Technical Sessions – Monday July 11

� TC-10
Tuesday, 15:00-16:30
Meeting Room 111
Railway Applications II
Stream: Public Transit
Invited session
Chair: Twan Dollevoet, Econometric Institute, Erasmus University
Rotterdam, 3000DR, Rotterdam, Netherlands, dollevoet@ese.eur.nl
1 - Knock, Knock, Knock-on on Railway Networks
Thijs Dewilde, Centre for Industrial Management/Traffic &
Infrastructure, University of Leuven, Celestijnenlaan 300A, bus
2422, B-3001, Leuven, Belgium,
Thijs.Dewilde@cib.kuleuven.be, Peter Sels, Dirk Cattrysse,
Pieter Vansteenwegen
In our study about robust railway timetables, knock-on delays or delay propagation
are of high importance. Interdependencies among trains and their paths
cause delays, however small, to propagate in space and time. Identifying possible
conflicts and accounting for the applied dispatching strategy, knock-on
delays can be modeled. We created a delay propagation model for one of Europe’s
major railway bottlenecks, the Brussels area. This model enables us to
improve the robustness of a timetable by avoiding knock-on delays as much as
possible. We validated our results using real-time data.
2 - Scheduling in Rapid Transit Networks. Quality of Service
vs Network Profitability
Eva Barrena, Applied Mathematics I, University of Sevilla, Avda.
Reina Mercedes s/n, 41012, Sevilla, Spain, ebarrena@us.es,
Alejandro Zarzo, Encarnación Algaba, David Canca
In the railway networks management context, set up of train schedules is a
topic which affects both the level of satisfaction of the users and the network
profitability. This double influence makes it a widely studied topic in the literature,
where the main lines of research tend to improve the solving methods of
the corresponding integer programming problems. However, literature about
methods that take both user and service provider point of view jointly into account
is sparse. The study of the trade-off between this two opposite aspects in
rapid transit networks is the main aim of this work.
3 - Finding a Passenger-optimal Revised Schedule for a
Rail Network
Todd Niven, Monash University, 3145, Caulfield East, Victoria,
Australia, todd.niven@monash.edu, Christopher Mears, Mark
Wallace, Ian Evans
In a suburban passenger railway network, a delay of a single train is likely to
affect subsequent trains as well. When a delay has occurred, the delayed train
and other nearby trains can be re-scheduled to minimise the effect on passengers’
travel time. A simple single-track train network with a single delay is
considered. We model and solve the problem using a constraint programming
system, and are able to find optimal revised schedules, with respect to time
spent waiting at the station and time aboard the train.
4 - Look-Ahead based Dynamic Ranking Heuristic as a
Contingency measure to handle disruptions in a Railway
Schedule.
Sundaravalli Narayanaswami, Information Technology Dept,
Higher Colleges of Technology, PB No 58855, Madinat Zayed,
58855, Abu Dhabi, United Arab Emirates,
sundaravalli@iitb.ac.in
Schedule disruptions occur due to deterministic and stochastic reasons. Many
published rescheduling models are based on optimal, heuristic or metaheuristic
approaches that effectively dispatch conflicting trains to optimize
rescheduling objective. A novel, dynamic, look-ahead based ranking heuristic
is proposed as a contingency measure to resolve deterministic disruptions
in this rescheduling model. The dispatch algorithm dynamically prioritizes
conflict trains using momentary train parameters with an objective of total
weighted delay of all trains at their respective destinations. Results are presented;
significant feature of the heuristics is reduction of problem complexity
by partitioning the problem space.
IFORS 2011 - Melbourne TC-11
� TC-11
Tuesday, 15:00-16:30
Meeting Room 112
Simulation for Operations Management
Stream: Simulation - Sponsored by I-SIM
Invited session
Chair: Walter Silva, Ingeniería Industrial, Pontificia Universidad
Católica del Perú, Calle Fray Angélico 443. San Borja, Lima, Peru,
walter.silva@pucp.edu.pe
1 - Dynamic Evaluation and Optimisation of an Urban Collective
Taxis System by Discrete-Event Simulation
Jennie Lioris, CERMICS_IMARA, ENPC-INRIA, Domaine de
Voluceau Rocquencourt, 78150, Le Chesnay, France,
jennie.lioris@cermics.enpc.fr, Guy Cohen
Our aim is to provide optimal strategies for the performance management of an
urban Collective-Taxis system, intelligently associating more than one passenger
to each vehicle, controlling detours, waits, operating with/without reservations
and door-to-door services, at low fares encouraging people its use. All
controls governing the system (e.g. client acceptance, dynamically constructed
vehicle itineraries, solutions for idle vehicles etc.) will be evaluated and system
performances will be optimised by a made to measure discrete-event simulator,
before any risky real-time application.
2 - Reducing Disturbance in Manufacturing System with
Vehicle Tracking System and Discrete-event Simulation
Norhanom Awang, Technology Management, University
Malaysia of Pahang, 26300, Kuantan, Pahang, Malaysia,
anumjp@yahoo.com
This study aims to present a modelling of production flow in automotive manufacturing
using discrete-event simulation (DES) model. The purpose of the
study is to reduce disturbance using Vehicle Tracking System (VTS) and DES.
The computer-based integrated approach successfully reduces the risk of inefficiency
cause of these problems in operation. The study makes a business
case that process improvement through reduction of disturbance can be effectively
accomplished with the integrated approach of VTS with widely available
inexpensive and user-friendly computer-based tools.
3 - A Simulation for Optimal Buffer Allocation in an In-line
System
Eishi Chiba, Hosei University, 3-7-2, Kajino-cho, Koganei-shi,
Tokyo 184-8584, Japan, e-chiba@hosei.ac.jp
A manufacturing system for Flat Panel Displays (FPDs) consists of series of
equipments, each of which is usually equipped with enough number of buffers
to avoid collision between glass substrates. However, they often contain redundant
buffers which are not actually used. In order to reduce the production
cost, the number of buffers should be minimized. In this paper, we try to find
a buffer allocation that achieves the smallest total number of buffers under an
arbitrarily specified collision probability. We also present some computational
results.
4 - Application of Discrete Simulation in the Optimal Allocation
of Ambulances Holding Points
Walter Silva, Ingeniería Industrial, Pontificia Universidad
Católica del Perú, Calle Fray Angélico 443. San Borja, Lima,
Peru, walter.silva@pucp.edu.pe, Gonzalo Raffo, Wilmer Atoche
The company operates 19 ambulances which are responsible for dealing with
190 calls per day. A valid assumption is that the waiting time is inversely proportional
to the quality of service. It is therefore understood that a shorter waiting
time, there would be a better quality of service. Given this situation, then
there is the decision - making by management and by the operator: What are
the sites of the stand by reducing waiting time? It is also important to answer
the question, due to an emergency call, what should I allocate ambulance?
53