Analyzing A Unified Ant System For The VRP and Some of Its Variants

Download as pdf or txt
Download as pdf or txt
You are on page 1of 11

Analyzing a Unified Ant System for the VRP

and Some of Its Variants

Marc Reimann, Karl Doerner, and Richard F. Hartl

Institute of Management Science, University of Vienna, Brünnerstrasse 72,


A-1210 Vienna, Austria
{marc.reimann, karl.doerner, richard.hartl}@univie.ac.at
http://www.bwl.univie.ac.at/bwl/prod/index.html

Abstract. In this paper we analyze the application of an Ant System


to different vehicle routing problems. More specifically, we study the
robustness of our Unified Ant System by evaluating its performance on
four different problem classes within the domain of vehicle routing.

1 Introduction
Vehicle routing and Scheduling has been of major interest to the scientific
community for the last 50 years due to the inherent complexity of the asso-
ciated problems. A large number of different approaches, in recent years mostly
meta-heuristics have been proposed for different variants of vehicle routing and
scheduling problems, see further Toth and Vigo [1].
On the other hand, goods distribution and more specifically vehicle rout-
ing represents an important area of an economy. Basically, in every supply chain
transportation occurs between member companies of the chain or from the chain
to final customers. Thus, firms have recognized the need to automatize this pro-
cess and use software to support their distribution process. Part of this software
is an optimization tool for routing and scheduling.
However, while it seems that scientific interest and company interest overlap
in this regard, this first impression is not completely true. Rather, the level of
development of tools in the scientific world is not matched in industry. Most
of the tools used in industry are based on very simple optimization techniques
developed many years ago.
The reasons for this are manyfold. First, it can not be expected that cutting
edge research will instantaneously be transferred to industry. Rather a certain lag
has to be accepted. Second, in the quest for ever more sophisticated optimization
techniques the scientific community has reached a level of specialization that
prevents the tools developed from being applicable to a wide range of problems.
Thus, while these tools solve particular problem scenarios much better than
the ’old’ simple techniques, they lack the flexibility of the latter approaches.
However, for practical purposes solution quality is not the only criterion for the
choice of an optimization tool.
This gap has been recognized by some researchers. For example Cordeau et
al.[2] present four measures of algorithm performance, namely solution quality,

S. Cagnoni et al. (Eds.): EvoWorkshops 2003, LNCS 2611, pp. 300–310, 2003.
c Springer-Verlag Berlin Heidelberg 2003
Analyzing a Unified Ant System for the VRP and Some of Its Variants 301

speed, flexibility and simplicity. According to their analysis most of the meta-
heuristics, while outperforming simple heuristics with respect to solution quality
can not compete with simple heuristics when it comes to speed or simplicity.
Simplicity is customarily measured in terms of the number of parameters of an
algorithm, while flexibility is concerned with the applicability of an algorithm
to different variants of a basic problem. This applicabilty covers also issues of
robustness, i.e. whether parameter changes are necessary if the problem instance
to be solved changes. On the other hand, a research group in Oslo works on a
formal representation and algorithms for rich vehicle routing problems, which
cover a wide range of problems with different constraints and objectives (c.f. [3]).
In light of these observations, the aim of our paper is to present an Ant
System algorithm that is capable of solving different variants of vehicle routing
problems reasonably well, with little or no adjustments to the implementation.
More specifically, we aim to solve the vehicle routing problem (VRP), the vehicle
routing problem with time windows (VRPTW), the vehicle routing problem with
backhauls (VRPB) and the vehicle routing problem with backhauls and time
windows (VRPBTW). The algorithm is based on the Insertion based AS we
presented in Reimann et al. [4].
The remainder of this paper is organized as follows. In the next section we
describe the characteristics of the different problems we aim to solve. After that
we briefly describe our Unified AS approach. In section 4 we present our rich
computational study, before we conclude in Section 5.

