A review of dynamic vehicle routing problems

Victor Pillac, Michel Gendreau, Christelle Guéret, Andrés Medaglia

To cite this version:

Victor Pillac, Michel Gendreau, Christelle Guéret, Andrés Medaglia. A review of dynamic ve-
hicle routing problems. European Journal of Operational Research, 2013, 225 (1), pp.1-11.
�10.1016/j.ejor.2012.08.015�. �hal-00739779�

HAL Id: hal-00739779

https://hal.science/hal-00739779v1
Submitted on 9 Oct 2012

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
 A Review of Dynamic Vehicle Routing
Problems
Victor Pillac1,2 , Michel Gendreau3,4 , Christelle Guéret∗,1 ,
and Andrés L. Medaglia2
1 École
des Mines de Nantes, IRCCyN, UMR CNRS 6597, Nantes, France
2 Universidad de los Andes, Industrial Engineering Department, Bogotá, Colombia
3 École Polytechnique de Montréal, Département de Mathématiques et de Génie

Industriel, Montréal, Canada

4 Centre Interuniversitaire de Recherche sur les Reseaux d’Entreprise, la

Logistique et le Transport (CIRRELT)

September 2012

The present work has been accepted for publication in European Journal of
Operational Research doi:10.1016/j.ejor.2012.08.015
The orginal publication is available at elsevier.com

Abstract
A number of technological advances have led to a renewed interest
on dynamic vehicle routing problems. This survey classifies routing
problems from the perspective of information quality and evolution.
After presenting a general description of dynamic routing, we intro-
duce the notion of degree of dynamism, and present a comprehensive
review of applications and solution methods for dynamic vehicle rout-
ing problems.
∗
Corresponding author: [email protected]

1
1 Introduction
The Vehicle Routing Problem (VRP) formulation was first introduced by
Dantzig and Ramser [35], as a generalization of the Traveling Salesman
Problem (TSP) presented by Flood [49]. The VRP is generally defined
on a graph G = (V, E, C), where V = {v0 , ..., vn } is the set of vertices;
E = {(vi , vj )|(vi , vj ) ∈ V 2 , i 6= j} the arc set; and C = (cij )(vi ,vj )∈E a cost
matrix defined over E, representing distances, travel times, or travel costs.
Traditionally, vertex v0 is called the depot, while the remaining vertices in V
represent customers (or requests) that need to be serviced. The VRP con-
sists in finding a set of routes for K identical vehicles based at the depot,
such that each of the vertices is visited exactly once, while minimizing the
overall routing cost.
Beyond this classical formulation, a number of variants have been studied.
Among the most common are the Capacitated VRP (CVRP), where each
customer has a demand for a good and vehicles have finite capacity; the
VRP with Time Windows (VRPTW), where each customer must be visited
during a specific time frame; the VRP with Pick-up and Delivery (PDP),
where goods have to be picked-up and delivered in specific amounts at the
vertices; and the Heterogeneous fleet VRP (HVRP), where vehicles have
different capacities. Routing problems that involve moving people between
locations are referred to as Dial-A-Ride-Problem (DARP) for land transport;
or Dial-A-Flight-Problem (DAFP), for air transport.
In contrast to the classical definition of the vehicle routing problem, real-
world applications often include two important dimensions: evolution and
quality of information [116]. Evolution of information relates to the fact that
in some problems the information available to the planner may change during
the execution of the routes, for example, with the arrival of new customer
requests. Quality of information reflects possible uncertainty on the available
data, for instance, when the demand of a customer is only known as a range
estimate of its real demand. In addition, depending on the problem and
the available technology, vehicle routes can either be designed statically (a-
priori) or dynamically. For instance, the VRP with Stochastic Demands
(VRPSD), can be seen from both perspectives. From a static perspective,
the problem is to design a set of robust routes a-priori, that will undergo
minor changes during their execution [16, 53]. From a dynamic perspective,
the problem consists in designing the vehicle routes in an online fashion,
communicating to the vehicle which customer to serve next as soon as it

2
becomes idle [104, 127, 128]. Based on these dimensions, Table 1 identifies
four categories of routing problems.

Information quality
Deterministic input Stochastic input
Input known Static and Static and
Information beforehand deterministic stochastic
evolution Input changes Dynamic and Dynamic and
over time deterministic stochastic

Table 1: Taxonomy of vehicle routing problems by information evolution and

quality.

In static and deterministic problems, all input is known beforehand and

vehicle routes do not change once they are in execution. This classical prob-
lem has been extensively studied in the literature, and we refer the interested
reader to the recent reviews of exact and approximate methods by Baldacci
et al. [4], Cordeau et al. [33], Laporte [83, 84], and Toth and Vigo [143].
Static and stochastic problems are characterized by input partially known
as random variables, which realizations are only revealed during the execu-
tion of the routes. Additionally, it is assumed that routes are designed a-
priori and only minor changes are allowed afterwards. For instance, allowable
changes include planning a trip back to the depot or skipping a customer.
Applications in this category do not require any technological support. Un-
certainty may affect any of the input data, yet the three most studied cases
are [33]: stochastic customers, where a customer needs to be serviced with a
given probability [15, 147]; stochastic times, in which either service or travel
times are modeled by random variables [79, 85, 146]; and lastly, stochastic
demands [31, 37, 86, 97, 98, 126, 128]. Further details on the static stochastic
vehicle routing can be found in the reviews by Bertsimas and Simchi-Levi
[16], Cordeau et al. [33], and Gendreau et al. [53].
In dynamic and deterministic problems, part or all of the input is un-
known and revealed dynamically during the design or execution of the routes.
For these problems, vehicle routes are redefined in an ongoing fashion, requir-
ing technological support for real-time communication between the vehicles
and the decision maker (e.g., mobile phones and global positioning systems).
This class of problems are also referred to as online or real time by some

3
authors [77].
Similarly, dynamic and stochastic problems have part or all of their input
unknown and revealed dynamically during the execution of the routes, but in
contrast with the latter category, exploitable stochastic knowledge is available
on the dynamically revealed information. As before, the vehicle routes can
be redefined in an ongoing fashion with the help of technological support.
Besides dynamic routing problems, where customer visits must be ex-
plicitly sequenced along the routes, there are other related vehicle dispatch-
ing problems, such as managing a fleet of emergency vehicles[23, 54, 66], or
the so-called dynamic allocation problems in the area of long haul truckload
trucking [60, 109, 134]. In this paper, we focus solely on dynamic problems
with an explicit routing dimension.
The remainder of this document is organized as follows. Section 2 presents
a general description of dynamic routing problems and introduce the notion
of degree of dynamism. Section 3 reviews different applications in which
dynamic routing problems arise, while Section 4 provides a comprehensive
survey of solution approaches. Finally, Section 5 concludes this paper and
gives directions for further research.

2 Dynamic vehicle routing problems

2.1 A general definition
The first reference to a dynamic vehicle routing problem is due to Wilson
and Colvin [148]. They studied a single vehicle DARP, in which customer
requests are trips from an origin to a destination that appear dynamically.
Their approach uses insertion heuristics able to perform well with low com-
putational effort. Later, Psaraftis [116] introduced the concept of immediate
request: a customer requesting service always wants to be serviced as early
as possible, requiring immediate replanning of the current vehicle route.
A number of technological advances have led to the multiplication of
real-time routing applications. With the introduction of the Global Position-
ing System (GPS) in 1996, the development and widespread use of mobile
and smart phones, combined with accurate Geographic Information Systems
(GIS), companies are now able to track and manage their fleet in real time and
cost effectively. While traditionally a two-step process (i.e., plan-execute),
vehicle routing can now be done dynamically, introducing greater oppor-

4
tunities to reduce operational costs, improve customer service, and reduce
environmental impact.
The most common source of dynamism in vehicle routing is the online
arrival of customer requests during the operation. More specifically, requests
can be a demand for goods [2, 62, 70, 71, 75, 99, 101, 144] or services [7,
11, 17, 52, 89, 141]. Travel time, a dynamic component of most real-world
applications, has been recently taken into account [1, 6, 28, 48, 64, 65, 94,
108, 136, 139, 153]; while service time has not been explicitly studied (but can
be added to travel time). Finally, some recent work considers dynamically
revealed demands for a set of known customers [104, 105, 126, 128] and
vehicle availability [92, 93, 103], in which case the source of dynamism is the
possible breakdown of vehicles. In the following we use the prefix “D-” to
label problems in which new requests appear dynamically.
To better understand what we mean by dynamic, Figure 1 illustrates
the route execution of a single vehicle D-VRP. Before the vehicle leaves the
depot (time t0 ), an initial route plans to visit the currently known requests
(A, B, C, D, E). While the vehicle executes its route, two new requests (X
and Y ) appear at time t1 and the initial route is adjusted to fulfill them.
Finally, at time tf the executed route is (A, B, C, D, Y, E, X).

t0 t1 tf
D E D E D E
Y Y

X X

C C C
A A A

B B B

Legend: A A Customer (unserved/served) Depot Planned arc Executed arc

Figure 1: Example of dynamic vehicle routing

This example reveals how dynamic routing inherently adjusts routes in

an ongoing fashion, which requires real-time communication between vehicles
and the dispatching center. Figure 2 illustrates this real-time communication
scheme, where the environment refers to the real-world while the dispatcher
is the agent that gives instructions to the vehicle. Once the vehicle is ready
(first dotted arrow), the dispatcher makes a decision and instructs the vehicle

5
to fulfill request A (first double-headed arrow). When the vehicle starts
(second dotted arrow) and ends (third dotted arrow) service at request A, it
notifies the dispatcher, which in turns updates the available information and
communicates the vehicle its next request (second double-headed arrow).

