Developing a Methodology for Investigating the Design of Award Winning Industrial

Anthony Holness & Bob Giddings

he aim of this research project is to Theme: Workspace Design II designer's method and the quality T identify possible common themes This paper reports on research cur• of the design proposal and, in the design methods used to generate rently being conducted to investi• • the form and quality of industrial a selection of exemplary industrial buil• gate the methodological structure of buildings. dings. Once identified, the next stage the design of industrial buildings. A of the research will be to establish the comparative analysis of a selection of The research project extends existing nature of these common approaches exemplary industrial buildings is be• studies in the area of general human and determine the influence they have problem-solving, by using the find• ing used to investigate the relation• both on the design process, as a whole, ings of previous studies to analyse and ship between the design process and and the quality of the building. For explain a specific example of problem- the quality of the design proposal as this purpose, four areas for comparison solving, namely the design of indust• realised. The evaluation of any reali• across the case-studies have been pro• rial buildings regarded as the best of posed. They are: sed design remains as tangible evi• their building-type. Figure i illustrates dence of the success or otherwise of how this research project relates to • the structure of the design process the design process that generated it. previous studies in this field of know• itself and the mechanisms utilised Therefore, to develop a conceptual ledge and extends the hierarchy of re• by the designers of each building to framework linking design methods search into design activity as a parti• generate and develop their design and building quality should contri• cular incident of the general pheno• proposal, bute to a greater understanding of menon of design. • concepts of what constitutes quality how the designer's interpretation of a Modern architecture generally and in the design of industrial buildings, design problem influences the struc• industrial architecture in particular, and the use of value-judgement in ture of the design process. The effec• has been thought of as (Richards 1950, design decision-making ', tiveness of the design process in pro• p. 14) deriving; • cause and effect in design; the na• ducing industrial buildings of high Its nature from the acceptance by archi• ture of the relationship between the quality can also be evaluated. tects of the principle that the process

HOLNESS & GIDDINGS: ... AWARD WINNING INDUSTRIAL ARCHITECTURE 79 THE RELATIONSHIP HIERARCHY BETWEEN THIS RESEARCH INVESTIGATION AND PREVIOUS STUDIES

1. Part of the general area of investigation into various HUMAN PROBLEM-SOLVING (e.g. Newell & Simon, 1972) aspects of cognitive activity.

DESIGN PROBLEM-SOLVING 2. Design as a specific example of problem-solving (e.g. Simon, 1973) activity.

3 At this level the study of design problem-solving mechanisms is placed into various contexts.

ARCHITECTURAL DESIGN DESIGN COMPUTER SOFTWARE DESIGN (e.g. Akin, 1978) (e.g. Stauffer & Uliman, 1991) (e.g. Malholra, et al 1987)

Industrial Building Design At the level of a particular building type, investigations can attempt to compare particular sets of design problem-solving strategies with the assessment of the proposed solution from a number of different aspects • quality attainment and quality measurement, • technical resolution, Building E>esign • information usage, etc. This is because the nature of the building type has a significant influence over how Leisure Building Design each of these aspects of a design are established and their relative importance Housing Design

Figure 1.

of designing a building begins wirh a This would suggest that the original teenth centuries saw the increase in close analysis of the needs it is to serve. nature of the industrial building design the number and importance of indust• process was not entirely based on deve• rial buildings. Prior to the mid nine• However, industrial building has a long loping forms logically derived from teenth century, it was engineers who history; some of the earliest examples the building's functional requirements, lead the way in "uncompromisingly being medieval granaries which drew but was in part based on a process of functional" design. Again Brockman heavily on the forms and details of other making analogies with existing buil• (1974, p. 20) has rationalised this po• contemporary building-types. With ding types. Broadbent (1988, pp. 30- sition by saying; this use of existing building-types as 35) has termed this "analogic design"; The services engineers were much in precedents, Brockman (1974, p. 15) and occurs when the form and detail demand for one very solid reason; commented that; of an established building archetype once the industrial revolution had got It is not, therefore, surprising that the (i.e. church) is used as the basis for the under way and steam became the forms and structures of the grear spans design of a new building type (i.e. normal source of power, it was the of medieval churches, supported by granaries). The designer makes an engineer who was employed to prov• stout walls, where rhe tendency to over• analogous connection between some ide the engines which produced it turn under the thrust of a timber roof specific aspect of the function of one and also the machinery it served. In was countered by buttressing, were with the other. mosr cases rhese engineers knew very adopted by rhe builders of rhese early The rapid industrialisation that took well how to provide rhe structures warehouses. place during the eighteenth and nine• which supported the long runs of

