UNIVERSITEIT VAN AMSTERDAM
GRADUATE SCHOOL OF SOCIAL SCIENCES | URBAN & REGIONAL PLANNING
Slimme Stad or Smartwashing?
Quantifying the congruence between
technocratic prophecy and the observed
experience of professionals in Amsterdam
Thesis MSc Urban and Regional Planning
Christopher Livett
Student no. 10862609
cdlivett@gmail.com
20 June 2017
Supervisor: Marco te Brömmelstroet
Second Reader: Bas Hissink Muller
Abstract
In recent years a new term has gained prevalence in many facets of the urban sphere - “smart cities”. This term has been applied to research institutes, government policies, private sector products, news media headlines, job titles, academic papers, and more. In the case of Amsterdam, the government launched the “Amsterdam Smart Cities Initiative” in 2009. The embrace of this term by so many sectors of society inspires further research. However, when a basic first step of research is undertaken – determining exactly what is it that is being studied – it becomes clear that the closest thing to a consensus about what “smart cities” means is that there is no consensus. Definitions for the term and opinions on what things fall under the smart cities umbrella vary widely. How is it that a concept with no clear meaning is able to inspire so much activity in society? This paper provides perspective on how the term “smart cities” has been defined, what has been considered to be part of the smart cities concept, and what are the effects of this concept on planning processes. It does this by analyzing at a large collection of academic literature and, with Amsterdam serving as a case study, interviewing professionals who self-identify as working within the framework of the smart cities concept. By doing this, this paper seeks to improve communicative planning by bringing additional clarity to what is currently an ambiguous term, to provide insight into the advantages and disadvantages of smart-cities-related approaches to planning, and to determine the level of agreement between practicing professionals and academic literature when it comes to conceptualizing smart cities.ABSTRACT 1 1. INTRODUCTION 3 1.1. BACKGROUND 3 1.2. PROBLEM STATEMENT 4 1.3. APPROACH AND RESEARCH QUESTIONS 4 1.4. THESIS OUTLINE 5 2. THEORY AND LITERATURE REVIEW 7 2.1. EMERGENCE OF THE SMART CITY CONCEPT 7 2.2. WHAT IS THE MEANING OF “SMART CITIES”? 10 2.2.1. SEEKING CONSENSUS ON SEMANTICS 10 2.2.2. LANGUAGE AND SYMBOLISM IN PLANNING 12 2.3. PLANNING OUTCOMES 13 3. CONCEPTUAL MODEL 15 4. DESIGN AND METHODOLOGY 17 4.1. CROSS-SECTIONAL RESEARCH DESIGN 17 4.2. LITERATURE REVIEW AND CODING 19 4.3. INTERVIEWS AND OBSERVATIONS 23 4.4. CROSS-SECTIONAL COMPARISON AND THEMATIC ANALYSIS 26 4.5. CONSIDERATIONS AND SCOPE 27 5. RESULTS 28 5.1. LITERATURE 28 5.1.1. DEFINITION OF SMART CITIES 29 5.1.2. OPERATIONALIZATIONS 31 5.1.3. PLANNING OUTCOMES 33 5.2. PRACTITIONER INTERVIEWS 35 5.2.1. DEFINITION OF SMART CITIES 35 5.2.2. OPERATIONALIZATIONS 36 5.2.3. PLANNING OUTCOMES 37 5.3. CROSS-COMPARISON 38 5.3.1. DEFINITION OF SMART CITIES 38 5.3.2. OPERATIONALIZATIONS 38 5.3.3. OUTCOMES 39 6. FINDINGS AND CONCLUSIONS 40 6.1. DEFINITIONS AND SYMBOLISM 40 6.2. OPERATIONALIZATIONS 41 6.3. OUTCOMES 42 7. DISCUSSION 44 7.1. FURTHER RESEARCH 44 7.2. REFLECTION 45 8. REFERENCES 46 9. APPENDICES 57 9.1. CODING SCHEMA 57 9.2. INTERVIEW SCHEMA 58 9.3. FULL CODING RESULTS 59 9.4. T 76
1. Introduction
1.1. Background A foremost challenge in any field of study or school of practice is that of remaining properly current in the knowledge of trends and developments in that field. It is only through the development of new ideas, and the testing of those ideas through experimentation and practice, that knowledge and society as a whole are able to progress. It is with that lens that I undertake an analysis of “smart cities” – an idea that has evolved to become a critical part of the discourse of urban planning, but has yet to be crystalized into a cohesive thought as to what it actually means. The modern era of cities has been characterized by urban growth and recognition of their existence as part of “complex socio-technical networks” which “are implicated in hierarchies of multi-level governance” (Hodson & Marvin, 2012, p. 422). The growing importance of cities, in turn, emphasizes the importance of the field of urban planning. Urban planning is hardly a field characterized by a lack of debate and change; research into approaches, outcomes, and practices continues unabated. One example of this debate is illustrated through the points of conflict between sociocratic and technocratic perspectives. Like the conflict of visions of the future of New York City put forward by Jane Jacobs and Robert Moses in the 1950s and 1960s that pitted the modernist philosophies that supported technological developments of automobiles and highways against community-based planning, this type of debate is an excellent demonstration of how different ideologies shape the development of cities, where the ideals of community-oriented spaces must go toe-to-toe with new technological developments that have the potential to revolutionize the way people live. Although serious proposals for expressway construction in dense urban locations are now rare, there continue to be questions about the role of technocratic approaches in urban planning processes. Finding its home in the social sciences, urban planning can feel worlds away from the fields of computer science and engineering. As such, research can easily fall into silos, leaving engineers and computer scientists to study technology and planners to focus on human factors, creating uncertainty about the interactions between these two approaches. This disconnect, and the discussion about the role of modernist and technocratic approaches in planning, is particularly obvious when looking at the discourse surrounding smart cities. This discourse is of particularHowever, despite the growing volumes of information on this subject, the precise definition of smart cities is illusive. The Smart Cities Council, based out of the USA, defines a smart city as one “that has digital technology embedded across all city functions” (Smart Cities Council, 2015). This broad definition (which includes planning, a core city function) helps explain why the smart cities concept has managed to capture the interest of cities and individuals around the world. As a result of this uncertainty, there is a need to better understand how planning practitioners actually experience smart cities and the influence that the concept has on planning practice. 1.2. Problem statement As noted above, there remains a lack of clarity on what “smart cities” actually means once a critical eye is applied. The lack of a clear definition has manifested itself, in turn, to inconsistent theories about the concept. As a result, there is a lack of understanding about how practitioners actually experience smart cities, whether their lived experience aligns with the expectations in the literature, and whether, in their views, processes and outcomes are improved by use of programs and processes that are categorized as being part of the “smart cities” concept. 1.3. Approach and research questions Fundamentally, if we don’t know what the benefits and disadvantages of smart city projects and approaches are, or even how to define “smart cities”, it will be much harder to foster the critical perspective needed for things to change and evolve. By increasing understanding of what people actually mean, people can understand how to approach the topic, language can evolve further, and benefits and issues can be clearer. Due to its commitment to the smart cities concept, the City of Amsterdam provides an excellent opportunity to gain a practical and relevant understanding of what a “smart city” actually means on the ground. As such, the research question for this project is “What are the conceptualizations of smart cities and its expected planning outcomes in the academic literature, and how do they align with the conceptualizations and experiences of practitioners in the case of Amsterdam?”