2 The Vehicle Routing Problem and Some of Its Variants


The classic VRP (c.f. e.g [1]) aims to find a set of minimum cost routes, each
starting and ending at a central depot. These routes have to satisfy the known
demands of a number of customers, where each customer must be served by
exactly one vehicle, i.e. order splitting is not an option. Of course vehicle capac-
ities and (possibly) maximum tour length restrictions have to be respected. It is
assumed that the available fleet is homogeneous, i.e. the vehicles are identical.
This basic setup can be extended in several directions to deal with more
realistic scenarios. The most important extensions are those dealing with more
complicated restrictions at the customers. More specifically, two characteristics
are dominant:
– time windows: Certain or all customers may require that service is performed
within a certain time span. In this case, vehicles arriving at a customer before
the beginning of the time window have to wait until the customer is willing
to accept the service. A vehicle that arrives at a customer location after the
end of the time window set by the customer may not serve the customer
anymore. However as service is obligatory, this means that any algorithm
has to make sure that all vehicles arrive at each customer before the end of
the respective time window. The VRPTW has been studied extensively in
the last decade; for a recent overview of metaheuristic approaches see (e.g.
[5]).
302 M. Reimann, K. Doerner, and R.F. Hartl

– backhauls: Intermediate companies of a supply chain often deal with sup-


pliers and customers. It has been recognized that a combination of the two
distribution areas leads to significant improvements. Thus, a vehicle that
delivers goods to customers may also pick up goods from the company’s
suppliers. This leads to a mixed problem where some locations need delivery
while others require pick up of goods. In the backhauling case an additional
restriction is that all the pickups be after the deliveries. The intuition for
this is that picking up goods before all deliveries are done may lead to the
necessity of rearranging goods in the truck. Doing these rearrangements en
route is expensive and thus to be avoided. The VRPB has also received a
lot of attention recently (see e.g. [6]).

Given these additional problem characteristics, we end up with 4 problem


classes. The simplest case is the vehicle routing problem (VRP) without time
windows or backhauls. Adding one of the two customer characteristics leads
to either the vehicle routing problem with time windows (VRPTW) if time
windows are considered, or the vehicle routing problem with backhauls (VRPB)
if backhauls are treated. Finally, both types of constraints are added in the
vehicle routing problem with backhauls and time windows (VRPBTW).
Compared to the first three problems, the VRPBTW has received only very
little attention. The only meta-heuristic approach is due to Duhamel et al. [8]
who proposed a Tabu Search algorithm to tackle the problem.
In the next section we will present an Ant System that can be applied to each
of the four problems. We will discuss the implementation with respect to the
characteristics of the different problems. In closing this section let us note that
other possible extensions are the consideration of vehicle heterogeneity, multiple
depots or multiple trips per vehicle. These extensions have been studied by
different authors (c.f. Gendreau et al. [9], Cordeau et al. [10]). Testing whether
our algorithm applies to these problems is left for further research.

3 Ant System Algorithms for the VRP and Some of Its


Variants

In this section we briefly describe our Ant System algorithm. In particular, we


focus on the constructive heuristics used.

3.1 Ant Systems

The Ant System approach, originally proposed by Colorni et al. (see e.g. [11])
is based on the behavior of real ants searching for food. Real ants communicate
with each other using an aromatic essence called pheromone, which they leave
on the paths they traverse. In the absence of pheromone trails ants more or less
perform a random walk. However, as soon as they sense a pheromone trail on
a path in their vicinity, they are likely to follow that path, thus reinforcing this
trail. More specifically, if ants at some point sense more than one pheromone trail,
Analyzing a Unified Ant System for the VRP and Some of Its Variants 303

they will choose one of these trails with a probability related to the strenghts
of the existing trails. This idea has first been applied to the TSP, where an ant
located in a city chooses the next city according to the strength of the artificial
trails.
Improved versions of the basic algorithm have been applied to a large num-
ber of different combinatorial optimization problems (for an overview see [12]).
Recently, a convergence proof for a generalized Ant System has been developed
by Gutjahr [13]. Generally, the Ant System algorithm consists of the iteration
of three steps:

– Generation of solutions by ants according to private and pheromone infor-


mation
– Application of a local search to the ants’ solutions
– Update of the pheromone information

Below we will discuss the first step in more detail. The second and third step
are reviewed only briefly (c.f. [4] for a more detailed description of these steps).