80 NORDISK ARKITEKTURFORSKNING 1996:4 machines, floor byfloor, an d the brick man hours, lead time and the like, or lastic in its lack of relation to the real and masonry walls and timber roofs a better product. In these terms a met• world, its concerns and motivations. which enclosed them. hod is only better than another (what• Of the few studies that have been car• ever it may be) if it produces a better It appears that from an early stage in the ried out in the area of quality assessment building or environment. The concept Industrial Revolution, an understan• in building design, Murta (1991) has "better" involves judgement - better ding of the potential of the new machine found some evidence to suggest that a than something else - and is related to technology, together with the know• non-contingent relationship does exist goals and values as well as knowledge. ledge of how to utilise new materials to between method and design solution provide suitable internal spaces to One reason for the lack of any sustained quality. He concluded that (Murta accommodate this technology was an attention being given to this question, 1991, p. 799); essential part of the design process. and implicit in Rapport's statement buildings of quality have a design However, such an approach to the design above, is that building design quality procedure which differs from the of industrial buildings was usually assessment is as complex and proble• conventional plan of work, particu• adopted by engineers. It was not before matic a question as that of the nature larly in the way in which basic crea• the mid nineteenth century that archi• of the design process itself2. If the ass• tive work is applied to the problem. tects began to adopt similar principles essment of design quality was mentio• in designing industrial buildings. Up ned at all then, not unexpectedly, it However, this relationship may not be to that point , when involved was in terms of quantifiable perfor• a simple linear causal one, as Murta in the design of industrial buildings, mance measures. Indeed, in response (1991, p. 801) goes on to imply; were pre-occupied with the application to criticism made about the lack of In the context of motivation an essen• of various styles and the aesthetic expres• interest in the evaluation of the quality tial ingredient is knowledge and under• sion derived from them, rather than of completed buildings, Moore (1970 standing of what constitutes quality, the pure functional requirements of p. Xll) conceded that the papers presen• and the pre-familiarisation with the . ted at the First International Confe• inherent attributes of quality should rence of the Design Group, took a Design Quality Assessment narrow behaviourist approach to the form part of the first stages in any in Design Methodology question of quality assessment. They methodology which is intended to produce fine buildings. The question of the evaluation of design focused exclusively on the measurable methods is one which has rarely been performance attributes of building If Murta is right, then ideas about what confronted by design methods resear• quality. However, the reflections of constitutes quality in building design 3 chers. It appears to have been taken as many design researchers affirm that does play an important role in design an act of faith that adopting systematic it had always been their primary objec• method formation, whether the desig• procedures would inevitably lead to bet• tive to improve the quality of designed ner is conscious of this or not. There• ter design solutions; but how are better artefacts and not just the efficiency of fore, a detailed description of recurring designed products to be judged? the process. They could also readily quality themes could certainly contri• Rapoport (1969, p.136) was among acknowledge the interpretation made bute a new dimension to our under• the first to draw attention to this by Rapoport (1969, p. 136), in his cri• standing of how quality achievement question, when he wrote; ticism of design methods research that; relates to various methodological approaches to building design. Nor has it been demonstrated that the In much discussion on methods, the new methods work better. We need, quality of the architecture or physical Design Methodology of course, to define what we mean by environment is nor discussed at all. in Industria! Building Design "better". We can look at this in two The methodology seems to become The study of industtial buildings has ways; increased efficiency in terms of an end in itself, sometimes almost scho• largely been confined within studies

