Tomaz Dentinho (p), Karina Sass (p)

Louafi Bouzouina

This paper tries to frame the complexity of spatial interactions using the structure of space; imagined as a network of spatial channels built up by interacting flows. The first assumption is that space is a productive system represented by flows that justify, built up, and maintain channels, the same way as income justifies, constructs and preserves capital. The second assumption is that either there is an organic evolution of the territory that maximizes the size of channels, like a city that puts most of its incomes in urban infrastructure, or there is human rationality that optimizes the size of the channels to maximize human interaction flows. Using an entropy function, we can explain the relation between channel sizes and the flows that move along them. On the other hand, we use the integral of the entropy function to explain how the size of channels are build-up through the accumulated flows they generate. Finally, entropy functions that explain flows and channels are open up to include distance constraints and external impacts. The demonstration of these conceptual and operational models of complex spatial systems comes in the last part of the chapter. First, with the calibration of a model that explains the spread of infectious diseases to exemplify the organic maximization of infection channels. Second, with the estimation of a regional economic model that maximizes flows subject to the spatial structure of channels, confined by policy measures.

Mehmet Guney Celbis (p), Cem Ozguzel, Karima Kourtit, Peter Nijkamp

Tomaz Dentinho

This chapter aims to explore the determinants of the reduction in workplace mobility during 2020 in 10 Canadian Provinces and 50 US states through the analysis of 29 variables representing regional industrial composition, remote working potential, government support policies alongside with other main socioeconomic and demographic attributes is used. The main empirical results are obtained through the use of an extreme gradient boosting machine algorithm followed by the analysis of Shapley Additive Explanations (SHAP) values. Spatial dependences are taken into account. Results suggest that regions where mobility was reduced more were those with a higher share of persons in jobs amenable to remote working and persons working in the information and communication and science and technology oriented sectors. Conversely, regions with larger shares of distributive trade, repairs, transport, accommodation, food services, finance and insurance sectors experienced increased workplace mobility.

Ouassim Manout (p), Thomas Leysens, Louafi Bouzouina (p), Karima Kourtit, Peter Nijkamp

Mehmet Guney Celbis

see extended abstract