National Technical University of Athens (NTUA)
Έλεγχος της κυκλοφορίας υπό συνθήκες δυναμικής φόρτισης με μεθόδους ελαχιστοποίησης του ρίσκου
Abstract
dc:descriptionThis thesis examines the complex problem of dynamic traffic management with mesoscopic approach methods. The users of a transport system experience different congestion levels during their everyday activities. The losses in travel time due to congestion is a consequence of the dynamic interdependence of travel demand served by the system and are influenced by the parameters that define its operating characteristics. The main objective is to investigate and formulate appropriate reasoning, methods and tools for efficient traffic management and tackling the congestion problem in urban networks, offering a different approach to the problem through the analysis and management of travel time risk. The approach is based on the concepts of risk and the loss of productive travel time. The travel time risk is defined as the combination of loss of productive time due to congestion and the likelihood of this loss to occur over the analysis horizon taking into account the dynamic traffic phenomena. The Conditional Value-at-Risk was used to quantify and predict the risk of high levels of congestion for both the transport system performance and the optimization of traffic control. The need for formulating a system that can handle the information, the complexity of the parameters of multiple problems and the breadth of the alternative solutions involved in traffic management led to the development of a Decision Support System. The proposed architecture of the System incorporates risk analysis and dynamic mesoscopic approaches for the representation of traffic conditions in its subsystems. Here, the Decision Support System is focused on the activities involved with the traffic control strategy formulation, incorporating the risk minimization methodology for finding the best traffic control plans as its main component. At first, the problem for the isolated intersection is defined and analyzed, formulating and describing the methodological framework, the risk model for minimizing delays and the optimization algorithms. Next, the extended problem of finding the best traffic control for a network by minimizing the risk of delays is formulated as a predictive off-line and on-line subsystem. The proposed approach is defined and described in detail with regard to the risk model for the analysis of the network performance through the dynamic mesoscopic simulation, the risk minimization model for finding the best signaling settings, the search algorithm and the objective functions of the optimization problem. The proposed approaches of traffic signal optimization by minimizing the risk of delay were applied to reference intersection and network from the literature. The benefits for integrating risk analysis into the design of traffic signals and the system’s performance are identified by comparing the proposed methodologies with standard methods and software widely used by researchers and practitioners. Furthermore, the basic hypothesis of the proposed approach is investigated, namely that traffic signals designed and optimized by minimizing the Conditional Value-at-Risk will offer users a lower risk of experiencing higher levels of congestion. The thesis examines an issue that has both research and practical interest. The results of the application show that its contribution can be maximized if the methodologies and procedures formulated here are implemented in real-world analysis and management conditions, improving the quality of life of users of the system and contributing positively to the economy and the environment.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2019
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Papatzikou, Eleni
- Παπατζίκου, Ελένη
Subjects
dc:subject × 28- Στρατηγικές ελέγχου κυκλοφορίας
- Σηματοδότηση μεμονωμένου κόμβου
- Σηματοδότηση δικτύου υπό δυναμική ισορροπία
- Δυναμικός καταμερισμός της κυκλοφορίας
- Μεσοσκοπικός καταμερισμός
- Ανάλυση ρίσκου
- Υπό συνθήκη αξία σε κίνδυνο
- Σύστημα υποστήριξης αποφάσεων
- Βελτιστοποίηση σηματοδότησης σταθερού χρόνου
- Βελτιστοποίηση σηματοδότησης προσαρμοστικού ελέγχου
- Αλγόριθμος διεσπαρμένης αναζήτησης
- Γενετικός αλγόριθμος
- Traffic control strategies
- Isolated intersection traffic control optimization
- Network traffic control under dynamic loading
- Dynamic traffic assignment
- Mesoscopic assignment
- Risk analysis
- Conditional value-at-risk
- Decision support system
- Fixed-time traffic control optimization
- Adaptive traffic control optimization
- Scatter search algorithm
- Genetic algorithm
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Πολιτικού Μηχανικού
- Engineering and Technology
- Civil Engineering
Rights
dc:rights- Statement dc:rights
-
- BY_NC
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/46515