HOLNESS S GIDDINGS: ... AWARD WINNING INDUSTRIAL ARCHITECTURE 81 of architectural design history and nology of machine production and and non-measurable quality attributes aesthetics or in relation to social and the new materials which came with it. is available at the start of the design economic developments. Symes (1984, This would suggest that the impact of process, the designer's own concepts p. 276) has also pointed out that; this new machine technology on the of quality must play a decisive role in design process was limited to develop• shaping the design process, and ultima• An important difficulty, both for ing and evaluating the detailed measur• tely the quality of the final building. rheory and research seems to be that able functional aspects of the design. As mentioned above, functional con• there is no agreement on the relative These would include: increasing the siderations have predominantly deter• importance of economic facrors and maximum span and allowing greater mined the success or otherwise of cultural qualifies in determining de• loadings of the floors at the same time; industrial building design. Therefore sign decisions. providing fire proof , and guidance on the design of industrial One would naturally assume that for integrating the building structure buildings, not surprisingly, has been industrial buildings the former would with the new technology of building concerned with the quantifiable aspects predominate. However, if decisions in services. The design of the building of the design task4. design depend crucially on values and envelope however, remained largely a The designer's position in determi• concepts of quality, then the situation matter of the application of suitable ning quality and where its attributes becomes problematic once more. Mas- design dogmas, if the design process lie in the design of industrial buildings, ser (1982) highlights this problem by was predominantly -led. is further reinforced when the basis of introducing the concept of "espoused In contrast, Kammerer (in Acker- decision-making about industtial theories". Symes (1984, p.277) cites an man 1991, p. 41) characterises the de• buildings is considered. A study con• example; sign process of contemporary indust• ducted by the Cranfield School of rial buildings by saying; Management (1980) found that most economic rationality, a theory which industrial clients base their decisions architects or potential clients may use Industrial buildings become signifi• about industrial buildings on measur• to justify decisions or choices which cant when an attempt is made to able requirements. This was the case an independent observer mighr have control and unify rhe rotal design even when difficult to quantify attri• suggested were made on intuitive or process, when the design of a larger butes, like flexibility, may be just as other irrational grounds. Alternatively span structure and rhe components important. Despite all these difficul• it is possible of course that those mak• used in its construction are all pulled ties, the relarionship between building- ing design choices might sometimes together into a recognisable and type, concepts of quality and the de• believe they are following cultural understandable whole. sign process needs to be clearly defined prejudices abour design quality when before an attempt to explain the way in fact they are using an economic The emphasis here is on the way a de• design quality in industrial architec• logic to which rhey do not admit. signer organises the process of design, and suggests a link between design tural design is achieved. During the early period of the Indust• method and quality of design solution. rial Revolution, in machine At the earliest stages of the design pro• Methodological Issues production and materials technology cess there is usually only a limited in the Study of Industrial played a major role in determining amount of explicit information on Architectural Design the structural and constructional deve• how quality is to be assessed. Any in• In another study of buildings for lopment of industrial building design. formation on quality assessment is industrial production, Manning However, it was not until the end of usually contained within the quanti• (1966A) implied that theoretical pro• the nineteenth century that the archi• fiable requirements made by the client. positions are significantly more decisive tecture of industrial buildings began As no project consensus on the appro• in determining industrial building de• to be more expressive of the new tech• priate relationship between measurable sign than the manufacturing process

82 NORDISK ARKITEKTURFORSKNING 1996:4 that the building is to accommodate. tise in industrial building design and of classes of events retrospectively. The Such a proposition, if confirmed, could compare it with the reports they give case-study method allows the connec• have a major effect on the way design of the actual design process followed tion between the data and its contexts methods (as realised in the design of during a specific project. The concep• to be maintained and locates the ana• industrial buildings) are perceived and tual model (01 petsonal construct) and lytical framework within the setting the role of various types of knowledge the method for obtaining it (repertory of the particular case-study. Other rea• used in the design process. This would grid technique) are discussed fully by sons for adopting this method include: undermine one of the central axiom Fransella and Bannister (1977). As to of modern architectural design that the question of its appropriateness to • the need to understand a specific the form of a building be determined the identification of commonality of phenomenon in great detail within by its functional requirements. As a approach among a group of individu• the context of the phenomenon, consequence the knowledge base on als, research conducted by Latta and • a large amount of information can which this interpretation of modern Swigger (1992) confirmed its effective• be obtained from a few examples, architecture was founded would have ness in identifying commonly held to be re-evaluated. This causal link expert knowledge constructs 5. From • restricted research where there are between building function and archi• their analysis of the sets of constructs only a small number of cases but tectural form is one that has domina• obtained in their experiment, Latta and the results have significance beyond ted much of modern architectural de• Swigger (1992, p. 126) concluded that; the cases studied. sign thinking. Richards (1958 p. 14) As the process of design problem-sol• The fact that one of these sets of con• makes this point quite strongly and goes ving itself still remains relatively structs was a common set provided on to say that modern architecture; poorly understood, this research credence to the idea that the grid can project is fundamentally descriptive has as its object the fulfilment of such be used to represent a consensus of and interpretive in nature. Its aims are needs as logically and economically as knowledge about concepts which lie in essence, to formulate a general possible, by taking full advantage of in the realm of communal knowledge. descriptive framework of a particular the means and materials available. Its Also, as a necessary consequence of design task by attempting to uncover aesthetic character is created as part of investigating quality achievement in tecurring patterns and/or themes in the the same process; nor does the exercise the design of industrial buildings, this approach to the design of industrial at every stage of personal preferences research project has to work backwards buildings, by a sample of designers. and aesthetic taste, which is inherent from the identification of suitable Only until a descriptive framework in all architectural activity, in any way award-winning industrial buildings, has been described can an attempt be invalidate the functional basis of mo• through to the design process which made to hypothesise an explanation. dern architecture. produced them in the first place. There• Figure 2 oudines the research method Therefore, a key methodological con• fore the study can only be based on a formulated for this study. cern will be how to devise a data col• retrospective analysis of the process lection and analytical mechanism itself, as reported by the designers Objectives which will allow a causal link to emerge, involved, during semi-structured of the Case-Study Analysis while safe-guarding against designers interview sessions. Thus, the research The initial data obtained during the post-rationalisation and selective me• process will essentially be a qualitative early stages of the research will be used mory which may distort the research study comprising a series of case studies. to formulate a descriptive framework findings. One way of attempting to This research technique has been used of the relationship between intuitive ensure validity in the information in recent investigations into design and and rational procedures in the design 6 elicited from designers is to obtain construction and is particularly suited of industrial buildings of high quality. to comparative analytical evaluations their conceptual models of their exper• The analysis process therefore, will

