Participative Design: User Driven Facility Creation
Mark James Tweddle
2452480A Thesis Submitted for the Degree of
MSc Operations and Supply Chain Management & MSc Technology and Operations Management
December 2013
I
Abstract
II
Table of Contents
1.1 Introduction ... 1 2.1 Theoretical Background ... 3 2.2 Facility Design ... 3 2.3 Social ‘Value’ ... 5 2.4 Participative design ... 6 3.1 Methodology ... 9 3.2 Problem statement ... 10 3.3 Sampling ... 11 3.4 Field Work ... 12 3.5 Data Analysis ... 16 4.1 Analysis ... 18 4.2 Results ... 184.3 Analysis & Discussion ... 22
5.1 Conclusion ... 28 References ... 32 Appendix ... 37
Figures
2.1 Construction Phases ... 5 4.1 PD Workshop ... 19 4.2 Final Renderings ... 211
Chapter 1
1.1 Introduction
The construction industry is often the subject of criticism regarding its performance. The process is difficult to manage; it involves thousands of decisions, with numerous interdependencies, under a highly uncertain environment (Tzortzopoulos et al. 2009). Initiatives have mainly focused on improving the efficiency of the construction process. The thinking being that improvements in operational efficiency should result in better layouts, adjacencies, business processes and information systems (Tzortzopoulos et al. 2009). However, the goal of construction is not only to provide a product efficiently, but to provide an appropriate product, which is of greatest value to the end-users. Focus is therefore shifting upstream in the process to the design stage (Jorgsen 2006). This is the most ambiguous stage of the process and requires complex integration of design, engineering and social considerations; it can also be seen as the most important with 80% of costs being fixed during design (Crawford & Di Benedetto 2008).
Traditional approaches to facility design are generally technically focused, considering only major human factors. The goal is usually quantitative i.e. to maximise profits or minimise circulation. However buildings are an important factor in enhancing the quality of our lives, creating the environment for living, working and socialising (Christiansson et al. 2008). This is exemplified within the majority of current research that focuses on the healthcare sector. Here facility design has focused on streamlining technical functions, such as reducing the number of beds required. Therefore social factors related to the quality of the environment have generally been neglected (Francis et al. 1999). These factors are vital as Ulrich et al. (2004) found in more than 600 studies linking aspects of the built environment to staff effectiveness and patient healing (Ulrich et al. 2004).
2 useful during the design phase, so that an optimal environment for end-users to participate and respond can be created. This should facilitate the translation of the needs of the end-users into design solutions (Christiansson et al. 2008).
Participatory design (PD) has been proposed as a solution for overcoming the limitations of traditional design, in which end-users are consulted at a distance and with limited contact to the designers (Greer & Lei 2012). PD treats users as experts and attempts to bring their implicit knowledge into the design process (Broberg & Edwards 2012). PD pursues active involvement from end-users in the design process, trying to bridge the gap with designers to avoid
generalization of needs (Park 2012). It therefore seems that when trying to incorporate the non-functional ‘soft’ factors ambience, colour, texture, light etc.), valued by end-users in facility design, PD can provide a solution. PD has the potential to build upon traditional ‘paper’ based techniques of information exchange, as well as embracing the latest developments in Building Information Modelling which allows all stakeholders to interact through one data rich Computer Aided Design (CAD) model(Steel et al. 2010).The following paper will aim to further address this area, looking at how PD can be applied to the design phase of facilities; thus ensuring that there is an adequate translation of the end-users needs. It is here when architects can exploit the social value of good design to enhance the environment for both functionality and enjoyment (D.
Thomson et al. 2003). The research will therefore focus on answering whether PD can accurately portray ‘soft’ factors valued by end-users and which PD method provides the best platform to do this with the following questions:
“How can Participative Design involve end-users in the facility design process to accurately portray the ‘soft’ factors valued by them?”
3
Chapter 2
2.1 Theoretical Background
In this section the background surrounding the research will be explored; thus building a picture of what has previously been done, what is being done now and why it is important to provide further research. Firstly facility design will be explored to identify current themes and the problems it faces. Secondly the notion of creating socio-technical designs in order to add end-user value will be discussed. This will establish what is meant by ‘value’ and why ‘soft’ factors are important in facility design. Participative design will then be reviewed in terms of how it is
currently being utilised and how research and practice indicate its applicability to this field. Finally the papers goals will be set, with relation to the specific research questions that are to be
answered.
2.2 Facility Design
The goal of facility design, to provide maximum value to end-users, is a well-established fact (D. Thomson et al. 2003; Macmillan 2006; Saxon 2002). Whilst the majority of research refers to ‘value’, there is a distinct lack of definition. Womack & Jones (2003) is one of the few examples of an explicit definition, found in their publication regarding lean and its applicability to the construction industry (amongst others):
“The critical starting point for lean thinking is value. Value can only be defined by the end-user. And it is only meaningful when expressed in terms of a specific product which meets the end-users needs at a specific price at a specific time.” (Womack & Jones 2003)
Thus it appears that end-users values must be identified for each facility. However as Jorgsen (2006) stated this causes a problem of only reflecting the values at one moment in time and may not necessarily reflect the true values of the ultimate end-users. The importance of representing value is further emphasised by Saxon 2002): “What society does not want from its built
4 The importance of the facility design stage has been further advocated by Freire Alarco n
(2002). They state that the design phase (Figure 1) is precisely when end-users values are transformed into technical specifications. Figure 1 shows a simplified view of the construction process, firstly there is a need for a new facility, it is then designed from concepts through to detailing (Jorgsen 2006). It is here where the majority of decisions are made and thus where the greatest influence on the design can be had. The final stages are concerned with technical building requirements and physical construction of the facility (Jorgsen 2006). However it seems that although the importance of the design stage is acknowledged, in practice the process is littered with improvisation and chaos, rather than being a controlled process. As many authors point out this leads to poor communication, inadequate documentation, lack of co-ordination and eratic choices (Lauri Koskela 2000; D. Thomson et al. 2003; Ballard 2008). Current practices within facility design therefore seem to fall short in addressing all the complexities of the design phase and restricting the overall quality of the output (Tzortzopoulos et al. 2009). However, it can be argued that by nature the facility design process is iterative and chaotic and can therefore justify some of the problems highlighted, although this hardly seems satisfactory Freire Alarco n 2002) .
5 Figure 1: Construction Phases (Jorgsen 2006)
2.3 Social ‘Value’
The concept of providing value to the end-user in facility design is not new; one of the earliest known attempts to define value was by the Roman architect Marcus Vitruvius Pollio. He stated that the value of architecture is in its strength, utility, beauty, commodity and delight, Vitruvius also mentions the importance of considering ‘the nature of the place’(Vitruvius Pollio 1914). This highlights the range and complexity involved in creating a facility design that all end-users will value. More recently Thomson et al (2003) has suggested there is a more basic positive and negative relationship. It states that value within facility design is derived from the trade-off between benefits (i.e. what you get) and sacrifices (i.e. what you put in). This view can be seen to be over simplified and as pointed out by Mills et al (2009) every stakeholder within the design has a different emphasis, values and role-frames; alignment needs to be created in order to deliver greatest value to the end-user.
