Dynamic Fleet Management: Concepts, Systems, Algorithms & Case Studies

249 Pages·2007·5.9 MB·English

by  Vasileios S. Zeimpekis

DYNAMIC FLEET MANAGEMENT
Concepts, Systems, Algorithms
& Case Studies
OPERATIONS RESEARCH/COMPUTER SCIENCE
INTERFACES SERIES
Professor Ramesh Sharda Prof. Dr. Stefan Voß
Oklahoma State University Universität Hamburg

Greenberg /A Computer-Assisted Analysis System for Mathematical Programming Models and Solutions:
A User’s Guide for ANALYZE
Greenberg / Modeling by Object-Driven Linear Elemental Relations: A Users Guide for MODLER
Brown & Scherer / Intelligent Scheduling Systems
Nash & Sofer / The Impact of Emerging Technologies on Computer Science & Operations Research
Barth / Logic-Based 0-1 Constraint Programming
Jones / Visualization and Optimization
Barr, Helgason & Kennington / Interfaces in Computer Science & Operations Research: Advances in
Metaheuristics, Optimization, & Stochastic Modeling Technologies
Ellacott, Mason & Anderson / Mathematics of Neural Networks: Models, Algorithms & Applications
Woodruff / Advances in Computational & Stochastic Optimization, Logic Programming, and Heuristic
Search
Klein / Scheduling of Resource-Constrained Projects
Bierwirth / Adaptive Search and the Management of Logistics Systems
Laguna & González-Velarde / Computing Tools for Modeling, Optimization and Simulation
Stilman / Linguistic Geometry: From Search to Construction
Sakawa / Genetic Algorithms and Fuzzy Multiobjective Optimization
Ribeiro & Hansen / Essays and Surveys in Metaheuristics
Holsapple, Jacob & Rao / Business Modelling: Multidisciplinary Approaches — Economics, Operational
and Information Systems Perspectives
Sleezer, Wentling & Cude/Human Resource Development And Information Technology: Making Global
Connections
Voß & Woodruff / Optimization Software Class Libraries
Upadhyaya et al / Mobile Computing: Implementing Pervasive Information and Communications
Technologies
Reeves & Rowe / Genetic Algorithms—Principles and Perspectives: A Guide to GA Theory
Bhargava & Ye / Computational Modeling And Problem Solving In The Networked World: Interfaces in
Computer Science & Operations Research
Woodruff / Network Interdiction And Stochastic Integer Programming
Anandalingam & Raghavan / Telecommunications Network Design And Management
Laguna & Martí / Scatter Search: Methodology And Implementations In C
Gosavi/ Simulation-Based Optimization: Parametric Optimization Techniques and Reinforcement
Learning
Koutsoukis & Mitra / Decision Modelling And Information Systems: The Information Value Chain
Milano / Constraint And Integer Programming: Toward a Unified Methodology
Wilson & Nuzzolo / Schedule-Based Dynamic Transit Modeling: Theory and Applications
Golden, Raghavan & Wasil / The Next Wave in Computing, Optimization, And Decision Technologies
Rego & Alidaee/ Metaheuristics Optimization via Memory and Evolution: Tabu Search and Scatter
Search
Kitamura & Kuwahara / Simulation Approaches in Transportation Analysis: Recent Advances and
Challenges
Ibaraki, Nonobe & Yagiura / Metaheuristics: Progress as Real Problem Solvers
Golumbic & Hartman / Graph Theory, Combinatorics, and Algorithms: Interdisciplinary Applications
Raghavan & Anandalingam / Telecommunications Planning: Innovations in Pricing, Network Design and
Management
Mattfeld / The Management of Transshipment Terminals: Decision Support for Terminal Operations in
Finished Vehicle Supply Chains
Alba & Martí/ Metaheuristic Procedures for Training Neural Networks
Alt, Fu & Golden/ Perspectives in Operations Research: Papers in honor of Saul Gass’ 80th Birthday
Baker et al/ Extending the Horizons: Adv. In Computing, Optimization, and Dec. Technologies
DYNAMIC FLEET
MANAGEMENT
Concepts, Systems, Algorithms
& Case Studies
Edited by

Vasileios Zeimpekis
Christos D. Tarantilis
George M. Giaglis
Ioannis Minis
Vasileios Zeimpekis Christos Tarantilis
Athens University of Economics & Business Athens University of Economics & Business
Athens, Greece Athens, Greece

George M. Giaglis Ioannis Minis
Athens University of Economics & Business University of Aegean
Athens, Greece Chios, Greece

Series Editors:
Ramesh Sharda Stefan Voß
Oklahoma State University Universität Hamburg
Stillwater, OK, USA Hamburg, Germany

Library of Congress Control Number: 2007924349

ISBN-13: 978-0-387-71721-0 e -ISBN-13: 978-0-387-71722-7

Printed on acid-free paper.