Environment
Vehicle Starts Finishes New customer:
is ready serving A serving A X

Next: A Next: B
Decision

Update

Decision

Update
Dispatcher

Figure 2: Timeline of events for the dynamic routing of a single vehicle

2.2 Differences with static routing

In contrast to their static counterparts, dynamic routing problems involve
new elements that increase the complexity of their decisions (more degrees
of freedom) and introduce new challenges while judging the merit of a given
route plan.
In some contexts, such as the pick-up of express courier [52], the transport
company may deny a customer request. As a consequence, it can reject a
request either because it is simply impossible to service it, or because the
cost of serving it is too high. This process of acceptance/denial has been
used in many approaches [2, 43, 52, 73–75, 92] and is referred to as service
guarantee [145].
In dynamic routing, the ability to redirect a moving vehicle to a new
request nearby allows for additional savings. Nevertheless, it requires real-
time knowledge of the vehicle position and being able to communicate quickly
with drivers to assign them new destinations. Thus, this strategy has received
limited interest, with the main contributions being the early work by Regan
et al. [120, 121, 122], the study of diversion issues by Ichoua et al. [73], and
the work by Branchini et al. [21].
Dynamic routing also frequently differs in the objective function [118]. In
particular, while a common objective in the static context is the minimiza-

6
tion of the routing cost, dynamic routing may introduce other notions such
as service level, throughput (number of serviced requests), or revenue max-
imization. Having to answer to dynamic customer requests also introduces
the notion of response time: a customer might request to be serviced as soon
as possible, in which case the main objective may become to minimize the
delay between the arrival of a request and its service.
Dynamic routing problems require making decisions in an online manner,
which often compromises reactiveness with decision quality. In other words,
the time invested searching for better decisions, comes at the price of a lower
reactiveness to input changes. This aspect is of particular importance in
contexts where customers call for a service and a good decision must be
made as fast as possible.

2.3 Measuring dynamism

Different problems (or instances of a same problem) can have different levels
of dynamism, which can be characterized according to two dimensions [76]:
the frequency of changes and the urgency of requests. The former is the rate
at which new information becomes available, while the latter is the time gap
between the disclosure of a new request and its expected service time. From
this observation three metrics have been proposed to measure the dynamism
of a problem (or instance).
Lund et al. [95] defined the degree of dynamism δ as the ratio between
the number of dynamic requests nd and the total number of requests ntot as
follows:
nd
δ= (1)
ntot
Based on the fact that the disclosure time of requests is also important
[117, 118], Larsen [87] proposed the effective degree of dynamism δ e . This
metric can be interpreted as the normalized average of the disclosure times.
Let T be the length of the planning horizon, R the set of requests, and ti the
disclosure time of request i ∈ R. Assuming that requests known beforehand
have a disclosure time equal to 0, δ e can be expressed as:
1 X ti
δe = (2)
ntot i∈R T

Larsen [87] also extended the effective degree of dynamism to problems

with time windows to reflect the level of urgency of requests. He defines the

7
reaction time as the difference between the disclosure time ti and the end
of the corresponding time window li , highlighting that longer reaction times
mean more flexibility to insert the request into the current routes. Thus, the
effective degree of dynamism measure is extended as follows:
 
e 1 X li − ti
δT W = 1− (3)
ntot i∈R T

It is worth noting that these three metrics only take values in the interval
[0, 1] and all increase with the level of dynamism of a problem. Larsen et al.
[88, 90] use the effective degree of dynamism to define a framework classifying
D-VRPs among weakly, moderately, and strongly dynamic problems, with
values of δ e being respectively lower than 0.3, comprised between 0.3 and
0.8, and higher than 0.8.
Although the effective degree of dynamism and its variations have proven
to capture well the time-related aspects of dynamism, it could be argued
that they do not take into account other possible sources of dynamism. In
particular, the geographical distribution of requests, or the traveling times
between requests, are also of great importance in applications aiming at the
minimization of response time. Although not considered, the frequency of
updates in problem information has a dramatical impact on the time available
for optimization.

3 A review of applications
Recent advances in technology have allowed the emergence of a wide new
range of applications for vehicle routing. In particular, the last decade has
seen the development of Intelligent Transport Systems (ITS), which are based
on a combination of geolocation technologies, with precise geographic infor-
mation systems, and increasingly efficient hardware and software for data
processing and operations planning. We refer the interested reader to the
study by Crainic et al. [34] for more details on ITS and the contributions of
operations research to this relatively new domain.
Among the ITS, the Advanced Fleet Management Systems (AFMS) are
specifically designed for managing a corporate vehicle fleet. The core prob-
lem is generally to deliver (pick-up) goods or persons to (from) locations
distributed in a given area. While customer requests can either be known
in advance or appear dynamically during the day, vehicles are dispatched

8
and routed in real time, potentially, by taking into account changing traffic
conditions, uncertain demands, or varying service times. A key technologi-
cal feature of AFMS is the optimization component. Traditionally, vehicle
routing relies on teams of human dispatchers, meaning a critical operational
process is bound to the competence and experience of dispatchers, as well
as the management costs that are directly linked to the size of the fleet [1].
Advances in computer science have allowed a technological transfer from op-
erational research to AFMS, as presented in the studies by Attanasio et al.
[1], Du et al. [38], Godfrey and Powell [60], Powell and Topaloglu [115], Roy
[125], Simao et al. [129], and Slater [130].
The remainder of this section presents applications where dynamic rout-
ing has been or can be implemented. The interested reader is also referred
to the work by Gendreau and Potvin [55] and Ichoua et al. [76] for comple-
mentary reviews.

3.1 Services
In this category of applications, a service request is defined by a customer
location and a possible time window; while vehicle routes just fulfill service
requests without considering side constraints such as capacity. Perhaps the
simplest, yet most illustrative case in this category is the dynamic traveling
salesman problem [89].
A common application of dynamic routing can be found in the area of
maintenance operations. Maintenance companies are often committed by
contract to their customers, which specify periodical or planned visits to per-
form preventive maintenance, and may also request corrective maintenance
on short notice. Therefore, each technician is first given a route with known
requests at the beginning of the day, while new urgent requests are inserted
dynamically throughout the day. An interesting feature of this problem is
the possible mix of skills, tools, and spare part requirements, which have to
be matched in order to service the request. This problem has been studied
by Borenstein et al. [19] with an application to British Telecom.
Another application of dynamic routing arises in the context of the French
non-profit organization SOS Médecins. This organization operates with a
crew of physicians, who are called on duty via a call center coordinated with
other emergency services. When a patient calls, the severity of the case
is evaluated, and a visit by a practitioner is planned accordingly. As in
other emergency services, having an efficient dispatching system reduces the

9
response time, thus improving service level for the society. On the other hand,
it is important to decide in real-time whether or not to send a physician, so
that it is possible to ensure a proper service level in areas where emergencies
are likely to appear.
Dynamic aspects can also appear on arc routing problems. This is for
instance the case in the study by Tagmouti et al. [136] on the operation
of a fleet of vehicles for winter gritting applications. Their work consider a
network of streets or road segments that need to be gritted when affected by a
moving storm. Depending on the movements of the storm, new segments may
have to be gritted, and the routing of vehicles has to be updated accordingly.

3.2 Transport of goods

Due to the fact that urban areas are often characterized by highly variable
traveling times, transport of goods in such areas have led to the definition of a
specific category of applications known as city logistics. City logistics can be
defined as an integrated vision of transport activities in urban areas, taking
into account factors such as traffic and competition or cooperation between
transport companies [140]. Barcelo et al. [6] developed a general framework
for city logistics applications. They describe the different modules ranging
from modeling the city road network and acquiring real-time traffic data to
the dynamic routing of a fleet of vehicles. Zeimpekis et al. [153] proposed
a Decision Support System (DSS) for city logistics which takes into account
dynamic travel and service times.
A typical application in city logistics is the courier service present in
most urban areas. Couriers are dispatched to customer locations to collect
packages, and either deliver them to their destination (short haul) or to a
unique depot (long haul). Depending on the level of service paid by the cus-
tomer, couriers may consolidate pick-ups from various customers, or provide
an expedited service. Companies offering courier services often have an het-
erogeneous fleet composed of bicycles, motorbikes, cars, and small vans. The
problem is then to dynamically route couriers, taking into account not only
the known requests, their type, pick-up and delivery locations, and time win-
dows, but also considering traffic conditions and varying travel times. A case
study by Attanasio et al. [1] outlines the benefits of using an optimization-
enabled AFMS at eCourier Ltd, a London based company offering courier
services. The authors illustrate that aside from the improvements in service
quality, response time, and courier efficiency, the use of an automated system