6 such as design quality and the voice of the users and thus led to a new focus on these issues as their importance became apparent (Whyte et al. 2003).
Therefore aligning all stakeholders within facility design and integrating their knowledge has been found to be a critical success factor (Yu et al. 2007). Gorse & Emmitt (2007) had similar findings determining that there must be meaningful interaction between parties and use of effective communication. However due to facility design containing many conflicting stakeholders, often meeting for the first time, they conclude this is a challenging task. Currently boundaries and objectives are most often set through meetings between multiple actors in the project team, and face-to-face meetings are preferred (Gorse & Emmitt 2007), this current approach seems to neglect involving end-users and their values within the design process.
End-user values within facility design have been mainly restricted to healthcare settings. Here studies unanimously show a positive relationship between patients health and wellbeing and the built environment (Corben 2013; BBH 2011). We can therefore surmise that to incorporate what end-users value can be seen to produce more appropriate facility designs. Architects Fosters + Partners (2009) took this approach, creating the healthcare facility CircleBath, which put end-users values at the centre of the design. Everyone involved, whether patient or porter, was regarded as a partner in the design, with the common goal of promoting patient well-being. They have found that a facility design which incorporates the values of end-users improves recovery times and contributes to better outcomes for patients, whilst providing a more enjoyable
workplace for medical staff (Fosters + Partners 2009). The values held by end-users are therefore starting to receive more serious attention, improvement initiatives are looking towards socio-technical designs rather than just socio-technical efficiencies (Tzortzopoulos et al. 2009). However literature around creating value in facility design, outside of healthcare, is still extremely sparse. Although ‘softer’ factors can be less critical outside of healthcare settings they still have an impact on people’s quality of life and are thus worth investigating (Christiansson et al. 2008).
2.4 Participative design
Participative design (PD) has been identified by many as the solution to involving the values of end-users in the design process (Broberg & Edwards 2012; Park 2012; Christiansson et al. 2008). PD aims to empower the end-users; it believes those affected by design should influence the design. The theory originated in Scandinavia during the 70’s when the benefits of
7 However, what is meant by PD and how it is conducted is rather ambiguous within literature, mostly due to the wide variety of contexts it appears in. Christiansson et al. (2008) stated that the PD process is highly iterative and many tools can be used depending on the circumstances, e.g. interviews, workshops, collaborative storytelling, commented virtual building model walkthroughs, and observations. Similarly, Greer & Lei (2012) stated that a multitude of techniques can be used when implementing PD, dependent upon the context. These methods aim to identify and collect data from end-users in a more scientific way so their needs can be more accurately translated into design objectives (Greer & Lei 2012).
Although some authors also stipulate that it is more than just data collection and the PD process needs to involve end-users in a co-creation scenario (Pallot et al. 2010; Park 2012). This puts end-users at the centre of the design process, not just as observed subjects, but rather as a design partner for co-creating value. At the extreme this can be observed with Wikipedia, which is a mass collaboration of users collectively creating content for the benefit of the wider society (Pallot et al. 2010).
The methods which are generally used and tested involve using physical models and interviews to engage the end users in design (Gorse & Emmitt 2007; Jorgsen 2006; Broberg & Edwards 2012). Whilst this has shown positive progression in the field, methods involving CAD systems have received relatively little attention. Especially when there is a definite trend in design and
construction away from traditional ‘paper’ based techniques towards 3D CAD systems, it seems appropriate to explore this opportunity in the context of user participation (Steel et al. 2010). The approach of using data rich 3D modelling is known as Building Information Modelling (BIM), and is defined as “a methodology to manage the essential building design and project data in digital format throughout the building’s lifecycle” (Penttila 2006), it is expected to form the primary means of collaboration between stakeholders in the future (Steel et al. 2010). BIM offers an open system of collaboration within one model and therefore has already been cited for its possibility to automate analysis done during the design phase whilst allowing collaborative input from all stakeholders, thus having great efficiency benefits (Azhar et al. 2007). In fact the UK government plans to make BIM compulsory on all public building projects by 2016 (CIOB 2011). It therefore seems logical to investigate how end-users could interact within this platform opposed to using ‘paper’ based techniques.
8 outcome. This shift of end-users, to centrally within the design process, is also advocated by Larson et al. (2004), as it can increase the speed and effectiveness of facility design by cutting out the trial and error iterations between designers and end-users. They go on to say that PD provides end-users with the tools to design and develop their own facility design. An advocated approach to this is a Lean tool called 3P, which considers people, products and processes. Its aim is to provide a participative approach that takes into account all stakeholders in order to deliver the greatest value in the final facility (Coletta 2012).
PD therefore seems an appropriate tool for integrating social values of end-users into facility designs. There is however a lack of specific research about using PD within facility design, most papers focusing on product development. Yet the designing of facilities is, in many ways, more challenging than product design, since the users of the facility have a wide range of age,
interests, skills, and cognitive ability. In addition facilities are a complex mix of less tangible ‘soft’ factors such as colour, light, form, and materials (Larson et al. 2004). Nevertheless the new stream of literature within healthcare facility design has started to address this area and shows its applicability and worth. Thus the rest of this paper will investigate how participative design should be conducted in order that it can include the ‘soft’ factors valued by end users within facility designs. This leads us to the papers research questions:
“How can Participative Design involve end-users in the facility design process to accurately portray the ‘soft’ factors valued by them?”
9
Chapter 3
3.1 Methodology
This section will set out the methods to be used during the research; these methods should ensure that the evidence obtained enables the research questions to be answered as unambiguously as possible. The paper will first discuss the type of research and the problem being addressed, before continuing onto how the data will be collected and finally analysed. In order to answer the research questions knowledge must be acquired, that is to say that the answer required cannot be judged in black and white but requires a judgement on its usefulness. The knowledge problem of this research is concerned with resolving the difference between the way end-users experience the ‘world’ and the way they would like to experience the
‘world’(Wieringa 2007).
The overall research will therefore have a large element of subjectivity regarding its analysis and usefulness. The subjective manner of the topic combined with the small sample sizes utilized within PD research (Frost & Warren 2000; Nasralden & Mandeli 2011; Bruseberg & McDonagh-Philp 2000) means that qualitative based research provides the best platform. It offers the ability for greater flexibility when exploring social paradigms; ideas can be clarified and developed when dealing with subjective topics. The major disadvantages are that the research is more time consuming and subjective than using a quantitative approach. Although these negatives are somewhat negated through the use of a small sample size and the already subjective nature of the research topic (Bell 2005). As Silverman (2004) stated the use of purely mathematical logic in quantitative methods can overlook common sense relationships in the data, where social paradigms are being explored. However the majority of PD studies use a mixed method approach to their research (Frost & Warren 2000; Nasralden & Mandeli 2011; Bruseberg & McDonagh-Philp 2000; Mobach 2008) and as Bryman & E. Bell (2007) identified qualitative methods are not without elements of quantification. There is often a ‘quasi-quantification’ in qualitative research through the classification terms such as many, frequently, rarely often and some. Although not as precise as quantification they make allusions to quantities and are useful in giving ideas of the relative frequencies within the research (Bryman & E. Bell 2007).