HOLNESS S GIDDINGS: ... AWARD WINNING INDUSTRIAL ARCHITECTURE 83 Diagram of the Research Method.

Data 1, Designers as Primary Source

1. Essential Characteristics of the Designers Objectives, Research Project Criteria and Concepts of quality. QUALITATIVE DATA > primarily qualitative Description of the Design Process 1 based on case comparisons i necessarily descriptive, Data 2, Published Text as Secondary Source < hypothesis generating, • Previous Experience in the Design i 2. Selection Criteria: of Industrial Buildings, THE ANALYTIC ' buildings built after 1970, • Existing articles and publications by PROCESS 1 recipient of a major design the Designers on their approach to award, Design and their Buildings, ' excludes design competition • Documents, sketches, models maps, 1 Data Reduction Summ.tr> of interviews and the schemes photographs coding of qaesliomiaires to identify themes and classify 3. Unit of Analysis: concepts • the individual design process Data 3; Repertory Grid formed from 1 & 2 that generated the case-study 2 Data Display: building. Use the information obtained from • matrices \,m to IdeirafS regularities in ihe lypes oi generating strategies .iml the previous data collection • network diagrams exercises to define the components concepis.'definitiojis used. Also the 4. Data Collection Method: •charts patterns and trends wifhut each of design knowledge semi-structured interviews, case study Establish how the designers classify repertory grids their design knowledge. 3 Conclusions & Verification: Devise a matrix relating the (Using cross-case comparisons to verify knowledge components (elements) similarities and recurring themes) with the means of classifying them (constructs), by a rating scale. Develop descriptive explanations of the design process and propositions about causal relationships between the methods The designers complete the matrix adopted by architects involved in the design of industrial by scoring each element against a set building :md ih. qimiii-. of the buildings Ihemsehvs of constructs. Figure 3 have to assume a "symmetry of choice" in the design processes that form the design quality, in current research in between the two approaches. A desig• case studies. If a characteristic pattern design methods. The way intuitive ner could be either scientifically ratio• in the way designers structure the de• and systematic approaches are nal or intuitive, or alternate between sign process, when designing industrial balanced in the design process may be the two approaches at different stages buildings of high quality, can be determined by prior concepts of throughout the process. Indeed, re• detected, and can also be shown to be quality. Conversely, the pattern of the search conducted by Stauffer et al absent in the design process adopted design process may just reflect (1987) showed that designers do vary by designers of other industrial build• programmatic and pragmatic between the two approaches as they ings, then this can be used to support concerns about design resourcing. progress through the process. How• the proposition that particular Therefore during the analysis of the ever, Stauffer et al (1987) made no relationships in the design process are data from each case-study, an attempt attempt to establish a link between this more likely to achieve high quality will be made to establish the pattern of the relationship between the pattern and the assessment of the quali• solutions than others in the design of industrial buildings. quality of the solution, pattern of the ty of the design solutions. Therefore, design process and concepts of quality an additional task of the data analysis There exists a significant gap in in the design of industrial buildings process will be to identify recurring establishing causal relationships (see figure 3). patterns of use of these two approaches between methods and concepts of

84 N0RDISK ARKITEKTURFORSKNING 1996:4 THE INTUITIVE/SYSTEMATIC BALANCE IN THE DESIGN PROCESS

The matrix is used to classify and analyse the interviews. Sections of the data which refer to the different tasks performed during the design process are used to infer what decision-making mechanisms were in operation at the time

communication

concept realisation

incisive/instrumental conjecturing

perspective on problem

speculation within frame of reference

establish frame of e.g. The Machine Aesthetic' as a principal concept provided the frame• reference work for the design process.

initial problem e.g. The rationalisation of the client's operations by« detailed analysis of the existing racking process interpretation.

