## A Novel Variable Neighborhood Particle Swarm Optimization for Multi-objective Flexible Job-shop Scheduling Problems

### Cached

### Download Links

Citations: | 3 - 0 self |

### BibTeX

@MISC{Liu_anovel,

author = {Hongbo Liu and Ajith Abraham and Crina Grosan and Ningning Li},

title = {A Novel Variable Neighborhood Particle Swarm Optimization for Multi-objective Flexible Job-shop Scheduling Problems},

year = {}

}

### OpenURL

### Abstract

This paper introduces a hybrid metaheuristic, the Variable Neighborhood Particle Swarm Optimization (VNPSO), consisting of a combination of the Variable Neighborhood Search (VNS) and Particle Swarm Optimization (PSO). The proposed VNPSO method is used for solving the multiobjective Flexible Job-shop Scheduling Problems (FJSP). The details of implementation for the multi-objective FJSP and the corresponding computational experiments are reported. The results indicate that the proposed algorithm is an efficient approach for the multi-objective FJSP, especially for large scale problems. 1

### Citations

517 | The particle swarm-explosion, stability, and convergence in a multidimensional complex space
- Clerc, Kennedy
- 2002
(Show Context)
Citation Context ...g all the particles towards it. Some previous studies has been shown that the trajectories of the particlessoscillate in different sinusoidal waves and converge quickly in the “gbest model” algorithm =-=[22, 23]-=-. During the iteration, the particle is attracted towards the location of the best fitness achieved so far by the particle itself and by the location of the best fitness achieved so far across the who... |

166 | Benchmarks for basic scheduling problems
- Taillard
- 1993
(Show Context)
Citation Context ... proposed algorithm, three representative instances based on practical data have been selected. Three problem instances ((J8, O27, M8), (J10, O30, M10) and (J15, O56, M10) are taken from Kacem et al. =-=[10, 11, 30]-=-. In our experiments, the algorithms used for comparison were MSPSO (Multi-start PSO) and VNPSO (Variable Neighborhood PSO). In VNPSO, ρ and vc were set to 2 and 1e-7 before 15,000 iterations, while t... |

147 | Recent Approaches to Global Optimization Problems Through Particle
- Parsopoulos, Vrahatis
- 2002
(Show Context)
Citation Context ...number of machines per operation; w1 and w2 are non-negative weights, and w1 + w2 = 1. These weights can be either fixed or adapt dynamically during the optimization. The dynamic weighted aggregation =-=[19]-=- was used in the paper. Alternatively, the weights can be changed gradually according to the Eqs. (2) and (3). The variation for different values of w1 and w2 (R = 200) are illustrated in Fig. 1. w1(t... |

91 |
Particle Swarm Optimization
- Clerc
- 2006
(Show Context)
Citation Context ...rticle Swarm Optimization (PSO) incorporates swarming behaviors observed in flocks of birds, schools of fish, or swarms of bees, and even human social behavior, from which the intelligence is emerged =-=[12, 13]-=-. It has become the new focus of research recently [14, 15, 16, 17]. However, its performance deteriorates as the dimensionality of the search space increases, especially for the multi-objective FJSP ... |

49 |
A tutorial survey of job-shop scheduling problems using genetic algorithms: Part II. Hybrid genetic search strategies
- Cheng, Gen, et al.
(Show Context)
Citation Context ... minimize the makespan. Liouane et al. [5] proposed a hybrid algorithm based on ant systems and local search optimization for FJSP. More researchers attempt to solve the FJSP using genetic algorithms =-=[6, 7]-=-. Ho et al. [8] proposed an architecture for learning and evolving of Flexible Job-Shop schedules to improve the computational time and quality of schedules. In many real-world FJSP, it is often neces... |

39 |
Variable neighbourhood search: Principles and applications
- Hansen, Mladenović
(Show Context)
Citation Context ...current one is found. These steps are repeated until a given termination condition is met. Hansen and Mladenović suggest various VNS strategies for a wide range of combinatorial optimization problems =-=[26, 27]-=-. The major difference between our approach and the VNS approach proposed by Hansen and Mladenović lies in the change of neighbourhoods by perturbing the particles effectively. The metaheuristic metho... |