10 (Bruseberg & McDonagh-Philp 2000; Nasralden & Mandeli 2011). As used by Bruseberg & McDonagh-Philp (2000) and Nasralden & Mandeli (2011) the data will be qualitatively evaluated and collected through structured interview questions, which aim to provoke discussion leading to more open ended questions. This has the advantage of firstly constructing an understanding of each respondents attitudes, evaluation and preferences before obtaining richer data and evidence of their opinions (Nasralden & Mandeli 2011). It is generally agreed that combining more objective and subjective indicators when studying user-environment relationships is
beneficial (Kahana et al. 2003; Carmona et al. 2003; Kallus 2001). Thus by using both objective and subjective indicators a greater understanding of how certain ‘soft’ factors affect the end users can be gained (Kallus 2001; B. Kahana et al. 2003).
The Methodology will follow the structure advocated by Oppenheim (1992) and Bryman & Bell (2007) when carrying out qualitative research:
1. Problem statement- formulation of the research questions; 2. Sampling- choice of the relevant site(s) and subjects; 3. Field work- collection of the relevant data;
4. Data Analysis- relating the findings to previous work, whilst drawing conclusions and interpretations;
5. Conclusion- summary of the findings and conclusions that address the research questions.
3.2 Problem statement
11 techniques (Steel et al. 2010). It is consequently important that the research investigates both methods in relation to PD.
The research questions aim to address these gaps in knowledge and are formulated as thus:
“How can Participative Design involve end-users in the facility design process to accurately portray the ‘soft’ factors valued by them?” and “Do traditional methods or CAD based methods for Participative Design provide the best platform for end-users to interact with the design and express their valued ‘soft’ factors?”. From previous research it is clear that PD creates more
valued facilities than the traditional design process (Fosters + Partners 2009; Park 2012; Pallot et al. 2010), the research is therefore not addressing PD against the traditional design process. More precisely the research will be concerned with how to utilize PD itself to incorporate the ‘soft’ factors valued by the end-users. ‘Soft’ factors and value are intrinsically vague and undoubtedly change from context to context. There will therefore be no attempt to produce a dictionary definition as this could stifle this and future research via limiting categorizations (Silverman 2006).
The goals of the research are to discover if PD will in fact improve facility design to bridge the void between designers and end-users, avoiding over the wall design (Hanyu 2000; Chesbrough 2003); can it accurately portray not only the hard factors (layout etc.), but more ambiguous ‘soft’ factors that create a valued design(Fosters + Partners 2009). Furthermore can advances in technology, such as BIM, be incorporated into PD and even enhance it. CAD systems have been touted to have the potential to be highly engaging and interactive, enhancing the creative process (Mobach 2008). However during the design phases of construction it is still common practice to use paper based techniques and some argue that CAD may constrain creative design as it limits actions (Burkhardt et al. 2008). The research will therefore test both methods in relation to portraying ‘soft’ factors.
3.3 Sampling
Firstly ,the context is to be established, this highlights what is to be influenced, then an artefact can be designed which aims to change the current state (Wieringa et al. 2009). An artefact influences or improves the problem context by contributing to the end-users goals. The context for the research must consist of a facility within which people interact as end-users, the majority of PD research to this date has entailed using healthcare facilities and patients (Cottam &
12 day-to-day basis by its design. The goal of the students is to achieve a University qualification and their critical success factors CSF’s) include an appropriate working environment. If the facility fails to satisfy the critical success factors it is likely that the facility will provide limitations to end-users in attaining their goals (Christiansson et al. 2008). However there are always trade-offs between CSF’s, for example between the wishes of the students and conforming to the Universities system or cost. The research population will be identified through theoretical sampling, this has the advantage of being able to select individuals whom promise the greatest insights into the research (Flick 2009). The case chosen should therefore be likely to match the research area in order to extend the emergent theory. As Bryman & E. Bell (2007) suggest, there is no definitive answer on sample sizes, previous PD studies show a range of 2-6 participants in the design workshops (Bonner & Porter 2000; Bruseberg & McDonagh-Philp 2000; Mobach 2008). This research will therefore recruit a sample within this range, largely due to the
practicality of participants working on one design which necessitates small groups in order to be viable. There are of course issues of representativeness of the population within such a small sample (Bryman & E. Bell 2007). However this is the generally accepted number within PD workshops, most likely due to the large amount of time and resources such workshops require.
3.4 Field Work
The structure of the field work is based upon previous PD studies by Mobach (2008), Frost & Warren (2000) and Broberg & Edwards (2012) which contain three key elements:
1. Context Investigation;
2. Participative Design Workshops; 3. Design Validation and Exploration. 3.4.1 Context Investigation
Within the context the problem will be investigated in greater depth. This will take the form of verifying end-users goals and CSF’s using primary data collected from a structured pre-interview. The research population will be Newcastle University business students as they represent the end-users within the investigative context. Emails will be sent to the relevant students outlining the research; respondents will be taken on a first come first serve basis to eliminate selection bias. All participants will be gathered in one room where a short introduction will be given addressing the main questions: ‘why are we here?’, ‘what are we doing’ and ‘what the value in
design is’. This will set the scene for the research, clarify terms and start them thinking about
13 sent prior to the research. This will allow participants to query information and/or withdraw from the research; hence the information will be both sent by email and verbally presented to all participants.
Participants will then be taken to complete a pre-interview (located in the appendix) which verifies the student’s goals, success factors and values. The interview will be structured, because the aims and workings of the study can be more convincingly explained than through a questionnaire and respondents can elaborate upon what they have said for greater clarity and detail
(Oppenheim 1992). The structured interview enables categories to be largely pre-determined; hence during the interview attention can be on the respondent rather than trying to record answers. Categorization, so responses can be simply circled, allows the outcomes to be easily and quickly recorded, summarised and analysed (Bell 2005). Interviews will follow the protocols set out by Oppenheim (1992), that is that the overall goal is to gather information from factual responses to attitudes and feelings. A structured interview must ensure an equality of stimulus to every participant, the interview will therefore always contain the same questions, worded in the same way, in identical sequence within the same setting (Oppenheim 1992). However it cannot be completely robotic like a questionnaire, the great advantage of an interview is its ability to gain responses closer to the truth. This is achieved through building a good rapport with the
respondent and allowing a degree of flexibility between participants to ensure every question means the same to every respondent (Oppenheim 1992). It is at the interviewers discretion to balance this, ensuring communication with the respondent is limited to the script so there is a limitation on biases introduced and each participant receives a stimulus equivalence (Oppenheim 1992). Interview questions will be prepared in advance with prompts to ensure everything is covered (Bell 2005). Categorized questions will be often followed by probing questions to get a greater clarity from the responses. Coding will be largely carried out through categorization of answers, probing questions will be written down verbatim at the time, this is possible due to the relatively short answers needed (Oppenheim 1992). In order to control biases, which can cause unreliable data in interview collection, the interviewer will always endeavour to stick to the interview structure, maintain a good rapport throughout the interview and record answers as verbatim (Oppenheim 1992). In order to ensure the interview schedule was feasible, the questions were clear, there was nothing offensive and that responses could be accurately recorded a pilot test was carried out amongst colleagues, any defects were then altered (Nasralden & Mandeli 2011; Bell 2005).