Through the use of a thorough review of academic literature, as well as fieldwork engaging professionals who work with the smart cities concept in Amsterdam, I intend to demonstrate that a greater level of coherence can be brought to the debate about smart cities. The framework that will be used to review the literature is based on the idea that current understanding of the relationship between theory, conceptualization, and application are in flux while directly influencing each other. Fieldwork, through the use of interviews, offers an opportunity to increase practical understanding of these concepts, and provides an insight into how smart cities unfold in an urban context. Due to the fact that the smart cities concept is multidisciplinary in nature, it makes sense to focus on a subset of the concept. As a result, this paper will focus on the use of the smart city concept in the transportation field of urban planning. The sub-questions that need to be answered in order to answer the primary research question are: 1. What are the different ways that “smart cities” has been conceptualized in academic literature? This includes definitions, operationalizations (consisting of processes, programs, and systems), and outcomes. 2. How do smart cities practitioners conceptualize “smart cities”, including definitions, operationalizations, and planning process outcomes? 1.4. Thesis outline This thesis is split into seven chapters: Introduction, Theory and Literature Review, Conceptual Model, Methodology, Results, Findings and Conclusions, and Discussion. This first chapter has briefly introduced the relevance of the subject this thesis explores, the problem statement that this paper aims to shed some light upon, and elaborates on the purpose and scope of this study. The theoretical framework of this paper is presented in the Theory and Literature Review (Chapter 2) and the Conceptual Model (Chapter 3). The Theory and Literature Review provides background on the smart cities concept, discusses theories about role of language in planning and the desired
Chapter 4 describes the cross-sectional research Design and Methodology approach that was used to answer the research questions. The results of the research are presented in Chapter 5. It will discuss the ways academic literature defines and operationalizes the smart cities concept, the planning outcomes that were identified, and the general themes that are described in the literature. This process will then be repeated for the interviews that were undertaken. This is followed by a cross-comparison of the results from the literature and the results from the interviews. Chapter 6 provides the Findings and Conclusions. The main research question is addressed through a comparing the content of the academic literature to the practitioner interviews. This includes comparisons to the theories about language and planning outcomes to determine if the theory was sufficient to explain the observed results. Finally, Chapter 7 offers a discussion on potential further research and a reflection on the development of this thesis. Full results of the literature and interview coding process can be found in the appendices along with interview transcripts.
2. Theory and Literature Review
2.1. Emergence of the smart city concept By 2015, the use of the term “smart cities” was well established. According to Harrison & Donnelly (2011, 2), the term had been in use by commercial businesses since at least 2004. Currently, the smart cities concept is an active part of current initiatives being led by both government and private business, as well as being present in the articles and lectures of schools of Engineering and Urban Planning. Finding examples of the term being used in popular contexts is not a challenge, with articles on the topic to be found in major newspapers and magazines. An example from the London-based newspaper The Guardian, which maintains a sub-site dedicated to the topic of smart cities is illustrated in Figure 2.1 below. Additionally, large, multi-national corporations such as IBM and Siemens take out full-page advertisements in major news publications and develop entire websites to the topic. Fig 2.1 – Smart Cities sub-site from the website of London’s The Guardian newspaper (Retrieved from: https://www.theguardian.com/cities/smart-cities)decision has been made throughout this project to refer to smart cities as a “concept”. The particular nature of this concept is the focus of this paper, studying how dialogue around the concept has helped to shape the development and implementation of it within the urban planning field. In the local context, the Amsterdam Smart City initiative was launched in 2009 to test innovative ideas and solutions for urban issues. Their website offers a listing of a number of projects that have been piloted or implemented in Amsterdam under the umbrella of working towards a “smart city”, which demonstrates that smart cities is an established concept in Amsterdam. Indeed, as illustrated by the Amsterdam Smart City initiative, the dozens of projects associated with that initiative, as well as the creation of the Amsterdam Institute for Advanced Metropolitan Solutions, Amsterdam has moved beyond embracing smart cities and has moved into actively promoting the concept through action. The use of the term has grown and, by all indicators, appears to still be increasing. To illustrate this growth in popularity, references to either “smart city” or “smart cities” were counted in major indices of published materials. The indices used for this exercise were: Google Scholar, which is a freely accessible search engine of academic literature; Web of Science, which is a subscription-based index of academic literature from a range of journals; and LexisNexis, which is a subscription-based database that includes popular media. In the case of LexisNexis, the search was restricted to major global newspapers and references to “street smart” and “Smart City Coupe” (a model of car) were excluded to remove records that clearly were not relevant to the topic. In the case of Google Scholar, patents and citations were excluded from the search results.
Fig 2.2 – Frequency of records containing “smart city” or “smart cities” in three major indices, 2000-2014. As can be seen from Figure 2.2, there has been a sharp increase in the use of the term “smart city” from 2000 to 2014. This illustrates an exponential growth of the use of the “smart cities” term in both academic and popular literature, with a major inflection point in 2008, with much greater growth after that point. The growth of this idea can be seen not just through newspapers and journal articles, but also in a number of other high-profile ways. The book Smart Cities by Anthony M. Townsend was published in 2013 and has been successful enough to be republished as a trade paperback. In addition, smart cities conferences have occurred in cities around the globe, as well as in Amsterdam. The Amsterdam Smart Cities Event served as an excellent opportunity to observe the discussion around the smart cities concept happening in real-time and has influenced the content of this thesis paper.