38 | A study of particle swarm optimization particle trajectories”, Information Sciences
- Bergh, Engelbrecht
- 2006
(Show Context)
Citation Context ...rally as a population of vectors whose trajectories oscillate around a region which is defined by each individual’s previous best success and the success of some other particle. Bergh and Engelbrecht =-=[21]-=- overviewed the theoretical studies, and extend these studies to investigate particle trajectories for general swarms to include the influence of the inertia term. They also provided a formal proof th... |

36 |
2002), Approach by localization and multiobjective evolutionary optimization for flexible job shop scheduling problems
- Kacem, Hammadi, et al.
(Show Context)
Citation Context ...cessary to optimize several criteria [9]. Minimization of makespan, lateness, tardiness, flow time, machine idle time, and such others are unusual the important criteria in the problems. Kacem et al. =-=[10, 11]-=- studied on modeling genetic algorithms for FJSP. Because of the intractable nature of the problem and its importance in both fields of production management and combinatorial optimization, it is desi... |

32 |
Routing and scheduling in a flexible job shop by tabu search
- Brandimarte
- 1993
(Show Context)
Citation Context ...lilli and Gambardella [1] proposed some neighborhood functions for FJSP. Ong et al. [2] applied the clonal selection principle of the human immune system to solve FJSP with recirculation. Brandimarte =-=[3]-=-, Saidi-Mehrabad and Fattahi [4] presented a tabu search algorithm that solves the flexible job shop scheduling problem to minimize the makespan. Liouane et al. [5] proposed a hybrid algorithm based o... |

28 |
Pareto-optimality approach for flexible job-shop scheduling problems: hybridization of genetic Algorithms with fuzzy logic
- Kacem, Hammadi, et al.
- 2002
(Show Context)
Citation Context ...cessary to optimize several criteria [9]. Minimization of makespan, lateness, tardiness, flow time, machine idle time, and such others are unusual the important criteria in the problems. Kacem et al. =-=[10, 11]-=- studied on modeling genetic algorithms for FJSP. Because of the intractable nature of the problem and its importance in both fields of production management and combinatorial optimization, it is desi... |

22 |
An effective hybrid optimization approach for multi-objective flexible job-shop scheduling problems
- Xia, Wu
- 2005
(Show Context)
Citation Context ...mber of generations is increased. Once the algorithm slows down, it is difficult to achieve better scheduling solutions. By hybridizing particle swarm optimization and simulated annealing, Xia and Wu =-=[18]-=- developed anshybrid approach for the multi-objective flexible job-shop scheduling problem (MFJSP). We introduce a novel hybrid meta-heuristic, the Variable Neighborhood Particle Swarm Optimization (V... |

20 | T.: Scheduling jobs on computational grids using fuzzy particle swarm algorithm
- Abraham, Liu, et al.
- 2006
(Show Context)
Citation Context ...iors observed in flocks of birds, schools of fish, or swarms of bees, and even human social behavior, from which the intelligence is emerged [12, 13]. It has become the new focus of research recently =-=[14, 15, 16, 17]-=-. However, its performance deteriorates as the dimensionality of the search space increases, especially for the multi-objective FJSP involving large scale. PSO often demonstrates faster convergence sp... |

18 | Swarm intelligence: Foundations, perspectives and applications, swarm intelligent systems
- Abraham, Guo, et al.
- 2006
(Show Context)
Citation Context ...iors observed in flocks of birds, schools of fish, or swarms of bees, and even human social behavior, from which the intelligence is emerged [12, 13]. It has become the new focus of research recently =-=[14, 15, 16, 17]-=-. However, its performance deteriorates as the dimensionality of the search space increases, especially for the multi-objective FJSP involving large scale. PSO often demonstrates faster convergence sp... |

