Never before has the analysis of bus travel times been so essential to service planning as it is today. Travellers and citizens complain about a decline in service quality compared to the past, urban congestion is on the rise, and public transport companies, struggling to find new drivers, need new tools to improve the efficiency and sustainability of their surface services with the same workforce. This thesis research aims to answer the urgent need to find new tools that can help increase commercial speed, average passenger travel performance, and extremely important variable for defining production costs for companies, thus preventing operators from having to invest “blindly” in excessively costly technological infrastructure, through a precise methodological analysis. This methodology offers a cost-neutral, agile and replicable first approximation model that allows for the immediate estimation of the potential benefits associated with Transport Signal Priority (TSP) logic, providing a preliminary decision-making support tool for transport service operators and planners based on transit data analysis. The main protagonist is the data recorded by Automatic Vehicle Monitoring (AVM) systems, which is elaborated and integrated into a transport network model defined by origin/destination arcs, defined starting from the stop pairs present at traffic light intersections, where future implementation of TSP systems is considered. The benefits of these systems are based on a comparison of the inter-times of crossing these arcs during peak and off-peak traffic times and are estimated through a two-level analysis process, which quantifies them in relation to the resources involved in the service planning process. The case study describes the urban transport service scenario in the city of Pavia, provided by Autoguidovie S.p.A. In this scenario, all the steps defined in the methodological analysis process are replicated, with a particular focus on the PV3 line, selected because of its high frequency and good potential for benefits resulting from the implementation of priority logic at the intersections it crosses. In fact, the first level of analysis shows a possible benefit of approximately 3h30'/day from the prioritisation calculated for the intersections that this line transit in both directions (-3.5% less driving time at the end of the day); the second shows a maximum of approximately 2'/trip recoverable along a corridor around 2km long (-6% of driving time). The analysis procedure allows for the innovative integration of both the infrastructural perspective (location of traffic light intersections where transit is most critical) and the service perspective (quantification of the impact of operational efficiency measures on lines/shifts) using a data-driven approach. Thanks to a methodology that can be replicated in different service scenarios, as it is based solely on data already collected by AVM/APC systems and on simple, cross-cutting network mapping procedures. This study aims to respond to the growing urgency to improve the efficiency and sustainability of surface transport services, preventing operators from having to invest “blindly” in excessively costly technological infrastructure. The research aims to develop a methodology to quantify the time lost by buses at traffic light junctions due to network congestion, to assess the potential benefits of introducing public transport priority (PTP) systems at traffic lights, and to provide a preliminary decision-making support tool for transport service operators and planners. The proposed methodology offers a simple, replicable and cost-neutral first approximation, which allows for the immediate estimation of the potential benefits associated with these Transit Signal Priority (TSP) systems and supports data-driven strategic decisions. In fact, the data analysis is based primarily on the large source of transit data (AVM) that transport companies themselves produce “in-house” on a daily basis. The entire process is based on the principles of: a simplified transport network, based on origin/destination arcs defined from stop pairs at traffic light intersections; a comparison between the inter-times of crossing these arcs during peak and off-peak traffic times; multi-level analysis to estimate average time differences and, therefore, margins for improving the efficiency of the transport service in the event of TSP system implementation. The application of the case study describes the urban transport service scenario in the city of Pavia, provided by Autoguidovie S.p.A. All the basic steps defined in the methodological analysis process are therefore replicated: from the research and definition of the analysis data, to the construction of the simplified network, to the identification of the critical areas and manoeuvres where priority interventions at intersections promise the best impact. In this sense, particular focus is placed on the PV3 line, selected for its high frequency and good potential for recovering dead time at intersections. The analysis procedure innovatively allows for the integration of both the infrastructural perspective (location of signalised intersections where traffic is most critical) and the service perspective (quantification of the impact of operational efficiency measures on lines/shifts) using a data-driven approach. This is done through a tool that is fully replicable in different service scenarios, as it is based solely on data already collected by AVM/APC systems and on simple, cross-cutting network mapping procedures.