decisions should be data-driven and that data will be plentiful and open. The ubiquity of technologies such as mobile computing and sensors and their declining cost have increased the spatial applicability of some technologies, allowing them to be leveraged by even moderately-sized businesses, university or hogeschool faculties, or government agencies, effectively spreading them from the private space into the public space of the city. Harrison and Donnnelly (2011, 5) speak of competitiveness as a driving factor behind the smart cities as cities and urban regions seek ways to set themselves apart in order to attract jobs, businesses, and investment. This view is similar to that of Hollands (2008, 308), who suggests that the smart cities concept is a natural evolution, influenced by the forces of neoliberalism and the resultant desire to attract the “creative class”. A core element of a neoliberal society is one with a reduced role of government. A weaker government would frequently be pressured to do more with less and could be expected to be particularly open to the idea of smart solutions that promise greater efficiency. It seems clear that theories about the motivations driving the interest in smart cities concept are numerous (almost as numerous as the definitions for term “smart cities”), the increasing presence of the smart cities concept within the wider discourse of urban governance is undeniable. 2.2. What is the meaning of “smart cities”? 2.2.1. Seeking consensus on semantics The growth of the smart cities concept has included it playing a growing influence in contemporary approaches to urban policy, suggesting a need to study and understand how it is influencing the decision-making process. However, to begin to understand the influence of a concept, one must understand what it is that he or she is attempting to study. In the case of this paper, studying smart cities requires a basic understanding of what the meaning of the term “smart cities” is. The high profile nature of the smart cities concept has resulted in a diverse range of literature, from technical documents and academic literature to commercial sales documentation and popular nonfiction paperbacks. With such a broad selection, it can hardly be surprising that attempts to define smart cities abound. In an article by IBM Smart Cities engineers Colin Harrison and Ian Abbot Donnelly (2011), smart cities
which seems to align with expectations for an engineering perspective. This can be directly compared to older theories about postwar modernism, defined by James Scott (1999) as “a particularly sweeping vision of how the benefits of technical and scientific progress might be applied - usually through the state - in every field of human activity” (p. 90). Application of technology to urban systems and governance certainly is not a new idea. This can be contrasted against a more social science focussed definition by Caragilu et al. (2011) who suggest that a smart city exists when “investments in human and social capital and traditional (transport) and modern [information and] communication infrastructure fuel sustainable economic growth and a high quality of life, with a wise management of natural resources, through participatory governance” (p. 70). Robert G. Hollands in one of the most comprehensive attempts to define smart cities, Will the Real Smart City Please Stand Up? (2008), illustrates that the term has been used to describe such broad concepts as technological innovation, e-governance, community and social learning, addressing urban growth and environmental sustainability (smart growth), the knowledge economy, and creative classes (p. 304). Sotiris Zygaris (2013) identified smart cities concepts as those that deal with growth in ways that are “green”, “interconnected”, “intelligent”, “innovating”, or containing “knowledge” (p. 218). This illustrates that definitions offered by those who have studied this question previously are broad and varied. So much so that a single definition that all sources agree upon, which would provide the clarity needed to provide a framework for what sources should be included in further study of smart cities, is simply not possible. A term which has often appeared alongside smart cities is “innovative cities”. However, this alternative term is so broad as to be useless in a practical sense while still managing to fail to properly account for a number of the concepts described above such as being sustainable or business-supportive. The concept has been used in so many different ways, that it is patently clear that a single word, like “innovative”, will never suffice as a definition. The projects and programs described as part of the emergence of the smart cities concept clearly emphasize how the concept is influencing behaviour on the part of business, government, and academia. While this chapter has illustrated the lack of consensus on what the term means and what
cities concept. Although a discourse about smart cities is ongoing, it’s not clear that everyone is engaged in the same discussion or is approaching it from the same direction. 2.2.2. Language and symbolism in planning The lack of a consistent definition represents a lack of a consistent conceptualization of what smart cities means. Without consensus on the meaning of the term, effective communication is rendered essentially impossible. This has the potential to serve as a barrier to meaningful and productive dialogue between actors involved in the urban planning process. Any type of collaborative or consensus-based planning requires mutual understanding of the topic and concepts being discussed. According to Dembski & Salet (2009), symbolism plays an important role in planning and urban policy. The use of symbolism “is to persuade the audience and shape public discourse, so that the projected meaning becomes reality.” (Dembski 2013, 2016). If smart cities can be considered as a symbolic term, interpretation is required for it to be used in communication. If interpretation must be part of the communicative process, great room for variation in understanding is created. The use of terms with a plainly positive connotation, such as “smart”, influences decision-making. The terms have the potential of creating vulnerabilities whereby opinions about projects or plans utilizing these positive connotations aren’t viewed with a sufficiently critical perspective. The fact that there are a great number of programs and technologies that, according to their supporters, should fall under the smart cities umbrella, as well as a large number of suggested definitions, suggests that there is an eagerness for cities to be considered “smart”, as opposed to it being something which has naturally evolved in the field of governance. These appear to be signs that the term may be thin, with essentially zero barrier to entry when it comes to its use. If the discourse is also popular, this leaves it open to being taken advantage of by anyone hoping to increase the respectability of his or her product or program. After all, “the appeal of a symbolic marker is more likely to be grounded in unconscious and non-rational processes of decision-making. (Friedland & Alford 1991, as cited in Dembski 2013, 2017). As such, there is a clear need to study how the concept is being presented by different actors, including both academic authors studying the topic as well as practitioners in the field. What does the term mean and how is it operationalized – that is, what types of programs, processes, and products are considered to be part of a smart city? The identification of potential disconnects that
A definitive definition will not be found as a result of the research in this paper, but I postulate that additional understanding of what is commonly meant has the potential to reduce communicative barriers and increase understanding between actors. This thesis also provides an opportunity to identify the particular barriers to understanding between the fields of planning, engineering and governance. Many of the authors of the books and papers reviewed do work in the field of urban planning, so this review does offer some insight into the definition of smart cities by practitioners. However, no paper has attempted to define the term by analyzing what it means to those in the field. In the end, even if symbolism is being utilized, common understanding and terminology is necessary for a formal discussion or through communication by words (discourse) to successfully occur. 2.3. Planning outcomes In addition to determining how the smart cities concept is conceptualized, the literature has presented a lack of insight into how the use of the smart cities concept influences the success of the planning process. Fundamentally, knowledge about how the use of smart cities approaches influences planning outcomes is necessary to determine whether it is worthwhile to embrace the smart cities concept. The influence of technological approaches on the smart cities concept does not necessarily mean that there is congruence of the desired outcomes of new technologies and of urban and regional planning processes, even though as outlined above there are cases where the two concepts intersect within the concept. For example, a positive outcome from the implementation may be greater efficiency, which could suggest that less time is needed for tasks or that less energy is utilized to run a system. However, in a planning context efficiency may not necessarily be seen as a positive as it can be associated with less public engagement (as engagement takes time and is an integrative process) and therefore a less liveable urban environment. According to Morozov (2014), technical solutions are often developed separately from an identified need. These are solutions looking for a problem, sometimes referred to as solutionism. By looking at planning outcomes, it will be possible to see to what extent smart cities are or are not having a
opportunity to understand the pervasiveness of solutionism within discussion surrounding the smart
3. Conceptual Model
The literature suggests that the understanding of smart cities is uncertain while simultaneously debate about their implications continues. Theories about how smart cities are experienced do exist, but there is a knowledge gap in methodological approaches to assessing and profiling these conceptualizations in a qualitative manner. The author’s conceptualization of how this project’s elements of research work together is displayed in the conceptual model below (Figure 3.1). It consists of how the conceptualization (consisting of definitions and operationalizations) of smart cities influences behaviour in a planning and governance context. Figure 3.1 – Conceptual model Description: 1. An actor conceptualizes their personal knowledge of the smart cities concept, which would include their perspective of the theory, definition, outcomes, program, and processes of smart cities.4. The actor experiences the outcomes of the operationalization, be they positive, negative, or neutral. 5. This, in turn, can influence their conceptualization of smart cities. At the same time, it could also cause the actor to adopt new conceptualizations, such as by increasing specificity (general “smart cities” solution to traffic may become a “traffic volume sensor” solution). The idea that conceptualizations, including definitions, can influence behaviour is key to explaining the relevance of this research. If there is a lack of congruence in conceptualizations between different actors, this would be expected to lead to challenges in planning due to a lack of common language and understanding.