13 |
Chaotic Dynamic Characteristics in Swarm Intelligence
- Liu, Abraham, et al.
(Show Context)
Citation Context ...tness achieved so far across the whole swarm. The “gbest model” has faster convergence. But very often for multi-modal problems involving high dimensions it tends to suffer from premature convergence =-=[24, 25]-=-. Algorithm 1 Particle Swarm Algorithm 01. Begin 02. Parameter settings and initialize swarm 03. Evaluation 04. Locate leader 05. g = 1 06. While (the end criterion is not met) do 07. For each particl... |

11 |
E¤ective Neighborhood Functions for the Flexible Job Shop Problem
- Mastrolilli, Gambardella
- 1998
(Show Context)
Citation Context ...tion with traditional optimization approaches owing to the high computational complexity. In the literature, different approaches have been proposed to solve this problem. Mastrolilli and Gambardella =-=[1]-=- proposed some neighborhood functions for FJSP. Ong et al. [2] applied the clonal selection principle of the human immune system to solve FJSP with recirculation. Brandimarte [3], Saidi-Mehrabad and F... |

11 | An Hybrid Fuzzy Variable Neighborhood Particle Swarm Optimization Algorithm for Solving Quadratic Assignment Problems
- Liu, Abraham
- 2007
(Show Context)
Citation Context ...rence between our approach and the VNS approach proposed by Hansen and Mladenović lies in the change of neighbourhoods by perturbing the particles effectively. The metaheuristic method we proposed in =-=[28]-=-, the VNPSO, was originally inspired by VNS. In PSO, if a particle’s velocity decreases to a threshold vc, a new velocity is assigned using Eq.(12): vij(t) =wˆv + c1r1(x # ij (t − 1) − xij(t − 1)) ˆv ... |

9 |
Nicoara Search Optimization Using Hybrid Particle Sub-Swarms and Evolutionary Algorithms
- Grosan, Abraham, et al.
- 2005
(Show Context)
Citation Context ...iors observed in flocks of birds, schools of fish, or swarms of bees, and even human social behavior, from which the intelligence is emerged [12, 13]. It has become the new focus of research recently =-=[14, 15, 16, 17]-=-. However, its performance deteriorates as the dimensionality of the search space increases, especially for the multi-objective FJSP involving large scale. PSO often demonstrates faster convergence sp... |

8 |
An Effective Architecture for Learning and Evolving Flexible Job-Shop Schedules
- Ho, Tay, et al.
- 2007
(Show Context)
Citation Context ...span. Liouane et al. [5] proposed a hybrid algorithm based on ant systems and local search optimization for FJSP. More researchers attempt to solve the FJSP using genetic algorithms [6, 7]. Ho et al. =-=[8]-=- proposed an architecture for learning and evolving of Flexible Job-Shop schedules to improve the computational time and quality of schedules. In many real-world FJSP, it is often necessary to optimiz... |

7 |
Solving the job shop scheduling problem using genetic algorithm
- Gen, Y, et al.
- 1994
(Show Context)
Citation Context ...e constraints). But there are no precedence constraints among operations of different jobs. If the job and the operations are ordered beforehand, there is not enough flexible between jobs. Gen et al. =-=[29]-=- proposed a perfect method: they name all operations for a job with the same symbol (for example, the corresponding job index) and then interpret them according to the order of occurrences in the sequ... |

6 |
Applying the Clonal Selection Principle to Find Flexible
- Ong, Tay, et al.
- 2005
(Show Context)
Citation Context ...h computational complexity. In the literature, different approaches have been proposed to solve this problem. Mastrolilli and Gambardella [1] proposed some neighborhood functions for FJSP. Ong et al. =-=[2]-=- applied the clonal selection principle of the human immune system to solve FJSP with recirculation. Brandimarte [3], Saidi-Mehrabad and Fattahi [4] presented a tabu search algorithm that solves the f... |

6 |
T.: Particle swarm optimisation from lbest to gbest, Applied Soft Computing Technologies: The Challenge of Complexity
- Liu, Li, et al.
(Show Context)
Citation Context ...tness achieved so far across the whole swarm. The “gbest model” has faster convergence. But very often for multi-modal problems involving high dimensions it tends to suffer from premature convergence =-=[24, 25]-=-. Algorithm 1 Particle Swarm Algorithm 01. Begin 02. Parameter settings and initialize swarm 03. Evaluation 04. Locate leader 05. g = 1 06. While (the end criterion is not met) do 07. For each particl... |

4 |
Fattahi P.: Flexible job shop scheduling with tabu search algorithms
- Saidi-mehrabad
- 2007
(Show Context)
Citation Context ...ed some neighborhood functions for FJSP. Ong et al. [2] applied the clonal selection principle of the human immune system to solve FJSP with recirculation. Brandimarte [3], Saidi-Mehrabad and Fattahi =-=[4]-=- presented a tabu search algorithm that solves the flexible job shop scheduling problem to minimize the makespan. Liouane et al. [5] proposed a hybrid algorithm based on ant systems and local search o... |

4 | Ant systems & local search optimization for flexible job shop scheduling production,” Int
- Liouane, Saad, et al.
- 2007
(Show Context)
Citation Context ...JSP with recirculation. Brandimarte [3], Saidi-Mehrabad and Fattahi [4] presented a tabu search algorithm that solves the flexible job shop scheduling problem to minimize the makespan. Liouane et al. =-=[5]-=- proposed a hybrid algorithm based on ant systems and local search optimization for FJSP. More researchers attempt to solve the FJSP using genetic algorithms [6, 7]. Ho et al. [8] proposed an architec... |

2 |
B.: Scheduling of flow-shop, job-shop, and combined scheduling problems using MOEAs with fixed and variable length chromosomes. Evolutionary Schedulling
- Kleeman, Lamont
- 2006
(Show Context)
Citation Context ...ecture for learning and evolving of Flexible Job-Shop schedules to improve the computational time and quality of schedules. In many real-world FJSP, it is often necessary to optimize several criteria =-=[9]-=-. Minimization of makespan, lateness, tardiness, flow time, machine idle time, and such others are unusual the important criteria in the problems. Kacem et al. [10, 11] studied on modeling genetic alg... |

2 |
Variable neighborhood search”, in Handbook of Applied Optimization
- Hansen, Mladenovic
(Show Context)
Citation Context ...current one is found. These steps are repeated until a given termination condition is met. Hansen and Mladenović suggest various VNS strategies for a wide range of combinatorial optimization problems =-=[26, 27]-=-. The major difference between our approach and the VNS approach proposed by Hansen and Mladenović lies in the change of neighbourhoods by perturbing the particles effectively. The metaheuristic metho... |

1 |
Multiobjective 0/1 Knapsack Problem using Adaptive ɛ -Dominance
- Grosan
(Show Context)
Citation Context |

1 |
An Evolutionary Approach for Solving the Multi-Objective Job-Shop Scheduling Problem
- Ripon, Tsang, et al.
(Show Context)
Citation Context ...· · , p − 1. (5) � ri,j,k ≤ 1, k ∈ M; t ≥ 0. (6) j∈A(t) Ci,j,k ≥ 0, j = 1, · · · , n. (7) The complexity of FJSP increases with the number of constraints imposed and the size of search space employed =-=[20]-=-. Except for some highly restricted special cases, very simple special cases of FJSP are already strongly NP-hard. For the standard FJSP, the size of search space is (n!) m , and for this reason, it i... |

1 |
Cristian “The Particle Swarm Optimization Algorithm: Convergence Analysis and Parameter Selection
- I
(Show Context)
Citation Context ...g all the particles towards it. Some previous studies has been shown that the trajectories of the particlessoscillate in different sinusoidal waves and converge quickly in the “gbest model” algorithm =-=[22, 23]-=-. During the iteration, the particle is attracted towards the location of the best fitness achieved so far by the particle itself and by the location of the best fitness achieved so far across the who... |