square-foot basis and allow tenants the freedom to organize the internal layout of their workspace to suit their particular operational schemes and choices of amenities and equipment. This strategy became particularly important in the Information Age with the need to accommodate a rapidly evolving assortment of technology in the workplace. Though sometimes elaborate to the point of absurdity on the outside, office buildings are designed with simple post and beam superstructures of as large a span as practical and organized into simple floor levels. This superstructure defines the primary routing for a networked utilities infrastructure. Hanging or ‘curtain’ exterior wall systems and large glass windows provide the basic environmental enclosure. Everything else is non-load-bearing partitions of light framing or sometimes modular panel systems which are all considered temporary or disposable.

This strategy affords a building a much longer life and great economy in use over time because of how freely the interior design can be adapted to suit the needs of changing tenants and unit space demands. Within the limits of its primary structure, it anticipates change and can evolve freely to suit the needs of its inhabitants, the relationship between exterior and interior forms not especially critical and interior design more readily adaptive to any overall form, even if not always efficient in materials use.

In this strategy we see an important demarcation in architectural responsibility. The building owner is concerned primarily with the rarely changing macro-scale architecture of the building while tenants are concerned with the frequently changing interior design or micro-scale architecture. The building owner may need to be consulted on some changes to the interior space, particularly where they involve the tear-down of built-up partition walls and modifications to non-modular utilities components as incorrect changes could impact other tenants or damage the superstructure. But for such things as the rearrangement of modular office partitions and the like, the building owner need have no concern.

This notion can be expanded to larger whole-community scales where the macro-scale architecture becomes the province of community-level management or collaboration while the micro-scale in-fill architecture becomes the province of the individual household or ‘building’. This can serve as an effective solution to the limitations in functional scale of wholly adaptive building systems, maintaining freedom of evolution at the discrete dwelling/facility scale even if freedom of macro-form is more limited by heavier structural systems. Thus we arrive at the concept of the conjoined or collective Community Macrostructure and the Urban Megastructure, the best example of this being Paolo Soleri’s Arcology; a whole city based on a single community-managed megastructure of vast size. Though often accused of ‘megalomaniaclemegabuild’, the Soleri arcology is actually a functionally generic structure that is only ‘designed’ at the macrostructural scale. Everything else is up to the inhabitants in the form of in-fill structure, which is essentially no different from how cities already work except that it’s organizing its ‘backplane’ in 3D. The notion of the independent building structure -and by extension independent property- is another one of those anachronisms perpetuated by contemporary architecture. Just turn any wide angle picture of a city sideways and you realize that most structures in our habitat are no more independent than the peripheral boards plugged into the backplane of a personal computer and thus real estate no less virtual than the domain name real estate of the Internet.

Now, there are much more advanced forms of functionally generic architecture that, for reasons unclear, seem to have been largely overlooked in commercial development and remain unexplored. Though designed for free adaptability, the contemporary office building or loft apartment building doesn’t include any integral systems of in-fill structure interface that would actually facilitate this spontaneous adaptability with the greatest convenience through some degree of smaller scale modular component interface. With the exception of hanging ceilings and raised ‘access’ flooring, these buildings typically rely on very wasteful methods of interior refitting borrowed from the primitive interior finishing methods common to suburban housing. As a result, the interior refitting of these buildings incurs large and unnecessary degrees of waste in labor, time, materials, and cost and much higher degrees of wear and potential damage to the superstructure during conversion. This author has often suggested the notion that the superstructures of large buildings -and especially community scale macrostructures- should include an integral plug-in socket grid over its entire surface area derived from the formed-in-place sockets used for climbing form systems in heavy concrete construction. Spaced in a dense grid akin to a raised floor system, this would allow the simple screw-in surface-mount attachment of an endless variety of fittings allowing for easy routing of all utilities hardware, the installation of mezzanine structures and other secondary support structures, and a standardized modular panel system for all finished walls, floors, ceilings, facade cladding, window framing, and large equipment and appliances. And all of this could be quickly removed and re-arranged as necessary without wear on the superstructure itself. It would seem that, in the context of contemporary trends, this would be a logical approach even for the production of conventional suburban housing.

Adaptive systems are building systems where whole structures are freely adaptable by virtue of easily demountable and manipulated modular components. This is the ideal form of adaptive architecture, where both the micro-scale structure of the discrete dwelling and the macro-scale structure of a whole community are freely and spontaneously evolvable at potentially the same very high rate of change if necessary. These systems are also potentially useful in the context of retrofit or in-fill structure in both the adaptive reuse and functional generic architecture contexts, providing the basis of light structures that can flesh-out the interior of other larger structures.

Such systems tend to fall into two categories; unit module systems and modular component systems. Unit module systems are based on relatively large modular units comprising one or more rooms which serve as complete prefabricated, sometimes pre-finished, structures akin to appliances that can be assembled into larger complexes, either directly or with the use of an external support superstructure.

One of the best and largest examples of such architecture is Moshe Safdi’s Habitat 67, built for the Montreal Expo and based on large interlocking stacked concrete modules forming a vast multi-storey complex.

Modular component systems are those where structures are built from relatively small size modular components, usually in some combination of frame, panel, and fixture modules all scaled for relatively easy assembly by hand and in some rare cases designed for robotic assembly. Space frame structures are the common example of this form of structure. Both of these types systems are sometimes referred to as ‘plug-in architecture’, though in general the term is more appropriately applied to modular component systems based on integrated component attachment methods needing few or no tools.

The chief limitation of adaptive systems is scale. As a general rule, given a particular structural material, the bigger the building the bigger its parts. In order to keep components within a manageable size, many building systems must compromise to some degree on maximum clear spans and maximum load bearing capacity. This is particularly the case with modular component building systems intended to be assembled by hand. However, even with such limits systems based on modern materials are still impressive in performance, usually topping-out in the area of ten storey high structures with spans under 40 feet. With the advent of future nanofiber composite and diamondoid materials we may see these dimensions increase dramatically but for truly large communal structures as those in the contemporary urban environment heavier construction systems may remain necessary. In such a situation the functionally generic architecture approach would supersede but these same wholly adaptive building systems are very likely to find roles as in-fill and finishing structural systems for macrostructures built with other methods.

Contemporary technology for adaptive systems is a relatively recent phenomenon commonly associated with Modernist design, though some vernaculars have exhibited characteristics of modulariity and were often inspiration to Modernist designers -traditional Japanese architecture based on the ‘ken’ system of geometry derived from the dimensions of tatami mats being particularly significant. One would imagine that the many virtues of modular construction, its high adaptability, and it’s potential for industrialized production would make such systems an inevitable evolutionary leader. But, in fact, while countless modular building systems have been devised across the 20th century, very few have survived to the present day or achieved any kind of broad use in the building industry. The chief reason for this seems to be the difficulty in matching the logistics of the real estate market to the logistics of Industrial Age mass production. Many modular building systems have been devised for the sake of one or a few building or home designs which their inventors/designers believed ideal in some way but which had no hope of enough market appeal to justify the cost of tooling for mass production. Early Modernist designers were particularly focused on the concept of industrializing housing as a means to making it accessible to all. But they commonly relied on a model of industrialization derived from the example of the automobile industry, regarding the house as a whole unit product like a car or appliance and seeking an idealized essential architecture for the house applicable to all housing needs akin to that of the car. (with the adoption of pressed steel welded unibody construction in the late 1930s, virtually all automobiles became manufactured in the exact same way and all car designs largely cosmetic derivatives of the same basic architecture) Many hundreds of designers across the 20th century sought the ideal universal – or at least one-size-fits-most – house architecture. They all failed. Though it may often seem as though the typical American suburbanite is subject to a habitat of tragically soul-crushing banality, sameness, and squalor little improved over the project tenement housing of urban areas, the sliding scale of economy by amenities and the spectrum of variations by climate and regional cultural aesthetics is still sufficient enough that a universal house design becomes impossible. No single contemporary prefab home design has ever sold more than a few thousand units in their entire production lifetime – not even those deliberately designed as mass production housing for the poorest and presumably least picky members of society.