4. Design and Methodology
4.1. Cross-sectional research design As illustrated in through the theory and literature review, this topic of study requires an approach that can provide valid results from a body of research which, by its very nature, can be unclear and even contradict itself at times, while also taking account for multiple perspectives. One of the primary issues affecting the discourse around smart cities is the difference between the theoretical constructs and the lived experience. This is reflected in the two independent research areas that were studied, resulting in two distinct datasets – academic literature and “smart city” practitioners. Since this thesis aims to compare two datasets (literature and practitioners), while still offering the flexibility to allow for both quantitative and qualitative analysis, a cross-sectional research design was chosen. According to Bryman (2008, 58), “A cross-sectional design entails the collection of data on more than one case (usually quite a lot more than one) and at a single point in time in order to collect a body of quantitative or quantifiable data in connection with two or more variables (usually many more than two), which are then examined to detect patterns of association”. For the purposes of this project, the cases will be academic literature and practitioners, with units of analysis being articles and interviews. In addition, understanding the conceptualizations of the smart cities concept and the planning outcomes of its operationalizations requires an analysis with varying levels of detail, with ideas such as definitions being quite broad while specific operationalizations can be very narrow. As such, this research constitutes meta-analysis, bringing together the conclusions of others who have considered the definitions, operationalizations, and planning outcomes that are part of the smart cities concept. These factors point to the need for an inductive approach, whereby the research will provide the potential of generating generalized conclusions about perspectives (Bryman, 2008, 25). In an area where so much meaning is still uncertain, case studies operationalize the theories providing an opportunity to loop back and refine definitions and theories further. This can occur either at a range of scales, from individual projects up to the global level. In the more applied sense, there are a large number of sources that focus on specific analyses ofprofile the use of smart cities approaches in practice it would make sense to limit analysis to a subset of the field in order to keep things manageable. As such, transportation has been used as a specific analysis frame. This will limit the scope of the analysis while ensuring that potential operationalizations of the concept are still numerous enough to be worthwhile for study. Harrison and Donnelly (2011) specifically mention the application of real-time data for response, planning, and to inform transportation users and managers (p. 4). The Amsterdam Smart City website lists thirteen examples of local projects, summarizing their aim being “to provide a multifaceted, efficient, safe and comfortable transport system, which is linked to ICT infrastructure and open data”. There are also concepts that are still under development, such as WiFi monitoring technology. There is a lack of literature that specifically profiles the use of smart cities concepts in a context that is localized to Amsterdam and subset to the field of transportation, so it is clear that there is the potential for such a detailed and descriptive exercise to be undertaken. Overall, the research design seeks to develop a generalizable framework for understanding the potential perspectives and outcomes of the smart city concept for transportation planning practice. This generalized categorization framework should be applicable to any case. This will be followed by applying this framework to the case of transportation in the City of Amsterdam. The research process has three main parts, outlined in table 4.1 below. Table 4.1 – Simplified outline of methodological approach The research results have external validity for the Literary Review/Content Analysis segment of the project as the design is based upon generating conclusions based on a comprehensive sample of literature related to both transportation and to smart cities. The results from this section will provide a framework that can be utilized for comparing against any local context, and therefore 1. In academic literature related to transportation, what are the conceptualizations of “smart cities” and what are the planning outcomes of those operationalizations? 2. In the case of practitioners connected to the smart cities discourse in the City of Amsterdam and who have been involved in projects related to transportation, what are the conceptualizations of “smart cities” and what are the planning outcomes of those operationalizations? 3. Compare the results from the literature and from interviews of the practitioners to generate a quantified and qualified understanding of how the perspectives of practitioners compare to the perspectives that exist within the literature.