14 acceptance as the standard, it is therefore readily understandable. Added to this the scale does not force participants into definitive answers but allows a degree of agreement as well as neutral feelings, which is important when dealing with subjective topics (Bell 2005). Furthermore due to the time and cost constraints on the research, the Likert scale offers a quick and inexpensive method of data collection, which can be administered face to face (Bryman & E. Bell 2007). However when exploring social attitudes the Likert scale can be seen as one-dimensional, limiting respondents to set responses (Oppenheim 1992). It is thus important to follow up the scale in the interview with more open ended questions to gain a richer insight, as exploited in preceding PD studies (Mobach 2008; Nasralden & Mandeli 2011; Frost & Warren 2000). As used by Bruseberg & McDonagh-Philp (2000), structured questions provoked thought and discussion about the ‘soft’ factors and what is valued in the facility before clarifying the answers in a more open ended question.
3.4.2 Participative Design Workshop
A PD workshop will be carried out. This will aim to identify how effectively the ‘soft’ factors valued by the end-users (identified in the pre-interview) can be integrated into a design. It will also help point out the current restrictions on the participative design process within facility design. The workshops are aimed at engaging the students in collaborative work to formulate a vision for a new facility in the University. The new facility in question will consist of two key spaces within the university (study room and computer room), using relatively small and familiar spaces will allow the workshops to explore these spaces in greater detail and focus on the ‘soft’ factors, within the relatively short time frame (Mobach 2008). There is no consensus on a defined list of ‘soft’ factors and they are heavily ambiguous and context dependent. Thus any ‘soft’ factor (that is to say non-functional element), which can be changed within the models, will be categorized into a ‘soft’ factor (natural Light, artificial light, textures, colour, furnishings, material type and
15 comparison to the current facilities. However it is important to keep the spaces neutral and simple so as not to imply any previous or personal design preferences, also because it can be confusing if too much information is presented at once (Mobach 2008). The spaces are situated without surroundings as they allow consistent representation between paper and digital formats (Frost & Warren 2000). Participants will have 1 hour to complete their designs, as this is feasible for both participant commitment and time to complete the designs as shown by previous studies (Broberg & Edwards 2012). Both methods will receive the same time limit to ensure a consistent comparison can be drawn between them. Each group should consist of 2-6 students to ensure there is a fair representation of the population within each group whilst still being able to collaborate effectively on relatively small spaces within the time constraints, as discussed in Sampling 3.3. All materials needed for the workshop will be provided, along with refreshments to ensure a good rapport is maintained during the research (Oppenheim 1992).
3.4.3 Design Validation & Exploration
The outcomes of the workshop will be converted by the researchers into virtual 3D images, as done previously by Broberg & Edwards (2012), this gives both designs an equal representation from which to be analysed and reduces biases to a particular representational method(Broberg & Edwards 2012). The rooms before the workshops are entirely neutral; hence any changes made can be easily recognised. It is of course a potential source of error when transferring data, thus extra care should be taken and all information double checked (Oppenheim 1992). Through providing spaces which are not overly complex and only require adaptation of the ‘soft’ elements it will eliminate ambiguity and variability when the workshops are converted into virtual images, any elements not addressed by the workshop should not be assumed and left neutral to eliminate biases. The images will then be presented back to the students for validation, along with a
representation of the current business school facility, for comparison to a standard facility which has not been PD designed. The current facilities were not reviewed or used as a design starting point, as this can limit participants to previous preconceptions and reduce creativity (Chesbrough 2003). Participants will be blinded to the origins of the three proposed space designs, so to reduce bias to a certain method.
16 Oppenheim (1992) identifies, gathering ratings to ranges which seem inappropriate serves to assess attitudes and feelings which are hard to define with labels. The use of both scales and follow up questions in the interview will allow a triangulation of data, thus adding to the reliability of the results obtained (Flick 2009). It will also help negate the disadvantage of categorizing the respondents feelings, which if done alone could lose the richness of the paradigm (Silverman 2006). Such responses are subject to the halo effect, where respondents are influenced by their general feelings and thus bias all answers accordingly (Oppenheim 1992). To reduce this bias within the interviews questions are deliberately formulated to encourage first responses as well as having a mix of scales to not allow respondents to become complacent (Oppenheim 1992). Furthermore as in the pre-interview ‘soft’ factors are ranked on a 5-point scale to assess whether pre-perceived important factors are met by the design i.e. does it meet the critical ‘soft’ factors identified? Each design will be shown in turn, at the appropriate question in the interview. These, as in the pre-interview, will be followed up with more probing questions to obtain a more detailed picture (Oppenheim 1992). Presenting images back to participants is a well-used method within PD studies (Mobach 2008; Frost & Warren 2000; Nasralden & Mandeli 2011) as they allow valuable direct observations to be made by the participants (Silverman 2006). Although as Flick (2009) recognizes the purposeful composition of images can lose a spaces expressiveness and feeling and is highly open to being manipulated by the researcher. The research will try to combat this by including two views of the spaces, which are identical for every design, as well as sticking to the protocol of only altering what participants changed and leaving the rest neutral. It is possible advancements in virtual reality, as used by Mobach (2008), can help towards creating a presentation closer to the reality of the spaces, however in this research there is neither the access or budget to access such techniques.
3.5 Data Analysis
As previously discussed the data is to be qualitatively analysed due to the combination of a small sample size and subjective nature of the topic (Silverman 2004). Similarly to Bruseberg &
McDonagh-Philp (2000), whom also evaluated the data qualitatively, the structured interview will be used as a basis which will facilitate the understanding of each end-users attitudes,
perceptions and preferences. Meanwhile the questions with greater flexibility will be utilised in obtaining greater details of the end-users opinions (Nasralden & Mandeli 2011).