© 2007 by Springer Science+Business Media, LLC
All rights reserved. This work may not be translated or copied in whole or in part
without the written permission of the publisher (Springer Science+Business Media,
LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in
connection with reviews or scholarly analysis. Use in connection with any form of
information storage and retrieval, electronic adaptation, computer software, or by
similar or dissimilar methodology now know or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks and similar
terms, even if the are not identified as such, is not to be taken as an expression of
opinion as to whether or not they are subject to proprietary rights.

9 8 7 6 5 4 3 2 1

springer.com
TABLE OF CONTENTS

Preface vii

Acknowledgments xiii

1. Planned route optimization for real-time vehicle routing 1
Soumia Ichoua, Michel Gendreau and Jean-Yves Potvin

2. Classification of dynamic vehicle routing systems 19
Allan Larsen, Oli B.G. Madsen and Marius M. Solomon

3. Dynamic and stochastic vehicle routing in practice
41
Truls Flatberg, Geir Hasle, Oddvar Kloster, Eivind J. Nilssen
and Atle Riise

4. A parallelizable and approximate dynamic programming-based
dynamic fleet management model with random travel times 65
and multiple vehicle types
Huseyin Topaloglu

5. Integrated model for the dynamic on-demand air
95
transportation operations
Yufeng Yao, Özlem Ergun and Ellis Johnson

6. An intermodal time-dependent minimum cost path algorithm 113
Elaine Chang, Evangelos Floros and Athanasios Ziliaskopoulos

7. Real-time emergency response fleet deployment: concepts,
133
systems, simulation & case studies
Ali Haghani and Saini Yang

8. Vehicle routing and scheduling models, simulation

and city logistics
163
Jaime Barceló, Hanna Grzybowska and Sara Pardo
vi Table of Contents
9. Dynamic management of a delayed delivery vehicle in a city 197
logistics environment
Vasileios Zeimpekis, Ioannis Minis, Kostas Mamassis
and George M. Giaglis