3.2 Generation of Solutions

As proposed in Reimann et al. [4], we use an Insertion algorithm derived from


the I1 insertion algorithm proposed by Solomon [14] for the VRPTW. This algo-
rithm works as follows: Routes are constructed one by one. First, the unrouted
customer farthest from the depot is selected as a seed customer for the current
route, that is, only this customer is served by the route. Sequentially other cus-
tomers are inserted into this route according to a cost criterion based on their
distance to the depot as well as the detour and delay caused by inserting them.
Once no more insertions are feasible with respect to time window, capacity or
tour length constraints, another route is initialized with a seed customer and
the insertion procedure is repeated with the remaining unrouted customers. The
algorithm stops when all customers are assigned to routes.
In order to use the algorithm described above within the framework of our
Ant System we need to adapt it to allow for a probabilistic choice in each decision
step. This is done in the following way. To initialize a tour, seed customers are
not chosen deterministically but probabilistically according to their distance
from the depot.
Inserting further customers on the current tour is done using a roulette wheel
selection over all unrouted customers with positive evaluation function κi . The
decision rule used can be written as

  κi
 h|κ >0 κh if κi > 0
h
Pi = (1)


0 otherwise.
The chosen customer i is then inserted into the current route at its best feasible
insertion position.
304 M. Reimann, K. Doerner, and R.F. Hartl

To compute the evaluation function κi for inserting an unrouted customer


i at its best insertion position on the current tour we first determine for each
unrouted customer i the attractiveness of insertion at any feasible insertion
position on the current tour. Formally the attractiveness of inserting customer
i immediately after customer j can be written as

ηij = max {0,


α · d0i − β · (dji + dik − djk ) − (1 − β) · (bik − bk )
+δ · (γ · typei + (1 − γ) · (1 − typei ))} ∀i ∈ Nu , ∀j ∈ Ri ,

where d0i denotes the distance between the depot and customer i and Nu denotes
the set of unrouted customers. Further, k is the customer visited immediately
after customer j in the current solution, bik is the actual arrival time at customer
k, if i is inserted between customers j and k, while bk is the arrival time at
customer k before the insertion of customer i and Ri denotes the set of customers
assigned to the current tour after which customer i could feasibly be inserted.
Finally, typei is a binary indicator variable denoting whether customer i is a
linehaul (typei = 0) or a backhaul customer (typei = 1). The intuition is that
we want to be able to discriminate between linehaul and backhaul customers.
Note, that given ’inappropriate’ values for the parameters α, β, γ and δ the
attractiveness ηij can become zero. Note further, that our attractiveness function
is an extension of the function proposed by Solomon [14] for the VRPTW. More
precisely, our function reduces to Solomon’s function if δ is set to δ = 0.
Given the attractiveness we then compute the evaluation function of the
best insertion position for each customer i on the current tour by

τji +τik
κi = max {0, maxj∈Ri [ηij · 2·τjk ]} ∀i ∈ Nu ,

where τji denotes the pheromone concentration on the arc connecting locations
(customers or depot) j and i. The pheromone concentration τji contains infor-
mation about how good visiting two customers i and j immediately after each
other was in previous iterations. Note, that the same pheromone utilization is
done for route initialization, thus augmenting the attractiveness of initializing a
route with an unrouted customer i by the search-historic information.
Computing the attractiveness ηij reflects the tradeoff between detour and
delay costs associated with inserting a customer. This tradeoff is weighted by the
parameter β. A higher value for β puts higher emphasis on the detour and lower
emphasis on the delay. The tradeoff is also influenced by customer characteristics
such as distance to the depot or type and these characteristics are weighted
by the parameters α and γ, δ respectively. More precisely, higher α will favor
customers far from the depot, while higher δ will put more emphasis on the
customer type. Finally, γ influences the discrimination between linehaul and
backhaul customers.
Analyzing a Unified Ant System for the VRP and Some of Its Variants 305

