Stochastic neural networks for solving job-shop scheduling. I - Problem representation

Yoon Pin Simon Foo, Yoshiyasu Takefuji

Research output: Chapter in Book/Report/Conference proceedingConference contribution

79 Citations (Scopus)

Abstract

An application of neural networks is presented for solving job-shop scheduling, an NP-complete optimization problem with constaint satisfaction. In particular, the authors introduce a neural computation architecture based on a stochastic Hopfield neural-network model. First, the job-shop problem is mapped into a two-dimensional matrix representation of neurons similar to those for solving the traveling salesman problem (TSP). Constant positive and negative current biases are applied to specific neurons as excitations and inhibitions, respectively, to enforce the operation precedence relationships. At the convergence of neural network, the solution to the job-shop problem is represented by a set of cost function trees encoded in the matrix of stable states. Each node in the set of trees represents a job, and each link represents the interdependency between jobs. A computation circuit computes the total completion times (costs) of all jobs, and the cost difference is added to the energy function of the stochastic neural network. Using a simulated annealing algorithm, the temperature of the system is slowly decreased according to an annealing schedule until the energy of the system is at a local or global minimum. By choosing an appropriate annealing schedule, near-optimal and optimal solutions to job-shop problems can be found.

Original languageEnglish
Title of host publicationIEEE Int Conf on Neural Networks
PublisherPubl by IEEE
Pages275-282
Number of pages8
Publication statusPublished - 1988
Externally publishedYes

Fingerprint

Neural networks
Neurons
Annealing
Hopfield neural networks
Traveling salesman problem
Bias currents
Simulated annealing
Networks (circuits)
Job shop scheduling
Temperature

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Foo, Y. P. S., & Takefuji, Y. (1988). Stochastic neural networks for solving job-shop scheduling. I - Problem representation. In IEEE Int Conf on Neural Networks (pp. 275-282). Publ by IEEE.

Stochastic neural networks for solving job-shop scheduling. I - Problem representation. / Foo, Yoon Pin Simon; Takefuji, Yoshiyasu.

IEEE Int Conf on Neural Networks. Publ by IEEE, 1988. p. 275-282.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Foo, YPS & Takefuji, Y 1988, Stochastic neural networks for solving job-shop scheduling. I - Problem representation. in IEEE Int Conf on Neural Networks. Publ by IEEE, pp. 275-282.
Foo YPS, Takefuji Y. Stochastic neural networks for solving job-shop scheduling. I - Problem representation. In IEEE Int Conf on Neural Networks. Publ by IEEE. 1988. p. 275-282
Foo, Yoon Pin Simon ; Takefuji, Yoshiyasu. / Stochastic neural networks for solving job-shop scheduling. I - Problem representation. IEEE Int Conf on Neural Networks. Publ by IEEE, 1988. pp. 275-282
@inproceedings{9d46a02a6901465d8adf777c90d2eee2,
title = "Stochastic neural networks for solving job-shop scheduling. I - Problem representation",
abstract = "An application of neural networks is presented for solving job-shop scheduling, an NP-complete optimization problem with constaint satisfaction. In particular, the authors introduce a neural computation architecture based on a stochastic Hopfield neural-network model. First, the job-shop problem is mapped into a two-dimensional matrix representation of neurons similar to those for solving the traveling salesman problem (TSP). Constant positive and negative current biases are applied to specific neurons as excitations and inhibitions, respectively, to enforce the operation precedence relationships. At the convergence of neural network, the solution to the job-shop problem is represented by a set of cost function trees encoded in the matrix of stable states. Each node in the set of trees represents a job, and each link represents the interdependency between jobs. A computation circuit computes the total completion times (costs) of all jobs, and the cost difference is added to the energy function of the stochastic neural network. Using a simulated annealing algorithm, the temperature of the system is slowly decreased according to an annealing schedule until the energy of the system is at a local or global minimum. By choosing an appropriate annealing schedule, near-optimal and optimal solutions to job-shop problems can be found.",
author = "Foo, {Yoon Pin Simon} and Yoshiyasu Takefuji",
year = "1988",
language = "English",
pages = "275--282",
booktitle = "IEEE Int Conf on Neural Networks",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - Stochastic neural networks for solving job-shop scheduling. I - Problem representation

AU - Foo, Yoon Pin Simon

AU - Takefuji, Yoshiyasu

PY - 1988

Y1 - 1988

N2 - An application of neural networks is presented for solving job-shop scheduling, an NP-complete optimization problem with constaint satisfaction. In particular, the authors introduce a neural computation architecture based on a stochastic Hopfield neural-network model. First, the job-shop problem is mapped into a two-dimensional matrix representation of neurons similar to those for solving the traveling salesman problem (TSP). Constant positive and negative current biases are applied to specific neurons as excitations and inhibitions, respectively, to enforce the operation precedence relationships. At the convergence of neural network, the solution to the job-shop problem is represented by a set of cost function trees encoded in the matrix of stable states. Each node in the set of trees represents a job, and each link represents the interdependency between jobs. A computation circuit computes the total completion times (costs) of all jobs, and the cost difference is added to the energy function of the stochastic neural network. Using a simulated annealing algorithm, the temperature of the system is slowly decreased according to an annealing schedule until the energy of the system is at a local or global minimum. By choosing an appropriate annealing schedule, near-optimal and optimal solutions to job-shop problems can be found.

AB - An application of neural networks is presented for solving job-shop scheduling, an NP-complete optimization problem with constaint satisfaction. In particular, the authors introduce a neural computation architecture based on a stochastic Hopfield neural-network model. First, the job-shop problem is mapped into a two-dimensional matrix representation of neurons similar to those for solving the traveling salesman problem (TSP). Constant positive and negative current biases are applied to specific neurons as excitations and inhibitions, respectively, to enforce the operation precedence relationships. At the convergence of neural network, the solution to the job-shop problem is represented by a set of cost function trees encoded in the matrix of stable states. Each node in the set of trees represents a job, and each link represents the interdependency between jobs. A computation circuit computes the total completion times (costs) of all jobs, and the cost difference is added to the energy function of the stochastic neural network. Using a simulated annealing algorithm, the temperature of the system is slowly decreased according to an annealing schedule until the energy of the system is at a local or global minimum. By choosing an appropriate annealing schedule, near-optimal and optimal solutions to job-shop problems can be found.

UR - http://www.scopus.com/inward/record.url?scp=0024122608&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024122608&partnerID=8YFLogxK

M3 - Conference contribution

SP - 275

EP - 282

BT - IEEE Int Conf on Neural Networks

PB - Publ by IEEE

ER -