ABSTRACT. The study of financial bubbles is a highly controversial topics in economics and finance. Despite a large number of economic analyses, anecdotal evidence and increased theoretical attention, the quantitative monitoring and modeling of financial bubbles still miss standards broadly accepted by scholars. Our study focuses on the famous dotcom bubble that inflated financial markets during the period 1995-2000. Specifically, we investigate Nokia share ownership during the onset of the bubble and during its aftermath up to 2010 by investigating a unique database that tracks the financial ownership of all Finnish legal entities. We document a persistent flow of investment from foreign investors in the Nokia company during the inflation period of the bubble. This is a typical anecdotical scenario observed in the setting of financial bubbles. A second fundamental observation concerns the number of Finnish investors having an open investment position in Nokia at a given day. This number increased more than exponentially during the 1998-2000, reflecting a dramatic raise of attention at a country-wise level during bubble inflation. We exploit the unique combination of studying a multinational company that was among worldwide protagonists during dotcom bubble and a complete coverage of daily financial ownerships for all Finnish investors. The distribution of investment gains and losses was strongly inhomogeneous across different categories of investors. Financial professionals were better equipped to obtain gains during bubble inflation and limit losses when the bubble bursts. On the contrary, investors with limited financial expertise gained during bubble inflation but incurred in significant losses — or struggled to limit them — after the bubble burst. Joint work with Federico Musciotto (University of Palermo, Italy) and Jyrki Piilo (University of Turku, Finland)

ABSTRACT. In this study, we compared the performance of four different methods for multi-label text classification using a specific imbalanced business dataset. The four methods we evaluated were Fine-tuned BERT, Binary Relevance, Classifier Chains, and Label Powerset. The results show that Fine-tuned BERT outperforms the other three methods by a significant margin, achieving high values of accuracy, F1-Score, Precision, and Recall. Binary Relevance also performs well on this dataset, while Classifier Chains and Label Powerset demonstrate relatively poor performance. These findings highlight the effectiveness of Fine-tuned BERT for multi-label text classification tasks, and suggest that it may be a useful tool for businesses seeking to analyze complex and multifaceted texts.

ABSTRACT. In this paper, we introduce the SINr Python package to train word and graph embeddings. The SINr approach is based on community detection: a vector for a node is basically the distribution of its connections through the communities detected on the graph at hand. Because of this, the algorithm runs very fast, and does not require GPUs to proceed. Furthermore, the dimensions of the embedding space are interpretable, those being based on the communities extracted. The package is distributed under Cecill-2.1 license and is available on Github and pypi.

ABSTRACT. In this work, starting from a massive data collection, we tested the possi- bility of infographing through a multilayer network, the various main topics while also taking into account the context and meaning of the connected node/topic. We collect data inherent to the 2022 Qatar FIFA World Cup event, and from the data obtained, we schematized the various layers in relation to the stages of the champi- onship. Using the software Gephi we have created the multilayer networks, being able to show topics-word and word-word relationships, while also showing the dynamics and change over time of the most discussed topics between layers.

ABSTRACT. Real-world multilayer networks can be very large and there can be multiple choices regarding what should be modeled as a layer. Therefore, there is a need for their effective storage and manipulation. Currently, multilayer network analysis software use different data structures and manipulation operators. We aim to categorize operators in order to assess which structures work best for certain operator classes and data features. In this work, we propose a preliminary taxonomy of layer and data manipulation operators. We also design and execute a benchmark of select software and operators to identify potential for optimization.

ABSTRACT. By using the mesoscopic network structure, this study proposes a general backbone extraction framework. In fact, many actual networks consist of dense parts of nodes known as communities, multi-core, or components. We suggest extracting the backbones from each of the different components of these groups separately and then fusing them to address the heterogeneity of these groupings. The effectiveness of the suggested approach versus classical techniques, which are agnostic to the mesoscopic structure of networks, is demonstrated by experimental studies on real-world networks.

ABSTRACT. NetBone is a new open-source Python package designed to simplify analyzing complex networks. With a wide range of techniques available, NetBone allows researchers to extract the backbone of a network while preserving its essential structure. The package includes nine structural methods and five statistical techniques, offering users a comprehensive solution to network analysis. It is user-friendly and straightforward to use, with easy installation. The package accepts different types of inputs, including data frames or Networkx graphs, and provides evaluation measures for comparative purposes. Additionally, NetBone offers an option to generate plots. Its versatility makes it a valuable tool for data scientists and social scientists, significantly enhancing their research and data analysis capabilities.