3.3 Local Search


After an ant has constructed its solution, we apply a local search algorithm
to improve the solution quality. In particular, we sequentially apply Move and
Swap operators to the solution. Generalized versions of both operators have been
proposed by Osman [15] for the VRP.
The Move operator tries to eject one or two adjacent customers from their
current position and insert it at another position. It is a special case of the 3-
opt operator. The elementary Swap operator, aims at improving the solution by
exchanging a customer i with a customer j. This operator is a special case of
the 4-opt operator. Both operators are used to move and exchange customers of
the same tour and between tours.

3.4 Pheromone Update


After all ants have constructed their solutions, the pheromone trails are up-
dated on the basis of the solutions found by the ants. According to the rank
based scheme proposed in [16] the pheromone level on all arcs is first decreased
according to the evaporation factor (1 − ρ). Second, only arcs belonging to either
the best solution found so far or to one of the E − 1 best solutions found in the
current iteration are receiving positive reinforcement. The amount of positive
reinforcement depends on the rank of the ant and is proportional to the inverse
solution quality found by the ant. That way the search is driven to promising
areas of the search space. For more details on this approach we refer to [16].

4 Numerical Analysis
As testing our algorithm on all instances in the four classes would have been
computationally too expensive we performed our numerical analysis on selected
problem instances. First, to eliminate size effects we only considered problems
with approximately 100 customers. In fact, only the VRPB instances have 90
and 94 customers respectively as for this class 100 customer instances are not
available. Also, in order not to bias the results by taking ’easy’ or ’hard’ instances
we randomly chose 19 instances. These are:
– VRP (from [17]): vrp3, vrp8, vrp12, vrp14
– VRPTW (from [14]): r101n, r206n, c105n, c207n, rc104n, rc203n
– VRPB (from [18]): lhbh-i1, lhbh-i3, lhbh-j1, lhbh-j3
– VRPBTW (from [7]): bhr101b, bhr102a, bhr103b, bhr104c, bhr105c

4.1 Parameter Settings


Let n be the problem size, i.e. the number of customers to be served, then the
Ant System parameters were: m = n/2 ants, ρ = 0.95 and E = 4 elitists.
These parameters are standard settings and were not tested systematically as
our experience suggests that the rank based Ant System is quite robust.
306 M. Reimann, K. Doerner, and R.F. Hartl

Generally, the objective for time constrained routing problems is to first


minimize the fleet size required to serve all customers and given a minimal fleet
size to minimize the total distance travelled. This objective was established by
minimization of the following objective function:

L = M · F S + T T, (2)
where L denotes the total costs of a solution, F S denotes the fleet size found,
and T T corresponds to the total travel time (or distance). The parameter M
has to be chosen in a way to ensure that a solution that saves a vehicle always
outperforms a solution with a higher fleet size. More precisely we set M = 10000
in our experiments.
Finally, as we are interested in the interplay of the parameters of our insertion
algorithm with the characteristics of the different problems we analyzed the four
parameters of the insertion algorithm in the ranges α ∈ {0.5, 1, 1.5, 2}, β ∈
{0, 0.33, 0.66, 1}, γ ∈ {0, 0.33, 0.66, 1} and δ ∈ {0, 0.33, 0.66, 1} on the instances
described above.
This means that we tested 256 parameter constellations for each instance.
For each constellation we performed 1 run of 10 minutes. All our computations
were performed on a Pentium 3 with 900MHz. The code was implemented in C.