Mai come in questo periodo storico, l’analisi dei tempi di viaggio degli autobus è un elemento essenziale per la pianificazione del servizio. Viaggiatori e cittadini lamentano un calo dell’offerta rispetto al passato, i fenomeni di congestione urbana sono sempre più in crescita e le aziende di trasporto pubblico, davanti ad una difficoltà senza precedenti nel reperire nuovi conducenti, necessitano di nuovi strumenti con cui migliorare efficienza e sostenibilità dei propri servizi di superficie a parità di forza lavoro. Questo studio di tesi vuole rispondere all’urgenza di ricercare nuovi strumenti che possano essere d’aiuto nell’innalzamento della velocità commerciale, performance media di viaggio per il passeggero, nonché variabile estremamente rilevante per la definizione dei costi di produzione per le aziende, evitando che le aziende esercenti debbano investire “al buio” in infrastrutture tecnologiche dai costi eccessivi, grazie ad una precisa metodologia di analisi. Questa metodologia offre un modello di prima approssimazione cost-neutral, agile e replicabile, che consente di stimare con immediatezza i benefici potenziali legati a logiche di Transport Signal Priority (TSP), fornendo uno strumento preliminare di supporto decisionale ad enti gestori e pianificatori del servizio di trasporto basato sull’analisi dei dati di transito. Principali protagonisti sono i dati registrati dai sistemi di Automatic Vehicle Monitoring (AVM), che vengono rielaborati e integrati ad un modello di rete di trasporto definita da archi di origine/destinazione, definiti a partire dalle coppie di fermate (“stop-pairs”) presenti in corrispondenza degli incroci semaforizzati, dove ipotizzare una futura implementazione di sistemi TSP. I benefici di questi sistemi si basano sul confronto tra gli intertempi di attraversamento di questi archi nelle fasce orarie di “punta” e di “non-punta” di traffico e sono stimati mediante un processo di analisi su due livelli, con cui vengono quantificati rispetto alle risorse che coinvolgono il processo di pianificazione del servizio. L’applicazione del caso studio descrive lo scenario del servizio del trasporto urbano della città di Pavia, erogato dall’azienda Autoguidovie S.p.A. In questo scenario vengono replicati tutti i passi definiti nel processo di analisi metodologica, con un particolare focus sulla linea PV3, selezionata perché ad alta frequenza e con un buon potenziale di beneficio conseguente all’implementazione di logiche di priorità presso gli incroci che incontra. Infatti, il primo livello di analisi mostra un possibile beneficio di circa 3h30’/giorno dal preferenziamento calcolato rispetto ai soli incroci per cui questa linea transita sulle sue due direzioni (-3,5% di tempo di guida in meno a fine giornata); il secondo, un massimo di circa 2’/corsa recuperabili lungo un corridoio lungo circa 2km (-6% del tempo di guida). La procedura di analisi consente in modo innovativo di integrare con un approccio “data-driven” sia la prospettiva infrastrutturale (localizzazione degli incroci semaforizzati dove i transiti sono più critici), sia la prospettiva di servizio (quantificazione dell’impatto di interventi di efficientamento operativo su linee/turni), grazie ad una metodologia replicabile in scenari di servizio differenti, perché fondata unicamente sui dati già raccolti dai sistemi AVM/APC e su procedure di mappatura di rete semplici e trasversali. Questo studio vuole rispondere alla crescente urgenza di migliorare efficienza e sostenibilità dei servizi di trasporto di superficie, evitando che le aziende esercenti debbano investire “al buio” in infrastrutture tecnologiche dai costi eccessivi. La ricerca mira a sviluppare una metodologia per quantificare i tempi persi dagli autobus agli incroci semaforizzati e imputabili a fenomeni di congestione della rete, a valutare potenziali benefici conseguenti l’introduzione di sistemi di priorità semaforica al trasporto pubblico (PTP) e a fornire uno strumento preliminare di supporto decisionale ad enti gestori e pianificatori del servizio di trasporto. La metodologia proposta offre una prima approssimazione semplice, replicabile e cost-netural, che consente di stimare con immediatezza i benefici potenziali legati a questi sistemi di Transit Signal Priority (TSP) e di supportare decisioni strategiche basate sui dati. Infatti, l’analisi dati ha come principale protagonista la grande fonte di dati di transito (AVM) che le stesse aziende di trasporto quotidianamente producono “in casa”. Su questa base si fonda l’intero processo, con: una rete di trasporto semplificata, basata su archi di origine/destinazione definiti a partire dalle coppie di fermate (“stop-pairs”) presenti in corrispondenza degli incroci semaforizzati; un confronto tra gli inter-tempi di attraversamento di questi archi nelle fasce orarie di “punta” e di “non-punta” di traffico; analisi multilivello per stimare scarti temporali medi, dunque, margini di efficientamento del servizio di trasporto nel caso di implementazione di sistemi TSP. L’applicazione del caso studio descrive lo scenario del servizio del trasporto urbano della città di Pavia, erogato dall’azienda Autoguidovie S.p.A. Vengono dunque replicati tutti i passi base definiti nel processo di analisi metodologica: dalla ricerca e definizione dei dati di analisi, alla costruzione della rete semplificata, fino all’identificazione delle aree e delle manovre critiche su cui interventi di priorità agli incroci promettono i migliori impatti. In questo senso, viene fatto un particolare focus sulla linea PV3, selezionata per alta frequenza e buon potenziale di recupero dei tempi morti agli incroci. La procedura di analisi consente in modo innovativo di integrare con un approccio “data-driven” sia la prospettiva infrastrutturale (localizzazione degli incroci semaforizzati dove i transiti sono più critici), sia la prospettiva di servizio (quantificazione dell’impatto di interventi di efficientamento operativo su linee/turni). Questo viene fatto attraverso uno strumento assolutamente replicabile in scenari di servizio differenti, perché fondato unicamente sui dati già raccolti dai sistemi AVM/APC e su procedure di mappatura di rete semplici e trasversali.