Amsterdam can serve as a useful and valid case study. The city has enthusiastically embraced the concept, as outlined above, and the existence of a smart cities policy and government initiative have also been announced in many cities, including Berlin and New York City. Therefore, any conclusions will have external validity through the ability to offer insight into the situation in other global cities that have embraced the smart cities concept. Fig. 4.1 – Methodological relationships and flow 4.2. Literature review and coding Through the use of a structured literature review process it will be possible to answer to the first research sub-question: What are the different ways that “smart cities” has been conceptualized in academic literature? This includes definitions, operationalizations (consisting of processes, programs,
transportation), the potential for sample bias is reduced or eliminated. As a result, it will be possible to offer a replicable and comprehensive profile of the smart cities concept. The online citation indexing service “Web of Science” was chosen as the source for the academic literature. This index was chosen due to its multi-disciplinary content, focus on journal articles, and positive reputation. As described above in Section 2.1, Emergence of the Smart City Concept, there was an inflection in the frequency with which smart cities were mentioned in literature, with the term growing in popularity starting in 2008. This inflection point was chosen as the start date for which literature would be considered for inclusion in this project because the rise in popularity after that point shows that the concept was becoming more popular and was maturing. As this paper seeks to increase understanding of how people are conceptualizing the smart cities concept, more mature data is more useful as it reduces the possibility of out-of-date results. On May 26, 2015, the author executed a query on Web of Science with the following attributes: • “Smart Cities” or “Smart City” in title or abstract • “Transportation”, “transport”, “transit”, or “mobility” as a topic • From January 1, 2008 to May 26, 2015. • Articles only This search returned 54 results, which would serve as the dataset for the literature review. The literature review generates quantified data through the use of content analysis, with a focus on objectivity and being systematic. “Content analysis is firmly rooted in the quantitative research strategy in that the aim is to produce quantitative accounts of the raw material in terms of the categories specified by the rules. The feature of quantification adds to the general sense of the systematic and objective application of neutral rules” (Bryman, 2008, 289). This approach to analysis will need to ensure that results are solid and defendable. Content analysis will be used to give structure to the literature and map out both the critical and complimentary schools of thought about the planning outcomes of smart city approaches and the conceptualization of smart cities. A content analysis of the literature will be undertaken, focussing on the themes of definitions and of operationalizations, which consists of programs, processes, products that are identified as the authors as being related to smart cities. For example, an author may describe the use of sensors in the road surface or about how citizens in a smart city would be need better outreach from the government in the context of smart cities. The content analysis will be done through the use of
coding, as described in Insch et al. (1997). The atlas.ti software will be utilized to assist with the coding process and to perform some basic analysis. Definitions and conceptualizations will be coded through a process of open coding, whereby codes are created on-the-go to accurately represent the definitions or conceptualizations as they are identified. In gain understanding about how the authors assessed the value of smart cities approaches, planning outcomes will also be coded. The criteria by which smart cities will be assessed are derived from the framework described by Marco te Brömmelstroet (2012). In that paper, a multidimensional framework for assessing the performance of planning support systems was outlined, which can also serve as a generalized framework for determining the success of any planning process. Early discussions with practitioners suggested that certain terms in the framework were harder to assess and provide a clear opinion that each term was not necessarily mutually exclusive from each other, making the provision of clear opinions more difficult. As such, the list was reduced to eight key outcomes, as outlined in table 4.2. Table 4.2 – Measures of Planning Performance Measures of Planning Process Measures of Planning Outcomes • Insight during the planning process (including insight into the problem and/or insight into assumptions) • Communication in the planning process • Achieving consensus in the planning process (including consensus about the problem, the goals, or the strategies) • Group cohesion (feeling that participants have of being a part of the group and the rate to which they can relate to other participants) • Efficiency of the planning process • Novelty (and originality) of the outcome or solution • Workability of the outcome or solution • Relevance of the outcome or solution As these indicators for planning outcomes are pre-determined, the coding process for these data will be more structured than for definitions or operationalizations. Each time a planning outcome is identified in the literature will be coded as one of: positive (the smart cities approach improved the outcome), negative (the approach reduced the quality of the outcome), or neutral (the approach had
Fig. 4.2 – Coding data schema Turning the results of the coding into a content analysis requires an iterative approach, whereby the codes are grouped into general categories. This is a manual process, involving “cleaning up” the data so that multiple incidences of the same information will be collapsed, generating a complete set of the concepts contained in the literature. As part of this process, consistency and accuracy of the codes will need to accurately fit into the categories to ensure semantic validity. In the end, these categories will comprise generalized descriptions of the definitions and operationalizations. This process is illustrated in Figure 4.3. As the codes for outcomes are derived from Table 4.2 above, this step will not be necessary for those data points.
Figure 4.3 – An example of the code abstraction process as it relates to the physical threats of the northern physical environment (Elo & Kyngäs, 2008) Overall, understanding will be gained about how practitioners define “smart cities”, what processes, projects, and technologies they perceive as being part of the smart cities concept, what they perceive as the purpose of these processes, projects, and technologies, and how whether the outcomes of these processes, projects, and technologies have merit. As this will be done in a quantitative manner, it will be possible to use the results to compare to other sources of data, such as interviews. 4.3. Interviews and observations The coding of a comprehensive dataset of literature helped to provide a refined and complete framework of the ways that the smart cities concept has been conceptualized, as well as its outcomes in academic literature. By applying the same coding method to interviews of practitioners, it will be possible to answer to the second research sub-question: “How do smart cities practitioners conceptualize “smart cities”, including definitions,
This question is of key importance as it provides additional points of reference, which are of both local relevance as well as being based upon their active, professional experience. By including this primary data in the form of interviews in the body of research presented here, a broader perspective will be gained while also making it possible to see how the conceptualizations of expert practitioners align and diverge with the literary sources, which is a requirement of answering the primary research question. A total of six interviews were undertaken, designed to capture a broad sampling of perspectives. The author determined that in order to ensure that responses provided sufficient coverage of the range of practitioners who were working in jobs that were related to the smart cities concept, interviewees would need to include the following perspectives: • Sectors: private and public • Points of view: advocative and critical • Roles: practical and theoretical • Geographic scopes: local and (Amsterdam-based) international The initial set of potential interviewees were identified through a combination of Internet research and through contacts made at the Amsterdam Smart City Event conference (June 3-5, 2015). In order to ensure that the range of perspectives described above would be met, more willing interviewees were identified than were interviewed in the end. In particular, private sector actors were much more willing to provide the time so they could be interviewed. The selected six expert interviews, chosen due to their level of involvement with the smart cities concept, are described in table 4.3.