DESIGN TASKS

SEARCH Intuitive (an even combination of both Systematic MECHANISM (experience/internal logic) approaches) (analytical/ external logic)

Figure 3

Case Study Example; tion submission, therefore reflecting went through to assess the Racking7 Plant for Greene the normal design situation for all the suitability of the design proposal, King Brewery, Bury St consultants involved. • to obtain the assessment criteria on Edmunds, UK Each interview consisted of a series the design of buildings to accommo• of open questions designed to elicit This section outlines one case-study date their production processes. which forms part of the data for the factual information and strategic des• investigation. A series of semi-struc• cription of the design process. This was For the architect: tured interviews were conducted with done by encouraging the respondents • to establish how the architects inter• the client, architect and building ser• to discuss at length their involvement preted the client's requirements to vices engineer to ascertain where the in the design development process. A form design priorities, specific set of objectives was identified major influences on the design concept • how they used these priorities to for each group of questions. These were: occurred during the design process. develop the solution, The building was selected as a suitable For the client: example of industrial building design • how the designers assessed the • to establish the priorities for the because it has received a number of suitability of the emerging solution building, design awards (see figure 4), and that during the design process, • to establish the process the client it had not been designed as a competi• • to evaluate the balance between intui-

HOLNESS & GIDDINGS: ... AWARD WINNING INDUSTRIAL ARCHITECTURE 85 BUILDING DESIGN AWARD SCHEMES

Building: Racking Plant for Green King Brewery;(Bury St. Edmunds, UK) Architects: Michael Hopkins and Partners

Award Area of Criteria for Method of Information Aim of Scheme Classes of Scheme Concern Assessment Assessment used Award Schemes for desiqn/environmental excellence RIBA Awards Architectural Design in Local Juries A summary of: To give public recognition to Regional & (1980) Design relation to: (architects and lay • client brief outstanding examples of (design) • purpose, members) • planning matters current architecture and National • surroundings, National Jury • construction thereby to achieve greater public appreciation of good • materials, (eminent architects) architectural design. • longevity. • photographs • visits Financial Times Major new buildings • architectural merit, Two eminent • key drawings to describe To encourage higher Main Award & that enhance both the • enhancement of the scheme, standards of architectural and Architecture at architects and one Commendations urban or rural the urban or rural • photographs, environmental design for Work Awards lay assessor. landscape of Britain landscape, • a written description of industrial and commercial (occasionally) (1980) and the lives of the • contribution to the the building covering the buildings, people who work in lives of the client's brief, town them. building users. planning context, materials and method of construction. Scheme for excellence in the use of materials Structural Steel The use of Structural • efficiency, Single Judging • drawings To recognise the high Single & Two Storey Steel • cost effectiveness standards of structural and Design Awards Panel (plans, sections etc.) Buildings • aesthetics architectural design attainable (1980) (ICE. RIBA. Government • innovation • photographs Organisation reps.) in the use of steel and its Buildings over Two • visits potential in terms of Storeys efficiency, cost effectiveness and innovation.

Figure 4

tive and rational approaches to the de• ber of key concepts in their approach (machine aesthetic for the total enclo• sign process for industrial buildings. to the design of the building. The deve• sure of all functional activities). This lopment of an understanding of the was used in two ways: For the building services engineer: racking process was significant in • how the engineering services priori• • as a generator of ideas and, determining: ties were identified and the implica• • as an evaluative mechanism to test tion for the design of the building's • that conceptually the industrial the appropriateness of the design enclosure, process was linear (a rationalisation proposal. of the previous process), • the process of developing the services A feature which appeared to permeate design with that of the building • and the identification of a concep• the design process was the establish• enclosure. tual design unit (the movement of ment of clear working definitions for casks through the racking process). From the responses obtained in each abstract concepts; for example effici• interview, an initial data display in the These were important because they ency in building design was defined as form of a conceptual matrix using formed the basis of a conceptual dia• using the minimum number of compo• descriprions to link design tasks to grammatic representation of the racking nents necessary to achieve the required conceptual units (codes) was under• process. The constant refinement of functional performance. However, taken, (see figure 5). This formed the this diagram was a feature of the archi• experience and professional judgement first step in the data analysis phase tects design process and mention of it was still reported as playing the decisive and is intended to identify the impor• often occurred during the interview. role in assessing the non-quantifiable tant themes within the case-study and Another important element of the aspects of the design proposal. The provide the basis for cross-case analysis. design process was the architect's expe• location of the building seemed to An initial analysis of the data indi• rience in designing buildings based on play a less prominent role in the design cated some interesting areas for further an on-going theme. This was the source process than either the architect's design analysis. The architects outlined a num• of the architect's principal concept theme or the racking process to be