17 factors in a post-interview, thus allowing a second form of analysis of how well ‘soft’ factors have been satisfied and whether PD has proved significantly better than the facilities already available. Furthermore the Affective Appraisal Scale (Hanyu 2000) will provide objective data on how participants perceive each design thus creating a reliability point for whether the design which satisfies the greater number of ‘soft’ factors is indeed the most favoured outcome by the end-users. By verifying the design proposals with the end-users in a post-interview, it will provide proof of its properties, validation that it satisfies the CSF’s and validation that the correct resources have been used (Wieringa 2008). The use of data triangulation by using multiple sources of appraisal within the interviews will increase the reliability of data collected (Karlsson 2009). It is also advocated by Kahana et al. (2003) to analyse using a mixed approach of more objective and subjective techniques in order to create a better understanding of the person-environment relationships. The small sample size chosen, whilst limiting the data available for analysis, allows a closer relationship with respondents both during the research and in any follow up work that is needed. There is a better understanding between researcher and participants, thus it serves to enhance the validity of inquiries in social settings (Crouch & McKenzie 2006).
18
Chapter 4
4.1 Analysis
This section will endeavour to firstly explain what happened during the fieldwork including the results and outcomes from the PD workshop and interviews. It will then analyse the results of the field work giving observations and interpretations of the data collected, with particular relation to work which has previously been done in the area. The goal of the overall discussion of this section is to be able to address the research questions as unambiguously as possible.
4.2 Results
Access was granted within the Newcastle University Business School, from which emails detailing the research were sent to students of three courses using the course group emails. This means it cannot be accurately gauged how many students received the research invitation; however it is assumed that there are sufficient students within each course to represent a sufficient number. Response rates were disappointingly low and as outlined by Bryman & Bell (2007) prompts were sent after the initial email. Due to time constraints on the project when the lower range of participants needed was reached (4), as identified through previous PD studies (Bonner & Porter 2000; Bruseberg & McDonagh-Philp 2000; Mobach 2008), the research was initiated. As per the research protocol, set out by Oppenheim (1992), all information required was included within the email along with incentives suggested, such as refreshments and conforming to the participants favoured time and place. Nevertheless the low response rates can be attributed mainly to two factors in this case: firstly the relatively large input required from participants including two separate meetings and secondly the fact that students had just commenced their courses and may not yet be using and/or checking their University email accounts. Greer & Lei (2012) similarly found that engaging participants in PD can be a major limiting factor if the topic is not critical and there are not ongoing incentives (e.g. monetary), it can be said that both are the case here. A room was booked within the Newcastle University Business School at a time convenient to the participants. The room offered all the facilities needed for the research and a quiet place where there would be no interruptions, thus ensuring a good rapport was created for the research (Oppenheim 1992). On arrival participants were given a presentation detailing the background to the research including key information on: ‘why are we here?’, ‘what are we doing’ and ‘what the
value in design is’. This set the scene for the research, clarified terms and allowed an opportunity
19
(located in the appendix) before being split into two groups for use of the CAD and model
techniques. After explaining the spaces, on which they would be working, it can be noted that the model group could immediately start whereas the CAD group required a basic explanation of how the software worked. The majority of previous studies using CAD systems, in PD, have used it mainly as a tool for presentation (Mobach 2008; Frost & Warren 2000), however continued technological advances mean it is becoming increasingly pronounced within the design stages of facilities (Azhar et al. 2007; CIOB 2011). The participants in this research were able to work independently after five minutes of introduction to the software, and as Penttila (2006) predicts; for future generations highly complex technology will be a normal part of their lives.
The process of the PD workshops, for both the CAD and model group, is shown in figure 4.1 below, from what was presented to what was completed:
Figure 4.1: PD Workshop: Pre, During and Post Design.
20 simply too time consuming. The CAD model proved easiest to represent as all the information was pre-stipulated by the participants, changes here were the addition of sunlight, shadows and realistic glazing. The same was true of the model; however this required the extra work of converting colours, textures, furnishings and material types into CAD representations. This is an area where researcher biases may have been introduced and certainly confirms the limitations of using traditional to CAD based methods used in this way (Flick 2009). Nonetheless by following the methods used by Frost & Warren (2000), this was limited as much as possible by leaving anything not specified by the participants white. Also through asking participants to write more specific instructions on the model if they felt they could not accurately represent them. This was also explained to the CAD group however they did not use any more specific notes in the model. The resulting images are shown below in figure 4.2, along with a representation of the current business school facility. The entirety of the rooms could be shown through two images from opposite corners of the room. In order to reduce participant biases to their design, resulting in a halo effect (Oppenheim 1992) the images were labelled: design ONE, design TWO and design THREE (CAD, model, current).
One week later, in the same time and place, participants were gathered for a second time and presented with the final renderings (Broberg & Edwards 2012). Then the post interviews were conducted on each participant’s perceptions of the designs, they followed the same structure as the pre-interviews and thus participants were easily and quickly able to grasp the concepts presented. The post interview (located in the appendix), contained a greater deal of open-ended questions, consequently the use of more prompts were needed to gain responses from
participants, especially when requiring both positive and negative reactions to the designs. This had two effects on the research, firstly it created greater departures from the interview structure therefore increasing the probability of biases being introduced by the interviewer (Oppenheim 1992). Secondly due to the decision to write answers verbatim during the interviews at this point, when the answers were relatively long, it was difficult to note down both what was being said whilst remaining engaged with the participant. This may have damaged the rapport with the interviewee and introduced error into the interview recordings (Oppenheim 1992). However, the majority of the interview could be recorded with pre-determined categories and these issues only affected a small proportion. Due to the iterative process of qualitative research (Silverman 2006) it was found necessary to return to participants after the research for additional information and clarification using further interview questions. This indeed confirms previous studies which require the use of more than 1 workshop in an evolving process to fully understand the
21 of their first design, thus the PD process cannot be seen as a linear progression but an iterative development towards a facility design (Park 2012).
Figure 4.2: Final Renderings
Design ONE (CAD)
Design TWO (Model)
22
4.3 Analysis & Discussion
4.3.1 Goals and Critical Success Factors
It was firstly important to establish and validate the participants’ goals and critical success factors within the chosen context, in order to ensure the correct narrative is constructed for the research (Wieringa 2007). To start with the participants overall goal of the facility was discovered; all participants referred to gaining both knowledge and experience. It should be noted at this point that to ensure participants confidentiality they have been given codes (P1, P2. P3 and P4), it is also relevant to state that all participants were studying internationally. This accounts for the greater emphasis on ‘experience’ in the goals of attending the facility. It also highlights a key problem found when researching PD, as previously identified by Broberg & Edwards (2012). It is difficult to always assemble a PD team which represents all the stakeholders being tested (Broberg & Edwards 2012). This research has focused upon end-users and thus students; however the facility has many other users which are not represented (academic staff, cleaning staff, maintenance staff etc.). This is clearly a limitation of the research; it would however be very difficult to assemble all end-users given both the “fictitious” nature of the research and the time constraints. Therefore as in the research of Broberg & Edwards (2012), the major end-users have been identified with which access is readily available.
