Renato Casagrandi (p), Lorenzo Mari, Davide Bogani

Zoltán Elekes

The COVID-19 pandemic has unequivocally demonstrated the vulnerabilities of global interconnections and their impact on local communities. The staggering number of deaths (7+ millions) and economic costs (estimated at around $14 trillion by the end of 2023) highlighted the utmost importance to project scenarios for the spread of emerging infectious diseases (EIDs) well before a possible next “Disease X” pandemic may occur. To prevent the potential spread of EIDs, it is crucial to model how diseases can be able to spread within the (typically pantropical) source regions and, from there, how they can reach remote connected nodes in sink regions, through large-scale networks (i.e., airlines). A key task for modellers becomes then to quantify and map how the hazard of potential pathogen emergence interplay with the exposure of people at source, due to local-scale human mobility, so as to assess the risk of disease transmission over longer distances toward sinks. A wealth of data can now be available to perform accurate analyses and to develop movement/network models that can serve as excellent bases for identifying vulnerabilities for global health security. As a prototypical example to illustrate the functioning of our operational framework, we show how the potential risk of pandemics could be estimated in the case of Indonesia by integrating two sets of data-driven information: a simple ecological mapping of the risk of pathogen emergence and a data-driven exposure mapping of human mobility derived from GPS and mobile phone connections. Integrating different regional-scale network models with potentially global-scale mobility network models is challenging, yet crucial to identify the nodes and the connections of greatest vulnerability and thus to work on strategies for assessing the pandemic risk. And to avert, or at least devise methods for monitoring, preventing or even controlling, any upcoming pandemics.

Federico Ninivaggi (p), Mrs Eleonora Cutrini

Renato Casagrandi

The objective of this work is to build a theoretical model that relates the connectivity architecture of a network with the type of output it provides and the related market institutional arrangement that derives from it. To this end, the relationships between information, know-how and entropy will be considered; in fact, these three elements will determine the degree of complexity of the output produced by the system which, in turn, will impact the consequent organizational form of the market. The market structure is an expression of the connectivity architecture of a complex system, such as to guarantee stability over time in the production of the same output. The work moves from the framework theorized by Hidalgo (2015) that focuses attention on the information and know-how that individuals and networks can store and use. Each node of the network and each network has a limited capacity to collect information. Comparing different networks with each other, the result is an unequal spatial accumulation of knowledge and know-how between the networks depending on their organization and extension. To be able to produce highly complex products, larger and more organized networks are needed. They can contain a lot of knowledge and at the same time exhibit high computation capacity. Therefore, the connectivity architecture with which a system is structured to produce complex products will be decisive. In this sense, central to the network and its hierarchy is the ability to organize a "Smithian" division of labor between the different nodes. The division of labor is also accompanied by the division of knowledge and know-how. Computing and information capacities are thus distributed in the network. The fragmentation of know-how leads to the emergence of a difficulty, introducing a combinatorial problem, that of connecting the nodes in a structure that allows us to reconstruct what was fragmented. This is where the vulnerability of a network arises. For complex products, it is necessary to fragment tasks, knowledge, and calculation capacity. By distributing these quantities along the network in a hierarchical manner, the risk that a shock could interrupt fundamental connections to produce output exposes the network to idiosyncratic vulnerabilities. The work aims to generalize and identify the relationships between knowledge, know-how, production complexity, entropy, and connectivity architecture with the different market forms. These different architectures imply fundamental features in terms of redundancy, efficiency, vulnerability, and resilience of the systems.