Table 4.3: Interviews
Name and organization Relevance to Smart Cities Date
Aart de Koning (Goudappel Groep) Private transportation consulting firm. States that it is a provider of “smart mobility” solutions. July 30, 2015 Robert Jan ter Kuile (Amsterdam GVB) Local public transportation operator. States that they working on ways to provide transportation for a “smart urban region”. July 31, 2015 Tamaas Erkelens (Eurocities/Gemeente Amsterdam) Municipal government and trans-European network of major cities. Amsterdam describes itself as a smart city and is a member of a trans-European network of smart cities August 5, 2015 Kees Jansen (Pluraal/PhD UvA) Smart cities consultant and PhD Smarter Cities and Regions. August 5, 2015 Janine Hogendoorn (Ring Ring) Cycling-related app developer. Use of a mobile app to influence mobility behaviour change is described as smart. Suggested by Amsterdam Smart Cities as a specific example of a smart cities solution. August 6, 2015 Hans van Lint (TU Delft/Amsterdam Institute for Advanced Metropolitan Solutions Urban Mobility Lab) Professor of Traffic Simulation and Computing and Project Leader of the Urban Mobility Lab. August 11, 2015 This set of interviewees ensured that all the desired perspectives were covered in order to provide sufficient coverage of the smart cities concept in Amsterdam. Private (Aart de Koning and Janine Hogendoorn) and public (Robert Jan ter Kuile and Tamaas Erkelens) sectors;Advocative (Aart de Koning and Tamaas Erkelens) and critical (Hans van Lint) points of view; Practical (Robert Jan ter Kuile and Janine Hogendoorn) and theoretical (Kees Jansen and Hans van Lint) roles; Local (Robert Jan ter Kuile and Janine Hogendoorn) and Amsterdam-based international (Tamaas Erkelens)
The interviews were designed to collect the same data as was collected from the literature in order to ensure comparability of results. The interview was semi-structured; questions were presented by the interviewer, who also steered the conversation to try and keep the interview on-topic. The complete responses from the interviewees were transcribed and were coded in the same manner as the literature, as described in the previous chapter. This qualitative interviewing approach was chosen to ensure that opportunities for interviewees to identify definitions, operationalizations, and outcomes were maximized through discussion, which could lead to additional relevant information. Interviewees were explicitly asked how they personally define the term “smart cities”, the first opportunity to identifying how they conceptualized smart cities. This was followed by asking the interviewee to self-identify projects they had been involved with that they would describe as being smart cities-related, which provided information on how they operationalized the concept of smart cities. A preference for projects that were transportation-related and located in Amsterdam was expressed. They were asked what made those projects smart, which served as a second opportunity to identify how they defined smart cities as well as potentially providing additional insight in to how the interviewee operationalized the concept. Finally, the interviewees were asked whether taking a “smart” approach to the project affected each of the planning outcome measures that have been described above. The results of the coding process will be presented in the same manner as was described in the previous chapter. A copy of the interview schema as well as full transcripts of the interviews are included in the appendices. 4.4. Cross-sectional comparison and thematic analysis The results of the coding process provide a quantified analysis of the conceptualization of the smart cities concept as well as the planning outcomes from smart cities operationalizations, as described in both the literature and in the interviews. In Chapter 6, Findings and Conclusions, the two sets of results will be cross-compared, identifying key similarities and differences. In addition, the use of thematic analysis will provide additional insight, over-and-above the coding process. Braun & Clarke (2006, 82) state that “a theme captures something important about the data in relation to the research question, and represented some level of patterned response or meaning within the dataset”. By identifying themes, additional value may be drawn from the
themes “are the product of a thorough reading and rereading of the transcripts or field notes that make up the data. This framework is then applied to the data, which are organized initially into core themes” (Bryman, 2008, 621). Identifying themes will also be useful for determining opportunities for further research, which are described in Chapter 7. Through these comparisons the primary research question of this thesis will be answered: “What is the conceptualization of smart cities and its expected planning outcomes in literature, and how do they align with the conceptualization and experience of practitioners in the case of Amsterdam?” 4.5. Considerations and scope With the resources of government and the corporate world behind it, the smart cities concept is broad and involves a large number of actors. To limit the work to a reasonable amount for a master’s thesis, while still maintaining external validity, the decision was made to limit this research to case studies of the transportation sector and, in the case of the expert practitioners, to the City of Amsterdam. A similar project, which did not limit its literature to transportation-related topics or to a single city, would be expected to offer different results. The comprehensive sample of academic literature is limited to the papers that have been identified by the Web of Science indexing service. Web of Science is a large database which covers a large number of disciplines, so it can be expected that the papers it has identified represent a relatively comprehensive dataset, but nevertheless the use of a single service means that the list of papers are unlikely to be perfectly comprehensive. Additionally, the process of coding requires the person doing the coding to apply their personal judgement, which leaves a lot of room for interpretation and variation in results. As such, validity is limited by the qualitative nature of coding. It is good to reiterate that this research project will not seek to empirically determine a narrow or comprehensive single definition for smart cities (since, as the literature has made clear, this is
5. Results
5.1. Literature The literature will enable me to answers research sub-question 1: What are the different ways that “smart cities” has been conceptualized in academic literature? This includes definitions, operationalizations (consisting of processes, programs, and systems), and outcomes. As described in Chapter 4, Design and Methodology, a total of 54 articles were identified by Web of Science when the following query was applied: • “Smart Cities” or “Smart City” in title or abstract • “Transportation”, “transport”, “transit”, or “mobility” as a topic • From January 1, 2008 to May 26, 2015. • Articles only The query resulted in a range of sources and papers. Generally, the papers tended to focus on engineering-based descriptions of technological developments, while many of the results approached the topic of smart cities from a social sciences perspective, analyzing the intersection between technology and society. All of the papers that took a critical perspective on the topic of smart cities fell into this category, with no consideration or criticism coming from the technical and engineering perspectives. Only three to four papers could be described as broaching the topic from an urban planning perspective, though since most of the papers did deal with the subject of society and individual behavior within urban environments, deriving information about planning outcomes was still very possible. This section will provide the results of the content analysis of the literature and interviews, including analyses comparing the two sets of data. These results will be interpreted in the following section, Findings and Conclusions.5.1.1. Definition of smart cities Table 5.1 – Definitions of smart cities within the literature Table 5.1 above presents the results of the content analysis of the literature related to how the literature defined smart cities. Interestingly, 43 out of 54 articles (80%) used a meta approach to defining as part of their definition. In these cases the definitions were based around the issues with defining the term, either by stating that there is no definition, listing a range of definitions, or by
In comparison to the definitions that were found as part of the literature review chapter, unsurprisingly most definitions dealt with the theme of technology. Notably, the content analysis contained multiple papers that presented sustainability or ubiquity of technology as key parts of what defines a smart city, specific characteristics that were not common topics in the literature review. A total of 13 articles (24%) did include social, human, and/or economic perspectives as part of their definition, as opposed to being limited to technology and urban settings. With a quarter of the papers including these aspects as part of their definition, this illustrates that any attempt to develop a conclusive definition that does not include these non-urban and non-technology aspects will exclude a substantive proportion of academic thought about the concept.
5.1.2. Operationalizations
Table 5.2 – Operationalizations of smart cities within the literature
Table 5.2 contains a long list due to the large number of ways that the literature described programs, processes, technologies, and other ways of operationalizing smart cities. The length of the list, along with the fact that this list there is a large number of codes with only one observation, illustrates the great variation in what is seen as being part of the smart cities concept.
Unsurprisingly, technology and data are well represented in the results. But other than technological themes, the most common theme is “social or economic”, with considerations of social or economic concepts appearing 57 times across the papers (21% of all coded observations). As with the definitions, a substantive number of the operationalizations include concepts such as governance, planning, people, and planning. Although it is in the minority, a large number of authors still agree that social and economic operationalizations are a key part of smart cities. The most common code is for “smart mobility”, which is not unexpected as one of the filters applied when selecting articles for analyzing was the inclusion of “transportation”, “transport”, “transit”, or “mobility” as a topic. The second most common was “smart grids”, which is noteworthy for being one of the few “smart” operationalizations with an accepted definition of what it takes for a grid to be smart: “Electricity networks able to take into account the behaviours of all the connected users in order to efficiently deliver sustainable, economic, and secure electricity supplies. Smart grids should be self-healing and resilient to system anomalies” (Netrotti et al, 2013, p. 27) 5.1.3. Planning outcomes Table 5.3 – Planning outcomes of smart cities within the literature Table 5.4 – Planning outcomes of smart cities within the literature as percentages
There is a clear result that the planning outcomes that are most commonly identified within the literature are insight during the planning process and communication. Based on these results, smart cities approaches have a largely positive impact on increasing insight into the problem at hand and/or insight into assumptions that are part of the planning work. As much of technology is focussed on communications and data collection, this result is not surprising. Other than insight, the analysis of the coded literature suggests that using a smart cities approach does not result in positive planning outcomes. Taking such an approach can have a positive effect on communication and cohesion half of the time, but otherwise all indicators are negative. For the outcome of reaching consensus, results were overwhelming negative, with the literature suggesting that the needs of meeting the goals of being a smart city can crowd out other concerns. This is summarized by Calzada and Cobo (2015) who describe the smart cities concept as “a top-down, master-planned vision shaped around the needs of suppliers rather than the needs of citizens” (p. 30). The literature offered little insight into the effect of smart cities on novelty and originality or workability any planning outcomes that may result from a planning exercise using these operationalizations.
5.2. Practitioner interviews The interviews will enable me to answers research sub-question 2: How do smart cities practitioners conceptualize “smart cities”, including definitions, operationalizations, and planning process outcomes? As described in Chapter 4, interviewees were chosen in order to ensure a broad sample of professionals working within the field of smart cities in Amsterdam. This included private and public sector perspectives, advocate and critical point of view perspectives, practical and theoretical roles perspectives, and local and (Amsterdam-based) international scope perspectives. As was the case in Section 5.1, this section will provide the results of the content analysis of the interviews and provide a summary of each table. 5.2.1. Definition of smart cities Table 5.5 - Definitions of smart cities according to the interviewees Even with a very small sample size of six, there is still a fair lack of consensus among the interviewees. A third of the interviewees explicitly stated that there is no definition, while technology-based definitions were identified four times. This highlights that even among professionals working within the smart cities concept, in the self-described smart city of Amsterdam, there is still a lack of agreement about exact what the concept means.
5.2.2. Operationalizations Table 5.6 - Operationalizations of smart cities according to the interviewees The six interviews provided a large number of ways that the concept of smart cities has been operationalized. The ability to ask six people this question and to get 29 answers, with no two responses being exactly the same, shows even less consensus than the author would have expected. Among this group, there is some agreement that sensors and/or automation is a way of operationalizing smart cities, but this group also focused on meta concepts, such as “making the wisdom of the crowds possible” or simply “apps”.
5.2.3. Planning outcomes Table 5.7 – Planning outcomes of smart cities according to the interviewees Table 5.8 – Planning outcomes of smart cities according to the interviewees Among the interviewees there was an overall impression that the smart cities concept has a positive to neutral effect on planning outcomes. There was a strong consensus that the use of smart cities processes and programs led to improved cohesion and novelty and originality in the planning process. In contrast, there was skepticism about smart cities approaches improving consensus, largely due to the challenges of big data. Large volume of information that can be generated through the use of technology has the potential to foster confusion and additional debate.
5.3. Cross-comparison 5.3.1. Definition of smart cities In both the academic literature and the among the interviewed practitioners, the approach to defining the smart cities concept by instead explicitly or implicitly discussing the issues associated with defining the term stands out. This takes the form of either explicitly stating that there is no definition, as was seen in both the literature and in the interviews, or of creating an extended list of the definitions used by others or ignoring the topic altogether, which was common in the literature. The next most common definition was based around the theme of combining technology with an urban environment. If “smart” is assumed to be a synonym for “technology”, this is clearly the most literal and obvious type of definition. That said, the fact that it only represents 16% of definitions provided in the academic literature and 38% of the definitions provided by the practitioners shows that “technology plus cities” only represents a small minority of views on how to define the term. Expanding on that definition, eight papers and one interviewee added an addition component for something to meet their definition of the smart city concept: people. In these cases, technology in a city is not sufficient, it must be applied in a way that relates to the citizens of the city. Suggesting a critical perspective that exists in both the academic literature and among practitioners, the smart cities concept was sometimes defined in socioeconomic terms, not related to technology at all. This included the criteria of concepts such as livable places and an improving economy. This appropriation of a term that, by and large, is typically associated with technology in some manner (although the manner may vary widely) shows that some authors and practitioners wish to ensure that social and economic concerns are not pushed aside by the smart cities concept. 5.3.2. Operationalizations Many of the operationalizations identified in the academic literature and in the practitioner interviews describe generalized concepts associated with technology and data. This could be as simple as “big data”, “information technology”, or “apps” as examples of the types of ways that smart cities manifest themselves. These operationalizations have nothing to do with cities in particular and potentially represent an attempt to use popularity of the smart cities concept to add credibility to their publications or projects.
A subset of this involves appending the term “smart”, “electronic”, or “intelligent” to well-established infrastructures. This occurred twice in the interviews, but 52 times across the academic literature. This included smart Internet, smart buildings, intelligent community services, smart public health, smart urban planning, electronic voting, and even smart garbage containers. The same problem as with the smart cities concept as a whole exists with these operationalizations: a lack of consensus about what the terms mean. Two themes that were common in the literature pertained to smart cities operationalizing themselves through sustainability and government. This is in stark contrast to the practitioner interviews, where sustainability was never mentioned and government was only mentioned once, in the form of neighbourhood budgeting. 5.3.3. Outcomes With regards to the planning outcomes of smart cities, the results from the academic literature and the practitioner interviews are quite different. Interviewees were much more positive with regards to how the smart cities concept influences planning outcomes. Of all responses, 71% were considered positive, while in the literature only 47% of identified planning outcomes were positive. The literature was far more likely to identify both the pros and cons of using a smart cities approach in planning. Particular outcomes where the results varied were cohesion as well as novelty and originality, which both received completely positive feedback from interviewees while responses were more balanced in the literature. In contrast, half of interviewees stated that a smart cities approach assisted with the workability of the planning process outcomes, while the influence of smart cities approaches on workability was simply not present in the literature, with only one article mentioning it. The greatest consensus when comparing the academic literature to the practitioner interviews were generally positive responses about insight (73% positive in the literature/80% positive in interviews), communication (50%/75%), and cohesion (50%/100%).
