Conference Paper

A Genetic Algorithm with Local Search for Solving Job Shop Problems.

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

This paper presents a genetic algorithm specially designed for job shop problems. The algorithm has a simple coding scheme and new crossover and mutation operators. A simple local search scheme is incorporated in the algorithm leading to a combined genetic algorithm(CGA). It is evaluated in three famous Muth and Thompson problems (i.e. MT6×6, MT10×10, MT20×5). The simulation study shows that this algorithm possesses high efficiency and is able to find out the optimal solutions for the job shop problems.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... The memetic algorithm, also known as the genetic local search, is a combination of the genetic algorithm and the local search-based approach in order to possess both the global search ability and search efficiency from these two kinds of approaches. Cai et al. (2000) proposed a memetic algorithm which embedded a hill climbing local search procedure to minimize the makespan in the job shop. The SA algorithm was adopted as the local search component to minimize the makespan in the job shop in Wang and Zheng (2001). ...
... The SA algorithm was adopted as the local search component to minimize the makespan in the job shop in Wang and Zheng (2001). In Goncalves et al. (2005), the hill climbing was also used but with a different neighborhood function from Cai et al. (2000). Other applications of the memetic algorithm on job sequencing problems can be found in Murata et al. (1998), Franca et al. (2001, and Sevaux and Dauzere-Peres (2003). ...
... In this scheme, best two individuals of the two selected parents and two offspring produced by mating survive to the next generation. This is called the 2/4 scheme, and was adopted in Cai et al. (2000). In our test, the premature convergence phenomenon still appeared, though later than in the second scheme. ...
Article
Full-text available
This paper addresses the job shop-scheduling problem with minimizing the number of tardy jobs as the objective. This problem is usually treated as a job-sequencing problem, and the permutation-based representation of solutions was commonly used in the existing search-based approaches. In this paper, the flaw of the permutation-based representation is discussed, and a rule-centric concept is proposed to deal with it. A memetic algorithm is then developed to realize the proposed idea by tailored genome encoding/decoding schemes and a local search procedure. Two benchmark approaches, a multi-start hill-climbing approach and a simulated annealing approach, are compared in the experiments. The results show that the proposed approach significantly outperforms the benchmarks.
... During the past decade, researches on meta-heuristic methods to solve the JSP have been widely studied, such as genetic algorithm [3],[4], simulated annealing [5], tabu search [6] and particle swarm optimization [7]. The majority of studies on JSP, however, are driven by production criteria, such as total flowtime, maximum complete time (makespan), maximum tardiness and number of tardy jobs, etc. ...
Article
Job-shop scheduling problem (JSP) is one of the most well-known machine scheduling problems and one of the strongly NP-hard combinatorial optimization problems. Cost optimization is an attractive and critical research and development area for both academic and industrial societies. This paper presents a cost driven model of the job-shop scheduling problem in which the solutions are driven by business inputs, such as the cost of the product transitions, revenue loss due to the machine idle time and earliness/tardiness penalty. And then, a new hybrid scatter search algorithm is proposed to solve the cost driven jobshop scheduling problem by introducing the simulated annealing (SA) into the improvement method of scatter search (SS). In order to illustrate the effectiveness of the hybrid method, some test problems are generated, and the performance of the proposed method is compared with other evolutionary algorithms such as genetic algorithm and simulated annealing. The experimental simulation tests show that the hybrid method is quite effective at solving the cost driven job-shop scheduling problem.
Article
Process planning and jobshop scheduling problems are both crucial functions in manufacturing. In reality, dynamic disruptions such as machine breakdown or rush order will affect the feasibility and optimality of the sequentially-generated process plans and machining schedules. With the approach of integrated process planning and scheduling (IPPS), the actual process plan and the schedule are determined dynamically in accordance with the order details and the status of the manufacturing system. In this paper, an object-coding genetic algorithm (OCGA) is proposed to resolve the IPPS problems in a jobshop type of flexible manufacturing systems. An effective object-coding representation and its corresponding genetic operations are suggested, where real objects like machining operations are directly used to represent genes. Based on the object-coding representation, customized methods are proposed to fulfill the genetic operations. An unusual selection and a replacement strategy are integrated systematically for the population evolution, aiming to achieve near-optimal solutions through gradually improving the overall quality of the population, instead of exploring neighborhoods of good individuals. Experiments show that the proposed genetic algorithm can generate outstanding outcomes for complex IPPS instances.
Article
Full-text available
In this paper, we propose a branch and bound method for solving the job-shop problem. It is based on one-machine scheduling problems and is made more efficient by several propositions which limit the search tree by using immediate selections. It solved for the first time the famous 10 × 10 job-shop problem proposed by Muth and Thompson in 1963.
Article
Full-text available