86 NORDISK ARKITEKTURFORSKNING 1996:4 CONCEPTUAL MATRIX; DESCRIPTION OF TASK DETERMINANTS IN INDUSTRIAL BUILDING DESIGN Building; Racking Plant for Greene King Brewery, Bury St. Edmunds, Suffolk, UK. Architectural Practice; Michael Hopkins & Partners, London, UK. Date of design; 1978

DESIGN TASKS PROBLEM INTERPRETATION FORMULATING A SOLUTION SOLUTION EVALUATION STRUCTURING THE DESIGN (Influences on the interpre• (How the initial interpretation (The criteria for assessing the PROCESS tation of the problem and the was used to generate a poten• emerging design solution (The balance between the use CONCEPTUAL UNITS development of design tial solution.) during the design process.) of rational and intuitive methods priorities.) in industrial building design.)

Accommodation of the existing Client defined the manufac• Continuous interactive process The manufacturing process CLIENT INPUTS process rather than re-organi• turing process. with the client; 'feed-back'. fixed by the client before con• & INFLUENCES sing the process itself. Specific Drawings & diagrams used to Must provide secure and sidering the enclosure to dimensional and equipment analyse the process. suitable environmental accommodate it. requirements to be accommo• No design input from the conditions. Cost& Improvement to the racking dated in the new building. client, tneyjustsetthe environmental requirements process, as it relates to a buil• Also, required sensitivity & performance parameters for prohibited initial concept of a ding enclosure, was limited by modernityforthis historic the required building glass box to put the racking the late inclusion of the desig• provincial town. process 'on show'. ners to the team.

SITE HISTORY There were serious conser• Building raised off the ground, Allow for the expansion of the The designers used their & CONSTRAINTS vation considerations linked to to be above flood-plain; building to double the capa• experience gained from the chosen site. Space developed as part of the city, if required and to previous work to get a 'fee• restrictions on both the city architectural expression. maintain the site as a flood- zing' of the scale of a new centre and the chosen site. A Restricted site requiring: lorry plain. project in relation to its site. 100 year old manufacturing parking and turning circle. process which has prompted Also room for future building piece-meal development expansion. through-out the town centre.

The 'Machine Aesthetic' as an Making buildings like 'pro• Established a 'principle con• The concept of an 'Industrial DESIGNER GENERATED PRE- Aesthetic' informed and gave CONCERNS appropriate expression for ducts' & using -made cept' (Machine Aesthetic) industrial architecture. components. which is used as test criterion direction to the design process. Universal enclosure of func• The building as 'Umbrella' for for the design proposal. The It provided a starting point for tional spaces. Strong the manufacturing process. building seen as a continuous the design process. Intuition precedent in previous work: Look at other buildings with a 'extrusion' by the designers. used to conceive the initial building form, aesthetics and machine aesthetic, as The building is a 'simple form', ideaa, then a rigorous process use of materials. precedent. merely acting as an 'umbrella' to develop the functional for the racking process. rationale and refine the details was adopted. Rationalisation of client's pre• Originally conceived as a U- A building finely tailored to INDUSTRIAL PROC ESS (actual Allowance for future expan• sent process. The basis of the shaped process with entry and sion precluded U-shaped orga• the client's existing and conceptual descriptions) process is the movement of exit at the same point. Final nisation of the process in the manufacturing process v barrels along the design creates a physically lin• building. flexible loose-fitting space. manufacturing system. ear movement of the barrels Limiting the number of The building design is all about Technology of the manufac• through the process to reflect components to the minimum understanding the racking turing process in harmony their conceptual movement necessary for the functional process. with the technology of the than the existing contorted objectives to be achieved. new building enclosure. one. Quality established through the Aim for a consistency in the Materials used to exploit their Use of materials, detailing and MATERIALS finishes strongly influenced by & COMPONENTS appropriate use of materials, use of materials in the external mechanical and visual proper• based on their mechanical and envelope. ties fully. 'machine aesthetic' (concepts of appropriate philosophy. usage) visual properties. Efficiency in Extendibility in two modes Components able to do a the design achieved by either by adding extra bays or number of functional tasks; Efficiency in a design proposal; materials and components by an extrusion of the 'material efficiency'. 'everything must earn its keep' being able to perform more building's cross-section. Focus on the details & finishes i.e. must have a functional rea• than one function. construction & assembly process. son for being used. Different functional areas The quality of the design is in The ability to devise ideas QUALITY CRITERIA the sophistication and ap• intuitively which then become (Operational concepts and required varying levels of quality. propriateness of materials used, the basis for rational methods of assessment) detailing and finishes. Quality in functional elements of the design is a product of the use of building. Use of professional appropriate materials, building judgement to identify form and spatial planning In architectural opportunities relation to required per• then developing functional formance levels. benefits from then.