ABSTRACT. Logistics management relies on the development of strategies that promote the fluidity of logistics operations and transactions, which in turn rely on the advancement of dematerialization and digital data sharing. However, a major obstacle to the implementation of these strategies is legal certifica- tion, especially when it comes to the execution of a contract between several partners involved in the entire supply chain. To overcome this obstacle, we propose the use of next-generation smart contracts, also known as smart legal contracts, which have legal standing. The objective is to create a digital trust between the different partners, which would allow to automate and speed up the logistic transactions.

ABSTRACT. Data access control is a crucial aspect of data management. As secure distributed ledger, blockchain is widely used today to manage consent for data access. However, a blockchain is not ideal for storing large volumes of data due to its characteristics. So, it's often coupled with off-chain systems to facilitate the storage of these kinds of data. Therefore, data located outside the blockchain requires security procedures. This article proposes a securing mechanism based on data encryption, to secure user data stored in off-chain systems. The protocol uses a symmetric key system, which prevents the reading of data by malicious actors who would have access to data stored outside the sphere of the blockchain.

ABSTRACT. The maritime sector is facing numerous challenges such as inefficient processes, lack of transparency, and security issues. Permissioned blockchain technology has emerged as a potential solution to these challenges. Unlike public blockchains, permissioned blockchains are private networks where only authorized participants have access to the network. We argue for the use of permissioned blockchain technology in the maritime sector. We highlight the potential benefits of permissioned blockchain technology, including increased efficiency, transparency, and security in the supply chain. We conclude that the adoption of permissioned blockchain technology in the maritime sector is likely to increase in the coming years as stakeholders realize the benefits of this technology.

ABSTRACT. This project aims to better understand structural differences between the preterm and term infant brain. To this end, we employ graph theory measures to compare structural brain imaging data. We aimed to establish whether differences exist between preterm and term infant brains through comparison of rich club organisation. Further we considered factors that may be influencing previous inconsistent results.

The present research uses diffusion tensor imaging (DTI) data from 426 infants through the Developing Human Connectome Project (dHCP).

In this study we found rich club organisation in both preterm and term infant brains at term normalised age. Beyond this, preliminary analysis found higher RC coefficient in preterm infants.

ABSTRACT. The last decades have been marked by an increasing number of scientific papers dedicated to agent-based modeling applied in various domains and especially the integration or coupling with GIS data for more realism. This dynamic approach of modeling involves autonomous agents cooperating to solve a complex problem underlined in a domain such as a land use-cover system where human agents consequently modify their environment to live. However, the spatial aspect of intelligence led by the representation and reasoning of an environment is not explicitly captured during the modeling. Nowadays, the rise of its technologies offers new research opportunities to think and work on the analysis, formalized representation, and the use of the physical space of agents using spatial data. This naturally brings the agent more information, increases its spatial knowledge, and thus strengthens its decision-making process. This paper proposes a conceptual model for spatial knowledge representation from spatial data in view to improve the cognitive dimension of agents in land-use simulations. To illustrate the design of spatial knowledge semantic net, some examples have been presented. Finally, some ways have also been investigated about knowledge acquisition by cognitive agents.

ABSTRACT. This research aims to advance scholarly understanding of how team shared knowledge contributes to founders’ entrepreneurial orientation and well-being in the startup pro-cesses. The purpose is to address the shortfall in the literature by examining team-level and individual-level characteristics that impact on innovative behaviors. Drawing on research on entrepreneurship and organizational communication, this research ex-amines ways in which entrepreneurs rely on each other’s specialized knowledge to promote innovation and to nurture growth during critical periods of development. Us-ing quantitative survey approach, this research will collect data from entrepreneurs in the knowledge-intensive industries in Japan.

ABSTRACT. In a near future, because of climate change and the necessary reduction of soil artificialization, cities will be likely to seek for free spaces for their development by reusing already artificialized areas. One possibility could be to look at infrastructures dedicated to car mobility, infrastructures that sprawl over very wide areas and that cover a substantial part of city surfaces. This work investigates the possibility of reducing cities streets-networks by converting several lanes streets into one-way streets with only one lane.

