Emerald Dilworth (p), Emmanouil Tranos, Daniel Lawson

Lenka Hasova

When individuals and companies struggle to keep up with new technologies, there can be a wide range of barriers stifling their access, and in turn, many consequences from failing to keep up with the cutting edge. As we find ourselves in the midst of the fourth industrial revolution, we look to firms involved in green and digital technologies in the UK to better understand how and where twin transition technologies are developing. Where there are developments in technologies to digitise and green the economy, this twin transition (TT) leads to sustainable economic growth and multiple environmental benefits.

Traditionally researchers have relied on surveys, patent data, and place attribute-based data to investigate such problems. Instead we utilise the “digital breadcrumbs” created simply from the use and adoption of the Web. From vast quantities of web crawl data, we construct directed, weighted networks of hyperlink connections at various levels of granularity, in addition to utilising the webtext to inform if a website has an interest in green and/or digital technologies. We do this by exploiting existing Large Language Model (LLM) semantics information to apply term frequency–inverse document frequency (tf-idf) in order to explore if firms are involved with twin transition technologies.

Such data can reveal the spatial structure of those involved with TT technologies, and importantly allow these relationships to be contextualised. We use these web data to create meaningful geographic knowledge at the postcode level, and highlight where and why there is big growth for twin transition technologies, as well as where there is a struggle. Our empirical findings have the potential to support and inform city and regional policies.

Andrew Schendl (p)

Emerald Dilworth

The ‘x-minute city’ concept has been widely used in regional studies and urban geography with the term driving recent urban planning policy across cities and regions. This concept is grounded in four key dimensions: proximity, ensuring the availability of nearby amenities; density, serving a significant population; diversity, fostering mixed-use neighborhoods and cultural integration; and digitalization, incorporating advanced technologies such as sensors and green infrastructure. Existing modelling techniques for the x-minute city often struggle to capture the complexities of accessibility, diversity, and individual mobility patterns. This paper seeks to integrate various theoretical frameworks to analyse urban movement within the x-minute city, with a particular focus on greenspace accessibility as a critical component. Proximity-based spatial interaction approaches, essential in spatial analysis, lay the groundwork for conceptualizing the x-minute city. However, these methods sometimes fall short in addressing the nuances of individual accessibility and the diversity of urban life. Behavioral ecology frameworks, like movement ecology, provide social perspectives on human movement motivations but lack specificity in accounting for individual daily routines. Network-based human mobility models are adept at simulating personal routines but need integration with broader frameworks to be effective. We propose Multi-context Inclusive City (MIC) models which combine elements from spatial interaction models, human mobility networks, and movement ecology to provide a comprehensive understanding of human movement within the x-minute city. By integrating these different perspectives, MIC models offer a more holistic and nuanced approach for urban planners, public health officials, and other stakeholders to address complex urban movement dynamics in solving real-world problems.

Xueqing Liu (p), Ben Derudder, Frank Witlox

Andrew Schendl

Abstract:
Despite growing scholarly attention on the role of urban networks for understanding the dynamics of regional integration, a more explicit focus on the multiplexity of networks is required for understanding the nuanced ways in which cities regionally integrate, particularly in labor markets. Applying a panel model to data for the Yangtze River Delta for the period 2014-2021, we analyze and compare the different impacts of transport and technology network connections on labor market integration. We then abstract these different types of connections as a two-layer multiplex network, and explore their (potential) interplay. Our analysis reveals a positive correlation between transport networks and labor market integration, contrasting with a U-shaped effect observed in technology networks. We also observe evidence for a nuanced interplay between cities interconnected through these transport and technology networks. Specifically: 1) cities’ overall connectivity in the multiplex network has a U-shaped effect on labor market integration, and 2) the interdependencies between cities (across different network layers) contribute to an increased integration level of labor markets. We reflect on the broader implications of our empirical findings for regional development strategies and discuss possible avenues for further research.

Lenka Hasova (p)

Xueqing Liu

Spatial Interaction Models (SIM) have been used in geographical, economic and social science research for over half a century.
One of the significant problems with SIMs is the inability of the gravity equation to account for the spatial structure of the interaction, implicitly related to human behaviour. Much of the research explores what variable should be added to the model to account for the spatial structure that exists in the model's errors. The major limitation of this research is that there is no method or framework to empirically validate whether the selected variables account for the model's spatial structure. This talk addresses this issue by rethinking the causal pathways in SIMs by using Structural Equation Modelling (SEM) to create a framework to validate the addition of independent variables to SIMs.

Thus far,
it was assumed that distance is independent of masses and dependent on flow volume (based on the gravity equation). However, there is an argument to be made for the case of distance being dependent on the masses and endogenous to the flow volume due to the human understanding and use of space, where the already attractive destination may become more attractive with easier access and vice versa. Changing this basic assumption allows us to rethink the SIM equation and define one more appropriate for the bivariate outcome of flow and distance. Compared to the traditional SIM, such new model is no better at predicting spatial interaction. However, it allows for the exploration of the latent space of the SIM and its covariance with any variables added to the base model. In the empirical part of this work, this new validation framework is used to explore the ability of the Competing Destination model (SIM enhanced for accessibility parameter) and network structure SIM (SIM enhanced for network structure measure parameter) to account for the variation that exists in the latent space of the model, possibly the variation belonging to spatial structure. Bayesian modelling framework is used here to provide precise estimates and, thus, a more reliable framework for variable validation in SIM.

This novel method has the potential to help anyone using SIM to identify influential variables for their model. Nevertheless, a broader discussion on the validity and relevance of this method is needed. This talk aims to spark such discussion and collect feedback from a wider audience.