WORKSPACES Focusing attention and avai• The designers had to work (concepts on suitable lable resources on areas having within parameters set by the conditions for workers In the greatest visual and client, so they had no input in industrial buildings) functional impact, i.e. where the analysis of the racking pro• employees spend most time cess requirements. and therefore of most benefit.

Figure 5 87 accommodated. The site set up con• process. They emphasised the need to tinued in the design of the building straints which the architects responded commission architects who would services, where the services engineers to in a pragmatic rather than idealistic develop the design in detailed consul• emphasised the importance of formu• way, i.e. raising the building up on piles tation with them, and spoke positively lating a diagrammatic representation to allow flood water to pass underneath about the intensive analysis the desig• of the client's racking operation. This the building. The architects characte• ners gave to the detailed aspects of the representation had two functions: rised the design process as being a very racking process. This forced the client firstly, as a means of locating the ma• rigorous process of testing ideas from into a continual process of justifying jor building services loads along the both functional and constructional the requirements for the building on racking process operation line; and perspectives. The design qualities of the entirely rational grounds. This pro• secondly as a means of maintaining building were suggested to have been cess of rational justification was itself the close integration between the derived from this highly rational pro• continued in the clients design building services and building fabric cess. The client's input into the design evaluation criteria which included: design within one overall concept. process was crucial to the quality of This enabled the engineers to devise a the building. Greene King outlined a • establishing a direct operational number of technically feasible design number of important aspects of their flow through the building, options within the objectives set by racking process which had a major • the effective compartmentalising of the design concept. These concepts effect in determining the buildings the activities to separate the dirty were: form. These included the following: returned empty casks from the clean • the building services systems as casks ready for packaging and • the client was developing a number expressions of the dynamics of the dispatch, of in the storage and racking process, and packaging of their products and • simultaneous loading and unloading • to expose the building services therefore required a building which operations, within the building to directly would assist in the implementation • and the flexibility to introduce communicate their functions. of these innovations, innovations at any point along the The main feature of the building ser• • they had developed a precise descrip• packaging process; i.e. the most vices design was that it was considered tion of their operational flow of acti• effective point to introduce^zwzwgy as an integral part of the overall design vities by utilising a series of detailed — a traditional methodological term of the building with both the architect studies of their existing process ope• for clearing fine particles from the and engineers sharing in the develop• rations, beers. ment of the design concept right from • the client was prompt in making The initial analysis of the client's the start. decisions about specific operations description of the role rhey played in which had a direct bearing on the the design process reinforced the rigo• Conclusions design, e.g. loading of lorries was to rous procedures outlined by the archi• The analysis of the case-study outlined be done by the drivers themselves. tects. By confining their input to above has suggested that there were a This required a loading bay that purely measurable functional matters, number of key components in the de• would allow each driver to select Greene King provided an explicir and sign process for this building. Perhaps the individual delivery items. detailed basis for the building's func• the most significant was the dual role Greene King saw the specification of tional assessment without necessarily of the design concept. Not only did it the building's performance and func• pre-determining its form or architec• provide the generating mechanism for tional requirements as an ongoing ture. the building's form, but significantly process which they allowed to evolve The description of a highly ratio• it acted as a means for effective commu• during the early stages of the design nalised design process was also con• nication between client, architect and

88 NORDISK ARKITEKTURFORSKNING 1996:4 engineer; helping to co-ordinate their Notes inputs into the process. The functional 1. For a detailed discussion on the relevance analysis of the requirements of the rac• of value judgement in determining ac• king process was set within the context tions and explaining patterns of events of this conceptual overview of the de• during the design process see Lera (1982). 2. For further information see Burt 1978, sign aims. Although functional conside• Seymour & Low 1990, Low 1993. rations played a significant role in deter• 3. For example see Alexander 1984, Archer the building's form, their main 1994. contribution to the process was as the 4. See Manning 1966A, Manning 1966B, mechanism for establishing the criteria Manning 1966C. 5. A construct is a personal model of the against which the developing design world or a particular aspect of it which a Anthony Holness proposal was constantly tested. The person uses to predict and organise their next stage of the investigation will be behaviour (see Fransella 8t Bannister The authors of the article are Anthony to describe the network of relationships 1977)- Holness, Doktoral student, and Bob within the design process itself and 6. See Sinister 1995. Giddings, Ph.D., Principal Lecturer 7. The term Racking refers to the process of then use this as a basis for the cross-case and Research Coordinator at the filling the casks and bottles with beer. comparisons, where conclusions about Department of Built Environment, the nature of high quality industrial University of Northumbria, building design can be obtained. Newcastle Upon Tyne, U K.

HOLNESS & GIDDINGS: ... AWARD WINNING INDUSTRIAL ARCHITECTURE 89 Rererence ACKERMAN, K. (1981) Building for American Society for Information A. and Brandon, P. (eds). Manage• , Watermark Publications Science, Vol. 43 (2) pp. 115-129. ment, Quality and Economics in (UK) Ltd. LERA, S. (1982) "Design Process and Building, Chapman Hall, London, ALEXANDER, C. (1984) "The State of the Value Theory" in Langdon, R. & UK pp. 793-804. Art in Design Methods" in Cross, GregoryS. (eds). The Proceedings of RAPOPORT, A. (1969) "Facts and Mo• N. (ed) Developments in Design the Evaluation Section of an Interna• dels" in Broadbent, B. & Ward, Methodology, John Wiley & Son, tional Conference on Design Policy A. (eds) Design Methods in Archi• Chichester, UK pp. 309-316. held at the Royal College of Art, Lon• tecture. Lund Humphries, London, u 2 ARCHER, L. B. (1984) "Whatever don 20-23 J fy !9$ The Design UK pp. 136-149. Became of Design Methodology" Council, London, pp.57—58. RICHARDS, J. M. (1950) The Functional in Cross, N. (ed) Developments in Low, S. P. (1993) "The Rationalization Tradition in Early Industrial Buil• Design Methodology, John Wiley of Quality in the Construction In• dings, The Architectural Press Ltd., & Son, Chichester, UK pp. 347- dustry; Some Empirical Findings", London, UK 349- in and SEYMOUR, D. & Low, S. A. (1990) "The BROADBENT, G. (1988) Design in Archi• Economics, Vol.11 (4) pp. 147-259. Quality Debate", in Construction tecture; Architecture and the Hu• MANNING, P. (1966A) "Industrial Pro• Management and Economics, Vol. 8 man Sciences. David Fulton duction Buildings; A General Ap• (1) pp. 13-29 Publishers, London, UK. proach ro Design" m Architects SINISTER, S. (1945) "Case Study Metho• BROCKMAN, H. A. N. (1974) The Bri• Journal06 April 1966, pp. 911-923. dology for Construction Manage• tish Architect in Industry 1841— MANNING, P. (1966B) "Industrial Pro• ment Research" in Thorpe, A. (ed) 1940. George Allen & Unwin duction Spaces, Briefing Guide" Association of Researchers in Con• Ltd., London, UK. in Architects Journal13 April 1966, struction Management, Proceedings BURT, M. E. (1978) A Survey of Quality pp. 953-9 84- of the Eleventh Annual Conference and Value in Building, Building MANNING, P. (1966C) "Industrial Pro• (18-20 September 1995) pp. 21-32. Research Establishment, Wat• duction Buildings, Information STAUFFER, L. A., ULLMAN, D. G. & ford, UK. Sheets" in Architects Journal, 08 DIETTERICH, T. G. (1987) "Proto•

CRANFIELD SCHOOL OF MANAGEMENT June 1966, pp. 1445—1456. col Analysis of Mechanical Engine• (1980) Industrial Decision Making MASSER, I. (1982) "The Analysis of Plan• ering Design"; in Proceedings of in Britain - Building New Indust• ning Processes: Some Methodolo• the International Conference on rialand Commercial Premises, Fi• gical Considerations" in Environ• Engineering Design (17-20 August nancial Times, London, UK. ment and PlanningB., Vol. 9, pp. 5—14. 1987) Boston, M.A. pp. 74-85. FRANSELLA, F. & BANNISTER, D. (1977) A MOORE, G. T. (ed) (1970) Emerging SYMES, M. (1984) "Case Studies of Ar• Manual for Repertory Grid Technique Methods in Environmental Design chitecture for Industry", in Academic Press, London, UK. and Planning, MIT Press, MA. Brandon, P. S. & Powell, A. (eds). LÄTTA, G. F. & SwiGGER, K. (1992) MURTA, K. H. (1991) "An Investigation Quality and Profit in Building De• "Validation of the Repertory into Procedures leading to Produc• sign, Spons, London, UK pp. Grid for use in Modelling tion of Quality Assutance in Build• 275-288. Knowledge" in Journal ofthe ings of High Quality"; in Bezelga,

90 NORDISK ARKITEKTURFORSKNING 1 996:4