Noureddine Bouhmala

Current Approaches for Solving Over-Constrained Problems

We summarize existing approaches to model and solve overconstrained problems. These problems are usually formulated as combinatorial optimization problems, and different specific and generic formalisms are discussed, including the special case of multi-objective optimization. Regarding solving methods, both systematic and local search approaches are considered. Finally we review a number of case studies on overconstrained problems taken from the specialized literature.

Crowd Models for Emergency Evacuation: A Review Targeting Human-Centered Sensing

Emergency evacuation of crowds is a fascinating phenomenon that has attracted researchers from various fields. Better understanding of this class of crowd behavior opens up for improving evacuation policies and smarter design of buildings, increasing safety. Recently, a new class of disruptive technology has appeared: Human-centered sensing which allows crowd behavior to be monitored in real-time, and provides the basis for real-time crowd control. The question then becomes: to what degree can previous crowd models incorporate this development, and what areas need further research? In this paper, we provide a survey that describes some widely used crowd models and discuss their advantages and shortages from the angle of human-centered sensing. Our review reveals important research opportunities that may contribute to an improved and more robust emergency management.

A multilevel memetic algorithm for large sat-encoded problems

Many researchers have focused on the satisfiability problem and on many of its variants due to its applicability in many areas of artificial intelligence. This NP-complete problem refers to the task of finding a satisfying assignment that makes a Boolean expression evaluate to True. In this work, we introduce a memetic algorithm that makes use of the multilevel paradigm. The multilevel paradigm refers to the process of dividing large and difficult problems into smaller ones, which are hopefully much easier to solve, and then work backward toward the solution of the original problem, using a solution from a previous level as a starting solution at the next level. Results comparing the memetic with and without the multilevel paradigm are presented using problem instances drawn from real industrial hardware designs.

Combining finite learning automata with GSAT for the satisfiability problem

A large number of problems that occur in knowledge-representation, learning, very large scale integration technology (VLSI-design), and other areas of artificial intelligence, are essentially satisfiability problems. The satisfiability problem refers to the task of finding a satisfying assignment that makes a Boolean expression evaluate to True. The growing need for more efficient and scalable algorithms has led to the development of a large number of SAT solvers. This paper reports the first approach that combines finite learning automata with the greedy satisfiability algorithm (GSAT). In brief, we introduce a new algorithm that integrates finite learning automata and traditional GSAT used with random walk. Furthermore, we present a detailed comparative analysis of the new algorithms performance, using a benchmark set containing randomized and structured problems from various domains.

A multilevel learning automata for MAX-SAT

The need to solve optimization problems of unprecedented sizes is becoming a challenging task. Utilizing classical methods of Operations Research often fail due to the exponentially growing computational effort. It is commonly accepted that these methods might be heavily penalized by the NP-Hard nature of the problems and consequently will then be unable to solve large size instances of a problem. Lacking the theoretical basis and guided by intuition, meta-heuristics are the techniques commonly used even if they are unable to guarantee an optimal solution. Meta-heuristics search techniques tend to spend most of the time exploring a restricted area of the search space preventing the search to visit more promising areas thereby leading to solutions of poor quality. In this paper, a multilevel learning automata and a multilevel WalkSAT algorithm are proposed as a paradigm for finding a tactical interplay between diversification and intensification for large scale optimization problems. The multilevel paradigm involves recursive coarsening to create a hierarchy of increasingly smaller and coarser versions of the original problem. This phase is repeated until the size of the smallest problem falls below a specified reduction threshold. A solution for the problem at the coarsest level is generated, and then successively projected back onto each of the intermediate levels in reverse order. The solution at each child level is improved before moving to the parent level. Benchmark including large MAX-SAT test cases are used to compare the effectiveness of the multilevel approach against its single counter part.

Towards Multilevel Ant Colony Optimisation for the Euclidean Symmetric Traveling Salesman Problem

Ant Colony Optimization ACO metaheuristic is one of the best known examples of swarm intelligence systems in which researchers study the foraging behavior of bees, ants and other social insects in order to solve combinatorial optimization problems. In this paper, a multilevel Ant Colony Optimization MLV-ACO for solving the traveling salesman problem is proposed, by using a multilevel process operating in a coarse-to-fine strategy. This strategy involves recursive coarsening to create a hierarchy of increasingly smaller and coarser versions of the original problem. The heart of the approach is grouping the variables that are part of the problem into clusters, which is repeated until the size of the smallest cluster falls below a specified reduction threshold. Subsequently, a solution for the problem at the coarsest level is generated, and then successively projected back onto each of the intermediate levels in reverse order. The solution at each level is improved using the ACO metaheuristic before moving to the parent level. The proposed solution has been tested both in circular and randomized environments, and outperform single level counterparts.

A Variable Neighborhood Walksat-Based Algorithm for MAX-SAT Problems

The simplicity of the maximum satisfiability problem (MAX-SAT) combined with its applicability in many areas of artificial intelligence and computing science made it one of the fundamental optimization problems. This NP-complete problem refers to the task of finding a variable assignment that satisfies the maximum number of clauses (or the sum of weights of satisfied clauses) in a Boolean formula. The Walksat algorithm is considered to be the main skeleton underlying almost all local search algorithms for MAX-SAT. Most local search algorithms including Walksat rely on the 1-flip neighborhood structure. This paper introduces a variable neighborhood walksat-based algorithm. The neighborhood structure can be combined easily using any local search algorithm. Its effectiveness is compared with existing algorithms using 1-flip neighborhood structure and solvers such as CCLS and Optimax from the eighth MAX-SAT evaluation.

A parallel variant of simulated annealing for optimizing mesh partitions on workstations

Abstract In this paper we introduce a new parallel variant of simulated annealing for optimizing mesh partitions. We view this work as a first step towards a parallel multilevel algorithm for the load balancing problem. The serial execution time of the algorithm and the quality of the partitions it generates is compared to the well known METIS for three unstructured meshes. Finally, some preliminary experimental results regarding the performance of the proposed parallel variant of simulated annealing on a cluster of workstations are presented.

Stochastic Learning for SAT-Encoded Graph Coloring Problems

The graph coloring problem (GCP) is a widely studied combinatorial optimization problem due to its numerous applications in many areas, including time tabling, frequency assignment, and register allocation. The need for more efficient algorithms has led to the development of several GC solvers. In this paper, the authors introduce a team of Finite Learning Automata, combined with the random walk algorithm, using Boolean satisfiability encoding for the GCP. The authors present an experimental analysis of the new algorithms performance compared to the random walk technique, using a benchmark set containing SAT-encoding graph coloring test sets.

A Generalized Variable Neighborhood Search For Combinatorial Optimization Problems

Abstract The VNS is a simple meta-heuristic that systematically changes the size and type of neighborhood during the search process in order to escape from local optima. In this paper, a generalized variable neighborhood search is proposed for combinatorial optimization problems. The set of constructed neighborhoods satisfies the property that each small neighborhood is a subset of a larger one. Most of the work published earlier on VNS starts from the first neighborhood and moves on to higher neighborhoods without controlling and adapting the ordering of neighborhood structures. The order in which the neighborhood structures have been selected in this paper during the search process offers a better mechanism for performing diversification and intensification. A set of industrial benchmark problem instances is used to test the effectiveness of the new variant of VNS using the maximum satisfying problem as a test case.