Arnaud Zinflou

GISMOO: A new hybrid genetic/immune strategy for multiple-objective optimization

In this paper, we propose a new Pareto generic algorithm, called GISMOO, which hybridizes genetic algorithm and artificial immune systems. GISMOO algorithm is generic in the sense that it can be used to solve both combinatorial and continuous optimization problems. The proposed approach offers an original iterative process in two phases: a Genetic Phase and an Immune Phase. The Immune Phase is used to identify and to emphasize the solutions located in less crowded regions found during the iterative process of the algorithm. Simulation results on difficult test problems, both in combinatorial and continuous optimization, show that the proposed approach, in most problems, is able to obtain better results than state of the art algorithms.

Pareto memetic algorithm for multiple objective optimization with an industrial application

Abstract Multiple objective combinatorial optimization problems are difficult to solve and often, exact algorithms are unable to produce optimal solutions. The development of multiple objective heuristics was inspired by the need to quickly produce acceptable solutions. In this paper, we present a new multiple objective Pareto memetic algorithm called PMSMO. The PMSMO algorithm incorporates an enhanced fine-grained fitness assignment, a double level archiving process and a local search procedure to improve performance. The performance of PMSMO is benchmarked against state-of-the-art algorithms using 0–1 multi-dimensional multiple objective knapsack problem from the literature and an industrial scheduling problem from the aluminum industry.

Design of an Efficient Genetic Algorithm to Solve the Industrial Car Sequencing Problem

In many industrial sectors, decision makers are faced with large and complex problems that are often multi-objective. Many of these problems may be expressed as a combinatorial optimization problem in which we define one or more objective functions that we are trying to optimize. Thus, the car sequencing problem in an assembly line is a well known combinatorial optimization problem that cars manufacturers face. This problem involves scheduling cars along an assembly line composed of three consecutive shops: body welding and construction, painting and assembly. In the literature, this problem is most often treated as a single objective problem and only the capacity constraints of the assembly shop are considered (Dincbas et al., 1988). In this workshop, each car is characterized by a set of different options and the workstations where each option is installed are designed to handle a certain percentage of cars requiring the same options. To smooth the workload at the critical assembly workstations, cars requiring high work content must be dispersed throughout the production sequence. Industrial car sequencing formulation subdivides the capacity constraints into two categories, that are the capacity constraints linked to the highpriority options and the capacity constraints linked to the low-priority options. However, the reality of industrial production does not only take into account the assembly shop requirements. The industrial formulation proposed by French automobile manufacturer Renault, in the context of the ROADEF 2005 Challenge, also takes into account the paint shop requirements. In this workshop, the minimization of the amount of solvent used to purge the painting nozzles for colour changeovers, or when a known maximum number of vehicle bodies of the same colour have been painted, is an important objective to consider. Indeed, long sequences of cars of the same colour tend to render visual quality controls inaccurate. To ensure this quality control, the number of cars of the same colour must not exceed an upper limit. The industrial car sequencing problem (ICSP) is thus a multi-objective problem in nature, with three conflicting objectives to minimize. In the assembly shop, one tries to minimize the number of violations of capacity constraints related to high-priority options (HPO) and to low-priority options (LPO). In the paint shop, one tries to minimize the number of colour changes (COLOUR). In the 2005 ROADEF Challenge, the Renault automobile manufacturer proposes to tackle the problem by treating the three objectives lexicographically. O pe n A cc es s D at ab as e w w w .ite ch on lin e. co m

Tackling the Industrial Car Sequencing Problem Using GISMOO Algorithm

In many industrial sectors, managers are confronted with problems of an ever-growing complexity. The problem could be bus route optimization for a public transporter, production cost minimization, decision-making support, electronic circuit performance enhancement, or computer system process scheduling. In many cases the situation can be expressed as a combinatorial optimization problem. Solving an optimization problem consists of determining the best solution(s) validating a set of user-defined constraints and goals. To determine if one solution is better than another, the problem must include at least one performance evaluation metric that allows solutions to be compared. The best (or optimal) solution, is thus the one with the best evaluation, with respect to the defined goal. When only one goal is specified (e.g. total distance minimization), the optimal solution is clearly defined (the one with the smallest distance). However, in many situations there are several contradictory goals that have to be satisfied simultaneously. In fact, real-world optimization problems rarely have a single goal. This is the case for the Industrial Car Sequencing Problem (ICSP) on an automobile assembly line. The ICSP consists of determining the order in which automobiles should be produced, taking into account the various model options, assembly line constraints, and production environment goals. In this context, the optimal solution is not a single point, but rather a set of compromise solutions called the Pareto-optimal front. We can thus define two main goals in multi-objective optimization: (i) Find a set of compromise solutions whose evaluation is as close as possible to the Pareto-optimal front; and (ii) Find a set of compromise solutions as diverse as possible. Attaining these two goals in realistic time is an important challenge for any multi-objective algorithm. However, in the literature, the ICSP, despite its multi-objective character, has been treated as a problem with a single goal or with several goals lexicographically ordered (Benoist, 2008; Briant et al., 2008; Cordeau et al., 2008; Estellon et al., 2008; Ribeiro et al., 2008). To our knowledge, the only references that treat the ICSP from a purely multi-objective viewpoint are those of Zinflou et al. (2009) and of de Oliveira dos Reis ( 2007); the latter only examines small instances (fewer than 60 automobiles). Most of the algorithms proposed recently for multi-objective problems are Evolutionary Algorithms (EA) (Deb, 2000; Knowles & Corne, 2000a; Knowles & Corne, 2000b; Zitzler et al., 2001). This is so, doubtlessly because EA’s can traverse a large search space to generate

