Placido, Antonio (2015) The definition of a model framework for the planning and the management phases of the rail system in any kind of service condition. [Tesi di dottorato]


Download (6MB) | Preview
[error in script] [error in script]
Item Type: Tesi di dottorato
Lingua: English
Title: The definition of a model framework for the planning and the management phases of the rail system in any kind of service condition.
Date: 28 March 2015
Number of Pages: 264
Institution: Università degli Studi di Napoli Federico II
Department: Ingegneria Civile, Edile e Ambientale
Scuola di dottorato: Ingegneria civile
Dottorato: Ingegneria dei sistemi idraulici, di trasporto e territoriali
Ciclo di dottorato: 27
Coordinatore del Corso di dottorato:
Date: 28 March 2015
Number of Pages: 264
Uncontrolled Keywords: Microscopic rail simulation, travel demand assignment, rail service management, service quality, rail disruption management
Settori scientifico-disciplinari del MIUR: Area 08 - Ingegneria civile e Architettura > ICAR/05 - Trasporti
Date Deposited: 07 Apr 2015 08:34
Last Modified: 29 Sep 2015 08:18
DOI: 10.6092/UNINA/FEDOA/10168


The target of this thesis is the definition of an off-line procedure for managing the rail network in any kind of service conditions focusing on failure events. In particular, the methodology is based on a microscopic simulation approach which considers both rail operations and passenger flows. Basically, the idea is to simulate the network with the higher level of details without neglecting travel demand which is indeed assigned to the service. The benefits provided by this approach are numerous, namely: 1. it is possible to look for intervention strategies, which optimise passenger satisfaction and do not focus just on operational aspects; 2. by simulating passenger behaviour on the platforms, the procedure enables the assessment of the dynamic interaction between service and user flows. In other words, the model estimates the dwell time at stations as flow dependent providing important information about the influence of passengers on the service; 3. another advantage is the possibility to evaluate crowding levels within the trains or at stations resulting in a more suitable planning of the service as well as a better estimation of the comfort experienced by travellers. 4. the dynamic assignment, although increases the complexity of the model, is extremely useful. In fact, in this way demand peaks, temporary capacity variations, temporary over-saturation of supply elements, and formation and dispersion of queues can be considered; 5. the adoption of proper sensitivity analyses provides the assessment of robustness and effectiveness of planned recovery solutions not only in terms of operational service but also considering customer satisfaction. The research work can be mainly divided in two phases. The first one concerns the specification of the decision support system and all models which are part of it. The second by contrast, is related to the definition of an application for the dynamic assignment of passenger flow to the rail service and the definition of dwell times depending on the number of passengers at station. As regards the first phase, the whole procedure is formulated as a bi-level multidimensional optimisation model which is composed of four sub-models: a Failure Model, a Service Simulation Model, a Supply Model and a Travel Demand Model. In order to increase the service quality, the objective function is expressed through the user generalised cost (Cascetta, 2009) perceived by customers during their travel and, evidently, it has to be minimised. The Failure Model evaluates the failure scenarios which are worth analysing. In particular, through the adoption of RAMS (Reliability, Availability, Maintainability and Safety) techniques (Cenelec, 1999), it gives the possibility to select the breakdown contexts with the higher probability of occurrence. The Service Simulation Model analyses rail traffic and system performance during both ordinary and perturbed conditions by means of a microscopic simulation of the network. According to the target of the analysis, the simulation can be either deterministic or stochastic. The Supply Model is instead dedicated to the definition of performances of all public transportation systems within the study area. In fact, rail and metro lines, particularly within cities, are part of the public transportation system and cannot be considered individually. Hence, knowing the characteristics of the other transport modes can also provide a better estimation of the arrival rate at each station. The Travel Demand Model is the most innovative part of the whole procedure. It is divided into other two sub-models, namely a Pre-Platform Model and On-Platform Model. The first one estimates the number of passenger arriving at stations as a result of the interaction with the Supply Model. This causes a fixed point problem which has been largely dealt with in the literature (see Cantarella, 1997; Cascetta, 2009). Basically, the Pre-Platform Model reproduces the choice process made by passengers who evaluate among all possible alternatives (i.e. different transport modes) the one which maximises their utility. The On-Platform Model works on the dynamic assignment of passenger flows to the rail service. In particular, the model simulates passenger behaviour on the platform considering the maximum capacity of each train and estimating the dwell time necessary to complete the boarding/alighting process. In this way, travel demand is simulated dynamically according to rail service performances which, in turn, are influenced by passenger flows. As a consequence, this interaction generates another fixed point problem. Hence, the resolution of the whole procedure consists in solving a double fixed point problem which has required an in-depth analysis about the mathematical assumptions and the resolution techniques to solve it. However, all railway microscopic simulation software packages focus just on the simulation of train movements within the network and neglects travel demand. Therefore, the second phase of the thesis has concerned the definition of an application developed in C++ language for assigning travel demand to the rail service working in combination with microscopic simulation software. To this purpose the architecture of the OPM 1.0 (On-Platform Model) tool and its internal module DwTE 1.0 (Dwell Time Estimation) has been presented. Both require input data as text files related to infrastructure, rolling stock, travel demand and operational service. In particular, OPM 1.0 is composed of the following modules: - a 'Travel demand module' for the definition of passenger flow on the platform at each station; - a 'Rolling stock module' which describes the main features of rail convoys in terms of fleet composition, number and capacity of coaches, number of doors per coach and so on; - a 'Rail service module' which includes information about the simulated rail service such headways, running times, empty movement etc. Additionally, in case also DwTE 1.0 is launched further modules must be considered, that is: - a 'Passenger flow module' which considers the number of passengers who can actually board the train according to trains' capacity (this information is obtained by OPM 1.0); - a 'Station configuration module' specifying station characteristics (i.e. location of stairs and elevators); - a 'Dwell time estimation module' which defines the time trains has to stop within the station as function of the number of boarding/alighting passengers per door. As outputs, the application provides information about passenger trips, load diagrams, platform congestion and crowding levels within trains as well as dwell time values at stations. In order to validate the tools and verify the benefits of the proposed procedure, in the last part of this thesis several applications are presented. The majority of them has concerned the Line 1 of Naples (Italy) metro system. Results have shown the importance of considering service quality during the management of the rail service, especially when failures or breakdowns occur.


Downloads per month over past year

Actions (login required)

View Item View Item