Job-shop scheduling is essentially an ordering problem. A new encoding scheme for a classic job-shop scheduling problem is presented, by which a schedule directly corresponds to an ordering string. For the new encoding, a simple but highly effective crossover operation is contrived, and the problem of infeasibility in genetic generation is naturally overcome. Within the framework of the newly designed genetic algorithm, the NP-hard classic job-shop scheduling problem can be efficiently solved with high quality. Moreover the optimal solutions of the two famous benchmarks, the Fisher and Thompson's 10 × 10 and 20 × 5 problems, are found.
Conference Paper
Full-text available
Genetic algorithms (GAs) have been designed as general purpose optimization methods. GAs can be uniquely characterized by their population-based search strategies and their operators: mutation, selection and crossover. In this paper, we propose a new crossover called multi-step crossover (MSX) which utilizes a neighborhood structure and a distance in the problem space. Given parents, MSX successively generates their descendents along the path connecting both of them. MSX was applied to the job-shop scheduling problem (JSSP) as a high-level crossover to work on the critical path. Preliminary experiments using JSSP benchmarks showed the promising performance of a GA with the proposed MSX
Article
Full-text available
This work compares six sequencing operators that have been developed for use with genetic algorithms. An improved version of the edge recombination operator is presented, the concepts of adjacency, order, and position are reviewed in the context of these operators, and results are compared for a 30 city "Blind" Traveling Salesman Problem and a real world warehouse/shipping scheduling application.
Article
This paper presents a mixed-variable evolutionary programming (MVEP) for solving mechanical design optimization problems which contain integer, discrete, zero-one and continuous variables. The MVEP provides an improvement in global search and convergence performance in a mixed-variable space. An approach to handle various kinds of variables and constraints is discussed. Two examples of mechanical design optimization are tested, which demonstrate that the proposed approach is superior to current methods for finding optimum solution, both in the quality of solution and convergence performance.
Article
This paper studies the problem of robust control design for a class of interconnected uncertain systems under sampled measurements. The class of system under consideration is described by a state space model containing unknown cone bounded nonlinear interaction and time-varying norm-bounded parameter uncertainties in both state and output equations. Our attention is focused on the design of linear dynamic output feedback controllers using sampled measurements. We address the problem of robust H∞ control in which both robust stability and a prescribed H∞ performance are required to be achieved irrespective of the uncertainties and nonlinearities. The H∞ performance measure involves both continuous-time and discrete-time signals. It has been shown that the above problems can be recast into H∞ syntheses for related N decoupled linear sampled-data systems without parameter uncertainties and unknown nonlinearities, which can be solved in terms of Riccati differential equations with finite discrete jumps. A numerical example is given to show the potential of the proposed technique.
Article
A family of algorithms is described for finding optimum schedules for job-shops. The algorithms are of a branch and bound type but have a complete schedule associated with each node of the search tree. Branching from nodes is based on important conflicts in the schedule. Some results are provided.
Article
In this paper we introduce a genetic algorithm whose peculiarities are the introduction of an encoding based on preference rules and an updating step which speeds up the evolutionary process. This method improves on the results gained previously with Genetic Algorithms and has shown itself to be competitive with other heuristics. The same algorithm has been applied to flow shop problems, revealing itself to be considerably more effective than Branch and Bound techniques.
Article
The sandwiching of YBa2Cu3O7−δ by two layers of continuous copper-coated carbon fibre Sn-Pb matrix composite was found to (i) increase the tensile and compressive strengths in the direction parallel to the fibres, such that both strengths increased with increasing fibre volume fraction, (ii) increase the resistance to thermal cycling between room temperature and 77 K and (iii) increase the stability in air. The sandwich composites were made by diffusion bonding at 110°C and 0.34 MPa, or by hot roll bonding at 240°C and 7.0 cm s−1.
Conference Paper
This paper presents a mixed-variable evolutionary programming (MVEP) for solving mechanical design optimization problems which contain integer, discrete, zero-one and continuous variables. The MVEP provides an improvement in global search and convergence performance in a mixed-variable space. An approach to handle various kinds of variables and constraints is discussed. Two examples of mechanical design optimization are tested, which demonstrate that the proposed approach is superior to current methods for finding optimum solution, both in the quality of solution and convergence performance
Conference Paper
The genetic search can be modeled as a controlled Markovian process, the transition of which depends on control parameters (probabilities of crossover and mutation). This paper proposes a stochastic gradient and develops a stochastic approximation algorithm to optimize control parameters of genetic algorithms (GAs). The optimal values of control parameters can be found from a recursive estimation of control parameters provided by the stochastic approximation algorithm. The algorithm performs in finding a stochastic gradient of a given performance index and adapting the control parameters in the direction of descent. Numerical results based on the classical multimodal functions are given to show the effectiveness of the proposed algorithm
Industrial scheduling
  • J F Muth
  • G L Thompson
  • J. F. Muth