4.2 Influence of the Problem Characteristics on the Parameter


Values

As stated in the last section, we have tested a large number of constellations


for each of the 19 problem instances. As our main interest lies in the robustness
of the final approach we have compared the different settings based on averages
over all instances. First, we select for each instance the best solution found by
our Unified Ant System (over all parameter settings) as a reference value. Each
individual setting was then evaluated by relating its performance to the reference
values.
deviation from best UAS solution in %

0,6

0,5

0,4

0,3

0,2

0,1

0
vrp3
vrp8
vrp12
vrp14
r101n
r206n
c105n
c207n
rc104n
rc203n
lhbh-i1
lhbh-i3
lhbh-j1
lhbh-j3
bhr101b
bhr102a
bhr103b
bhr104c
bhr105c

problem instances

Fig. 1. Robustness of the ’best’ setting over the 19 test instances


Analyzing a Unified Ant System for the VRP and Some of Its Variants 307

Using this approach, we found the following ’best’ setting:


α = 2, β = 1, γ = 0.33, δ = 1
The average deviation of this setting over the best solution found for each
instance is equal to 0.03%. Let us look more precisely at the behavior of this
setting over all instances. Figure 1 shows the percentage deviation from the best
solution found by our Unified AS.
Clearly, for 17 out of the 19 instances this setting shows excellent behavior
with virtually 0% deviation. For the remaining two instances the deviations of
0.1% and 0.5% respectively still are more than reasonable.
Given these encouraging results, let us now perform some sensitivity tests.
More precisely, we are interested how changing one parameter from its ’best’
setting influences the performance of the Unified AS given that all other param-
eters remain unchanged. Table 1 shows the deviation in % from the best solution
found by our Unified AS, for different settings of the 4 parameters.

Table 1. Sensitivity analysis of the parameter settings

Parameter Parameter
setting 0.5 1 1.5 2 setting 0 0.33 0.66 1
α 0.14% 0.07% 0.05% 0.03% β 0.03% 0.04% 0.04% 0.03%
γ 0.03% 0.03% 0.03% 0.03%
δ 0.03% 0.03% 0.03% 0.03%

Clearly, these results show that the parameter α has the strongest influence
on solution quality. The other three parameters show basically no sensitivity to
their actual setting. To understand this result let us consider the effects of the
parameters. Both the parameters α and δ are free parameters, i.e. they can take
any values. Moreover, higher levels for these parameters may implicitly increase
the number of positive evaluations κi . Thus, the number of customers on each
truck may increase which in turn leads to a possible reduction in fleet sizes. As
this is the main goal, higher values for α and δ should be expected to improve
the solution quality. However, both of the free parameters α and δ are at the
highest level we tested. This suggests that considering even higher values for
these parameters might further improve the performance of the algorithm. This
question is not considered in the remainder of this paper, but rather left for
future research.
Finally, the parameters β and γ have more indirect effects on the fleet size
such that their actual values will influence mainly the second objective namely
the total distance travelled.

4.3 Comparison of Our Unified AS with Existing Results


Now that we have analyzed the influence of the actual parameter settings on the
solution quality let us evaluate the performance of the ’best’ setting we found
308 M. Reimann, K. Doerner, and R.F. Hartl

against state of the art results. More precisely, we consider for each instance the
best known solution.1 Apart from the VRP instances - for those the standard
objective is to minimize total distance travelled only - all the other best known
results are based on the same lexicographic ordering of objectives we use.
In our comparison, we split the solutions into the fleet size and the total
distance travelled. Table 2 shows the results for all of the 19 instances.