ABSTRACT. The main objective of this research is to analyze the connectivity of cities in a coupled network made of planar (railways) and non-planar (maritime) topologies. It takes the state of the network during the period 1880-1925, namely the context of the First Globalization wave (1880-1914), when trade and urban development were closely tied to progress in communications systems and especially steam propulsion. Edges represent intercity physical infrastructure on land, and inter-port ship voyages at sea. Main results underline a stronger relationship between railways and steam shipping compared with sail across ports and port cities of the world.

ABSTRACT. Agent-based modelling has been used in many studies of urban expansion, land use and land cover change patterns. While representing a powerful tool for depicting and formulating predictions about the evolution of interconnected complex systems, this method also poses a series of challenges to the researcher community, most notably in terms of model calibration and validation, and output data visualisation. Based on these findings, we present an agent-based model developed in GAMA, coupled with a data exploration platform coded in python, for analysing commuting patterns in the canton of Geneva, Switzerland. Output datasets generated from a set of simple evolution rules for the agents, are distributed in open access together with the code for the associated data visualisation platform. This prototype is early work in developing a series of agent-based models for simulating urban expansion and land cover change dynamics, together with their own data exploration platforms for calibration, validation and output data analysis purposes. These toolboxes will be developed with the intent of addressing the various shortcomings in agent-based modelling research discussed in this paper.

ABSTRACT. Using a nationwide dataset with millions of real estate transactions, this paper investigates the relationship between housing prices and city size through a radial (center-periphery) analysis. We find that housing price radial profiles scale in three dimensions with the power 1/5 of city population. Nonetheless, housing prices in the city center seem to be more sensitive to city population, raising the question of housing affordability in the center for low-income households.

ABSTRACT. We present in this paper a formal model that focuses on relational complexity. The model enables to connect concepts from the physical world and concepts from the realm of ideas both topological and geometrically. The formal properties obtained in this model, such as identification, adjunction, copy and deletion are coherent with the phenomena that are modelled.

ABSTRACT. Proposed by Bak, Tang and Wiesenfeld in 1987, the Abelian sandpile was the first mathematical model to describe the phenomenon of Self-Organized Criticality (SOC). In its canonical form, the model is defined as a cellular automaton with cells arranged in a regular grid structure that can also be represented as a graph: nodes representing cells and edges the neighborhood of a cell. By introducing random modifications to the regular grid structure, this paper aims to explore the impact that other regular but non-grid topologies can have on the dynamics of the system. Starting with the canonical topology as a baseline, edges are rewired at random with the only constraint of keeping constant the indegree and outdegree of nodes. The different graphs studied start from a regular grid structures (when no edge is rewired), going through small-world graphs (for small rewiring coefficients), to end with random topology. The obtained results show that, among the wide range of used metrics, the signature of SOC is preserved despite different graph structures have an impact on the dynamics of the model.

ABSTRACT. In this work, we investigate the analysis of generators for dynamic graphs, which are defined as graphs whose topology changes over time. We focus on generated graphs whose order (number of nodes) varies over time. We introduce a novel concept, called "sustainability," to qualify the long-term evolution of dynamic graphs. A dynamic graph is considered sustainable if its evolution does not result in a static, empty, or periodic graph. To illustrate how the analysis can be conducted, a parameterized generator, named D3G3 (Degree-Driven Dynamic Geometric Graphs Generator), that generates dynamic graph instances from an initial geometric graph, has been introduced in \cite{bridonneau_2023}. The evolution of instances obtained from this model is driven by two rules that operate on the vertices based solely on their degree. In this work, we focus on particular sets of parameter leading the generator to produce graphs whose order increase or decrease exponentially or remain almost constant.

ABSTRACT. In this paper, we investigate the problem of inferring the network coupling strengths from partially observed time series data in an Ising model on scale-free networks. By assuming that only a certain fraction of observations for spin states are available, we study how an observer, who wants to maximise the accuracy of the network inference, should distribute a limited number of observations. Along with the benchmark case of randomly-chosen hidden nodes, we propose degree-dependent heuristics for observation allocations. We observe two regimes for the best observation strategies based on varying amounts of missing data. If only a small proportion of data cannot be observed, then one should focus on the observation of the states of periphery nodes. Otherwise, if a large number of states cannot be observed, allocating more observations to the high-degree nodes is preferable.