From delays to opportunities: data-driven strategies for bus priority at signalized intersections
Giani, Alessandro
2024/2025
Abstract
Never before has the analysis of bus travel times been so essential to service planning as it is today. Travellers and citizens complain about a decline in service quality compared to the past, urban congestion is on the rise, and public transport companies, struggling to find new drivers, need new tools to improve the efficiency and sustainability of their surface services with the same workforce. This thesis research aims to answer the urgent need to find new tools that can help increase commercial speed, average passenger travel performance, and extremely important variable for defining production costs for companies, thus preventing operators from having to invest “blindly” in excessively costly technological infrastructure, through a precise methodological analysis. This methodology offers a cost-neutral, agile and replicable first approximation model that allows for the immediate estimation of the potential benefits associated with Transport Signal Priority (TSP) logic, providing a preliminary decision-making support tool for transport service operators and planners based on transit data analysis. The main protagonist is the data recorded by Automatic Vehicle Monitoring (AVM) systems, which is elaborated and integrated into a transport network model defined by origin/destination arcs, defined starting from the stop pairs present at traffic light intersections, where future implementation of TSP systems is considered. The benefits of these systems are based on a comparison of the inter-times of crossing these arcs during peak and off-peak traffic times and are estimated through a two-level analysis process, which quantifies them in relation to the resources involved in the service planning process. The case study describes the urban transport service scenario in the city of Pavia, provided by Autoguidovie S.p.A. In this scenario, all the steps defined in the methodological analysis process are replicated, with a particular focus on the PV3 line, selected because of its high frequency and good potential for benefits resulting from the implementation of priority logic at the intersections it crosses. In fact, the first level of analysis shows a possible benefit of approximately 3h30'/day from the prioritisation calculated for the intersections that this line transit in both directions (-3.5% less driving time at the end of the day); the second shows a maximum of approximately 2'/trip recoverable along a corridor around 2km long (-6% of driving time). The analysis procedure allows for the innovative integration of both the infrastructural perspective (location of traffic light intersections where transit is most critical) and the service perspective (quantification of the impact of operational efficiency measures on lines/shifts) using a data-driven approach. Thanks to a methodology that can be replicated in different service scenarios, as it is based solely on data already collected by AVM/APC systems and on simple, cross-cutting network mapping procedures. This study aims to respond to the growing urgency to improve the efficiency and sustainability of surface transport services, preventing operators from having to invest “blindly” in excessively costly technological infrastructure. The research aims to develop a methodology to quantify the time lost by buses at traffic light junctions due to network congestion, to assess the potential benefits of introducing public transport priority (PTP) systems at traffic lights, and to provide a preliminary decision-making support tool for transport service operators and planners. The proposed methodology offers a simple, replicable and cost-neutral first approximation, which allows for the immediate estimation of the potential benefits associated with these Transit Signal Priority (TSP) systems and supports data-driven strategic decisions. In fact, the data analysis is based primarily on the large source of transit data (AVM) that transport companies themselves produce “in-house” on a daily basis. The entire process is based on the principles of: a simplified transport network, based on origin/destination arcs defined from stop pairs at traffic light intersections; a comparison between the inter-times of crossing these arcs during peak and off-peak traffic times; multi-level analysis to estimate average time differences and, therefore, margins for improving the efficiency of the transport service in the event of TSP system implementation. The application of the case study describes the urban transport service scenario in the city of Pavia, provided by Autoguidovie S.p.A. All the basic steps defined in the methodological analysis process are therefore replicated: from the research and definition of the analysis data, to the construction of the simplified network, to the identification of the critical areas and manoeuvres where priority interventions at intersections promise the best impact. In this sense, particular focus is placed on the PV3 line, selected for its high frequency and good potential for recovering dead time at intersections. The analysis procedure innovatively allows for the integration of both the infrastructural perspective (location of signalised intersections where traffic is most critical) and the service perspective (quantification of the impact of operational efficiency measures on lines/shifts) using a data-driven approach. This is done through a tool that is fully replicable in different service scenarios, as it is based solely on data already collected by AVM/APC systems and on simple, cross-cutting network mapping procedures.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_09_Giani_Tesi.pdf
accessibile in internet per tutti
Descrizione: Tesi
Dimensione
5.74 MB
Formato
Adobe PDF
|
5.74 MB | Adobe PDF | Visualizza/Apri |
|
2025_09_Giani_Executive_summary.pdf
accessibile in internet per tutti
Descrizione: Executive Summary
Dimensione
1.14 MB
Formato
Adobe PDF
|
1.14 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/10589/243843