Abstract
dc:descriptionOne of the major issues in the design of transport networks is related to the identification of solutions and measures, which improve the performance and the operation of transport networks. These solutions and measures are identified through an assessment that concludes to solutions, mainly in the form of alternative scenarios, which are assessed against the current situation. The solution chosen is usually the one that satisfies better the goals originally set. The final solution chosen, although improving certain network characteristics, may be neither optimal nor near optimal. Traffic models used for the assessment of such scenarios may provide answers to “what if…” questions.This thesis presents a new methodological and computational framework, which can provide general rules for the identification of an optimal solution to transport network design problems, with an emphasis on traffic signal control. This framework relies on the formulation of the transport network design problem as a hierarchical, bi-level optimization problem with properties of a strategic Stackelberg game. Using methods of bilevel mathematical programming with equilibrium constraints, the upper level problem seeks to optimize a function, which is related to the network design problem that is being studied, subject to the equilibrium constraints of the lower level problem.The upper level may examine the problems of capacity maximization, traffic control, road pricing, origin-destination matrix estimation and calibration. Hence, the traffic models used in an embedded way in the proposed method may provide answer “what to…” questions.The application of the proposed bi-level mathematical programming method considers the traffic signal control problem, using the methodological and computation framework developed. Several control strategies are assessed on test networks from the related scientific literature, together with a detailed application on a large-scale urban road network of the city of Thessaloniki with the use of Smith’s P0 policy and a dynamic traffic assignment model, which circumvents the limitations and weaknesses of the traditional static traffic assignment models.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2011
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Mitsakis, Evangelos
- Μητσάκης, Ευάγγελος
Subjects
dc:subject × 12- Πρόβλημα σχεδιασμού δικτύων
- Δυναμικός καταμερισμός της κυκλοφορίας
- Μεθοδολογικό και υπολογιστικό πλαίσιο για τον έλεγχο της κυκλοφορίας
- Θεωρία παιγνίων
- Network design problem
- Dynamic traffic assignment
- Methodological and computational framework for traffic control
- Game theory
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Πολιτικού Μηχανικού
- Engineering and Technology
- Civil Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/30371