Table 2. Comparison of our Unified AS with best known results

P roblem Unified AS Best known results


instance fleet size total distance fleet size total distance
vrp3 8 859.68 8 826.14
vrp8 9 886.47 9 865.94
vrp12 10 848.64 10 819.56
vrp14 10 971.82 11 866.37
r101n 19 1655.41 19 1650.80
r206n 3 979.45 3 912.00
c105n 10 828.94 10 828.94
c207n 3 589.20 3 588.29
rc104n 10 1196.05 10 1135.48
rc203n 3 1175.88 3 1060.45
lhbh − i1 9 361.08 10 360.00
lhbh − i3 5 317.07 5 306.00
lhbh − j1 10 352.93 10 352.00
lhbh − j3 7 301.67 7 302.00
bhr101b 23 1975.07 23 1999.20
bhr102a 19 1681.34 19 1677.60
bhr103b 15 1405.02 16 1395.88
bhr104c 12 1216.32 12 1208.50
bhr105c 17 1642.70 17 1633.01
T otal 202 19244.75 205 18788.16

Our results show that considering the first goal, namely the fleet size, our
algorithm improves the best known results in three cases, leading to a reduction
in fleet sizes equal to 1.46%. With respect to the second objective, our algorithm
deviates by 2.43% from the best known solution. Thus, the reduction in fleet
size is achieved through giving up some solution quality associated with the
total distance travelled. Overall, our algorithm being the first unified approach
for the four classes studied shows very promising behavior.

1
Note, that these best known results are taken from different sources as to date there
is no unified approach for the studied problem classes. Additional information about
the corresponding sources can be found at
http://www.bwl.univie.ac.at/bwl/prod/reimann/vrpresults.html.
Analyzing a Unified Ant System for the VRP and Some of Its Variants 309

4.4 Flexibility and Simplicity of Our Unified Ant System


The approach presented and evaluated above was developed to deal with the
issues of flexibility and simplicity, which become more and more important as
additional criteria to compare heuristic algorithms as pointed out in the intro-
duction. The best known results presented above come from a number of re-
searchers using different techniques and (presumably) none of these approaches
was designed to be flexible in terms of applicability for different variants of the
VRP. Thus, a comparison of flexibility and simplicity is impossible.
However, we believe that our results enable us to make some statements for
our approach. First, our algorithm can be applied to (at least) the four types of
problems studied without modification. In fact, once the format of the input data
is fixed, our algorithm does not make a distinction between the types of problems.
Further, while the Ant System features a significant number of parameters, our
results suggest that the performance of the algorithm is quite robust over the
different types of problems and instances. Finally, the basic principle of the Ant
System meta-heuristic is quite simple.
Overall, our algorithm seems to be rather flexible, simple and robust. How-
ever, these careful statements show that a much more thorough analysis of these
issues is necessary.

5 Conclusions and Future Research


The results presented in this paper suggest that our Unified AS is capable of
finding high quality solutions for four important problem classes within the do-
main of vehicle routing. While some solution quality has to be sacrificed to gain
this flexibility, the overall performance of the algorithm is still very reasonable.
However, we clearly have to gather more experimental evidence to support
our thesis about the robust performance of our approach. First, we need to
extend the parameter ranges for α and δ as we found that these parameters
should take the maximum possible values in the domains we currently specified.
Second, we have to evaluate our Unified AS on an extended problem set that also
includes larger instances. Finally, we aim to extend our model further by giving
up the linehaul backhaul precedence relationship and by considering, among
others, soft time windows and multiple depots.
Apart from that the tradeoff between the two objectives suggests, that a true
multi-objective approach should be applied to further enhance the applicability
of the approach to real world problems. An existing approach by Jozefowiez et
al.[19] studies this possibility for the basic VRP. Their objectives are the total
distance travelled and a load balancing criterion that favors solutions with tours
that differ little in their total distance.