Solving the industrial car sequencing problem in a Pareto sense

Until now, the industrial car sequencing problem, as defined during the ROADEF 2005 Challenge, has been tackled by organizing objectives in a hierarchy. In this paper, we suggest tackling this problem in a Pareto sense for the first time. We thus suggest the adaptation of the PMSMO, an elitist evolutionary algorithm which distinguishes itself through a fitness calculation that takes into account the history of solutions found so as to diversify the compromise solutions along the Pareto frontier. A comparison of the performance is carried out using a well-known published algorithm, the NSGAII, and proves an advantage for the PMSMO. As well, we aim to demonstrate the relevance of handling applied problems such as the car sequencing problem using a multi-objective approach.

Crossover operators for the car sequencing problem

The car sequencing problem involves scheduling cars along an assembly line while satisfying as many assembly line requirements as possible. The car sequencing problem is NP-hard and is applied in industry as shown by the 2005 ROADEF Challenge. In this paper, we introduce three new crossover operators for solving this problem efficiently using a genetic algorithm. A computational experiment compares these three operators on standard car sequencing benchmark problems. The best operator is then compared with state of the art approach for this problem. The results show that the proposed operator consistently produces competitive solutions for most instances.

An hybrid algorithm for the industrial car sequecing problem

In most research papers, the industrial car sequencing problem, as defined during the ROADEF 2005 Challenge, has been tackled by organizing objectives in a hierarchy. However from a decision-making viewpoint it would be interesting to tackle this problem in a Pareto sense. Indeed, tackling the problem in Pareto sense can offer greater latitude to a manager by presenting him several alternative solutions. In this paper, we suggest to adapt the GISMOO algorithm to solve the industrial car sequencing problem. A comparison of the performance is carried out using well-known published algorithms and proves an advantage for GISMOO. As well, we aim to demonstrate the relevance of handling applied problems such as the industrial car sequencing problem using a Pareto multi-objective approach.

An ontology-based reasoning approach for electric power utilities

Because of the semantic conflicts, the exchange of information between heterogeneous applications remains a complex task. One way to address this problem is to use ontologies for the identification and association of semantically corresponding information concepts. In the electric power industry, the IEC/CIM represents the most complete and widely accepted ontology. We attempt to show through three concrete examples how the CIM can reap advantages from a formal representation of knowledge in order to support complex processes. We present a semantic approach for finding ringlets in the distribution network, for checking specific data inconsistencies and finally for identifying CIM topological nodes. We conclude by stating that the combination of CIM and RDF has the main advantage of offering valuable flexibility in processing complex tasks.

A hybrid genetic/immune strategy to tackle the multiobjective quadratic assignment problem

The Genetic Immune Strategy for Multiple Objective Optimization (GISMOO) is a hybrid algorithm for solving multiobjective problems. The performance of this approach has been assessed using a classical combinatorial multiobjective optimization benchmark: the multiobjective 0/1 knapsack problem (MOKP) [1] and two-dimensional unconstrained multiobjective problems (ZDT) [2]. This paper shows that the GISMOO algorithm can also efficiently solve the multiobjective quadratic assignment problem (mQAP). A performance comparison carried out using well-known published algorithms and shows GISMOO to advantage.

Solving multiple-objective optimization problems using GISMOO algorithm

In this paper, we proposed a new Pareto generic algorithm which hybridizes genetic algorithm and artificial immune systems. Numerical experiments were made using a classical benchmark in multiple-objective optimization (MOKP). Results show that our approach is able to obtain better performance than two state of the art approaches: NSGAII and PMSMO.