10
allows decoupling the fleet size from the need for more dispatchers. Further
results motivated by a similar application can be found in Gendreau et al.
[51] and Ghiani et al. [59].
The delivery of newspapers and magazines is a domain in which customer
satisfaction is of first importance. When a magazine or newspaper is not
delivered, a subscriber contacts a call center and is offered to choose between
a voucher or a future delivery. In the latter case, the request is then forwarded
to the delivery company, which assigns it to a driver that will do a priority
delivery. Traditionally, this process relies on an exchange of phone calls,
faxes, and printed documents, that ultimately communicate the driver about
the pending delivery, once he/she comes back to the depot. As an alternative,
Bieding et al. [18] propose a centralized application that makes use of mobile
phones to communicate with drivers and intelligently perform the routing in
real time, reducing costs and improving customer satisfaction. More recently,
Ferrucci et al. [44] developed an approach that makes use historical data to
anticipate future requests.
Another application in which customer requests need to be answered with
short delays can be found in companies with a direct service model, such as
grocery delivery services. In general, the customer selects products on a
website, and then chooses a time frame for the delivery at his home. Tra-
ditionally, the vendor defines an arbitrary number of customers that can be
serviced within a time window, and the time window is made unavailable
to customers as soon as the capacity is reached. Campbell and Savelsbergh
[24] defined the Home Delivery Problem, in which the goal is to maximize
the total expected revenue by dynamically deciding whether or not to accept
a customer request within a specific time window. In comparison with the
traditional approach, this means that the time windows available for a cus-
tomer are dynamically defined taking into consideration the possible future
requests. The authors propose a Greedy Randomized Adaptive Search Pro-
cedure (GRASP) and compare different cost functions to capture the problem
uncertainty. Later, Azi et al. [3] proposed an Adaptive Large Neighborhood
Search (ALNS) that take into account uncertainty by generating scenarios
containing possible demand realizations.
Apart from classical routing problems, related operational problems also
arise in many organizations. The review by Stahlbock and Voss [135] on op-
erations research applications in container terminals describes the dynamic
stacker crane problem [5, 14], which considers the routing of container car-
riers loading and unloading ships in a terminal. Other applications include

11
transport of goods inside warehouses [132], factories, and hospitals, where
documents or expensive medical instruments must be transferred efficiently
between services [45].

3.3 Transport of persons

The transport of persons is in general–and by many aspects–similar to the
transport of goods, yet it is characterized by additional constraints such as
regulation on waiting, travel, and service times.
Taxis are arguably the most common on-demand individual transport
systems. Requests are composed of a pick-up location and time, possibly
coupled with a destination. They can be either known in advance, for in-
stance when a customer books a cab for the next day, or they can arrive
dynamically, in which case a taxi must be dispatched in the shortest time.
When customers cannot share a vehicle, the closest free taxi is generally
the one which takes the ride, leaving limited space for optimization. The
study by Caramia et al. [25], generalized by Fabri and Recht [42], focuses on
a multi-cab metropolitan transportation system, where a taxi can transport
more than one passenger at the same time. In this case the online algorithms
minimize the total traveled distance, while assigning requests to vehicles and
computing the taxi routes. This multi-cab transportation system can be
generalized as an on-demand or door-to-door transport service.
Many applications involve the transport of children, the elderly, disabled
people, or patients, from their home to schools, place of work, or medical
centers. Xiang et al. [149] studied a DARP with changing travel speeds,
vehicle breakdowns, and traffic congestion; while Dial [36], followed by Horn
[67, 68, 69], studied demand-responsive transport systems. An extensive
review of this class of problems can be found in the studies by Cordeau et al.
[32] and Berbeglia et al. [14].
A singular application of on-demand transportation systems can be found
in major hospitals, with services possibly spread across various buildings on
several branches. Depending on the medical procedure or facility capacity,
a patient may need to be transferred on short notice from one service to
another, possibly requiring trained staff or specific equipment for his/her
care. This application has been studied by Beaudry et al. [7], Kergosien
et al. [80], and Melachrinoudis et al. [96].
Air taxis developed as a flexible response to the limitations of traditional
airlines. Air taxis offer passengers the opportunity to travel through smaller

12
airports, avoiding waiting lines at check-in and security checks. Air taxi
companies offer an on-demand service: customers book a flight a few days
in advance, specifying whether they are willing to share the aircraft, stop
at an intermediate airport, or have flexible traveling hours. Then, the com-
pany accommodates these requests, trying to consolidate flights whenever
possible. The underlying optimization problems have not been subject to
much attention, except in the studies by Cordeau et al. [32], Espinoza et al.
[40, 41], Fagerholt et al. [43], and Yao et al. [152]. Similar problems arises in
helicopter transportation systems, typically used by oil and gas companies
to transport personnel between offshore petroleum platforms [63, 124].

4 Solution Methods
Few research was conducted on dynamic routing between the work of Psaraftis
[116] in 1980 and the late 1990s. However, the last decade has seen a renewed
interest for this class of problems [39], with solution techniques ranging from
linear programming to metaheuristics. This section presents the major con-
tributions in this field, and the reader is referred to the reviews, books,
and special issues by Gendreau and Potvin [55, 56], Ghiani et al. [57], Goel
[61], Ichoua [72], Ichoua et al. [75, 76], Jaillet and Wagner [78], Larsen et al.
[91], and Zeimpekis et al. [154], to complement our review.

4.1 Dynamic and deterministic routing problems

This section presents approaches that have been successfully applied to dy-
namic routing, in the absence of stochastic information. In this context,
critical information is revealed over time, meaning that the complete in-
stance is only known at the end of the planning horizon. As a consequence,
exact methods only provide an optimal solution for the current state, but
do not guarantee that the solution will remain optimal once new data be-
comes available. Therefore, most dynamic approaches rely on heuristics that
quickly compute a solution to the current state of the problem. Approaches
for dynamic and deterministic vehicle routing problems can be divided into
two categories: those based on periodic reoptimization, and those based on
continuous reoptimization.

13
4.1.1 Periodic reoptimization
To the best of our knowledge, the first periodic reoptimization approach is
due to Psaraftis [116], with the development of a dynamic programming ap-
proach. His research focuses on the DARP and consists in finding the optimal
route each time a new request is known. The main drawback of dynamic pro-
gramming is the well-known curse of dimensionality [110, Chap. 1], which
prevents its application to large instances.
More generally, periodic reoptimization approaches start at the beginning
of the day with a first optimization that produces an initial set of routes.
Then, an optimization procedure periodically solves a static problem corre-
sponding to the current state, either whenever the available data changes, or
at fixed intervals of time –referred to as decision epochs [29] or time slices
[81]. The advantage of periodic reoptimization is that it can be based on al-
gorithms developed for static routing, for which extensive research has been
carried out. The main drawback is that all the optimization needs to be
performed before updating the routing plan, thus increasing delays for the
dispatcher.
Yang et al. [150] addressed the real-time truckload PDP, in which a fleet of
trucks has to service point-to-point transport requests arriving dynamically.
Important assumptions are that all trucks can only handle one request at
a time, with no possible preemption, and they travel at the same constant
speed. The authors propose MYOPT, a rolling horizon approach based on
a linear program (LP) that is solved whenever a new request arrives. Along
the same line of linear programming, Chen and Xu [29] designed a dynamic
column generation algorithm (DYCOL) for the D-VRPTW. The authors
propose the concept of decision epochs over the planning horizon, which are
the dates when the optimization process runs. The novelty of their approach
relies on dynamically generating columns for a set-partitioning model, using
columns from the previous decision epoch. The authors compared DYCOL to
a traditional column generation with no time limit (COL). Computational
results based on the Solomon benchmark [133] demonstrate that DYCOL
yields comparable results in terms of objective function, but with running
times limited to 10 seconds, opposed to the various hours consumed by COL.
Montemanni et al. [102] developed an Ant Colony System (ACS) to solve
the D-VRP. Similar to Kilby et al. [81], their approach uses time slices, that
is, they divide the day in periods of equal duration. A request arriving during
a time slice is not handled until the end of the time bucket, thus the problem

14
solved during a time slice only considers the requests known at its beginning.
Hence, the optimization is run statically and independently during each time
slice. The main advantage of this time partition is that similar computational
effort is allowed for each time slice. This discretization is also possible by
the nature of the requests, which are never urgent, and can be postponed.
An interesting feature of their approach is the use of the pheromone trace to
transfer characteristics of a good solution to the next time slice. A similar
approach was also used by Gambardella et al. [50] and Rizzoli et al. [123].

4.1.2 Continuous reoptimization

Continuous reoptimization approaches perform the optimization throughout
the day and maintain information on good solutions in an adaptive memory
[138]. Whenever the available data changes, a decision procedure aggregates
the information from the memory to update the current routing. The ad-
vantage is that the computational capacity is maximized, possibly at the
expense of a more complex implementation. It is worth noting that because
the current routing is subject to change at any time, vehicles do not know
their next destination until they finish the service of a request.
To the best of our knowledge, the first continuous reoptimization ap-
proach is due Gendreau et al. [52] with the adaptation of the parallel Tabu
Search (TS) framework introduced by Taillard et al. [137] to a D-VRPTW
problem arising in the local operation of long distance express courier ser-
vices. Their approach maintains a pool of good routes–the adaptive memory–
which is used to generate initial solutions for a parallel TS. The parallelized
search is done by partitioning the routes of the current solution, and op-
timizing them in independent threads. Whenever a new customer request
arrives, it is checked against all the solutions from the adaptive memory to
decide whether it should be accepted or rejected. This framework was also
implemented for the D-VRP [73, 74], while other variations of TS have been
applied to the D-PDP [6, 27] and the DARP [2, 7].
Bent and Van Hentenryck [9] introduced the Multiple Plan Approach
(MPA) as a generalization of the TS with adaptive memory [52]. The general
idea is to populate and maintain a solution pool (the routing plans) that are
used to generate a distinguished solution. Whenever a new request arrives,
a procedure is called to check whether it can be serviced or not; if it can be
serviced, then the request is inserted in the solution pool and incompatible
solutions are discarded. Pool updates are performed periodically or whenever

15
a vehicle finishes servicing a customer. This pool-update phase is crucial and
ensures that all solutions are coherent with the current state of vehicles and
customers. The pool can be seen as an adaptive memory that maintains a
set of alternative solutions.
In an early work, Benyahia and Potvin [13] studied the D-PDP and pro-
posed a Genetic Algorithm (GA) that models the decision process of a human
dispatcher. More recently, GAs were also used for the same problem [30, 65]
and for the D-VRP [144]. Genetic algorithms in dynamic contexts are very
similar to those designed for static problems, although they generally run
throughout the planning horizon and solutions are constantly adapting to
the changes made to the input.

4.2 Dynamic and stochastic routing problems

Dynamic and stochastic routing problems can be seen as an extension of their
deterministic counterparts, where additional (stochastic) knowledge is avail-
able in the dynamically revealed input. Approaches for this class of problems
can be divided in two categories: those based on sampling and those based
on stochastic modeling. As their name suggests, sampling strategies incor-
porate stochastic knowledge by generating scenarios based on realizations
drawn from random variable distributions. Each scenario is then optimized
by solving the static and deterministic problem they define. On the other
hand, approaches based on stochastic modeling integrate stochastic knowl-
edge analytically. The advantage of sampling is its relative simplicity and
flexibility on distributional assumptions, while its drawback is the massive
generation of scenarios to accurately reflect reality. Alternatively, stochastic
modeling strategies formally capture the stochastic nature of the problem,
but they are highly technical in their formulation and require to efficiently
compute possibly complex expected values. Examples of these two strategies
follow.

4.2.1 Stochastic modeling

Powell et al. [113] formulated a truckload PDP as a Markov Decision Process
(MDP). Later, MDPs were used by Thomas and White [142] and Thomas
[141] to solve a VRP in which known customers may ask for service with a
known probability. Kim et al. [82] also used MDPs to tackle the VRP with
dynamic travel times. Unfortunately, the curse of dimensionality and the

16
simplifying assumptions make this approach unsuitable in most real-world
applications. Nonetheless, it allowed new insights in the field of dynamic
programming.
To cope with the scalability problems of traditional dynamic program-
ming, Approximate Dynamic Programming (ADP) steps forward in time,
approximates the value function, and ultimately avoids the evaluation of all
possible states. We refer the interested reader to Powell [110, 111] for a
more detailed description of the ADP framework. ADP has been success-
fully applied to freight transport [112, 114] and fleet management problems
[60, 115, 129]. In particular, Novoa and Storer [104] propose an ADP algo-
rithm to dynamically solve the VRPSD.
Linear programming has also been adapted to the dynamic and stochas-
tic context. The OPTUN approach, proposed by Yang et al. [150] as an
extension of MYOPT (see § 4.1.1), considers opportunity costs on each arc
to reflect the expected cost of traveling to isolated areas. Consequently,
the optimization tends to reject isolated requests, and avoids traversing arcs
that are far away from potential requests. Later, Yang et al. [151] studied
the emergency vehicle dispatching and routing and proposed a mathemati-
cal formulation that was later used by Haghani and Yang [66] on a similar
problem.

4.2.2 Sampling
Sampling approaches rely on the generation of scenarios containing possible
realizations of the random variables. Figure 3 illustrates how scenarios are
generated for the D-VRP. Solely based on the current customers, the opti-
mal tour would be (A, B, E, D, C) (3a.), which ignores two zones (gray areas)
where customers are likely to appear. By sampling the customer spatial dis-
tributions, customers X, Y , and Z are generated, and the new optimal tour is
(C, X, Y, B, A, Z, E, D) (3b.). Removing the sampled (potential) customers
leads to the tour (C, B, A, E, D) (3c.) which is suboptimal regarding a my-
opic cost evaluation, but leaves room to accommodate new customers at a
lower cost.
The Multiple Scenario Approach (MSA) is a predictive adaptation of the
MPA framework discussed in § 4.1.2. The idea behind MSA is to take ad-
vantage of the time between decisions to continuously improve the current
scenario pool. During the initialization, the algorithm, generates a first set
of scenarios based on the requests known beforehand. Throughout the day,

17
 D C D C D C

X X
E E E
A A A
Z Z
B B Y B Y

a. A-priori optimal tour b. Optimal tour with sam- c. Optimized scenario with-
pled customers out sampled customers
Legend: A A Customer (actual/sampled) Depot Tour Customer distribution density

Figure 3: Scenario generation in sampling approaches.

scenarios are then reoptimized and new ones are generated and added to the
pool. When a decision is required, the scenario optimization procedure is
suspended, and MSA uses the scenario pool to select the request to service
next. MSA then discards the scenarios that are incompatible with the cur-
rent routing, and resumes the optimization. Computational experiments on
instances adapted from the Solomon benchmark [133] showed that MSA out-
performs MPA both in terms of serviced customers and traveled distances,
especially for instances with high degrees of dynamism [9]. Flatberg et al.
[47] adapted the SPIDER commercial solver to use multiple scenarios and
a consensus algorithm to tackle the D-VRP, while Pillac et al. [107] imple-
mented an event-driven optimization framework based on MSA and showed
significant improvements over state-of-the-art algorithms for the D-VRPSD.
An important component of scenario based-approaches such as MSA is
the decision process, which defines how the information from the scenario pool
is used to reach upon a decision regarding the next customer to visit. The
most common algorithms used to reach a decision in MSA are: consensus,
expectation, and regret. The consensus algorithm [9, 10] selects the customer
appearing first with the highest frequency among scenarios. Expectation
[8, 10, 26] consists in evaluating the cost of visiting each customer first by
forcing its visit in all scenarios and performing a complete optimization.
Finally, regret [8] approximates the expectation algorithm and avoids the
reoptimization of all scenarios. Even though these algorithms were initially
designed for the routing of a single vehicle, they can be extended to the
multi-vehicle case [145].

18
 Hvattum et al. [70] developed the Dynamic Sample Scenario Hedge Heuris-
tic (DSHH), an approach similar to the consensus algorithm for D-VRP. This
method divides the planning horizon into time intervals. At the beginning of
each interval, DSHH revises the routing by assigning a subset of promising
requests to the vehicles, depending on the frequency of their assignment over
all scenarios. DSHH later led to the development of the Branch and Regret
Heuristic (BRH), where scenarios are merged to build a unique solution.
Various local search approaches have been developed for the stochastic
and dynamic problems. Ghiani et al. [59] developed an algorithm for the
D-PDP that only samples the near future to reduce the computational ef-
fort. The main difference with MSA is that no scenario pool is used and
the selection of the distinguished solution is based on the expected penalty
of accommodating requests in the near future. Azi et al. [3] developed an
Adaptive Large Neighborhood Search (ALNS) for a dynamic routing problem
with multiple delivery routes, in which the dynamic decision is the acceptance
of a new request. The approach maintains a pool of scenarios, optimized by
an ALNS, that are used to evaluate the opportunity value of an incoming
request.
Tabu search has also been adapted to dynamic and stochastic problems.
Ichoua et al. [75] and Attanasio et al. [1] tackled with tabu search the D-
VRPTW and the D-PDP, respectively.

4.2.3 Other strategies

In addition to the general frameworks described previously, the use of stochas-
tic knowledge allows for the design and implementation of other strategies
that try to adequately respond to upcoming events.
The waiting strategy consists in deciding whether a vehicle should wait
after servicing a request, before heading toward the next customer; or plan-
ning a waiting period on a strategic location. This strategy is particularly
important in problems with time windows, where time lags appear between
requests. Mitrović-Minić et al. [100] proved that in all cases it is better to
wait after servicing a customer, but a more refined strategy can lead to fur-
ther improvements. The problem is in general to evaluate the likelihood of a
new request in the neighborhood of a serviced request and to plan a waiting
period accordingly. The waiting strategy has been implemented in various
frameworks for the D-VRP [22, 141], D-VRPTW [12, 21, 75, 145], D-PDP
[59, 100], and Dynamic and Stochastic TSP [58]. The strategy has shown

19
good results, especially in the case of a limited fleet facing a high request
rate [145].
Aside from the waiting after or before servicing a customer, a vehicle can
be relocated to a strategic position, where new requests are likely to arrive.
This strategy is the keystone of emergency fleet deployment, also known
as Emergency Vehicle Dispatching–or Redeployment–Problem [54, 66]. The
relocation strategy has also been applied to other vehicle routing problems,
such as the D-VRP [87],D-VRPTW [12, 21, 75, 145], D-TSPTW [89], D-PDP
[59, 119], and the Resource Allocation Problem (RAP) [60].
Request buffering, introduced by Pureza and Laporte [119], consists in
delaying the assignment of some requests to vehicles in a priority buffer, so
that more urgent requests can be handled first.