Acknowledgments. The authors wish to thank three anonymous referees for


valuable comments on the paper and Harald Andree for performing the numerical
tests. This work was supported by the Oesterreichische Nationalbank (OeNB)
under grant #8630.
310 M. Reimann, K. Doerner, and R.F. Hartl

References
1. Toth, P. and Vigo, D. (Eds.): The Vehicle Routing Problem. Siam Monographs on
Discrete Mathematics and Applications, Philadelphia (2002)
2. Cordeau, J. F., Gendreau, M., Laporte, G., Potvin, J. Y. and Semet, F.: A guide
to vehicle routing heuristics. Journal of the Operational Research Society 53 (5)
(2002) 512–522
3. http://www.top.sintef.no/
4. Reimann, M., Doerner, K., Hartl, R. F.: Insertion based Ants for Vehicle Routing
Problems with Backhauls and Time Windows. In: Dorigo, M. et al. (Eds.): Ant
Algorithms, Springer LNCS 2463, Berlin/Heidelberg (2002) 135–147
5. Bräysy, O. and Gendreau, M.: Metaheuristics for the Vehicle Routing Problem
with Time Windows. Sintef Technical Report STF42 A01025 (2001)
6. Toth, P. and Vigo, D.: VRP with Backhauls. In Toth, P. and Vigo, D. (Eds.):
The Vehicle Routing Problem. Siam Monographs on Discrete Mathematics and
Applications, Philadelphia (2002) 195–224
7. Gelinas, S., Desrochers, M., Desrosiers, J. and Solomon, M. M.: A new branching
strategy for time constrained routing problems with application to backhauling.
Annals of Operations Research. 61 (1995) 91–109
8. Duhamel, C., Potvin, J. Y. and Rousseau, J. M.: A Tabu Search Heuristic for
the Vehicle Routing Problem with Backhauls and Time Windows. Transportation
Science. 31 (1997) 49–59
9. Gendreau, M., Laporte, G., Musaraganyi, C. Taillard, E. D.: A tabu search heuris-
tic for the heterogeneous fleet vehicle routing problem. Computers and Operation
Research 26 (1999) 1153–1173
10. Cordeau, J. F., Gendreau, M., Laporte, G.: A Tabu Search Heuristic for Periodic
and Multi-Depot Vehicle Routing Problems. Networks 30 (1997) 105–119
11. Colorni, A., Dorigo, M. and Maniezzo, V.: Distributed Optimization by Ant
Colonies. In: Varela, F. and Bourgine, P. (Eds.): Proc. Europ. Conf. Artificial
Life. Elsevier, Amsterdam (1991) 134–142
12. Bonabeau, E., Dorigo, M. and Theraulaz, G.: Swarm Intelligence. Oxford Univer-
sity Press, New York (1999)
13. Gutjahr, W. J.: ACO algorithms with guaranteed convergence to the optimal so-
lution. Information Processing Letters. 82 (2002) 145–153
14. Solomon, M. M.: Algorithms for the Vehicle Routing and Scheduling Problems
with Time Window Constraints. Operations Research. 35 (1987) 254–265
15. Osman, I. H.: Metastrategy simulated annealing and tabu search algorithms for
the vehicle routing problem. Annals of Operations Research. 41 (1993) 421–451
16. Bullnheimer, B., Hartl, R. F. and Strauss, Ch.: A new rank based version of the ant
system: a computational study. Central European Journal of Operations Research
7(1) (1999) 25–38
17. Christofides, N., Mingozzi, A. and Toth, P.: The vehicle routing problem. In:
Christofides, N. et al. (Eds.): Combinatorial Optimization. Wiley, Chicester (1979)
18. Jacobs-Blecha, C., Goetschalckx, M.: The vehicle routing problem with backhauls:
properties and solution algorithms. Technical Report, School of Industrial and
Systems Engineering, Georgia Institute of Technology, Atlanta, Georgia (1992)
19. Jozefowiez, N., Semet, F. and Talbi, E.: Parallel and Hybrid Models for Multi-
objective Optimization: Application to the Vehicle Routing Problem. In Guervos,
J. M. et al. (Eds.): Parallel Problem Solving from Nature – PPSN VII, Springer
LNCS 2439, Berlin/Heidelberg (2002) 271–280

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy