Genetic local search the graph partitioning problem under cardinality constraints

For the graph partitioning problem under cardinality constraints, a genetic local search method is developed. At each iteration of the method, there is a set of local optima of the problem. This set is used to search for new local optima with a smaller error. The local search problem with certain polynomially searchable neighborhoods is proved to be tight PLS-complete. It is shown that, in the worst case, number of local improvements can be exponentially large for any pivoting rule. Numerical experiments are performed in the special case of edge weights equal to unity, when local search is a polynomial-time procedure. The results of the experiments indicate that the method is highly efficient and can be applied to large-scale problems.     

Interactive model-based search with reactive resource allocation

We revisit the interactive model-based approach to global optimization proposed in Wang and Garcia (J Glob Optim 61(3):479–495, 2015) in which parallel threads independently execute a model-based search method and periodically interact through a simple acceptance-rejection rule aimed at preventing duplication of search efforts. In that paper it was assumed that each thread successfully identifies a locally optimal solution every time the acceptance-rejection rule is implemented. Under this stylized model of computational time, the rate of convergence to a globally optimal solution was shown to increase exponentially in the number of threads. In practice however, the computational time required to identify a locally optimal solution varies greatly. Therefore, when the acceptance-rejection rule is implemented, several threads may fail to identify a locally optimal solution. This situation calls for reallocation of computational resources in order to speed up the identification of local optima when one or more threads repeatedly fail to do so. In this paper we consider an implementation of the interactive model-based approach that accounts for real time, that is, it takes into account the possibility that several threads may fail to identify a locally optimal solution whenever the acceptance-rejection rule is implemented. We propose a modified acceptance-rejection rule that alternates between enforcing diverse search (in order to prevent duplication) and reallocation of computational effort (in order to speed up the identification of local optima). We show that the rate of convergence in real-time increases with the number of threads. This result formalizes the idea that in parallel computing, exploitation and exploration can be complementary provided relatively simple rules for interaction are implemented. We report the results from extensive numerical experiments which are illustrate the theoretical analysis of performance.     

PageRank centrality for performance prediction: the impact of the local optima network model

A local optima network (LON) compresses relevant features of fitness landscapes in a complex network, where nodes are local optima and edges represent transition probabilities between different basins of attraction. Previous work has found that the PageRank centrality of local optima can be used to predict the success rate and average fitness achieved by local search based metaheuristics. Results are available for LONs where edges describe either basin transition probabilities or escape edges. This paper studies the interplay between the type of LON edges and the ability of the PageRank centrality for the resulting LON to predict the performance of local search based metaheuristics. It finds that LONs are stochastic models of the search heuristic. Thus, to achieve an accurate prediction, the definition of the LON edges must properly reflect the type of diversification steps used in the metaheuristic. LONs with edges representing basin transition probabilities capture well the diversification mechanism of simulated annealing which sometimes also accepts worse solutions that allow the search process to pass between basins. In contrast, LONs with escape edges capture well the diversification step of iterated local search, which escapes from local optima by applying a larger perturbation step.     

Statistical analysis of local search landscapes

This paper discusses the application of some statistical estimation tools in trying to understand the nature of the combinatorial landscapes induced by local search methods. One interesting property of a landscape is the number of optima that are present. In this paper we show that it is possible to compute a confidence interval on the number of independent local searches needed to find all optima. By extension, this also expresses the confidence that the global optimum has been found. In many cases, this confidence may be too low to be acceptable, but it is also possible to estimate the number of optima that exist. Theoretical analysis and empirical studies are discussed, which show that it may be possible to obtain a fairly accurate picture of this property of a combinatorial landscape. The approach is illustrated by analysis of an instance of the flowshop scheduling problem.     

A Comparison of Global Optimization Methods for the Design of a High-speed Civil Transport

The conceptual design of aircraft often entails a large number of nonlinear constraints that result in a nonconvex feasible design space and multiple local optima. The design of the high-speed civil transport (HSCT) is used as an example of a highly complex conceptual design with 26 design variables and 68 constraints. This paper compares three global optimization techniques on the HSCT problem and two test problems containing thousands of local optima and noise: multistart local optimizations using either sequential quadratic programming (SQP) as implemented in the design optimization tools (DOT) program or Snyman's dynamic search method, and a modified form of Jones' DIRECT global optimization algorithm. SQP is a local optimizer, while Snyman's algorithm is capable of moving through shallow local minima. The modified DIRECT algorithm is a global search method based on Lipschitzian optimization that locates small promising regions of design space and then uses a local optimizer to converge to the optimum. DOT and the dynamic search algorithms proved to be superior for finding a single optimum masked by noise of trigonometric form. The modified DIRECT algorithm was found to be better for locating the global optimum of functions with many widely separated true local optima.     

Path relinking for the vehicle routing problem

This paper describes a tabu search heuristic with path relinking for the vehicle routing problem. Tabu search is a local search method that explores the solution space more thoroughly than other local search based methods by overcoming local optima. Path relinking is a method to integrate intensification and diversification in the search. It explores paths that connect previously found elite solutions. Computational results show that tabu search with path relinking is superior to pure tabu search on the vehicle routing problem.     

Iterated tabu search for the car sequencing problem

This paper introduces an iterated tabu search heuristic for the daily car sequencing problem in which a set of cars must be sequenced so as to satisfy requirements from the paint shop and the assembly line. The iterated tabu search heuristic combines a classical tabu search with perturbation operators that help escape from local optima. The resulting heuristic is flexible, easy to implement, and fast. It has produced very good results on a set of test instances provided by the French car manufacturer Renault.     

Application of graph-theoretic approaches to the random landscapes of the three-dimensional assignment problem

We study local optima of combinatorial optimization problems. We show that a local search algorithm can be represented as a digraph and apply recent results for spanning forests of a diagraph. We establish a correspondence between the number of local optima and the algebraic multiplicities of eigenvalues of digraph laplacians. We apply our finding to the three-dimensional assignment problem.     

Avoiding local optima in thep-hub location problem using tabu search and GRASP

In the discretep-hub location problem, various nodes interact with each other by sending and receiving given levels of traffic (such as telecommunications traffic, data transmissions, airline passengers, packages, etc.). It is necessary to choosep of the given nodes to act as hubs, which are fully interconnected; it is also necessary to connect each other node to one of these hubs so that traffic can be sent between any pair of nodes by using the hubs as switching points. The objective is to minimize the sum of the costs for sending traffic along the links connecting the various nodes. Like many combinatorial problems, thep-hub location problem has many local optima. Heuristics, such as exchange methods, can terminate once such a local optimum is encountered. In this paper, we describe new heuristics for thep-hub location problem, based on tabu search and on a greedy randomized adaptive search procedure (GRASP). These recently developed approaches to combinatorial optimization are capable of examining several local optima, so that, overall, superior solutions are found. Computational experience is reported in which both tabu search and GRASP found optimal hub locations (subject to the assumption that nodes must be assigned to the nearest hub) in over 90% of test problems. For problems for which such optima are not known, tabu search and GRASP generated new best-known solutions.     

A local search method for permutation flow shop scheduling

It is well known that a local search method, a widely used approach for solving the permutation flow shop scheduling problem, can easily be trapped at a local optimum. In this paper, we propose two escape-from-trap procedures to move away from local optima. Computational experiments carried out on a standard set of instances show that this heuristic algorithm generally outperforms an effective approximation algorithm.