6. Findings and Conclusions
The comparisons in this chapter will provide the answer the research question of this thesis: “What are the conceptualizations of smart cities and its expected planning outcomes in the academic literature, and how do they align with the conceptualizations and experiences of practitioners in the case of Amsterdam?” By comparing the results of the content analysis of the two datasets, the alignment of the literature and of the practitioners in Amsterdam, we can form conclusions about whether a difference in alignment exists and, if so, the form of the difference in conceptualizations. 6.1. Definitions and symbolism The goal of this research was not to develop a specific definition for what the term “smart cities” means, but instead to see how different actors define it and, therefore, how the symbolic marker of “smartness” is conceptualized by article authors and professionals. The biggest consensus is around a definition that explicitly states that there is no definition. The lack of a clear definition is identified as a problem across both examined sets. In the case of the academic literature, the most common code was for articles that simply made no attempt at defining the term, avoiding the topic altogether. Awareness about the definitional issues seems fairly high among individuals who would be involved in writing or being interviewed about smart cities. This aligns with what would be expected if smart cities concept was being used as a symbolic marker. Although it lacks a literal, concrete definition, its frequent use even by those familiar with the definitional issue suggests a different, symbolic meaning instead. The role of symbolism is reinforced by an unexpected result of the content analysis: a full 15 out of the 54 papers, or 28%, simply never mentioned smart cities anywhere in the text beyond the introduction. Some did not even mention it past the title. This clear case of making use of the smart cities symbolism while not actually taking part in any form of further discussion about the concept further supports the idea that in some cases the use of the smart cities concept can be considered a case of solutionism, as described in section 2.3. Taking a top-down approach whereby a technology is developed, and only after is presented as being associated with the smart cities concept is a case of a technological solution in search for a problem. The evidence that smart cities is a symbolic term could be compared to the concept of “greenwashing”, whereby utilizing the language of environmental sustainability can add a veneer ofrespectability. Delmas and Burbano (2011, 66) explain that “greenwashing is the act of misleading consumers regarding the environmental practices of a company (firm-level greenwashing) or the environmental benefits of a product or service (product-level greenwashing).” The results of this research suggest that a similar process can occur with practitioners and businesses hoping to be increase their profile by using the language of smartness and smart cities, by simply applying the term “smart” without requiring that any criteria be met to earn that term. As such, the results suggest that within the smart cities concept there is a minority of actors who have utilized the uncertainty around the concept to engage in a form of “smartwashing”. Other than these symbolism-based definitions, the most common definitions could be generalized as “innovation and technology for any type of an urban issue”. The issue with this is that this generalized concept is that it is broad enough that it can be applied to an extremely wide range of concepts. Other than the cases where the definition is completely technology-focussed, an interdisciplinary perspective does seem to an essential element for the definitions of smart cities among both academic writers and practitioners. Two examples of this are, the intersection of society and technology or the intersection of commercial investment and technology. One attempt at defining the smart cities concept sticks out in particular. Kees Jansen, who offered a perspective of an academic studying the topic of smart cities, defined it as “the point at which technology, sustainability, and the new economy intersect”. This definition captures the most common themes in a clear and succinct manner. Overall, the defining of the term smart cities by both the academic literature and the practitioners align. This alignment takes the form of lack of consistency and consensus, but with themes of technology first and foremost, often expanded upon through the addition of language describing concerts pertaining to cities and people. Both the literature and interviewees also included a small subset of critical definitions, which utilize the smart cities term them transpose it to highlight non-technological urban issues, such as social services.
automation as common operationalizations while also recognizing how social elements were also part of the concept. Interviewees were more likely to use broad concepts in this part of their conceptualization compared to the literature, though this may be a result of a bias resulting from data that is collected face-to-face in a discussion, where meta ideas can be elaborated upon, as opposed to in print. This included ideas such as “open data” or “algorithms”. A result that could be valuable to consider is that there is a lack of consensus across the literature and across the interview responses at the level of individual codes. There were a number of concepts that were described in the literature on multiple occasions, but with the sole exceptions of “smart mobility” and “smart grids”, no operationalization was coded more than 10 times across 54 papers. In the case of the interviews, there wasn’t a single response that was coded more than once. (As mentioned in the results section, the frequency of “smart mobility”, is not unexpected as one of the filters applied when selecting articles for analyzing was the inclusion of “transportation”, “transport”, “transit”, or “mobility” as a topic. In addition, the frequency of “smart grids” could be explained by the fact that it is one of the few “smart” operationalizations with an accepted definition of what it takes for a grid to be smart.) It is clear that different people and different pieces of literature have very different ideas of what a smart cities approach looks like in practice. Much like with definitions, the only consensus is a lack of consensus. 6.3. Outcomes The greatest disagreement between the two datasets related to the topic of planning outcomes. Interviewed practitioners were generally positive about the contribution of smart cities operationalizations to planning processes, with 71% of responses being positive. There was a strong consensus that the use of smart cities processes and programs led to improved cohesion and novelty and originality in the planning process. This is in contrast to the academic literature, which was primarily negative, with only 25% of responses being positive and indifferent views towards how the smart cities concept can influence cohesion or novelty and originality. The only outcome which had primarily positive results was insight, which should be expected for a concept where so many of the conceptualizations related to data collection and sensors.
In contrast to the general thematic alignment between literature and practitioners with regards to definitions and operationalizations, there is a clearer divergence with regards to planning outcomes. One potential explanation is that practitioners may be applying a solutions-based perspective, focusing upon how smart cities processes, programs, and systems can assist them in doing their jobs and meeting their goals, while academic literature is more likely to take a critical perspective. Fundamentally, the interview responses were opinions based upon previous experiences, while academic articles are held to a more rigorous standard. This may suggest that there is a danger that practitioners are failing to identify the negative outcomes of the ways they have chosen to operationalize the planning process.
