Dalessandro Soares Vianna

A parallel evolutionary algorithm for the vehicle routing problem with heterogeneous fleet

Nowadays genetic algorithms stand as a trend to solve NPcomplete and NP-hard problems. In this paper, we present a new hybrid metaheuristic which uses Parallel Genetic Algorithms and Scatter Search coupled with a decomposition-into-petals procedure for solving a class of Vehicle Routing and Scheduling Problems. The parallel genetic algorithm presented is based on the island model and was run on a cluster of workstations. Its performance is evaluated for a heterogeneous fleet problem, which is considered a problem much harder to solve than the homogeneous vehicle routing problem.

An asynchronous parallel metaheuristic for the period vehicle routing problem

This paper presents an asynchronous parallel metaheuristic for the period vehicle routing problem (PVRP). The PVRP generalizes the classical vehicle routing problem by extending the planning period from a single day to M days. The algorithm proposed is based on concepts used in parallel genetic algorithms and local search heuristics. The algorithm employs the Island model in which the migration frequency must not be very high. The results of computational experiments carried out on problems taken from the literature indicate that the proposed approach outperforms existing heuristics in most cases.

A GRASP algorithm for the multi-criteria minimum spanning tree problem

Abstract This paper proposes a GRASP (Greedy Randomized Adaptive Search Procedure) algorithm for the multi-criteria minimum spanning tree problem, which is NP-hard. In this problem a vector of costs is defined for each edge of the graph and the problem is to find all Pareto optimal or efficient spanning trees (solutions). The algorithm is based on the optimization of different weighted utility functions. In each iteration, a weight vector is defined and a solution is built using a greedy randomized constructive procedure. The found solution is submitted to a local search trying to improve the value of the weighted utility function. We use a drop-and-add neighborhood where the spanning trees are represented by Prufer numbers. In order to find a variety of efficient solutions, we use different weight vectors, which are distributed uniformly on the Pareto frontier. The proposed algorithm is tested on problems with r=2 and 3 criteria. For non-complete graphs with n=10, 20 and 30 nodes, the performance of the algorithm is tested against a complete enumeration. For complete graphs with n=20, 30 and 50 nodes the performance of the algorithm is tested using two types of weighted utility functions. The algorithm is also compared with the multi-criteria version of the Kruskal’s algorithm, which generates supported efficient solutions.

An Evolutionary Hybrid Metaheuristic for Solving the Vehicle Routing Problem with Heterogeneous Fleet

Nowadays genetic algorithms stand as a trend to solve NP-complete and NP-hard problems. In this paper, we present a new hybrid metaheuristic which combines Genetic Algorithms and Scatter Search coupled with a decomposition-into-petals procedure for solving a class of Vehicle Routing and Scheduling Problems. Its performance is evaluated for a heterogeneous fleet model, which is considered a problem much harder to solve than the homogeneous vehicle routing problem.

A Parallel Hybrid Evolutionary Metaheuristic for the Period Vehicle Routing Problem

This paper presents a Parallel Hybrid Evolutionary Metaheuristic for the Period Vehicle Routing Problem (PVRP). The PRVP generalizes the classical Vehicle Routing Problem by extending the planning period from a single day to M days. The algorithm proposed is based on concepts used in Parallel Genetic Algorithms and Local Search Heuristics. The algorithm employs the island model in which the migration frequency must not be very high. The results of computational experiments carried out on problems taken from the literature indicate that the proposed approach outperforms existing heuristics in most cases.

A simulation model to evaluate sugarcane supply systems

We present, in this paper, a simulation model to evaluate the sugarcane supply system to mills. The model addressed, on the whole, harvest operations (cutting and shipping), transportation and unloading at the mill (also considering the reception system of sugarcane within the mill). The model could adequately assess the relation of the freight, the lead time, the fleet of trucks and discount (opposite of agio), apart from the cost of cutting and shipping, related to the amount to be paid by the sugarcane load furnished to the mill.

Local search-based heuristics for the multiobjective multidimensional knapsack problem

In real optimization problems it is generally desirable to optimize more than one performance criterion (or objective) at the same time. The goal of the multiobjective combinatorial optimization (MOCO) is to optimize simultaneously r > 1 objectives. As in the single-objective case, the use of heuristic/metaheuristic techniques seems to be the most promising approach to MOCO problems because of their efficiency, generality and relative simplicity of implementation. In this work, we develop algorithms based on Greedy Randomized Adaptive Search Procedure (GRASP) and Iterated Local Search (ILS) metaheuristics for the multiobjective knapsack problem. Computational experiments on benchmark instances show that the proposed algorithms are very robust and outperform other heuristics in terms of solution quality and running times.

Hybrid GRASP heuristics for the phylogeny problem combining path-relinking and genetic algorithm as an intensification strategy

A phylogeny is a tree that relates taxonomic units based on their similarity over a set of characteristics. The phylogeny problem under the parsimony criterion consists in finding a phylogeny with a minimum number of evolutionary steps. We propose hybrid heuristic methods – based on GRASP, path-relinking and genetic algorithm methodologies – to build a phylogeny while minimizing parsimony. Computational experiments using benchmark conditions are reported, and the results obtained by the proposed hybrid heuristics are compared with the solutions obtained by a traditional GRASP (without hybridization) heuristic and with previously reported solutions in the literature. The experimental results illustrate that the proposed heuristics are efficient in terms of solution quality and time-to-target-value.

GRASP heuristic for p-median problem applied to the location of concentrators

Várias situações práticas reais, tais como localizações de depósitos, hospitais e dispositivos de telecomu-nicações (concentradores, torres de celulares, etc), podem ser vistas como um problema de p-medianas. Este trabalho apresenta uma proposta para a solução do problema de p-medianas baseado no backbone da rede de computadores que será instalado no Instituto Federal Fluminense (IFF). Esse tipo de ocorrência é conhecida na literatura como problema de localização de concentradores. Para resolver a questão citada foi proposta uma heurística GRASP. Testes computacionais realizados, mostram que a heurística elaborada neste trabalho atingiu resultados satisfatórios.

Heurística GRASP para o problema de p-medianas aplicado à localização de concentradores

Several real practical situations, such as location of depots, hospitals and telecommunications devices (hubs, cellular towers, etc.), can be seen as a p-median problem. This paper presents a proposal for solving the p-median problem based on the backbone network of computers that will be installed at the Federal Fluminense Institute (IFF). This type of problem is known in literature as a problem of locating concentrators. To solve the problem cited was proposed a GRASP heuristic. Computational tests performed show that the heuristic developed in this work has reached satisfactory results.