This thesis studies the topological analysis of pedestrian crashes in Città Studi, Milano, over four years (2017-2020). All pedestrian-related crashes occurred in the four years 4,654 crashes were reported across Milan, in our study area Città Studi, 112 crash reports were studied thoroughly. The project will investigate spatial patterns, identify high-risk areas, and explore potential features that can help to improve pedestrian safety in urban areas. Using Geographic Information System (GIS) and MATLAB software to analyse and visualize the data, this study provides a detailed analysis of pedestrian crashes and their relationship to factor characteristics in the city. In this thesis, we extensively explore the dynamics of pedestrian safety in the context of Città Studi, Milano. Our investigation focuses on how pedestrian accidents, vehicle incidents, and network topology, within the road infrastructure interact. We introduce Icenter, a centrality index designed to identify relevant nodes in the road network. It provides insights based on path metrics because Icentr combines node and edge weights with distance scaling considerations. This index offers perspectives on node importance and helps us understand their specific roles in improving pedestrian safety. The investigation begins with a Network analysis plots to understand the complicated connections between crash sites, with different node sizes and colors representing the severity and quantity of crashes, respectively. This aspect of the analysis identifies potential areas where interventions could be most effective in reducing crash incidence. The research then moves on to a cluster analysis after the network analysis. This phase makes use of bar graphs to show the frequency of incidents within identified clusters, focusing specifically on the concentration of crash occurrences and revealing the presence of notable hotspots that present increased risks to pedestrians. The dendrogram analysis is a method that uses hierarchical data clustering to identify similarities and proximity between crash sites. This helps in determining the structure of crash patterns and identifying common features. Heatmaps are used to compare clusters, providing a more critical perspective on cluster overlap and highlighting shared risk factors that transcend individual crash sites. The thesis suggests that strategic urban safety interventions should be based on data-driven insights, with vehicular traffic and spatial configurations being key predictors of pedestrian crash patterns. It emphasizes the importance of integrated urban design, emphasizing pedestrian well-being and the significance of vehicular presence. The thesis also suggests future research into pedestrian behaviour, urban design's role, and socio-economic factors influencing pedestrian safety in urban areas.
Questa tesi sviluppa l'analisi topologica degli incidenti pedonali in Città Studi, Milano, nell'arco di quattro anni (2017-2020). Gli incidenti legati ai pedoni verificati si nei quattro anni in tutta Milano sono stati 4.654, nell’area di studio Città Studi sono stati 112. Il progetto esaminerà i modelli spaziali, identificherà le aree ad alto rischio ed esplorerà le potenziali caratteristiche che possono aiutare a migliorare la sicurezza dei pedoni nelle aree urbane. Utilizzando il Geographic Information System (GIS) e il software MATLAB per analizzare e visualizzare i dati, questo studio fornisce un'analisi degli incidenti pedonali e della loro relazione con le caratteristiche topologiche della città. La nostra indagine si concentra su come interagiscono gli incidenti pedonali, gli incidenti automobilistici e la topologia della rete all’interno dell’infrastruttura stradale. Introduciamo il indice di centralita Icenter progettato per identificare i nodi rilevanti della rete stradale. Icentr combina i pesi dei nodi e dei lati in funzione della loro distanza dal nodo analizzato. L'indagine inizia con l’analisi della rete per comprendere le complicate connessioni tra i luoghi degli incidenti, con diverse dimensioni e colori dei nodi per rappresentare la gravità e la quantità di incidenti. Questo aspetto dell'analisi identifica le aree potenziali in cui gli interventi potrebbero essere più efficaci nel ridurre l'incidenza degli incidenti. La ricerca effettua poi un'analisi dei cluster dei risultati otenuti con l’indice di centraliata’. Questa fase utilizza grafici a barre per mostrare la frequenza degli incidenti all'interno dei cluster identificati, concentrandosi specificamente sui cluster piu numerosi e rivelando la presenza di punti neri che possono presentare maggiori rischi per i pedoni. L'analisi del dendrogramma è un metodo che utilizza il clustering gerarchico dei dati per identificare somiglianze e vicinanza tra i luoghi dell'incidente. Ciò aiuta a determinare la struttura dei modelli di incidente e a identificare le caratteristiche comuni. Le mappe di calore vengono utilizzate per confrontare i cluster, fornendo una prospettiva più critica sulla sovrapposizione dei cluster ed evidenziando i fattori di rischio condivisi che trascendono i singoli luoghi di incidente. La tesi suggerisce che gli interventi strategici per la sicurezza urbana dovrebbero essere basati su intuizioni guidate dai dati, con il traffico veicolare e le configurazioni spaziali che sono predittori chiave dei modelli di incidente pedonale. Sottolinea l’importanza della progettazione urbana integrata, rivolta al benessere dei pedoni. La tesi suggerisce inoltre ricerche future sul comportamento dei pedoni, sul ruolo della progettazione urbana e sui fattori socioeconomici che influenzano la sicurezza dei pedoni nelle aree urbane.
A topological analysis of pedestrian crashes in Citta' Studi, Milano
Gonuguntla, Sai Krishna
2022/2023
Abstract
This thesis studies the topological analysis of pedestrian crashes in Città Studi, Milano, over four years (2017-2020). All pedestrian-related crashes occurred in the four years 4,654 crashes were reported across Milan, in our study area Città Studi, 112 crash reports were studied thoroughly. The project will investigate spatial patterns, identify high-risk areas, and explore potential features that can help to improve pedestrian safety in urban areas. Using Geographic Information System (GIS) and MATLAB software to analyse and visualize the data, this study provides a detailed analysis of pedestrian crashes and their relationship to factor characteristics in the city. In this thesis, we extensively explore the dynamics of pedestrian safety in the context of Città Studi, Milano. Our investigation focuses on how pedestrian accidents, vehicle incidents, and network topology, within the road infrastructure interact. We introduce Icenter, a centrality index designed to identify relevant nodes in the road network. It provides insights based on path metrics because Icentr combines node and edge weights with distance scaling considerations. This index offers perspectives on node importance and helps us understand their specific roles in improving pedestrian safety. The investigation begins with a Network analysis plots to understand the complicated connections between crash sites, with different node sizes and colors representing the severity and quantity of crashes, respectively. This aspect of the analysis identifies potential areas where interventions could be most effective in reducing crash incidence. The research then moves on to a cluster analysis after the network analysis. This phase makes use of bar graphs to show the frequency of incidents within identified clusters, focusing specifically on the concentration of crash occurrences and revealing the presence of notable hotspots that present increased risks to pedestrians. The dendrogram analysis is a method that uses hierarchical data clustering to identify similarities and proximity between crash sites. This helps in determining the structure of crash patterns and identifying common features. Heatmaps are used to compare clusters, providing a more critical perspective on cluster overlap and highlighting shared risk factors that transcend individual crash sites. The thesis suggests that strategic urban safety interventions should be based on data-driven insights, with vehicular traffic and spatial configurations being key predictors of pedestrian crash patterns. It emphasizes the importance of integrated urban design, emphasizing pedestrian well-being and the significance of vehicular presence. The thesis also suggests future research into pedestrian behaviour, urban design's role, and socio-economic factors influencing pedestrian safety in urban areas.File | Dimensione | Formato | |
---|---|---|---|
2023_12_Gonuguntla.pdf
accessibile in internet per tutti
Descrizione: A TOPOLOGICAL ANALYSIS OF PEDESTRIAN CRASHES IN CITTA STUDI, MILANO
Dimensione
4.09 MB
Formato
Adobe PDF
|
4.09 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/214902