4.3 Performance evaluation

In contrast to static problems, where measuring the performance of an algo-
rithm is straightforward (i.e., running time and solution quality), dynamic
problems require the introduction of new metrics to assess the performance
of a particular method.
Sleator and Tarjan [131] introduced the competitive analysis [77, 90]. Let
P be a minimization problem and I the set of all instances of P . Let z ∗ (Ioff )
be the optimal cost for the offline instance Ioff corresponding to I ∈ I. For
offline instance Ioff , all input data from instance I, either static or dynamic,
is available when building the solution. In contrast, the data of the online
version I is revealed in real time, thus an algorithm A has to take into
account new information as it is revealed and produce a solution relevant to
the current state of knowledge. Let zA (I) = z(xA (I)) be the cost of the final
solution xA (I) found by the online algorithm A on instance I. Algorithm A
is said to be c-competitive, or equivalently to have a competitive ratio of c, if
there exists a constant α such that

zA (I) ≤ c · z ∗ (Ioff ) + α , ∀I∈I (4)

In the case where α = 0, the algorithm is said to be strictly c-competitive,

meaning that in all cases the objective value of the solution found by A will
be at most of c times the optimal value. The competitive ratio metric allows
a worst-case absolute measure of an algorithm performance in terms of the
objective value. We refer the reader to Borodin and El-Yaniv [20] for an

20
in-depth analysis of this measure, and to Jaillet and Wagner [77] and Fink
et al. [46] for results on various routing problems.
The main drawback of the competitive analysis is that it requires to prove
the previously stated inequality analytically, which may be complex for real-
world applications. The value of information proposed by Mitrović-Minić
et al. [100] constitutes a more flexible and practical metric. We denote by
zA (Ioff ) the value of the objective function returned by algorithm A for the
offline instance Ioff . The value of information VA (I) for algorithm A on
instance I is then defined as
zA (I) − zA (Ioff )
VA (I) = (5)
zA (Ioff )

The value of information can be interpreted as the gap between the solution
returned by an algorithm A on a instance I and the solution returned by the
same algorithm when all information from I is known beforehand. In contrast
with the competitive ratio, the value of information gives information on the
performance of an algorithm based on empirical results, without requiring
optimal solutions for the offline instances. It captures the impact of the dy-
namism on the solution yield by the algorithm under analysis. For instance,
Gendreau et al. [52] report a value of information between 2.5% and 4.1% for
their tabu search algorithm for the D-VRPTW, while Tagmouti et al. [136]
reports values between 10% and 26.7% for a variable neighborhood search
descent applied to a dynamic arc routing problem.

4.4 Benchmarks
To date, there is no reference benchmark for dynamic routing problems.
Although, it is worth noting that various authors based their computational
experiments on adaptations of the Solomon [133] instances for static routing
[8, 9, 28, 29, 52]. Van Hentenryck and Bent [145, Chap. 10] describe how the
original benchmark by Solomon [133] can be adapted to dynamic problems.
The interested reader is referred to the website of Pankratz and Krypczyk
[106] for an updated list of publicly available instances sets for dynamic
vehicle routing problems.

21
5 Conclusions
Recent technological advances provide companies with the right tools to man-
age their fleet in real time. Nonetheless, these new technologies also introduce
more complexity in fleet management tasks, unveiling the need for decision
support systems adapted to dynamic contexts. Consequently, during the
last decade, the research community have shown a growing interest for the
underlying optimization problems, leading to a new family of approaches
specifically designed to efficiently address dynamism and uncertainty. By
analyzing the current state of the art, some directions can be drawn for
future research in this relatively new field.
First, further work should aim at creating a taxonomy of dynamic vehicle
routing problem, possibly by extending existing research on static routing
[39]. This would allow a more precise classification of approaches, evaluate
similarities between problems, and foster the development of generic frame-
works.
Second, there is currently no reference benchmark for dynamic vehicle
routing problems. Therefore, there is a strong need for the development of
publicly available benchmarks for the most common dynamic vehicle routing
problems.
Third, with the advent of multi-core processors on desktop computers,
and low-cost graphical processing units (GPU), parallel computing is now
readily available for time-consuming methods such as those based on sam-
pling. Although early studies considered distributed optimization [52], most
approaches reviewed in this document do not take advantage of parallel ar-
chitectures. The development of parallel algorithms is a challenge that could
reduce the time needed for optimization and provide decision makers with
highly reactive tools.
Fourth, our review of the existing literature revealed that a large fraction
of work done in the area of dynamic routing does not consider stochastic
aspects. We are convinced that developing algorithms that make use of
stochastic information will improve the fleet performance and reduce oper-
ating costs. Thus this line of research should become a priority in the near
future.
Finally, researchers have mainly focused on the routing aspect of the
dynamic fleet management. However, in some applications there is more
that can be done to improve performance and service level. For instance,
in equipment maintenance services, the call center has a certain degree of

22
freedom in fixing service appointments. In other words, it means that the
customer time windows can be defined, or influenced, by the call center
operator. As a consequence, a system in which aside from giving a yes/no
answer to a customer request, suggests convenient times for the company
would be highly desirable in such contexts.

Acknowledgements
Financial support for this work was provided by the CPER (Contrat de
Projet Etat Region) Vallée du Libre; and the Centro de Estudios Interdis-
ciplinarios Básicos y Aplicados en Complejidad (CEIBA, Colombia). This
support is gratefully acknowledged.

References
[1] Attanasio, A., Bregman, J., Ghiani, G., and Manni, E. (2007). Real-time fleet man-
agement at Ecourier Ltd. In Zeimpekis, V., Tarantilis, C. D., Giaglis, G. M., and Minis,
I., editors, Dynamic Fleet Management, volume 38 of Operations Research/Computer
Science Interfaces, chapter 10, pages 219–238. Springer US.

[2] Attanasio, A., Cordeau, J. F., Ghiani, G., and Laporte, G. (2004). Parallel tabu
search heuristics for the dynamic multi-vehicle dial-a-ride problem. Parrallel Computing,
30(3):377–387.

[3] Azi, N., Gendreau, M., and Potvin, J. Y. (2011). A dynamic vehicle routing problem
with multiple delivery routes. Annals of Operations Research, In-press.

[4] Baldacci, R., Toth, P., and Vigo, D. (2007). Recent advances in vehicle routing exact
algorithms. 4OR: A Quarterly Journal of Operations Research, 5(4):269–298.

[5] Balev, S., Guinand, F., Lesauvage, G., and Olivier, D. (2009). Dynamical handling of
straddle carriers activities on a container terminal in uncertain environment - a swarm
intelligence approach -. In Proceedings of the 2009 International Conference on Complex
Systems and Applications (ICCSA 2009), Le Havre, France. University of Le Havre.

[6] Barcelo, J., Grzybowska, H., and Pardo, S. (2007). Vehicle routing and scheduling
models, simulation and city logistics. In Zeimpekis, V., Tarantilis, C. D., Giaglis,
G. M., and Minis, I., editors, Dynamic Fleet Management, volume 38 of Operations
Research/Computer Science Interfaces, pages 163–195. Springer US.

[7] Beaudry, A., Laporte, G., Melo, T., and Nickel, S. (2010). Dynamic transportation of
patients in hospitals. OR Spectrum, 32:77–107.

23
[8] Bent, R. and Van Hentenryck, P. (2004a). Regrets only! online stochastic optimization
under time constraints. In Proceedings of the 19th National Conference on Artificial
Intelligence (AAAI-04), pages 501–506. AAAI Press.

[9] Bent, R. and Van Hentenryck, P. (2004b). Scenario-based planning for partially dy-
namic vehicle routing with stochastic customers. Operations Research, 52(6):977–987.

[10] Bent, R. and Van Hentenryck, P. (2004c). The value of consensus in online stochastic
scheduling. In Proceedings of the 14th International Conference on Automated Planning
and Scheduling (ICAPS-04). AAAI Press.

[11] Bent, R. and Van Hentenryck, P. (2005). Online stochastic and robust optimization.
In Maher, M., editor, Advances in Computer Science – ASIAN 2009, volume 3321 of
Lecture Notes in Computer Science, pages 286–300. Springer Berlin / Heidelberg.

[12] Bent, R. and Van Hentenryck, P. (2007). Waiting and relocation strategies in online
stochastic vehicle routing. In Veloso, M., editor, Proceedings of ohe 20th International
Joint Conference on Artifical Intelligence (IJCAI-07), pages 1816–1821.

[13] Benyahia, I. and Potvin, J. Y. (1998). Decision support for vehicle dispatching using
genetic programming. IEEE Transactions on Systems Man and Cybernetics Part A -
Systems and Humans, 28(3):306–314.

[14] Berbeglia, G., Cordeau, J.-F., and Laporte, G. (2010). Dynamic pickup and delivery
problems. European Journal of Operational Research, 202(1):8 – 15.

[15] Bertsimas, D. (1988). Probabilistic combinatorial optimization problems. PhD thesis,

Massachusetts Institute of Technology, Dept. of Mathematics.

[16] Bertsimas, D. and Simchi-Levi, D. (1996). A new generation of vehicle routing re-
search: robust algorithms, addressing uncertainty. Operations Research, 44(2):286–304.

[17] Bertsimas, D. and Van Ryzin, G. (1991). A stochastic and dynamic vehicle-routing
problem in the Euclidean plane. Operations Research, 39(4):601–615.

[18] Bieding, T., Görtz, S., and Klose, A. (2009). On-line routing per mobile phone : A
case on subsequent deliveries of newspapers. In Beckmann, M., Künzi, H. P., Fandel,
G., Trockel, W., Basile, A., Drexl, A., Dawid, H., Inderfurth, K., Kürsten, W., Nunen,
J. A., Speranza, M. G., and Bertazzi, L., editors, Innovations in Distribution Logistics,
volume 619 of Lecture Notes in Economics and Mathematical Systems, pages 29–51.
Springer Berlin Heidelberg.

[19] Borenstein, Y., Shah, N., Tsang, E., Dorne, R., Alsheddy, A., and Voudouris, C.
(2010). On the partitioning of dynamic workforce scheduling problems. Journal of
Scheduling, 13(4):411–425.

[20] Borodin, A. and El-Yaniv, R. (2005). Online Computation and Competitive Analysis.
Cambridge University Press, Cambridge.

24
[21] Branchini, R. M., Armentano, V. A., and Lokketangen, A. (2009). Adaptive granular
local search heuristic for a dynamic vehicle routing problem. Computers & Operations
Research, 36(11):2955–2968.