ABSTRACT. The concept of keystone species was originally referred to an ecosystem level, defined as species that are disproportionally important for ecosystem functioning, which greatly surpasses that of other more abundant species in the community. Yet, their detection has so far been restricted to examining one or only a few interaction types in a community (e.g. parasitism, pollination, seed dispersal, herbivory). However, all species are involved in a myriad of interactions with other coexisting species, playing therefore multiple ecological roles that together define the multiple dimensions of their Eltonian niche [1, 2].

Given the multidimensional nature of ecosystem interactions, any attempt to fully capture their richness requires a representation that meaningfully incorporates such complexity. Thus, in order to gain a more comprehensive understanding of a species’ functional importance, it is necessary to move beyond considering only its single role to considering its multiple ecological roles, i.e. to shift from unifunctionality to multifunctionality [3]. To our knowledge, only one study has empirically estimated the weight of edges between layers by quantifying the role of the same individual in two ecological processes [4].

Our approach follows the consumer-resource paradigm, where plants are seen as “resources”, and “consumers” encapsulate different types of animals or fungi. In this work, interactions between both resources and consumers develop along six different observed functions: pollination, herbivory, seed dispersion, and pathogen-, saprotrophic-, and symbiotic-fungal interaction. The precise span of dimensions is here driven by the extent of our data, based on observations recently collected in the islet Na Redona, which includes direct observation of 16 plant species, 675 animal/fungus species, interacting across six fundamentally different ecological functions. Incorporating the functional dimension, the complete relational dataset is thus formalized in terms of a rank-3 tensor that we call the Resource-Consumer Function tensor (RCF).

We interpret the architecture of this tensor as a weighted, multipartite, multilayer network and effectively visualise it as a multipartite edge-colored weighted network (Figure 1) [5]. The network displays two types of nodes: resources (plant species) and consumers (animals and fungi), with interactions (links) taking place between groups but no direct intragroup links. Each layer of the network represents a specific function and the strength of each interaction is represented by a link weight. Consumers (animals, fungi) are often centered around a single plant species and thus form clusters (see however the cluster formed by Lavatera Marı́tima and Geranium Molle). Interestingly, cross-cluster links are also present, thereby the ecosystem is entangled.

A first question worth addressing from the RCF is to quantify the relationship between the resources of the ecosystem and the functions the system embodies. This is achieved by integrating out the consumer index and thereby building a Resource-Function Matrix (RFM), which is a multifunctional ecological network. For that purpose, we define the participation strength of a plant species in an ecological function. This matrix shows a stylized nested structure commonly found in e.g. mutualistic interaction networks, world-trade, inter-organizational relations, and others [6]. The complex nested pattern observed suggests that certain ecological functions and plant species can be classified as ”generalists” or ”specialists”, thereby highlighting the hierarchical nature of multifunction keystonenes. We formulate the novel concept of function keystonness, which focuses on the robustness of ecosystems with respect to perturbations to (instead of species as commonly assumed).

To further understand the multifunctional species keystone-ness and the role of plant species as ecosystem assemblers, we project the RFM into the function class and extract a Function-Function Interaction Network (FFIN). We quantify the ecosystem robustness against perturbations (extinctions) of plant species by sequentially pruning edges in FFIN. Additionally, we rank plant species, based on their multifunctional keystoneness, by conditioning the FFIN to each single plant species and thus obtaining a set of 16 6-node networks. By properly normalizing sets of node and edge weights, we rank based on the specific role of resources (plant species) as brokers of functions.

The dual concept of function keystonness can be addressed by following a similar mathematical manipulation: initially starting again from RFM we project now on the plant class and thus construct a Resource-Resource (i.e. plant-plant) interaction network (PPIN). We address two complementary questions: (i) how robust is the ecosystem against perturbations of functions? And (ii) how to quantify the heterogeneous roles and impacts of each different function in the ecosystem? The first question is answered by performing a robustness analysis in the PPIN, by sequentially pruning edges (i.e. functions) and analysing the response. We subsequently proceed to disaggregate PPIN by conditioning on each function, and extract 6 different plant-plant networks. The resulting ranking certifies the heterogeneity of roles and impacts of the different functions.

[1] Anna Eklöf et al. “The dimensionality of ecological networks”. In: Ecology letters 16.5 (2013), pp. 577–583.

