JOHN P. R. FALCONER
JOHN P.R. FALCONER is associate professor of architecture and associate director, International Devolopment Technology Center, Washington University, Saint Louis, Missouri, United States.
This paper provides some background to a World Consultation on the Use of Wood in Housing to be held in Vancouver, Canada, in July 1971.
The Consultation has been convened on the initiative of FAO and will give emphasis to the needs of developing countries. It has been organized by the Government of Canada and its Forestry Service in collaboration with FAO, the United Nations Housing, Building and Planning Centre, the United Nations Industrial Development Organization, and the International Union of Forestry Research Organizations.
THE TRADITIONAL FORMS of rural housing in the tropics are most commonly earth, timber or bamboo constructions, or a combination of these, with materials such as grass, leaves or cane used for roofing and filling in the frame. The availability of timber is not necessarily an index of its use for housing; even in heavily forested areas, and where soils are not particularly suitable, earth construction may predominate. Maintenance problems with such traditional construction are not important considerations in the rural context where labour and materials may be had at little or no cost. However, in urbanizing areas, traditional planning and building methods conflict with and are restricted by land-use requirements, the availability of materials, the new demands placed on the time of the urban migrant and his family, and by the intensified health and safety standards of high density living. Thus, in tropical areas where the population pressure is greatest, lack of suitable materials and related building methods are important factors contributing to high construction costs and to delays in meeting housing needs.
One approach to this problem lies in the development of nontraditional building materials, such as concrete, fired-clay products or even high performance materials utilizing advanced technology. There are often severe local restrictions on such new development, however, such as the lack of exploitable limestone deposits for cement, unsuitability or difficult localization of clay soils for bricks and tiles, or the high cost of energy for manufacturing.
Another approach to the materials problem lies in the improvement of indigenous materials and related building methods. Earth construction, for example, has received considerable study in the last two decades, and new stabilizing and waterproofing techniques have resulted in improved rural housing, although it has not yet found favour for use in most urban areas. In the case of timber most of the technical obstacles to its use in the tropics have been the object of research and experiment for some time so that there now exists sound knowledge on which timber housing design and construction can be based. Many species suitable for building construction have been identified. Their strengths, durability, and other performance characteristics, as well as problems associated with their use, are a matter of record. Technical information is not complete and studies must continue, but the major problems lie in areas which are economic rather than technical and associated with the supply of a usable product.
One of the fundamental objectives of housing is that of protection from climatic extremes. In temperate areas, climatic control first centred on providing a refuge from extreme cold and later turned to the relatively less important considerations of heat and humidity. Their control has been accomplished by the creation of indoor environments by mechanical means, made possible, both technically and economically, by industrialization. The implication here is that as tropical areas of the world industrialize their economies will increasingly support mechanical solutions to problems in environmental control, including the research necessary to induce such technology.
Housing, of course, is much more than a controlled climatic environment. The social and psychological needs met by housing outweigh marginal considerations of physical comfort. In considering problems of tropical housing, technicians from industrialized countries may lay too much stress on purely physical considerations and material comfort, and be too quick to introduce technological improvements. It is clear that housing solutions in developing countries, except for the elite minority, cannot afford the luxury of air conditioning or expensive architectural devices such as double roofs, brise-soleil, or heat absorbing glass. Uncomfortable gain, storage, and emission of heat must therefore be kept within reasonable limits by the application of climatic and biological principles to planning and design, and by the selection and use of building materials based on their thermal characteristics.
Tropical climates range from hot-humid to hot-dry extremes and include transitional climates which exhibit characteristics of both. Timber suitable for building occurs mainly in the hot-humid climates - those areas ranging from about 15° north to 15° south of the equator. They are characterized by high relative humidity (upward of 75 percent), low diurnal temperature variations of about 1015°F (5-8°C), and low annual variation. Heavy cloud cover limits the average maximum temperature generally to 90°F (32°C) in the shade. Rainfall can be intense in rainy seasons -up to 2 or 3 inches (5 or 8 centimetres) per hour. Wind speeds are low - about 5 miles (8 kilometres) per hour- but can reach 60 to 80 miles (95 to 130 kilometres) per hour during thunderstorms. Brightness of the sky can be intense, and sky glare caused by the diffuse radiation from cloud cover can be painful.
THERMAL AND CONSTRUCTIONAL CHARACTERISTICS OF TIMBER
Traditional variables in tropical housing have been the subject of study for some time. They are the conflict between privacy and the need for openness to allow air movement, the use of outdoor space formalized as the tropical courtyard and veranda, the shape of interiors and openings to optimize air movement, and orientation to solar radiation and wind direction. Research in these areas must be intensified to find planning alternatives which are compatible with the demands of high density urban living. However, except for high-rise construction, present and future criteria in these aspects of housing design are as likely to be met with timber as with other materials. The thermal and constructional characteristics of timber are, therefore, of more particular concern to us here.
When radiant energy strikes a surface, part will be absorbed by the material and part will be reflected. A material simultaneously reflects, absorbs, and emits (by reradiation) heat to its cooler surroundings. The surface characteristics of the material largely determine the relative amount of heat energy reflected, absorbed, and emitted. Heat movement through a material is a function of temperature differential, the resistance of the material to heat flow, its thickness, and its thermal capacity-that is, its ability to store heat before showing a rise in temperature. Resistance and capacity are directly related to density: lightweight materials have high resistance and low thermal capacity, while massive materials offer low resistance but have high thermal capacity. Heat flow through building elements is non-uniform due to diurnal variations in temperature. It is said to be periodic, and the interior temperature of a building in response to variations is also periodic. Heat transfer is delayed in relation to the thermal capacity of the element, and this delay is referred to as "time-lag," measured in hours.
In hot-dry climates, which are characterized by large diurnal temperature variations, the thermal capacity of materials may be put to use by selecting those which produce a time-lag calculated to control the arrival of heat in the building interior during the uncomfortably cool hours of the early morning. Massive materials suit this purpose very well. However, in hot-humid areas the objective is to maintain interior air temperatures at or below outside shade temperature and especially to avoid temperature rises at night. For this purpose materials with high time-lag should be avoided, for even though diurnal temperature variations are lower than in hot-dry climates, massive materials will absorb considerable heat energy during the day and reradiate it during the night. The most satisfactory enclosure in hot-humid climates will be light in colour and weight: light in colour to reflect as much energy as possible and low in density so that accumulated heat will be dissipated as quickly as possible by the movement of air. Timber is a low-density material containing entrapped air in its cells after seasoning. It has, therefore, high resistance to heat flow -that is, it is a good insulator - and low thermal capacity. These characteristics make it an inherently suitable material for building in hot-humid climates.
Of all building elements the roof is the most costly and difficult to construct, and has the greatest effect on thermal comfort in a building. Traditional rural roofing materials such as leaves, grasses and canes are no longer suitable in urban centres in the tropics due to their short life and flammability. Manufactured alternatives are few, costly, and perform poorly from a thermal point of view. They were developed for other climates. A manufactured roofing material which is satisfactory for tropical conditions has yet to be developed.
The ideal roof will absorb as little heat as possible and will offer almost complete resistance to heat flow. Studies have shown that with roofing materials commonly available in the tropics this is not possible without the additional benefit of a ceiling (1). Timber is commonly used as the structural support for tropical roofs, but this of course adds little to thermal performance. It is also a fairly common ceiling material for the more expensive dwellings. In past years timber shingles, usually of hand-split manufacture, were successfully used for house roofs in the tropics. But with the introduction of metal roof materials, shingles have gone out of fashion and are now considered " bush" material. This is unfortunate, for shingles from suitable species of timber can provide an attractive and comfortable roof with greater durability than some common manufactured materials. In Ghana, for example, a few buildings with wood shingle roofs about 30 years old are still in excellent condition. Efforts are now under way at the Department of Housing and Planning Research, of the University of Science and Technology, Kumasi, to reintroduce the use of shingles. A building was constructed in 1968 near Kumasi with a wood-shingled roof which did not include a ceiling (2). Tests showed that interior temperatures remained at or only a few degrees above shade temperature during periods of the most intense radiation. Shingles in this case were machine-sawn, a fairly expensive special process in Ghana, and efforts are being made to develop split shingles, which it is felt could be produced from log off-cuts by small local industries at considerably less cost than sawn shingles or imported metal roof materials.
Another Kumasi experiment points toward possible future development. In 1970 a prototype timber construction was built, utilizing prefabricated stressed-skin plywood roof panels 4 inches (10 centimetres) thick, filled with impregnated wood planer shavings and covered with bitumen and stone chips. Testing had not been completed at the time of writing, but preliminary results show little if any heat gain through the roof. The cost of this complete roof system, in which the bottom plywood skin formed a finished ceiling, was slightly less than a sheet aluminium roof without a ceiling.
Although timber roofs are not easily ignited and do not present a fire hazard like thatch or other plant materials, they will burn in a serious fire. They are not effective, therefore, in preventing the spread of flames between buildings, and their use in urban areas must be controlled, although not necessarily prevented, by building codes which place limits on flammability and flame spread. For housing of lower density and in rural areas roofs of this type appear to have good potential.
WALLS IN RELATION TO THE ROOF
Generous roof overhangs are important for several reasons. When combined with proper orientation to the sun they can be designed to cast shade on exterior walls, on wall openings and on the adjacent ground surface. They reduce sky glare and protect walls and openings from rain. The importance of these contributions to comfort and maintenance make it evident that the exterior walls should be considered not as independent elements, but rather in their relationship to the roof.
From a constructional point of view the hollow walls produced with conventional wood studs create problems in the tropics. They provide ideal breeding places for insects and vermin, they may retain moisture which causes rot, and they do not perform well in the case of fire unless special and difficult fire-stopping precautions are taken. Their insulating value is less important than in cold climates. For these reasons, a better case can be made for walls of a single thickness of timber or timber composites or plywood, and research should be directed toward alternatives which will satisfy psychological and practical demands for security when using these materials.
TRADITIONAL RAISED FLOORS
Wood-frame floor construction which is raised substantially off the ground to form a platform is traditional in many hot-humid areas. It permits air circulation through, around and under the living areas, at the same time increasing security and privacy, and is particularly suitable for lightweight timber construction. Apart from this remarkably satisfactory regional expression, choice of ground floor material has less to do with thermal than with constructional considerations. Either suspended wood-framed floors, concrete ground slabs, or stabilized earth can be used. Wood-framed floors should be raised at least high enough to protect from dampness and termites. If concrete or earth floors are used, critical details will occur at the bottom of wooden walls or supporting posts, which must be designed to stand free of surface water and ground moisture.
In addition to its thermal advantages, other properties of timber contribute to its suitability for housing construction. Unlike concrete and other masonry materials, timber is strong in relation to its weight, more easily and cheaply handled and transported, and works well with simple hand tools. Properly constructed timber buildings will withstand the pressure of storm winds and, being lightweight and resilient, have a greater capacity to withstand earthquake shocks than structures built of massive materials, unless costly precautions are taken with the latter. Light timber construction lends itself to flexible planning, to the production of prefabricated components and therefore to industrialized building methods.
The main technical problems with the use of timber lie in the area of preservation: against the attack of insects and fungus, against the effects of weather, and against fire.
The use of metal or concrete shields for protection against termites is common practice. Chemical treatment of the soil at foundations has been used successfully alone or in conjunction with mechanical barriers, but its effective life may be uncertain. Where subterranean termites abound, emphasis has traditionally been placed on the selection of the most resistant species of timber for construction. While the concern is legitimate, this emphasis has had unfortunate effects on the development of timber construction. The most resistant species are not often those which are most suitable from a constructional point of view; they are apt to be unnecessarily dense and strong for housing, therefore difficult to work with, and they may well have undesirable seasoning characteristics. Most tropical species have been tested for chemical permeability and effectiveness of treatment. There are tropical timbers suitable for construction which are either sufficiently resistant to termite attack with little or no preservative treatment necessary or are nonresistant with good permeability characteristics. Preference, therefore, should be for species with good constructional characteristics and reasonable natural resistance to termites (which may be increased artificially if conditions warrant the extra expense) and according to availability and cost. Since green timber is especially attractive to insects the use of seasoned timber should be mandatory.
VARIATION IN DURABILITY OF TROPICAL TIMBERS
The term "natural durability" of timber refers to its ability to resist fungal attack. There is a wide variation in durability of tropical timbers. Although fungal and insect attack are two separate problems, preservative treatments are usually sought which guard against both. Generally speaking this is possible, but it should be kept in mind that a single preservative which provides complete immunity against all forms of fungal and insect attack as yet does not exist. Decay from fungal attack can occur only if the moisture content of the wood is over 20 percent, within a temperature range of approximately 40-100°F (4.5-38°C), and with an adequate supply of oxygen. While these conditions are more likely to occur simultaneously in the tropics than elsewhere, it is quite possible to obtain an initial moisture content of less than 20 percent by means of air seasoning. Variations in building usage and conditions of service sometimes make it difficult to guarantee that moisture content will remain below 20 percent in all parts of the building at all times. Because of this, architectural detailing which prevents the collection and retention of moisture at joints and connexions is particularly important in the tropics. If there is any possibility that the moisture content will rise above 20 percent for a sustained period of time additional precautions such as preservative treatment should be taken.