10. Real-time fleet management at eCourier Ltd
219
Andrea Attanasio, Jay Bregman, Gianpaolo Ghiani
and Emanuele Manni

Index 239
PREFACE

The challenges of contemporary fleet management are moving beyond cost
efficiency towards superior customer service, agility, and responsiveness to
requirements that vary at a time scale unthinkable even a decade ago. Over
the last forty years classical methods of fleet management have addressed
extensively the issue of cost efficiency by developing a priori routing plans
in a wide spectrum of practical problems. However, the use of an initial plan,
although necessary, is by no means sufficient to address events that are
likely to occur during plan execution and significantly affect system per-
formance. Typical examples of such evens are customer orders that arrive in
real time and should be served by vehicles already on route, as well as
disturbances intrinsic to urban environments, such as traffic delays, parking
unavailability, and breakdowns. The ability to deal with such cases in a satis-
factory manner is increasingly important to the competitiveness of logistics
and transport related operations.
Dynamic fleet management refers to environments in which information
is dynamically revealed to the decision maker. This information may not be
known at the initial planning stage, and/or may change after the construction
of the initial fleet routes during plan execution. In addition, there are
significant cases in which no routing exists and the system responds to
requests that arrive dynamically.
Methods that address the critical issues of dynamic fleet management may
be implemented in practical systems by taking advantage of recent advances in
satellite and mobile communication technologies. Specifically, satellite
location identification systems that use the Global Positioning System (GPS)
and terrestrial mobile communication systems, such as the General Packet
Radio Service (GPRS) or Terrestrial Trunked Radio (TETRA), enable fleet
operators to monitor the execution of a plan and to manage operations in real
time, thus improving fleet performance.
This edited volume aims to highlight important advances in the emerging
field of Dynamic Fleet Management. The fundamental problem of real time
vehicle routing is defined and solution methods are presented and classified.
Emphasis is also given to algorithmic approaches that are able to process
dynamic information and produce solutions of acceptable quality for
significant dynamic fleet management problems in almost real time. Finally,
the volume includes case studies that address actual dynamic problems by
combining systemic and algorithmic approaches.
The first three chapters survey important aspects of the dynamic vehicle
routing problem.
viii Preface
Chapter 1 reviews and classifies solution methodologies that address
real-time vehicle routing problems where customer requests are dynamically
revealed over time. Each service request either has a combined pick-up and
delivery location or only a single pick-up (or delivery) location. Different
algorithmic methodologies are presented that handle the occurrence of new
requests through the construction of the part of the route that has not yet
been executed by the vehicle (i.e. planned route). To construct the planned
routes, adaptations of methods originally employed to solve the static problem
are presented. Issues of diverting a vehicle away from its current planned
destination to serve a new request that has just occurred are then discussed.
Solution methodologies that anticipate future requests to efficiently satisfy
future demands are also reviewed. Chapter 1 concludes by proposing future
directions for research, such as the development of solutions approaches for
handling the occurrence of vehicle breakdowns and unexpected congestion,
formal modeling frameworks that integrate the uncertainty associated with
future requests, as well as other theoretical and practical issues of research.
Chapter 2 discusses important characteristics and properties of the dynamic
vehicle routing problem from a temporal point of view. Differences between
the static and the dynamic vehicle routing problems are also demonstrated by
analyzing critical issues from previous published papers. The importance of
measuring the performance of a dynamic vehicle routing system is then
highlighted, and measures for dynamism in systems with and without time
windows are discussed. Methods for evaluating the performance of on-line
routing algorithms are presented and important issues to include in the system
objective are reported. A three-echelon classification of dynamic vehicle
routing systems is also proposed based on a) their degree of dynamism and b)
the objective of the system. Finally, Chapter 2 emphasizes the significance of
considering the volume and the temporal composition of immediate requests
along with the system objective when developing an algorithmic methodology
for a dynamic vehicle routing system.
Chapter 3 discusses the experience gained in practical issues in stochastic
and dynamic routing in the context of developing a VRP solver at a Norwegian
research institute. First, a review of the literature on dynamic and stochastic
vehicle routing problem (DSVRP) is presented. To illustrate the need for
dynamic and stochastic models in real world applications, two examples
involving transportation of goods and persons, respectively, are demons-
trated. Modelling and formal description of DSVRPs is also proposed to
create the platform upon which new computational methods will be tested and
evaluated. Chapter 3 also considers the context in which a VRP solver
operates and proposes solution approaches based on scenario generation. The
way to exploit dynamic events to produce more robust plans to the VRP is
discussed, as well as the role of generating statistical knowledge of events
Preface ix
automatically from past experience. Chapter 3 concludes with suggestions for
further research in these areas.
Chapters 4 to 6 provide algorithmic approaches of significant appli-
cability to practical dynamic fleet management problems.
Chapter 4 proposes a dynamic programming-based approach to address a
general fleet dispatching problem. In this problem the vehicles are dispatched
to serve load requests, which arise randomly during each time period of a
finite time horizon at different locations in a transportation network. The fleet
comprises vehicles of multiple types. An additional critical (and practical)
complication is that the travel times between the network nodes are random.
The approach presented in this chapter uses modelling and methodological
concepts from the deterministic travel time case to present a novel approach
for random travel times. Major contributions of this chapter include the
formulation of the problem as a dynamic program; the way of approximating
the value function of the resulting subproblem for each time period by
separable piecewise linear concave functions; the proof that the approximate
subproblem is a min-cost network flow problem; the further decomposition of
the latter to multiple problem instances by location, which can be solved in
parallel. Furthermore, an updating method is employed to improve the value
function approximations, and a comprehensive algorithm is proposed to obtain
solutions of superior quality, as evidenced by the experimental results of three
classes of problems included in this work. Chapter 4 concludes by proposing
challenging new opportunities for research in the dynamic fleet management
area, such as the introduction of load pick up and delivery windows when the
load requests arrive randomly, as well as other unresolved practical issues.
Chapter 5 proposes a column generation-based approach to plan the
operations of an on-demand air transport system. The problem consists of
determining the fleet assignment, aircraft routing and crew pairing in an
integrated fashion for a system that provides point-to-point service at
customer request. The dynamic elements of the problem are twofold: i) The
demand for service is not known in advance, and is dynamically received.
ii) There are unscheduled maintenance requirements that are also raised
dynamically. The proposed model is based on a three-day planning period
within a rolling horizon setting. It uses a crew duty network and a fleet-
station time line in order to embed the crew and aircraft information in the
fleet assignment problem, while keeping the crew and aircraft separate
during planning. In addition to the model, the major contributions of this
chapter include: The use of column generation and identification of good
pairings by solving special shortest path problems; the dynamic adjustment
of the plan when new requests for service or unscheduled maintenance are
revealed without relying on demand forecasts; the comparison between cases
with fixed (immovable) requests, and cases in which (limited) freedom is