Following this the participants were asked more specifically about their goals and preferences for rooms they would be designing. Regarding the computer room all participants reflected a need to study as their main goal and three out of four for a quiet space in order to do so. Furthermore when asked about the rooms most important aspects, the features expressed concerned its facilities (printers, computers, chairs etc.). The goals for the study room were very similar to the computer room, in that the participants stated the ability to study and focus in a quiet space. However when asked about the most important factors there are no mentions of facilities, instead the participants valued a relaxing, light and quiet space. Therefore the goal of each room is very similar, but both have conflicting success factors with which they will be assessed by the
participants. These findings are in alignment with the study by CABE (2005), which indicated that students valued three main aspects in education facilities: aiding study, creating an inspirational ambience and providing the correct facilities. These findings support the underpinnings of the research in the belief that high quality education requires good quality facilities (CABE 2005). Additionally, even though a facility may have an overall goal, every context within it can have differing success factors and thus there can never be a generalization for all situations within facility design.
4.3.2 ‘Soft’ Factors
As ‘soft’ factors are one of the key elements with which this research is trying to expand
23 reflected the intangible nature of ‘soft’ factors, responses alluding to them improving ambiences and feelings, without actually making concrete statements about their value. This was picked up through the common use of ‘I think’ and asking for assurance at the end of the responses. However when asked definitively if they are an important factor all respondents answered yes. It is important to the research that participants were motivated and enthusiastic towards PD in order to create commitment to a good design. Previous work has also found respondents to be positive in their approach towards PD (Mobach 2008). Indeed the CABE (2005) study found that 60% of students identified cosmetic and environmental features as being key to an educational facility. It can therefore be seen that end-users are aware of the value of these factors and thus are willing to help influence them to their benefit.
The research reviewed which ‘soft’ factors the participants valued most for both the computer and study room (ambience, material types, artificial lighting, natural light, textures, colours and furnishings). Regarding the computer room participants highly valued the majority of the ‘soft’ factors stipulated, rating them 3 or above on the 5point Likert scale, it is therefore more
beneficial
to state that natural light was consistently scored at 3 or below by all participants. For the study room similar results were found, in as much as participants had a varied spread of most valued factors, in this case no factor was scored below 3 on the scale. Interestingly the factors that were ranked first by the participants were the same for each room (Artificial light, Ambience, Furnishings and Material Types). This corresponds with the participant’s goals and most valued aspects of each room which were also near identical. Consequently participants seem togenerally value all the ‘soft’ factors within the designs and even though the rooms have different functions similar attributes are valued in each.
The perception of ‘soft’ factors was also obtained in relation to each of the three designs. It can then be seen if what was initially valued has in fact been translated into the designs via the PD workshop. For this section each participant will be analysed in turn:
24 failed to meet the soft factors which he valued whereas the model was a more accurate
representation.
For P2 furnishings were the most important ‘soft’ factor within the computer room, stating that he needed ‘good chairs for his bad back’. From the results, the CAD design ranked equal to or above the ‘soft’ factors identified as most important. P2 was also relatively satisfied with the model design although it did not meet his expectations for furnishings. The same is true of the study room; however both designs did not meet his expectations of furnishings and the model was found as unsatisfactory. It is important to note here a discrepancy, as P2 criticized the CAD design for its lack of furnishings (printers and places to sit), yet ranked his perceptions as very satisfied and greater than the model in both cases.
P3 was satisfied with all aspects of both the CAD and model computer room and was prompted in the interview for negative comments. The same was also true for the CAD design in the study room; conversely the study room was felt to be unsatisfactory in both colour and ambience within the model design. This was further elaborated as being due to lack of colour in comparison to the CAD design. Overall therefore the CAD effectively meets the values of P3, and whilst the model is satisfactory it fails to accurately portray all the factors valued by P3.
P4 regarded ambience as the most important ‘soft’ factor within the computer room, both CAD and model designs were found to be satisfactory for this factor. They also both ranked between neutral and satisfactory for every other factor. The study room however is found to be generally unsatisfactory on all points for both designs. This is reflected in the negative comments about the study room of the model having no colour and the lack of furnishings within the CAD design. Overall P4 sees both methods as relatively equal in meeting the valued ‘soft’ factors, although relatively good in the computer room and relatively bad in the study room.
In summary CAD generally proved more successful at representing the ‘soft’ factors which were initially perceived important by the participants. Nevertheless both methods had varying
25 paradigms. Greer & Lei (2012) identify that end-users often have the knowledge to improve designs and identify flaws, but that this must be coupled with expertise in the area in order to release the end-users knowledge to create feasible ideas. Hence the research indicates that the same is also true with the use of end-users within facility design, and end-users working
independently cannot always express their wishes to create feasible designs.
During the research the limitation of the halo effect has been observed, to which studies dealing with perceptions are highly susceptive (Oppenheim 1992). The halo effect of participants general feelings, towards the design they worked on, was combated by blinding the participants to the origins of the final renderings (Oppenheim 1992). Nonetheless in reality it appeared to prove relatively easy for participants to identify each design, P1, P2 and P3 all favouring the method they used. This can explain some discrepancies in the data gathered, between ratings given and what was said. That being said P4 was more balanced and showed positive tendencies to the opposite method.
Figure 4.3: Number of ‘Soft’ Factor Alterations made during each method
4.3.3 Design Perception
This section will try to identify the participant’s perceptions of the differing designs and the methods behind them. P2, P3 & P4 preferred the CAD design whereas P1 preferred the model based design. The affective appraisal scale (Hanyu 2000) used backed up the perceptions
0 2 4 6 8 10 12
Natural Light Artificial Lighting
Textures Colours Material Types Furnishings N o. o f C ha nges Soft Factor
'Soft' Factor Alterations
26 expressed by the participants. When more specifically questioned, about which room design would best allow them to meet their goals for that room, P1 and P3 chose the same designs as previously stated. Conversely P4 changed to the model design, which is not surprising considering the close satisfaction exhibited between the two methods. More interestingly P2 chose the
current design, which received overall low satisfaction levels regarding its ‘soft’ factors.
Nevertheless when probed to clarify this P2 stated that whilst it was not aesthetically pleasing it served its function well. This provokes the question form or function? Nasralden & Mandeli (2011) found that there was often conflict between the ‘softer’ values of form and the ‘harder’ elements of function. Nevertheless it is not in this researches remit to go into this debate, although it is worth noting that there is a tangible link between the two factors. This can often be conflicting or overriding, as seen in P2’s choice of the current design for purely functional reasons over aesthetic reasons. Further research exploring end-users designing both ‘hard’ and ‘soft’ factors in conjunction would be beneficial in the advancement of understanding in this area. Participants were invited to further elaborate on the reasons for their choices; all four mentioned colours as a deciding factor, two mentioned textures and furnishings and P1 made a reference to the light. This is similar to previous findings by Ulrich et al. (2004), whom discovered positive relationships with end-users to colour, artwork and light within facilities. It thus seems that there are certain factors which receive greater attention from end-users. Other factors, such as artificial lighting, which have been expressed as being important are not realised in the designs. This somewhat contradicts findings by Frost & Warren (2000), whom express CAD methods as
allowing end-users to comprehend things they would not have done with modelling methods. This research suggests that neither CAD nor modelling offer the ability for end-users to express all the factors important to them. Undeniably, as Frost & Warren (2000) also identify, PD helps users by providing methods through which they can express their ideas, but used in isolation does not convey all aspects valued by end-users. All in all however results show that both PD designs are favoured over the current rooms. This backs up the findings of Mobach (2008), and reiterates the value of this research in highlighting the importance of both ‘soft’ factors and utilizing the
knowledge held by end users.
Participants were then questioned regarding their satisfaction with the design method they used. P1 and P4 both worked with the model and both were unsatisfied with the ease of which they could design and the level of detail achievable. They both went on to state that they felt limited by the materials available and that the method could be improved through allowing greater
27 of representation between model and CAD. Regarding the materials, a range of coloured pencils and cardboard was supplied, it was also stated that anything difficult to represent could be written onto the model (for example wooden floor was written by participants onto the model). Unlike the findings of Frost & Warren (2000), whom observed that participants were happy to leave things sketchy and unfinished, in this case participants were dissatisfied with the inability to properly represent and specify all details. The most likely cause of this is the fact only one
workshop was conducted in which everything was to be finished; with larger time constraints it would be beneficial to create a greater iterative process of design and reflection (Frost & Warren 2000).
28
Chapter 5
5.1 Conclusion
This research has endeavoured to expand upon traditional approaches to facility design, which generally neglect the ‘soft’ factors valued by the end-users (Francis et al. 1999). Studies, such as Ulrich et al. (2004), have already identified that these factors are vital in a facilities
effectiveness. PD has proved a viable solution to this problem within product design, making end-users co-creators in the process (Greer & Lei 2012), however research addressing PD within facility design is sparse (Tzortzopoulos et al. 2009). Hence this research has been concerned with bridging the gap in knowledge, to identify if PD is a viable solution to integrating the social values of end-users into facility design. As well as exploring the key methods that have been proposed for its facilitation (Mobach 2008; Broberg & Edwards 2012; Jorgsen 2006). The paper will first conclude with references to the research questions set out for the paper, before addressing the implications of the study.
5.1.1 Research Question 1
“How can Participative Design involve end-users in the facility design process to accurately portray the ‘soft’ factors valued by them?”
29 state that end-users without guidance will be limited in their considerations. Experts are needed, in order to guide and cooperate with end-users, to aid in translating what they desire into reality (Broberg & Edwards 2012). Consequently the PD approach applied here with a multi-disciplinary team may provide fruitful.
Through highlighting the importance of ‘soft’ factors and the previous studies into their value (CABE 2005), it is hoped that the danger of them being suffocated has been reversed. Amidst the modern drives for efficiency and cost that favour attention on tangible factors there is a need to reiterate the value created in paying attention to more intangible elements (CABE 2005). One can highlight the Pruitt Igoe project, once celebrated for solving the financial problems within public housing was consequently demolished because it did not meet the social needs of the end-users (Yu et al. 2007). However given the rapidly evolving nature of technology and society a buildings use can change a number of times over its life-cycle. Future research must investigate the advantages and disadvantages of using PD around the values of first use end-users, as opposed to designing for flexibility and adaptability (Macmillan 2006). This will aid understanding in balancing short and long term end-user values in facility design and perhaps highlight the need for differing forms of PD at different stages of a facilities life-cycle.
5.1.2 Research Question 2
“Do traditional methods or CAD based methods for PD provide the best platform for end-users to interact with the design and express their valued ‘soft’ factors?”
This research has demonstrated that neither method is complete in allowing end-users to express their valued ‘soft’ factors. However the use of CAD produced more favourable results in this case, and adds to the building research demonstrating its applicability to both PD and facility design (Mobach 2008; Frost & Warren 2000). Advancements in Virtual Reality (VR) and Building Information Modelling (BIM) show great potential for collaboration in the early design phases (Azhar et al. 2007). As Mobach (2008) found participants understood the consequences of the design more easily in CAD. Elements could be tested and changed more readily than in the model which allowed a greater deal of refinement in the final designs. This research shows that limiting these advancements in technology to industry experts would be a waste of the valuable
30 receiving the attention it deserves in facility design (Clegg 2000). This research shows that
humans should not be designed out of technical systems but balanced with them to co-create valuable design solutions. Although we must also note that CAD did not provide a complete solution, and much like Clegg (2000) has identified within other design realms, methods should be created that are context and user specific. Therefore the results here show that much the same is true within facility design; there is no one-fits all method and PD must be tailored to individual cases.
5.1.3 Implications
The research presented has further highlighted the value created through PD in facility design, in accessing the implicit knowledge held by end-users, further to this it has reviewed the two key methods used in practice. Indeed as is the case in relatively new research areas many issues have been highlighted which require further attention in order to establish a clearer picture. Due to research of this area concerning almost exclusively healthcare settings (Cottam & Leadbeater 2004; Mobach 2008; Frost & Warren 2000), the research has highlighted its applicability to a wider context and thus paved the way for future research. As the CABE (2005) report also
suggests research should not just be limited to PD within facilities, there is also potential for PD to be explored within wider urban design, including local communities. Furthermore this research has focused upon only ‘soft’ factors; the next logical step therefore would be to incorporate ‘hard’ elements into the design process and monitor its effects.
It has been concluded from this research that a mixed method approach would be more
beneficial that includes a greater range of stakeholders. Therefore future research should expand the range of stakeholders considered, including for example a greater number of end-users and professionals such as architects. Another important issue which has emerged from the research is the subjectivity of perception and values of ‘soft’ factors. Whilst looking at the positive impacts ‘soft’ factors can have on a facility, negative influences arising from such factors have been neglected. Therefore further research in this area should consider positive and negative impacts of such factors and what is perceived as highly desirable as appose to acceptable. Furthermore would it prove more beneficial to eliminate negative aspects within the design or create new positive factors?
32
References
Azhar, S., Hein, M. & Sketo, B., 2007. Building Information Modeling ( BIM ): Benefits , Risks and Challenges. Building Science, 18(9), pp.2–5. Available at:
http://ascpro.ascweb.org/chair/paper/CPGT182002008.pdf.
Ballard, G., 2008. The Lean Project Delivery System : An Update. Construction, 8(Revision 1), pp.1–19. Available at: http://www.leanconstruction.org/pdf/WP_9_ProjectDefinition.pdf. BBH, 2011. The importance of design in the creation of end-of-life hospital environments.
Building Better Healthcare.
Bell, J., 2005. Doing your research project: a guide for first-time researchers in education, health
and social science, Open University Press. Available at:
http://www.worldcat.org/title/doing- your-research-project-a-guide-for-first-time-researchers-in-education-and-social-science/oclc/318177771?referer=di&ht=edition.
Bonner, J.V.H. & Porter, J.M., 2000. Introducing User Participative Design Methods to Industrial Designers. Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 44(38), pp.787–790. Available at:
http://pro.sagepub.com/lookup/doi/10.1177/154193120004403828 [Accessed November 6, 2013].
Broberg, O. & Edwards, K., 2012. User-driven innovation of an outpatient department. Work A
Journal of Prevention Assessment and Rehabilitation, 41(0), pp.101–106. Available at:
http://dx.doi.org/10.3233/WOR-2012-0142-101.
Bruseberg, A. & McDonagh-Philp, D., 2000. User-Centred Design Research Methods: The
Designer’s Perspective. In P. R. N. Childs E. Brodhurst, eds. Integrating Design Education
Beyond 2000 Conference. University of Sussex, pp. 179–184. Available at:
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.99.7523&rep=rep1&t ype=pdf.
Bryman, A. & Bell, E., 2007. Business Research Methods, Oxford University Press.
Burkhardt, J.-M. et al., 2008. Multimodal collaborative activity among architectural designers using an augmented desktop at distance or in collocation. In J. Abascal, I. Fajardo, & I. Oakley, eds. Proceedings of the 15th European conference on Cognitive ergonomics the
ergonomics of cool interaction. ACM New York, NY, USA.
CABE, 2005. Design with DIstinction: The Value of Good Building Design in Higher Education, London: CABE.
Carmona, M. et al., 2003. Public places-urban spaces: the dimensions of urban design, Architectural Press. Available at:
http://books.google.com/books?hl=en&lr=&id=GTQqshLjwCoC&oi=fnd&
33 Chesbrough, H.W., 2003. Open Innovation: The New Imperative for Creating and Profiting from
Technology H. W. Chesbrough, W. Vanhaverbeke, & J. West, eds., Harvard Business School
Press. Available at: http://books.google.com/books?id=OeLIH89YiMcC&pgis=1.
Christiansson, P. et al., 2008. User Driven Innovation in the Building Process. Tsinghua Science
Technology, 13(October), pp.248–254. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S1007021408701573.
CIOB, 2011. BIM to be compulsory on all government projects. Magazine of the Chartered
Institute of Building. Available at:
http://www.construction-manager.co.uk/news/bim-be-compulsory-all-projects/ [Accessed September 10, 2013].
Clegg, C.W., 2000. Sociotechnical principles for system design. W. S. Bainbridge, ed. Applied
Ergonomics, 31(5), pp.463–477. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/11059460.
Coletta, A., 2012. The Lean 3P Advantage: A Practitioner’s Guide to the Production Preparation
Process, Productivity Press.
Corben, S., 2013. Making Hospitals More Welcoming. Building Better Healthcare.
Cottam, H. & Leadbeater, C., 2004. HEALTH: Co-creating Services D. C. Uk, ed. Red Paper 01, (RED PAPER 01).
Crawford, M. & Di Benedetto, A., 2008. New Product Management 9th ed., McGraw-Hill Higher Education.
Crouch, M. & McKenzie, H., 2006. The logic of small samples in interview-based qualitative research. Social Science Information, 45(4), pp.483–499. Available at:
http://ssi.sagepub.com/cgi/doi/10.1177/0539018406069584 [Accessed November 11, 2013].
Flick, U., 2009. An introduction to qualitative research J. Heigham & R. C. Croker, eds., Sage. Available at: http://books.google.cz/books?id=sFv1oWX2DoEC.
Fosters + Partners, 2009. CircleBath. Available at:
http://www.fosterandpartners.com/projects/circlebath/ [Accessed May 28, 2013]. Francis, S. et al., 1999. 50 Years of ideas in health care buildings. , p.68.
Freire, . Alarco n, L.F., 2002. Achieving Lean Design Process: Improvement Methodology.
Journal of Construction Engineering and Management, 128(3), p.248. Available at:
http://link.aip.org/link/JCEMD4/v128/i3/p248/s1&Agg=doi.
Frost, P. & Warren, P., 2000. Virtual reality used in a collaborative architectural design process, IEEE Comput. Soc. Available at:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=859814.
Gorse, C.A. & Emmitt, S., 2007. Communication behaviour during management and design team meetings: a comparison of group interaction. Construction Management and Economics, 25(11), pp.1197–1213. Available at:
34 Greer, C.R. & Lei, D., 2012. Collaborative Innovation with Customers: A Review of the Literature
and Suggestions for Future Research. International Journal of Management Reviews, 14(1), pp.63–84. Available at: http://doi.wiley.com/10.1111/j.1468-2370.2011.00310.x.
Hanyu, K., 2000. Visual Properties and Affective Appraisals in Residential Areas in Daylight.
Journal of Environmental Psychology, 20(3), pp.273–284. Available at:
http://linkinghub.elsevier.com/retrieve/doi/10.1006/jevp.1999.0163.
Howell, G.A., 1999. WHAT IS LEAN CONSTRUCTION - 1999. Construction, 7(July), pp.1–10. Available at: http://www.leanconstruction.org/pdf/Howell.pdf.
Jorgsen, B., 2006. Intergrating Lean Design and Lean Construction: Process and Methods. The Technical University of Denmark.
KAHANA, B. et al., 2003. Person, Environment, and Person-Environment Fit as Influences on Residential Satisfaction of Elders. Environment & Behavior, 35(3), pp.434–453.
Kallus, R., 2001. From Abstract to Concrete: Subjective Reading of Urban Space. Journal of Urban
Design, 6(2), pp.129–150.
Karlsson, C., 2009. Researching Operations Management, New York: Routledge. Koskela, Lauri, 2000. An exploration towards a production theory and its application to
construction An exploration towards a production theory and. Construction, 408, p.298. Available at:
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.6.7319&rep=rep1&ty pe=pdf.
Larson, K. et al., 2004. Open Source Building — Reinventing Places of Living. BT Technology
Journal, 22(4), pp.187–200. Available at:
http://www.springerlink.com/openurl.asp?id=doi:10.1023/B:BTTJ.0000047598.66712.e0. Macmillan, S., 2006. Added value of good design. Building Research Information, 34(3), pp.257–
271. Available at: http://www.tandfonline.com/doi/abs/10.1080/09613210600590074. Mattelmäki, T. & Sleeswijk Visser, F., 2011. Lost in X: Interpretations of design and
Co-creation. In N. F. M. Roozenburg, L. L. Chen, & P. J. Stappers, eds. In Proceedings of
IASDR2011 the 4th World Conference on Design Research. IASDR, pp. 1–12. Available at:
http://designresearch.fi/courses/ds11/wp-content/uploads/2011/10/Mattelmaki_Lost-in-CoX_fin1.pdf.
Mills, G.R. et al., 2009. Applying a Universal Content and Structure of Values in Construction Management. Journal of Business Ethics, 90(4), pp.473–501. Available at:
http://www.springerlink.com/index/10.1007/s10551-009-0055-7.
Mobach, M.P., 2008. Do virtual worlds create better real worlds? Virtual Reality, 12(3), pp.163– 179. Available at: http://www.springerlink.com/index/10.1007/s10055-008-0081-2 [Accessed March 1, 2013].