Like all natural materials, timber is subject to change in appearance with age and exposure to weather. Weathering of timber is caused by the mechanical breakdown of the surface as the result of exposure to wind, rain and sunlight. Not all species weather in the same way, and in some weathering makes the surface more susceptible to wood-rotting fungi. This is a secondary condition, however, and weathering can and usually does occur without decay. The chief effect is the loss of a fresh-cut appearance. Many tropical timbers, including those with excellent constructional characteristics, weather very well, taking on a pleasant patina with exposure. However, public opinion generally objects to this phenomenon, an attitude which is more appropriate to synthetic materials. The solution to this problem lies in education, and experience suggests that this will take time and considerable effort.
Weathering may be controlled in two ways. One is by the application of a mechanical barrier to the surface, such as paint or varnish, in an attempt to prevent weathering entirely. Widespread acceptance of this solution is deplorable, especially in the tropics where intense radiation and fungi are natural enemies of paint and varnish. Furthermore, "hard" finishes are expensive and require constant maintenance in near-perfect condition to ensure the effectiveness of the mechanical barrier. This is difficult because when surface breaks and imperfections occur they are hard to detect. This creates a false sense of security; in fact the film may increase the likelihood of decay by trapping moisture under the surface, thereby maintaining the moisture content of the outer layers high enough to encourage the progressive growth of fungi. Finally the hard finishes, especially paint, conceal the natural beauty of timber.
An alternative to paint and varnish is the use of chemical preservatives which impart toxic properties to timber, at the same time reducing mechanical breakdown of the surface. These are the so-called "natural" finishes. They permit the timber to age naturally but control the effects of weathering. Preservatives of this type may be colour-less, or colour may be added to obtain a staining effect. Periodic reapplication is necessary-the frequency will vary with the exposure - but preservative finishes are much less costly and require far less skill to apply than do paints and varnishes.
Wood is a combustible material, but is not easily ignited. When heated to about 250°C it will decompose, producing inflammable gas and charcoal. When these gases are produced in sufficient quantity and are ignited, combustion results. This further heats the wood, producing more gases which ignite, and the fire is kept going until extinguished or the wood is consumed. As timber increases in dimension it loses its ability to support combustion by itself. Formation of charcoal on the surfaces of timber acts as an insulating barrier against the source of heat and this in turn retards the formation of combustible gases. In this way timber is "self-fireproofing," and timber of large dimension often survives when the rest of a building is completely gutted by fire.
There is no such thing as a fireproof building - only varying degrees of fire resistance. This is because the behaviour of a building in a fire is less a matter of the structure than of its contents and function. Housing provides opportunities for fire to start; contents and furnishings, more flammable than the timber structure, contribute to its support and spread. It is debatable whether the danger of fire is less or greater in the tropics than in cold climates. The traditional use of charcoal or open wood fires for cooking in the tropics is a factor which must be taken into account at the planning stage and when selecting materials. Design objectives should be to reduce the flammability of construction as much as possible so as to permit egress in the event of an uncontrollable fire, and to limit the spread of flames both within and between buildings. The latter is particularly important in urban areas where municipal fire control services are still lacking. These objectives must be met by a combination of sound space planning, the choice of appropriate materials, good construction detailing, and the enforcement of building codes which place reasonable limits on the spread of fire. Hollow elements which result from conventional stud wall construction act as natural flues in the case of fire, thereby increasing the spread of flames within a building. The case for solid timber building elements, for this and other reasons, has been made earlier.
The designer should not hesitate to depart from timber where a more satisfactory solution will be obtained with other materials. For example, cement plaster on some form of lath will provide moisture and fire resistance in bath or cooking areas. In attached housing, masonry party walls may be the most practical method of limiting the spread of fire and providing acoustic privacy between units, thus legitimizing the use of timber as the principal building material.
FIRE RETARDANTS
The use of fire-retarding chemicals may be considered. The wood must be thoroughly impregnated for them to be effective. In areas where this service is commonly provided cost has not proved to be particularly high, but this does not yet apply to developing tropical areas. Chemicals exist which are both wood preservatives and fire retardants, and it can be assumed that the permeability of a species by wood preservatives holds for fire-retardant chemicals too. However, little research has been done on the effectiveness of these on the various tropical species, nor have economic studies been conducted.
It was stated earlier that the major impediments to the use of timber are not technical problems, but rather those of cost and supply. In spite of the general abundance of timber in the tropics its cost remains high for several reasons. Tropical timber-producing countries depend upon log exports as a source of foreign exchange. Extraction and conversion are thus directed at export rather than domestic markets. Species and grades felled are those most in demand on the world market, while others are often ignored or wasted. Such selective extraction results in a low yield per hectare and this is costly and wasteful of forest resources. If extraction were less selective the cost of all timber produced would tend to be lowered and resources would be better utilized. In most areas there are certain secondary species well suited to building construction which are extracted inconsistently if at all. Unfortunately their utilization is not a simple matter. For one thing, they may not be uniformly available. Many timber mills would have difficulty expanding their operations to include the processing of additional species.
DETERRENTS TO EXPANSION
In the developing countries there is frequently a large foreign exchange component in the timber industry, caused by costly equipment and imported management personnel. Increased earnings from expansion and improvement of the industry will therefore be reduced by the extent to which such a component exists.
Two other factors are particularly serious deterrents to growth and improvement of the industry. One is the problem of transport, which can cause serious delays in the movement of forest products from source to market areas. The other is the policy which often places taxes on imported equipment as well as on exported products. The combined effect of such factors is to reduce output and net income as well as the incentive to modernize and rationalize the industry. In spite of these problems, and particularly in view of the depletion rate of tropical forest resources, it would be in the interest of all concerned with the industry-governments, extractors, sawmillers, and consumers-if policy were to develop maximum exploitation of the secondary species. A major step in this direction would be the creation of demand by an expanded building industry utilizing timber.
Other problems related to the supply of a usable product are those of seasoning, and standardization of quality and dimension. Most tropical timber is sold for local construction shortly after conversion, with a moisture content frequently over the fibre-saturation point-that is, about 25 to 30 percent of the dry weight. Such wood shrinks and warps as it approaches the equilibrium moisture content where it approximates the humidity of the surrounding atmosphere. Green wood is generally inferior in most strength properties to dry wood and is much more susceptible to insects and fungal attack.
The marketing of green timber undoubtedly has contributed in a major way to the poor reputation timber may have in some tropical areas for building construction. There are several reasons for this practice. Sawmilling yards are often concerned mainly with export contracts. After conversion and selection of contract grades, overhead costs of storing the off-cut timber are avoided by marketing it as quickly as possible. Small operators simply may not have the space for drying yards. There may be widespread disregard of the importance of using well-seasoned timber or, conversely, misdirected emphasis on the kiln-drying of timber which requires costly equipment and scarce technical skill. Fortunately it is not necessary to kiln-dry tropical timber for most constructional purposes, though the moisture control afforded by kiln-drying is important for furniture and some built-in wooden fixtures. Many tropical species suitable for general construction possess good air-drying characteristics, and it is quite possible to reduce moisture content to below 20 percent by air-drying, which renders it safe from rot. The main problem with air-drying is the space and time consumed; it takes up to four times as long to air-dry as it does to kiln-dry. However, if sufficient quantities of good constructional timber were made available for the local market, drying time could be programmed with little effect on the cost.
Uniform grading rules are the basis for marketing timber. They establish for the producer the needs of the building industry according to different uses, and they indicate to the consumer what he can reasonably expect to be available. The formulation of grading rules based on local supplies and needs has been retarded in the tropics. Lack of this form of quality control has been another factor contributing to the mistrust of timber as a building material. Without uniform grading rules it is very difficult for the architect to specify timber according to purpose or stress requirements, or to have confidence that specifications can or will be observed. This situation will continue until such time as producers and consumers agree to workable, written rules for those species suitable for local construction. The problem of grading is complicated in the tropics by the great diversity of species. However, relatively recent Australian timber development experience, from which has evolved the practice of assigning species to a series of strength groups, promises to help simplify the grading problem and is receiving considerable attention in developing tropical areas.
Dimensional standardization has also been retarded in tropical areas, partly due to lack of local demand and partly to the orientation of the industry to export contracts for timber which is resawn to overseas requirements after shipment. Cross-sectional dimensions and lengths available locally are therefore inconsistent. Standard sizes which reflect the most common and rational needs of the building designer often do not exist. Contractors are forced to instal additional plant and equipment to produce large quantities of material from a variety of rough sizes and lengths. Resawing, off-cutting, and planing to dimension under these conditions are wasteful and inefficient. On smaller contracts non-specified sizes are often substituted for reasons of availability, expediency, or economy, causing unfortunate changes in building detail. Greater standardization at the sawmill would avoid many of these problems.