Other designers/inventors more pragmatically sought to devise multi-purpose building systems rather than specific building designs -as was the case with the many developers of the various space frame building systems. But many of these designer/inventors refused to assume the personal responsibility for establishing manufacturing industries for them, assuming this to be the role of already established large companies, and putting themselves, again, in the position of having to prove the pre-existence of a sufficiently large market. Those that did assume this responsibility – because no one else would – quickly found themselves limited to the use of very high cost low volume component production for demonstrating their technology. To bootstrap production, they would then seek to focus on commercial ‘glamour’ architecture where the premium cost of their systems was more tolerable than in other construction markets. For most this proved to be a trap, their small businesses never able to get enough building projects to bridge them to continuous production of standardized component lines that could bring their prices down to something the mainstream market could tolerate. This was often exacerbated by a failure of these ventures to actively pursue cultivation of broad spectrums of boilerplate designs that could expand their market. Most settled into this vertical market niche, waiting for the architects and the projects to come to them, and abandoned their original ideals.

Historically, most space frame manufacturers have proven business failures, either being ruined outright with shifts in architectural fashion or surviving by being absorbed into other commercial building products companies. The more successful companies have survived by going international in marketing but even the single largest and oldest space frame producer in the world -MERO-TSK -still, after nearly 70 years in business, cannot get enough work to move beyond on-demand production and now assumes, as a business policy, that it’s simply not possible. Once regarded as the epitome of Industrial Age and High Tech building technology, modular space frame systems have been in existence now for almost a century yet no standardized commercially manufactured component sets currently exist (with the exception of very light systems for store display and theatrical uses) and building a modest home with them can still cost millions.

More recently, designers and inventors have begun exploring the possibilities of repurposing industrial building systems that are already in production as the basis of multi-purpose architectural building systems. Repurposing prefab modular industrial structures was common among Modernist designers throughout the 20th century but limited to discrete building designs, largely because the building systems being repurposed were themselves limited to a few kinds of structures. But the late 20th century saw the emergence of a number of industrial building systems of much more generic aspect intended for such applications as industrial automation, custom shop-floor furnishings, and prototype machine tools. Leader among these are the aluminum T-slot profiles based on extruded aluminum beams with T-shaped slots formed in their sides. Manufactured by many companies around the world, this building system has produced a huge family of standardized industrial components offering many possibilities for repurposing to architectural applications. Several companies are now developing housing based on these parts. This strategy offers the potential to overcome the problems associated with past modular building systems by virtue of the fact that the primary components are already in mass production worldwide, eliminating the need to prove the pre-existence of a market sufficient to justify tooling-up production. These new architectural uses simply present a new extension of an already existing market. However, components specialized to the architectural application are still very necessary, though thankfully limited largely to finished panels much easier to fabricate and incurring no added costs for production on demand when compared to conventional on-site building finishing -typically the most expensive and labor-intensive part of conventional construction. Also, many designers and inventors have been caught up in the current fad of repurposing the seemingly unlikely ISO modular shipping container and, though mostly employing traditional adaptive reuse to them, some are exploring them as the basis of more standardized unit module building systems based on standardized modifications. Again, the advantage here is that one is repurposing a component that is already in mass production worldwide and thus needs no pre-justification for its production.

As new digital fabrication technologies are coming on-line, the cost premium for on-demand production is beginning to drop. Minimum necessary volumes for production are steadily shrinking. designers can now employ modular and other alternative building systems at their own convenience rather than aspiring toward meeting the demands of mass production. This is one of the key forces behind the recent surge in interest in Modernist Prefab housing. It was once unthinkable for most designers and inventors to actually experiment on the scale of entire buildings. But today, this is becoming increasingly practical, resulting in a remarkable explosion in cottage-scale architectural experimentation which, in turn, is spawning new cottage-scale entrepreneurship. This has only just begun to impact modular construction technology, largely because so many of the modular building concepts of the past were forgotten and remain to be rediscovered by contemporary designers/inventors and because, for the most part, they continue to operate largely in isolation of each other because of professional competitiveness. But its clear the virtues of modularity are coming through in these new designs and we may soon find ourselves in the midst of a new modular building technology boom. In the immediate future, production of sophisticated modular building systems will be as practical for owner-builders as other more conventional building methods and may become the basis of significant development movements. We are already seeing hints of this with the T-slot technology.”

1 Comment Eric Hunting on Adaptive Architecture (2): Types of Adaptive Architecture

Leave A Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.