Roberto Patuelli (p), Caterina Malandri, Rabasco Michele, Aura Reggiani, Rebecca Rossetti

Federico Ninivaggi

Bike-sharing systems are attracting considerable interest in the literature for their potential key role in encouraging the transition from car-based private transport to more sustainable mobility, particularly in the urban context. The success rate of bike-sharing schemes depends on many factors, including demographics, morphology, and service design. When designing a bike-sharing service, one of the first issues that arise is where to locate the bike stations.
This paper aims to propose a new location model for bike stations, based not only on spatial economic factors but also on factors related to the robustness of the urban public transport network. Such a model can be applied to a set of feasible alternative sites to install bike-sharing docks. The hypothesis behind this choice is that, given the complementarity between public and shared transport, implementing bike-sharing stations near public transport stops would increase the integration between the two modes.
Our aim is to guide decision-makers in ranking these alternative sites. Since this is an optimization problem, we use the Analytic Hierarchy Process (AHP) within a multi-criteria analysis (MCA) to identify the most suitable locations for new bike-sharing stations. AHP is a method grounded in mathematical and psychological principles that offers a systematic approach to handling intricate decision-making scenarios through pairwise comparisons (Saaty, 2008).
In particular, we perform our analysis by considering the following criteria (derived from the literature): a) proximity of bike-sharing stations to points of interest/amenities; b) socio-demographic characteristics and pollution rates of the surrounding areas; and c) network-based features to explore the robustness of public transport networks.
Concerning the proximity criteria, we adopted four (quantitative) measures: proximity to green areas, proximity to sports/entertainment centres, proximity to schools, and proximity to tourism areas.
It should be noted that the weights of all the criteria are based on the answers from a questionnaire addressed to a group of experts on sustainable mobility. These expert opinions allow us to establish different scenarios, and, consequently, the hierarchical importance of the alternatives ensuring, by means of AHP, a comprehensive and balanced evaluation.
This approach will be applied to the transport stations of selected European cities. It can be considered a prototype model for further applications in urban areas.

Zoltán Elekes (p), Gergő Tóth, Rikard Eriksson

Roberto Patuelli

While regional economies can be regarded as webs of specialized production units, largely dependent on the technologies, skills and tacit knowledge integrated in the process of value creation, the role that explicit network structures play in regional economic resilience is still poorly understood. In this paper we use information on three decades of inter-industry labour flows within regions to construct networks of labour reallocation for 72 local labour markets in Sweden. Drawing on advancements in network science that are novel in regional science, we stress-test these networks against the sequential elimination of their nodes, finding substantial heterogeneity in network robustness across regions. Our previous work indicates that this network robustness conditions regional economic resilience in the cross-sectional setting of the 2008 crisis. In this work we extend on this basic finding by first exploring how local labour flow network structure changes over time. Second, using established measures of node contribution to overall network robustness, we quantify the individual contribution of an industry to the labour flow network robustness of its corresponding regions. With this approach we are able to explore the connection between robustness and industry-region-specific resistance to and recovery from idiosyncratic and systemic economic shocks. Finally, as regional resilience in an evolutionary perspective entails the capacity to withstand shocks as well as the ability to create new growth paths, we combine the standard measurement of industry relatedness to the local economy (relatedness density) with our measure of industry contribution to overall network robustness. This allows us to study how the (related) diversification of regional economies brings about more or less robust labour flow structures. These findings elaborate on how variation in the self-organisation of regional economies as complex systems makes for more or less resilient regions over time, thereby forging new connections between the literatures on regional economic resilience, evolutionary economic geography and network science.

Jelena Stankovic, John Östh, Umut Türk, Marija Dzunic, Marina Stanojevic, Ivana Marjanović (p), Aura Reggiani

This research investigates the relationship between spatial connectivity and resilience, focusing on the context of Swedish municipalities. The research on resilience (conceptualized as the capacity of a system to withstand shocks or absorb perturbations) often lack considerations for connectivity and interactions.
Our primary research question revolves around understanding the interplay between connectivity and resilience development. Connectivity, defined as the establishment and maintenance of spatial connections, is examined in the context of spatial labour market networks characterized by main nodes and corridors. Additionally, the study delves into the complexity of networks, considering their irregular and dynamically evolving structures.
The research specifically explores how spatial connectivity is linked to community resilience. Topological indexes derived from network connectivity measures are used to estimate the impact of network connectivity on resilience. The analysis employs Ordinary Least Squares (OLS) regression and Quantile regression.
Data from the PLACE database, encompassing labor market, education, and social sector information, along with the Swedish Electoral Authority database, is employed for the year 2017 at the municipality level.
We identify network connectivity, age, and education as the most influential variables in explaining resilience variations.