The lack of a timber building tradition does not necessarily mean a shortage of woodworking skills, for timber window frames, and doors, louvres and their frames are common components of most buildings. Carpenters and joiners are available; a much more serious problem is the lack of supervisory personnel to organize and manage their efforts. When this lack is combined with problems of quality control and supply, construction overhead contributes heavily to building costs.
In one sense, a low state of timber development can be regarded as an asset. With no traditional methods to overcome, with labour less dedicated to obsolete construction methods, and with an opportunity to write building codes based on reasonable performance standards rather than direct specification or restriction of materials, an opportunity exists for regional design and construction methods whose goals are better performance and cost reduction.
In areas with a dependable supply of raw materials and a continuing demand, prefabrication can increase building efficiency and reduce cost below on-site construction. It achieves these gains in two ways. First, it brings the worker out of the elements into the workshop, organizes and supervises his efforts, keeps him constantly supplied with materials, and makes less demand on skill and responsibility. Second, it significantly reduces problems of quality control and supply by directing the flow of raw materials to a central location rather than to scattered building sites. These problems are more severe in developing countries, and their solution holds proportionately greater potential for increasing efficiency and lowering cost than in the more industrialized countries. Successful prefabrication in timber requires neither sophisticated nor highly mechanized techniques. The greatest savings are those achieved initially under workshop conditions with a rationalized system of supply. A factory for the production of prefabricated timber building components can in fact be a highly labour-intensive operation, permitting low initial plant and equipment cost, and allowing for progressive industrialization as required by a developing housing market.
The prefabricated timber housing constructed in the summer of 1970 by the Department of Housing and Planning Research in Kumasi (3) was an experiment in part directed at testing the prevalent idea that timber cannot compete on a cost basis with concrete for construction. Machinery was kept to a minimum, and glue-nailed components from which the building frame could be assembled on the site were built by hand, utilizing simple wooden jigs to ensure dimensional uniformity. The house was all timber with the exception of foundation posts, waterproofing and flashing of roof panels, and hardware and service installations. Accurate records kept of time and materials consumed showed that the completed unit cost $3.60 per square foot ($38 per square metre). Contractual costs would raise this figure to roughly $5 per square foot ($54 per square metre). In this particular experiment the objective had not been the most economical plan form, otherwise unit costs could have been reduced by about 25 percent, and further economies will result from the experience gained on this project. Agency-built housing in Ghana utilizing concrete now costs $6 to $7 per square foot (about $64 to $76 per square metre). Thus one conclusion from this project was that timber can compete with or better the cost of other materials when rationalized methods of production and assembly are used in construction.
When housing production includes the interrelated aspects of land, finance, design, production and marketing, a "housing system" may be said to exist. A "building system" is that part of the total housing system concerned with design, production and assembly of units. Prefabrication itself is not system building, although it is an essential part of all building systems. Systems which require the transportation and hoisting of large components would not be appropriate in most developing areas. The most feasible are those based on modular panels or structural components easily handled by a few men or with simple hoisting equipment. These systems also have potential value for self-help and mutual-aid methods of assembly and for the progressive completion of core housing, keyed to serviced-site schemes. Capital costs are low compared with more highly industrialized systems, since labour-intensive prefabrication and progressive industrialization are possible.
Design and construction are inextricably related in system building as the designer must be concerned not only with the end product but also with production and assembly methods. Under these conditions the building process will influence design, and there must be complete and continuous collaboration between the design, manufacturing and assembly phases for the total process to operate effectively. It is important that this collaboration be reflected in the research and development of system building.
System building will eventually make a major contribution to the housing needs of developing countries, with timber materials forming the basis of a variety of lightweight, easily erected and thermally effective systems. However, before the building process can industrialize to that extent, solutions must be found to problems which exist all along the continuum of processes, from forest science to construction, which make up the timber building industry. This can only be accomplished effectively by the coordination of in-country research and development efforts, directed not only at timber design and construction methods, but more importantly toward the rationalized supply of a standardized raw material.
(1) KOENIGSBERGER, OTTO & LYNN, ROBERT. 1965 Roofs in the warm humid tropics. London, The Architectural Association.
(2) Forest Research Nursery of the Forest Products Research Institute, Council for Scientific and Industrial Research, Kumasi, Ghana.
(3) Industrialized Timber Housing. A Joint Research and Development Project sponsored by the Department of Housing and Planning Research, University of Science and Technology, Kumasi, Ghana, and the International Development Technology Center, Washington University, St. Louis, Missouri, U.S.A.