[22] Branke, J., Middendorf, M., Noeth, G., and Dessouky, M. (2005). Waiting strategies
for dynamic vehicle routing. Transportation Science, 39(3):298–312.

[23] Brotcorne, L., Laporte, G., and Semet, F. (2003). Ambulance location and relocation
models. European Journal of Operational Research, 147(3):451–463.

[24] Campbell, A. and Savelsbergh, M. (2005). Decision support for consumer direct
grocery initiatives. Transportation Science, 39(3):313–327.

[25] Caramia, M., Italiano, G., Oriolo, G., Pacifici, A., and Perugia, A. (2002). Routing
a fleet of vehicles for dynamic combined pick-up and deliveries services. In Proceedings
of the Symposium on Operation Research 2001, pages 3–5, Duisburg, Germany.

[26] Chang, H., Givan, R., and Chong, E. (2000). On-line scheduling via sampling. In
Proceedings of the Artificial Intelligence Planning and Scheduling (AIPS 2000), pages
62–71.

[27] Chang, M. S., Chen, S., and Hsueh, C. (2003). Real-time vehicle routing problem
with time windows and simultaneous delivery/pickup demands. Journal of the Eastern
Asia Society for Transportation Studies, 5:2273–2286.

[28] Chen, H.-K., Hsueh, C.-F., and Chang, M.-S. (2006). The real-time time-dependent
vehicle routing problem. Transportation Research Part E: Logistics and Transportation
Review, 42(5):383–408.

[29] Chen, Z. and Xu, H. (2006). Dynamic column generation for dynamic vehicle routing
with time windows. Transportation Science, 40(1):74–88.

[30] Cheung, B. K. S., Choy, K. L., Li, C.-L., Shi, W., and Tang, J. (2008). Dynamic
routing model and solution methods for fleet management with mobile technologies.
International Journal of Production Economics, 113(2):694–705.

[31] Christiansen, C. and Lysgaard, J. (2007). A branch-and-price algorithm for the

capacitated vehicle routing problem with stochastic demands. Operations Research
Letters, 35(6):773–781.

[32] Cordeau, J.-F., Laporte, G., Potvin, J.-Y., and Savelsbergh, M. W. (2007a). Trans-
portation on demand. In Barnhart, C. and Laporte, G., editors, Transportation, vol-
ume 14 of Handbooks in Operations Research and Management Science, chapter 7, pages
429–466. Elsevier.

[33] Cordeau, J.-F., Laporte, G., Savelsbergh, M. W., and Vigo, D. (2007b). Vehicle rout-
ing. In Barnhart, C. and Laporte, G., editors, Transportation, volume 14 of Handbooks
in Operations Research and Management Science, chapter 6, pages 367–428. Elsevier.

25
[34] Crainic, T. G., Gendreau, M., and Potvin, J.-Y. (2009). Intelligent freight-
transportation systems: Assessment and the contribution of operations research. Trans-
portation Research Part C: Emerging Technologies, 17(6):541–557.

[35] Dantzig, G. and Ramser, J. (1959). The truck dispatching problem. Management
Science, 6(1):80–91.

[36] Dial, R. B. (1995). Autonomous dial-a-ride transit introductory overview. Trans-

portation Research Part C: Emerging Technologies, 3(5):261–275.

[37] Dror, M., Laporte, G., and Trudeau, P. (1989). Vehicle routing with stochastic
demands: Properties and solution frameworks. Transportation Science, 23(3):166–176.

[38] Du, T., Wang, F. K., and Lu, P.-Y. (2007). A real-time vehicle-dispatching system for
consolidating milk runs. Transportation Research Part E: Logistics and Transportation
Review, 43(5):565–577.

[39] Eksioglu, B., Vural, A. V., and Reisman, A. (2009). The vehicle routing problem: A
taxonomic review. Computers & Industrial Engineering, 57(4):1472 – 1483.

[40] Espinoza, D., Garcia, R., Goycoolea, M., Nemhauser, G. L., and Savelsbergh, M.
W. P. (2008a). Per-seat, on-demand air transportation part I: Problem description and
an integer multicommodity flow model. Transportation Science, 42(3):263–278.

[41] Espinoza, D., Garcia, R., Goycoolea, M., Nemhauser, G. L., and Savelsbergh, M.
W. P. (2008b). Per-seat, on-demand air transportation part II: Parallel local search.
Transportation Science, 42(3):279–291.

[42] Fabri, A. and Recht, P. (2006). On dynamic pickup and delivery vehicle routing with
several time windows and waiting times. Transportation Research Part B: Methodolog-
ical, 40(4):335 – 350.

[43] Fagerholt, K., Foss, B. A., and Horgen, O. J. (2009). A decision support model for
establishing an air taxi service: a case study. Journal of The Operational Research
Society, 60(9):1173–1182.

[44] Ferrucci, F., Block, S., and Gendreau, M. (2011). Real-time distribution of perishable
goods using past request information to anticipate future requests. Operations Research,
Under review:34.

[45] Fiegl, C. and Pontow, C. (2009). Online scheduling of pick-up and delivery tasks in
hospitals. Journal of Biomedical Informatics, 42(4):624 – 632.

[46] Fink, I., Krumke, S. O., and Westphal, S. (2009). New lower bounds for online
k-server routing problems. Information Processing Letters, 109(11):563 – 567.

26
[47] Flatberg, T., Hasle, G., Kloster, O., Nilssen, E. J., and Riise, A. (2007). Dynamic
and stochastic vehicle routing in practice. In Zeimpekis, V., Tarantilis, C. D., Giaglis,
G. M., and Minis, I., editors, Dynamic Fleet Management, volume 38 of Operations
Research/Computer Science Interfaces, pages 41–63. Springer US.

[48] Fleischmann, B., Gnutzmann, S., and Sandvoss, E. (2004). Dynamic vehicle routing
based on online traffic information. Transportation Science, 38(4):420–433.

[49] Flood, M. (1956). The traveling-salesman problem. Operations Research, 4(1):61–75.

[50] Gambardella, L., Rizzoli, A., Oliverio, F., Casagrande, N., Donati, A., Montemanni,
R., and Lucibello, E. (2003). Ant colony optimization for vehicle routing in advanced lo-
gistics systems. In Proceedings of the International Workshop on Modelling and Applied
Simulation (MAS 2003), pages 3–9.

[51] Gendreau, M., Guertin, F., Potvin, J.-Y., and Séguin, R. (2006). Neighborhood
search heuristics for a dynamic vehicle dispatching problem with pick-ups and deliveries.
Transportation Research Part C: Emerging Technologies, 14(3):157–174.

[52] Gendreau, M., Guertin, F., Potvin, J.-Y., and Taillard, E. (1999). Parallel tabu search
for real-time vehicle routing and dispatching. Transportation Science, 33(4):381–390.

[53] Gendreau, M., Laporte, G., and Séguin, R. (1996). Stochastic vehicle routing. Euro-
pean Journal of Operational Research, 88(1):3 – 12.

[54] Gendreau, M., Laporte, G., and Semet, F. (2001). A dynamic model and parallel tabu
search heuristic for real-time ambulance relocation. Parallel Computing, 27(12):1641 –
1653.

[55] Gendreau, M. and Potvin, J.-Y. (1998). Dynamic vehicle routing and dispatching.
In Crainic, Teodor G. and Laporte, Gilbert, editors, Fleet management and logistics,
chapter 5, pages 115–126. Kluwer, Boston.

[56] Gendreau, M. and Potvin, J.-Y., editors (2004). Transportation Science. Number 38.
(special issue on real-time fleet management).

[57] Ghiani, G., Guerriero, F., Laporte, G., and Musmanno, R. (2003). Real-time vehicle
routing: Solution concepts, algorithms and parallel computing strategies. European
Journal of Operational Research, 151(1):1 – 11.

[58] Ghiani, G., Laporte, G., Manni, E., and Musmanno, R. (2008). Waiting strategies
for the dynamic and stochastic traveling salesman problem. International Journal of
Operations Research, 5(4):233–241.

[59] Ghiani, G., Manni, E., Quaranta, A., and Triki, C. (2009). Anticipatory algorithms
for same-day courier dispatching. Transportation Research Part E: Logistics and Trans-
portation Review, 45(1):96 – 106.

27
[60] Godfrey, G. and Powell, W. B. (2002). An adaptive dynamic programming algorithm
for dynamic fleet management, I: Single period travel times. Transportation Science,
36(1):21–39.

[61] Goel, A. (2008). Fleet Telematics: Real-time management and planning of commer-
cial vehicle operations, volume 40 of Operations Research Computer Science Interfaces
Series. Springer Verlag.

[62] Goel, A. and Gruhn, V. (2008). A general vehicle routing problem. European Journal
of Operational Research, 191(3):650–660.

[63] Gribkovskaia, I., Laporte, G., and Shlopak, A. (2008). A tabu search heuristic for a
routing problem arising in servicing of offshore oil and gas platforms. Journal of the
Operational Research Society, 59(11):1449–1459.

[64] Güner, A. R., Murat, A., and Chinnam, R. B. (2012). Dynamic routing under re-
current and non-recurrent congestion using real-time its information. Computers &
Operations Research, 39(2):358 – 373.

[65] Haghani, A. and Jung, S. (2005). A dynamic vehicle routing problem with time-
dependent travel times. Computers & Operations Research, 32(11):2959 – 2986.

[66] Haghani, A. and Yang, S. (2007). Real-time emergency response fleet deployment:
Concepts, systems, simulation & case studies. In Zeimpekis, V., Tarantilis, C. D., Gi-
aglis, G. M., and Minis, I., editors, Dynamic Fleet Management, volume 38 of Operations
Research/Computer Science Interfaces, pages 133–162. Springer US.

[67] Horn, M. E. T. (2002a). Fleet scheduling and dispatching for demand-responsive

passenger services. Transportation Research Part C: Emerging Technologies, 10(1):35 –
63.

[68] Horn, M. E. T. (2002b). Multi-modal and demand-responsive passenger transport

systems: a modelling framework with embedded control systems. Transportation Re-
search Part A: Policy and Practice, 36(2):167 – 188.

[69] Horn, M. E. T. (2004). Procedures for planning multi-leg journeys with fixed-route
and demand-responsive passenger transport services. Transportation Research Part C:
Emerging Technologies, 12(1):33–55.

[70] Hvattum, L. M., Lokketangen, A., and Laporte, G. (2006). Solving a dynamic and
stochastic vehicle routing problem with a sample scenario hedging heuristic. Trans-
portation Science, 40(4):421–438.

[71] Hvattum, L. M., Lokketangen, A., and Laporte, G. (2007). A branch-and-regret

heuristic for stochastic and dynamic vehicle routing problems. Networks, 49(4):330–
340.

28
[72] Ichoua, S. (2001). Problèmes de gestion de flottes de véhicules en temps réel. PhD
thesis, Universite de Montreal, Montreal, Canada.

[73] Ichoua, S., Gendreau, M., and Potvin, J.-Y. (2000). Diversion issues in real-time
vehicle dispatching. Transportation Science, 34(4):426–438.

[74] Ichoua, S., Gendreau, M., and Potvin, J.-Y. (2003). Vehicle dispatching with time-
dependent travel times. European Journal of Operational Research, 144(2):379 – 396.

[75] Ichoua, S., Gendreau, M., and Potvin, J.-Y. (2006). Exploiting knowledge about
future demands for real-time vehicle dispatching. Transportation Science, 40(2):211–
225.

[76] Ichoua, S., Gendreau, M., and Potvin, J.-Y. (2007). Planned route optimization for
real-time vehicle routing. In Zeimpekis, V., Tarantilis, C. D., Giaglis, G. M., and Minis,
I., editors, Dynamic Fleet Management, volume 38 of Operations Research/Computer
Science Interfaces, pages 1–18. Springer US.

[77] Jaillet, P. and Wagner, M. R. (2008a). Generalized online routing: New compet-
itive ratios, resource augmentation, and asymptotic analyses. Operations Research,
56(3):745–757.

[78] Jaillet, P. and Wagner, M. R. (2008b). Online vehicle routing problems: A survey.
In The Vehicle Routing Problem: Latest Advances and New Challenges, volume 43 of
Operations Research/Computer Science Interfaces Series, pages 221–237. Springer US.

[79] Kenyon, A. S. and Morton, D. (2003). Stochastic vehicle routing with random travel
times. Transportation Science, 37(1):69.

[80] Kergosien, Y., Lenté, C., Piton, D., and Billaut, J.-C. (2011). A tabu search heuristic
for the dynamic transportation of patients between care units. European Journal of
Operational Research, In Press, doi:10.1016/j.ejor.2011.04.033:–.

[81] Kilby, P., Prosser, P., and Shaw, P. (1998). Dynamic VRPs: a study of scenarios.
Technical Report APES-06-1998, University of Strathclyde, Glasgow, Scotland.

[82] Kim, S., Lewis, M., and C., W. C. (2005). Optimal vehicle routing with real-time
traffic information. IEEE Transactions on Intelligent Transportation Systems, 6(2):178–
188.

[83] Laporte, G. (2007). What you should know about the vehicle routing problem. Naval
Research Logistics, 54(8):811–819.

[84] Laporte, G. (2009). Fifty years of vehicle routing. Transportation Science, 43(4):408–
416.

[85] Laporte, G., Louveaux, F., and Mercure, H. (1992). The vehicle routing problem
with stochastic travel times. Transportation Science, 26(3):161–170.

29
[86] Laporte, G., Louveaux, F., and Van Hamme, L. (2002). An integer L-shaped algo-
rithm for the capacitated vehicle routing problem with stochastic demands. Operations
Research, 50(3):415–423.

[87] Larsen, A. (2001). The Dynamic Vehicle Routing Problem. PhD thesis, Technical
University of Denmark (DTU).

[88] Larsen, A., Madsen, O. B. G., and Solomon, M. M. (2002). Partially dynamic vehicle
routing - models and algorithms. The Journal of the Operational Research Society,
53(6):637–646.

[89] Larsen, A., Madsen, O. B. G., and Solomon, M. M. (2004). The a priori dynamic
traveling salesman problem with time windows. Transportation Science, 38(4):459–472.

[90] Larsen, A., Madsen, O. B. G., and Solomon, M. M. (2007). Classification of dynamic
vehicle routing systems. In Zeimpekis, V., Tarantilis, C. D., Giaglis, G. M., and Minis,
I., editors, Dynamic Fleet Management, volume 38 of Operations Research/Computer
Science Interfaces Series, chapter 2, pages 19–40. Springer US.

[91] Larsen, A., Madsen, O. B. G., and Solomon, M. M. (2008). Recent developments
in dynamic vehicle routing systems. In The Vehicle Routing Problem: Latest Advances
and New Challenges, volume 43 of Operations Research/Computer Science Interfaces
Series, pages 199–218. Springer US.

[92] Li, J.-Q., Mirchandani, P. B., and Borenstein, D. (2009a). A lagrangian heuristic for
the real-time vehicle rescheduling problem. Transportation Research Part E: Logistics
and Transportation Review, 45(3):419–433.

[93] Li, J.-Q., Mirchandani, P. B., and Borenstein, D. (2009b). Real-time vehicle rerouting
problems with time windows. European Journal of Operational Research, 194(3):711 –
727.

[94] Lorini, S., Potvin, J.-Y., and Zufferey, N. (2011). Online vehicle routing and schedul-
ing with dynamic travel times. Computers & Operations Research, 38(7):1086 – 1090.

[95] Lund, K., Madsen, O. B. G., and Rygaard, J. M. (1996). Vehicle routing problems
with varying degrees of dynamism. Technical report, IMM Institute of Mathematical
Modelling.

[96] Melachrinoudis, E., Ilhan, A. B., and Min, H. (2007). A dial-a-ride problem for
client transportation in a health-care organization. Computers & Operations Research,
34(3):742–759.

[97] Mendoza, J. E., Castanier, B., Guéret, C., Medaglia, A. L., and Velasco, N. (2010). A
memetic algorithm for the multi-compartment vehicle routing problem with stochastic
demands. Computers & Operations Research, 37(11):1886–1898.

30
[98] Mendoza, J. E., Castanier, B., Guéret, C., Medaglia, A. L., and Velasco, N. (2011).
Constructive heuristics for the multicompartment vehicle routing problem with stochas-
tic demands. Transportation Science, 45(3):346–363.

[99] Mes, M., Van der Heijden, M., and Van Harten, A. (2007). Comparison of agent-based
scheduling to look-ahead heuristics for real-time transportation problems. European
Journal of Operational Research, 181(1):59–75.

[100] Mitrović-Minić, S., Krishnamurti, R., and Laporte, G. (2004). Double-horizon based
heuristics for the dynamic pickup and delivery problem with time windows. Transporta-
tion Research Part B: Methodological, 38(8):669 – 685.

[101] Mitrović-Minić, S. and Laporte, G. (2004). Waiting strategies for the dynamic
pickup and delivery problem with time windows. Transportation Research Part B:
Methodological, 38(7):635–655.

[102] Montemanni, R., Gambardella, L. M., Rizzoli, A. E., and Donati, A. V. (2005).
Ant colony system for a dynamic vehicle routing problem. Journal of Combinatorial
Optimization, 10(4):327–343.

[103] Mu, Q., Fu, Z., Lysgaard, J., and Eglese, R. (2011). Disruption management of the
vehicle routing problem with vehicle breakdown. Journal of the Operational Research
Society, 62(4):742–749.

[104] Novoa, C. and Storer, R. (2009). An approximate dynamic programming approach

for the vehicle routing problem with stochastic demands. European Journal of Opera-
tional Research, 196(2):509–515.

[105] Novoa, C. M. (2005). Static and dynamic approaches for solving the vehicle routing
problem with stochastic demands. PhD thesis, Lehigh University, Pennsylvania, United
States. AAT 3188502.

[106] Pankratz, G. and Krypczyk, V. (2009). Benchmark data sets for dynamic vehicle
routing problems. http://www.fernuni-hagen.de/WINF/inhalte/benchmark_data.
htm.

[107] Pillac, V., Guéret, C., and Medaglia, A. L. (2012). An event-driven optimiza-
tion framework for dynamic vehicle routing. Decision Support Systems, Accepted
manuscript.

[108] Potvin, J. Y., Xu, Y., and Benyahia, I. (2006). Vehicle routing and scheduling with
dynamic travel times. Computers & Operations Research, 33(4):1129–1137.

[109] Powell, W., Shapiro, J., and Simao, H. (2002). An adaptive dynamic programming
algorithm for the heterogeneous resource allocation problem. Transportation Science,
36(2):231–249.

31
[110] Powell, W. B. (2007). Approximate dynamic programming: solving the curses
of dimensionality, volume 703 of Wiley Series in Probability and Statistics. Wiley-
Interscience, Hoboken, New Jersey.

[111] Powell, W. B. (2009). What you should know about approximate dynamic program-
ming. Naval Research Logistics, 56(3):239–249.

[112] Powell, W. B., Bouzaiene-Ayari, B., and Simao, H. (2007). Dynamic models for
freight transportation. In Barnhart, C. and Laporte, G., editors, Transportation, vol-
ume 14 of Handbooks in Operations Research and Management Science, chapter 5, pages
285–365. North-Holland.

[113] Powell, W. B., Sheffi, Y., Nickerson, K. S., Butterbaugh, K., and Atherton, S.
(1988). Maximizing profits for North American Van Lines’ truckload division: A new
framework for pricing and operation. Interfaces, 18(1):21–41.

[114] Powell, W. B. and Topaloglu, H. (2003). Stochastic programming in transportation

and logistics. Handbooks in Operations Research and Management Science, 10:555–636.

[115] Powell, W. B. and Topaloglu, H. (2005). Fleet management. In Wallace, S. and

Ziemba, W., editors, Applications of Stochastic Programming, volume 5 of MPS-SIAM
series on Optimization, chapter 12, pages 185–215. SIAM.

[116] Psaraftis, H. (1980). A dynamic-programming solution to the single vehicle many-

to-many immediate request dial-a-ride problem. Transportation Science, 14(2):130–154.

[117] Psaraftis, H. (1988). Dynamic vehicle routing problems. In Golden, B. and Assas,
A., editors, Vehicle Routing: Methods and Studies, pages 223–248. Elsevier Science
Publishers B.V.

[118] Psaraftis, H. N. (1995). Dynamic vehicle routing: Status and prospects. Annals of
Operations Research, 61(1):143–164.

[119] Pureza, V. and Laporte, G. (2008). Waiting and buffering strategies for the dynamic
pickup and delivery problem with time windows. INFOR, 46(3):165–175.

[120] Regan, A., Mahmassani, H., and Jaillet, P. (1995). Improving efficiency of com-
mercial vehicle operations using real-time information: potential uses and assignment
strategies. Transportation Research Record: Journal of the Transportation Research
Board, 1493:188–198.

[121] Regan, A., Mahmassani, H., and Jaillet, P. (1998). Evaluation of dynamic fleet
management systems - simulation framework. In Forecasting, Travel Behavior, And
Network Modeling, number 1645 in Transportation Research Record, pages 176–184.

[122] Regan, A. C., Mahmassani, H. S., and Jaillet, P. (1996). Dynamic decision making
for commercial fleet operations using real-time information. Transportation Research
Record: Journal of the Transportation Research Board, 1537:91–97.

32
[123] Rizzoli, A., Montemanni, R., Lucibello, E., and Gambardella, L. (2007). Ant colony
optimization for real-world vehicle routing problems. Swarm Intelligence, 1:135–151.

[124] Romero, M., Sheremetov, L., and Soriano, A. (2007). A genetic algorithm for the
pickup and delivery problem: An application to the helicopter offshore transportation.
In Theoretical Advances and Applications of Fuzzy Logic and Soft Computing, volume 42
of Advances in Soft Computing, pages 435–444. Springer Berlin / Heidelberg.

[125] Roy, J. (2001). Recent trends in logistics and the need for real-time decision tools in
the trucking industry. In System Sciences, 2001. Proceedings of the 34th Annual Hawaii
International Conference on.

[126] Secomandi, N. (2000). Comparing neuro-dynamic programming algorithms for the

vehicle routing problem with stochastic demands. Computers & Operations Research,
27(11-12):1201 – 1225.

[127] Secomandi, N. (2001). A rollout policy for the vehicle routing problem with stochas-
tic demands. Operations Research, 49(5):796–802.

[128] Secomandi, N. and Margot, F. (2009). Reoptimization approaches for the vehicle-
routing problem with stochastic demands. Operations Research, 57(1):214–230.

[129] Simao, H., Day, J., George, A., Gifford, T., Nienow, J., and Powell, W. B. (2009).
An approximate dynamic programming algorithm for large-scale fleet management: A
case application. Transportation Science, 43(2):178–197.

[130] Slater, A. (2002). Specification for a dynamic vehicle routing and scheduling system.
International Journal of Transport Management, 1(1):29 – 40.

[131] Sleator, D. and Tarjan, R. (1985). Amortized efficiency of list update and paging
rules. Communications of the ACM, 28(2):202–208.

[132] Smolic-Rocak, N., Bogdan, S., Kovacic, Z., and Petrovic, T. (2010). Time windows
based dynamic routing in multi-agv systems. IEEE Transactions on Automation Science
and Engineering, 7(1):151–155.

[133] Solomon, M. M. (1987). Algorithms for the vehicle-routing and scheduling problems
with time window constraints. Operations Research, 35(2):254–265.

[134] Spivey, M. and Powell, W. B. (2004). The dynamic assignment problem. Trans-
portation Science, 38(4):399–419.

[135] Stahlbock, R. and Voss, S. (2008). Operations research at container terminals: a

literature update. OR Spectrum, 30(1):1–52.

[136] Tagmouti, M., Gendreau, M., and Potvin, J.-Y. (2011). A dynamic capacitated arc
routing problem with time-dependent service costs. Transportation Research Part C:
Emerging Technologies, 19(1):20 – 28.

33
[137] Taillard, E., Badeau, P., Gendreau, M., Guertin, F., and Potvin, J. (1997). A tabu
search heuristic for the vehicle routing problem with soft time windows. Transportation
Science, 31(2):170–186.

[138] Taillard, E. D., Gambardella, L. M., Gendreau, M., and Potvin, J.-Y. (2001). Adap-
tive memory programming: A unified view of metaheuristics. European Journal of
Operational Research, 135(1):1 – 16.

[139] Taniguchi, E. and Shimamoto, H. (2004). Intelligent transportation system based

dynamic vehicle routing and scheduling with variable travel times. Transportation Re-
search Part C: Emerging Technologies, 12(3-4):235–250.

[140] Taniguchi, E. and Thompson, R. (2002). Modeling city logistics. Transportation

Research Record: Journal of the Transportation Research Board, 1790(1):45–51.

[141] Thomas, B. W. (2007). Waiting strategies for anticipating service requests from
known customer locations. Transportation Science, 41(3):319–331.

[142] Thomas, B. W. and White, Chelsea C., I. (2004). Anticipatory route selection.
Transportation Science, 38(4):473–487.

[143] Toth, P. and Vigo, D., editors (2002). The vehicle routing problem, volume 9 of
SIAM Monographs on Discrete Mathematics. SIAM Philadelphia.

[144] Van Hemert, J. I. and Poutré, J. L. (2004). Dynamic routing problems with fruitful
regions: Models and evolutionary computation. In Yao, X., Burke, E., Lozano, J. A.,
Smith, J., Merelo-Guervós, J. J., Bullinaria, J. A., Rowe, J., Tino, P., Kabán, A., and
Schwefel, H.-P., editors, Parallel Problem Solving from Nature, volume 3242 of Lecture
Notes in Computer Science, pages 692–701. Springer Berlin / Heidelberg.

[145] Van Hentenryck, P. and Bent, R. (2006). Online stochastic combinatorial optimiza-
tion. MIT Press.

[146] Verweij, B., Ahmed, S., Kleywegt, A., Nemhauser, G., and Shapiro, A. (2003).
The sample average approximation method applied to stochastic routing problems: a
computational study. Computational Optimization and Applications, 24(2):289–333.

[147] Waters, C. (1989). Vehicle-scheduling problems with uncertainty and omitted cus-
tomer. The Journal of the Operational Research Society, 40(12riz):1099–1108.

[148] Wilson, N. and Colvin, N. (1977). Computer control of the Rochester dial-a-ride
system. Technical Report Report R77-31, Dept. of Civil Engineering, Massachusetts
Institute of Technology, Cambridge, Massachusetts.

[149] Xiang, Z., Chu, C., and Chen, H. (2008). The study of a dynamic dial-a-ride problem
under time-dependent and stochastic environments. European Journal of Operational
Research, 185(2):534–551.

34
[150] Yang, J., Jaillet, P., and Mahmassani, H. (2004). Real-time multivehicle truckload
pickup and delivery problems. Transportation Science, 38(2):135–148.

[151] Yang, S., Hamedi, M., and Haghani, A. (2005). Online dispatching and routing
model for emergency vehicles with area coverage constraints. In Network Modeling
2005, number 1923 in Transportation Research Record, pages 1–8.

[152] Yao, Y., Ergun, O., and Johnson, E. (2007). Integrated model for the dynamic
on-demand air transportation operations. In Zeimpekis, V., Tarantilis, C. D., Giaglis,
G. M., and Minis, I., editors, Dynamic Fleet Management, volume 38 of Operations
Research/Computer Science Interfaces Series, pages 95–111. Springer US. 10.1007/978-
0-387-71722-7 5.

[153] Zeimpekis, V., Minis, I., Mamassis, K., and Giaglis, G. M. (2007a). Dynamic man-
agement of a delayed delivery vehicle in a city logistics environment. In Zeimpekis, V.,
Tarantilis, C. D., Giaglis, G. M., and Minis, I., editors, Dynamic Fleet Management,
volume 38 of Operations Research/Computer Science Interfaces Series, chapter 9, pages
197–217. Springer US.

[154] Zeimpekis, V., Tarantilis, C. D., Giaglis, G. M., and Minis, I., editors (2007b).
Dynamic Fleet Management, volume 38 of Operations Research Computer Science In-
terfaces Series. Springer US.

35