1 Introduction

1.1 Research Context and Problem Statement

Cities are complex systems characterised by tensions arising from socio-economic and political agents, which lead to patterns of urban development that are interconnected with future urban challenges. Rapid urbanisation, climate change, and resource scarcity pose significant challenges for policymakers, planners, researchers, and other stakeholders in developing sustainable strategies to address these issues. One such strategy is the “smart city” approach.

The term “smart” has a broad range of interpretations among different actors responsible for creating cities, each with their own unique purposes and connotations. Some scholars tend to focus on a technology-driven approach, while others prioritise a human-centred method (Letaifa 2015). The smart city concept can be categorised using the 3RC Framework, encompassing three schools of thought: restrictive, reflective, and rationalistic or pragmatic, and a fourth known as the critical school of thought (Kummitha and Crutzen 2017). This framework considers both the social dimension of the city and the social aspects of sustainable development.

The restrictive school of thought views Information and Communication Technology (ICT) as the main component of smart city development, with limited consideration given to social inclusion and justice in implementation. The reflective school emphasises that smart city development must address human capital and use technology to improve social order, while still considering ICT as the main component. The rationalistic approach prioritises people over technology, seeking to enhance individuals’ capabilities and empower communities (Kummitha and Crutzen 2017). The critical school views the concept of “smartness” as a neoliberal marketing strategy implemented by large companies to privatise urban spaces (Letaifa 2015).

Despite the varying views on the epistemology of smart cities (Kummitha and Crutzen 2017), these approaches agree on their primary goal: to utilise, implement, and develop Information and Communication Technologies (ICTs) to enhance cultural, social, and urban development (Hollands 2008; Letaifa 2015).

In the design domain, urban development is the responsibility of urban designers (Carmona 2021). Urban design is a complex problem-solving process (Punter and Carmona 1997; Carmona 2021) that involves managing multiple dimensions (spatial, morphological, contextual, visual, perceptual, social, functional, and sustainable) (Carmona 2002). Due to the co-evolutionary nature of the design problem-solution process (Dorst and Cross 2001), different interpretations and approaches may be valid for the same design problem (William J. Mitchell 1975; Cross et al. 1984; Dorst 2003). Recently, urban development design has been explored through a data-driven approach, with the emergence of smart infrastructure, smart technologies, data science methods, and the availability of open city data (Karakiewicz 2010; Deutsch 2015; Kvan 2020; Gösta et al. 2020). Designers use computational tools and service platforms to assist the urban design process, including data gathering, real-time analysis, evidence-based design proposition generation, and visualisation (Picon 2015; Kvan 2020).

However, there is a lack of systematisation in the use of computational data-driven urban design tools, in terms of their relationship to design processes, data-driven cycles, and available computational technologies, as well as the integration of multiple urban design dimensions. Additionally, there is a shortage of free and open-source data-driven urban design tools that support community-based software development, which is crucial for empowering citizens as active users and decision-makers in the development and implementation of future smart cities (Stallman 2002; Busch 2014; Sadoway, Shekhar, et al. 2014; Ratti and Claudel 2015, 2016; Hernàndez 2021).

1.2 Research Aim and Objectives

This research aims to develop, demonstrate, and validate an integrated data-driven approach to assisting computational urban design processes. This approach will be developed as two research artefacts: (a) a framework, including the theoretical and conceptual systematisation of urban design processes, data-driven cycles and enabling computational technologies, and (b) a computational toolkit prototype as an implementation of the framework to validate its feasibility among urban designers.

The integrated approach seeks to facilitate the exploratory process of data-driven design across multiple urban design and data dimensions. The approach will be demonstrated through a scenario combining multiple data dimensions, using public available resources from open portals and open-source services. The computational toolkit prototype seeks to provide an experimental and iterative exploratory environment to assist urban design processes, facilitating data-driven design explorations across multiple data dimensions, enhancing evidenced-based decisions in data-driven urban design processes.

To achieve the above aim, the following five objectives have been identified for further investigation in this research:

  1. Understanding current data-driven urban design processes in practice and their enabling computational smart technologies;

  2. Developing a framework to systematise an approach for data-driven urban design processes;

  3. Developing a computational design toolkit prototype to assist urban design processes;

  4. Demonstrating the computational toolkit prototype through a scenario;

  5. Validating the integrated data-driven urban design approach amongst designers.

1.3 Research Design

This research aims to develop, demonstrate, and validate an integrated data-driven approach for assisting computational urban design processes. For this reason, this study’s follows a Design Science Research (DSR) methodology. In contrast with other methodologies in natural and social sciences, which aim to comprehend and clarify reality, design science research aims to create artefacts that seek to solve “real world problems” from a specific domain (Dresh, Pacheco, and Valle 2015). One area in which these methodologies are extensively utilised is software engineering and product design, with DSR being particularly prominent. For instance, Peffers et al. (2007) presented a comprehensive framework for conducting Design Science Research (DSR) in the realm of information technology.

As stated by Dresh, Pacheco, and Valle (2015), DSR is the most appropriate research methodology when the objectives centre on designing and creating, rather than describing, investigating, or interpreting a phenomenon. The primary objective of a DSR study is to design an artefact that advances the production of knowledge (Peffers et al. 2007).

The Design Science Research (DSR) methodology is well-suited for this study, as it proposes the development of an approach that aims to address a practical challenge for data-driven urban designers; specifically, the integration of data and design tools, as well as multiple datasets, in a cohesive design environment. Furthermore, this approach is composed of two research artefacts: a conceptual framework for data-driven urban design processes, and a computational toolkit prototype, which serves as an implementation of the framework.

In line with the six steps of DSR (1-Identification of the problem and Motivation, 2-Defining objectives of a solution, 3-Design and Development, 4-Demonstration, 5-Evaluation, and 6-Communication) and the five research objectives, this study is organised into three main phases:

  1. Conceptual Development Phase (Chapters 2 and 4): this phase includes creating an understanding of current data-driven urban design processes and their relationship to enabling smart technologies, as well as developing a framework to systematise an approach for data-driven urban design processes (Objectives 1 and 2).

  2. Computational Development Phase (Chapter 5): this phase includes developing a computational toolkit prototype to assist data-driven urban design processes, based on the developed framework. The Framework and Computational Toolkit compose an integrated data-driven approach for supporting computational urban design processes, and demonstrate the proposed approach through a scenario (Objectives 3 and 4).

  3. Validation Phase (Chapter 6): this phase includes validating the approach among designers (Objective 5).

1.4 Research Outcomes and Significance

In achieving the research aim and objectives, the research presents an integrated data-driven urban design approach that assists designers during the urban design process in making better evidence-based decisions, providing a systematic conceptual structure of data-driven urban design processes, and a free and open-source computational toolkit prototype that integrates data and design tools in a visual programming environment. The significance of this research is demonstrated through its outcomes, which are comprehensively outlined:

  • The research systematises data-driven urban design processes, considering the steps of the urban design process, data-driven cycles, and computational technologies. The proposed systematisation of data-driven urban design processes provides guidance for urban design professionals, and better support during evidence-based urban design decision-making. Also, it can be used as conceptual structure for the development of new data-driven urban design tools.

  • The research offers an integrated approach for data-design in an environment that enhances the overall workflow of data-driven urban design processes. This integration improves designers’ experiences regarding interoperability and data accessibility during the urban design process.

  • The research establishes the socio-political implications of the process of digital toolmaking for data-driven urban design in the context of future smart cities, advancing the awareness of the social and political implications of the use and development of computational tools in designing future (smart) cities.

  • The research initiates a discussion about the potential differences and similarities of the steps of the design process between traditional and data-driven designers.

1.5 Thesis Structure

This thesis is structured into eight chapters.

  • Chapter 1 introduces this PhD, proving the relevant background information and justification for this research in order to provide a proper understanding of its context and purpose. Also, it presents the aim, objectives, and overall structure of the research design, followed by the outcomes and significance of the research, and the structure of the thesis.

  • Chapter 2 presents a literature review, examining the existing theoretical foundations of research related to the development of Smart Cities, Data-Driven Urban Design, and Computational Design-Toolmaking, providing an understanding that informs the subsequent stages of this research.

  • Chapter 3 presents the methodology and research design. The chapter details the three phases of this research.

  • Chapter 4 presents the findings and a discussion of the semi-structured interviews conducted with the sixteen designers. These semi-structured interviews have the purpose of providing a snapshot of the current urban design practice, providing guidance for the development of the integrated data-driven urban design approach, which includes both a framework and a computational toolkit prototype.

  • Chapter 5 presents the strategies used for the approach development and its implementation as computational toolkit prototyping.

  • Chapter 6 presents the validation process carried through rounds of focus group interviews.

  • Chapter 7 discusses the key findings described in Chapters 4, 5, and 6, providing explanations and evidence to address the five research objectives.

  • Chapter 8 concludes the PhD and presents the contribution of this research in this research field. Finally, the chapter presents the limitations and future research directions.