[2]D Matthias Dehling and Daniel B Stouffer. “Bringing the Eltonian niche into functional diversity”. In: Oikos 127.12 (2018), pp. 1711–1723.

[3] Peter Manning et al. “Redefining ecosystem multifunctionality”. In: Nature ecology & evolution 2.3 (2018), pp. 427–436.

[4] Sandra Hervı́as-Parejo et al. “Species functional traits and abundance as drivers of multiplex ecological networks: first empirical quantification of inter-layer edge weights”. In: Proceedings of the Royal Society B 287.1939 (2020), p. 20202127.

[5] Manlio De Domenico et al. “Mathematical formulation of multilayer networks”. In: Physical Review X 3.4 (2013), p. 041022.

[6] Manuel Sebastian Mariani et al. “Nestedness in complex networks: observation, emergence, and implications”. In: Physics Reports 813 (2019), pp. 1–90.

Acknowledgement: We acknowledge support from the Spanish Agency of Research (AEI) through Research and Development Projects Program (Grant PID2020-114324GB-C22 funded by MCIN/AEI/10.13039/501100011033). In particular, MCB thanks financial support Grant PID2020-114324GB-C22 funded by MCIN/AEI/10.13039/501100011033 and by “ESF Investing in your future”.

ABSTRACT. Temporal graphs have edges that bear labels according to the time of the interactions between the nodes. Betweenness centrality has been extended to the temporal graph settings, and the notion of paths has been extended to temporal paths. Recent results by Buss et al. and Rymar et al. showed that the betweenness centrality of all nodes in a temporal graph can be computed in $O(n^3\,T^2)$ or $O(n^2\,m\,T^2)$, where $T$ is the number of time units and $m$ the number of temporal edges. In this paper, we improve the running time of these previous approaches to compute the betweenness centrality of all nodes in a temporal graph. We give an algorithm that runs in $O(n\,m\,T+ n^2\,T)$.

ABSTRACT. Social relationships are the key determinant of crucial societal outcomes, including diffusion of innovation, productivity, happiness, and life expectancy. To better attain such outcomes at scale, it is therefore paramount to have technologies that can effectively capture the type of social relationships from digital data. NLP researchers have tried to do so from conversational text but mostly focusing on sentiment or topic mining, techniques that fall short on either conciseness or exhaustiveness. We propose a theoretical model of 10 dimensions (colors) of social relationships that is backed by decades of research in social sciences and that captures most of the common relationship types. We trained a deep-learning model to accurately classify text along these ten dimensions. By applying this tool on large-scale conversational data, we show that the combination of the predicted dimensions suggests both the types of relationships people entertain and the types of real-world communities they shape. We believe that the ability of capturing interpretable social dimensions from language using AI will help closing the gap between the oversimplified social constructs that existing social network analysis methods can measure and the multifaceted understanding of social dynamics that has been developed by decades of theoretical research.

ABSTRACT. Information and mobile technology have become essential in modern life, transforming the way we communicate, access information, entertain ourselves, and do business. Thereby we have the opportunity to access new datasets that did not exist for previous generations of scholars. Traditionally urban commute was studied based on census datasets, which represent the commute flow as a static number and get updated once in a few years. Over the last two decades, mobile phone datasets enabled new research avenues. In this paper, we utilize mobile phone mobility data to define a signature of urban districts in the city of Brno in the Czech Republic and leverage it for urban zoning.

ABSTRACT. Understanding the behavior of commuters is crucial as the number of commuters steadily rises, causing significant traffic congestion in many cities. Indeed, commuter behavior is vital in city and transport planning and policy-making. Previous studies have investigated various factors that may impact commuting decisions. Still, these studies are often limited by the scale of data examined, including time duration, space, and the number of commuters. To address this gap, we gathered large-scale inter-city commuting data in Germany and analyzed the weighted commuting network from 2013 to 2021. This work relies on publicly available data so that the results can be reproduced.

ABSTRACT. pSTEM fields (Physical Sciences, Technology, Engineering and Mathematics) are known for showing a gender imbalance favouring men. Such imbalance can be seen at several levels, including jobs in university and industry, where men fill the majority of posts. In academia, it is well known that success comes partially from the value of the researchers' collaboration networks. One of the mechanisms to enrich one's network relates to academic movement; the change of institutions in search of better opportunities within the same country or internationally. In this paper, we look at the data for one specific pSTEM field, Computer Science, and describe the productivity and collaboration patterns that emerge as a function of academic mobility. Our results show that women and men indeed benefit from national and international mobility, women who never changed affiliations over their career are rarely well-cited or highly productive, and women are not well-represented in the overall top-ranking researchers.

ABSTRACT. Modeling socioeconomic dynamics has always been an area of focus for urban scientists and policymakers, who aim to better understand and predict the well-being of local neighborhoods. Such models can inform decision-makers early on about expected neighborhood performance under normal conditions, as well as in response to considered interventions before official statistical data is collected. While features such as population and job density, employment characteristics, and other neighborhood variables have been studied and evaluated extensively, research on using the underlying networks of human interactions and urban structures is less common in modeling techniques. We propose using the structure of the local urban mobility network (weighted by commute flows among a city's geographical units) as a signature of the neighborhood and as a source of features to model its socioeconomic quantities. The network structure is quantified through node embedding generated using a graph neural network representation learning model. In the proof-of-concept task of modeling the location's median income and housing profile in two different cities, such network structure features provide a noticeable performance advantage compared to using only the other available social features. This work can thus inform researchers and stakeholders about the utility of mobility network structure in a complex urban system for modeling various quantities of interest.

ABSTRACT. Cities play an important role in various contexts such as society and economics but also sustainability and ecology. Regarding all the various aspects of a cities composition and development over time, cities can be regarded as complex systems. In recent years urban development and design shifted towards a more pedestrian and cycling-friendly approach. One of the most popular models hereby is the 15-minute city concept or its more generalized version, the X-minute city concept. The Covid-19 pandemic triggered the emergence of a total rethinking of the functionality, composition and design of cities towards rather social , sustainable and resilient concepts. The aim is to provide access to a broad range of amenities within x minutes walking distance in order to recreate social links between the inhabitants and to avoid car journeys for public health and environmental reasons. X-minutes cities is an urban planning objective defined by Carlos Moreno and made public by Anne Hidalgo during the 2020 municipal campaign and since then, Paris serves as an example city for many existing studies in the field. However, we did not find an extensive study showing to which extent Paris can be seen as a 15-minute city. In this project, we analyze the city of Paris , including its composition, accessibility to amenities and usage as an example for the 15- or X-minute city concept.

ABSTRACT. Urban networks are the foundation of modern cities, ensuring most mobilities, such as flows of products and people. While complex networks are known for their vulnerability to attacks targeting most important nodes, urban networks, which possess an homogeneous node distribution, exhibit a greater robustness to nodes failure. Still, demonstrations and protests can lead to massive perturbations. Many protesters gathering in a given place can overload its capacity. This impacts locally the traffic flow, but can also induce a cascade of congestion in the traffic network. While some articles studied the interactions between protesters and counter forces with agent based modeling, to our knowledge none have evaluated the direct impact of a given protest over a whole city. We do not pretend here to reproduce realistic large scale protests. What we intend is to evaluate if an important network disruption can emerge from a simple collective tactic, which rules would be comprehensive at a single human level. Moreover, we seek to identify what rules are sufficient to generate those behaviors. In our study, we present how agents following a common walking tactic perform at gathering and walking in aggregates on a given city. Our first results indicate large-scale impact is reachable with simple rules.

ABSTRACT. This abstract outlines ongoing research that employs spatially-explicit agent-based simulation to explore households’ decision-making on adopting different energy transition measures in Amsterdam. These measures include insulation of walls, roof, and floor and adopting solar panels and heat pumps. This study is conceptually developed based on a comprehensive meta-model Consumat that offers a theoretical framework with macro and micro-level factors affecting consumers’ behavior and a set of behavioral rules for an artificial agent. We use the combination of agent-based modeling and spatial microsimulation to examine the households' adoption decisions across the Amsterdam districts and their contribution to the energy transition by 2030. The preliminary findings suggest differences in the adoption of the measures based on their type, the differences in adoption rates that vary across the city districts, and population groups such as homeowners and tenants. Additionally, we examine different scenarios of possible policy interventions that can inform policymakers on households' observed energy-related decision patterns, examine factors affecting these patterns, and contribute to achieving Positive Energy Districts goals in the city. The findings of this study can also be useful for other cities.