www.tjprc.org editor@tjprc.org
International Journal of Applied Engineering
Research and Development (IJAERD)
ISSN(P): 2250-1584 ISSN(E): 2278-9383
Vol. 4, Issue 5, Oct 2014, 33-42
© TJPRC Pvt. Ltd.
A FUZZY-LOGIC TRAINED NEURAL NETWORK APPROACH FOR MACHINE
SCHEDULE OPTIMIZATION
D. KAVITHA
1
, K. G. DHARANI
2
& M. PRIYADARSINI
3
Assistant Professors, Department of E.C.E., M.V.J College of Engineering, Bangalore, India
ABSTRACT
The competitive manufacturing global scenario today needs multi-job, multi-machine and multi criterion shop
floor machine scheduling concept for higher production. Achieving key performance measures is important for the
successful operation of the manufacturing systems, which are complex and dynamic in nature. To have a closer look at
real-world problems, neural network concept becomes its solutions. The problem discussed in this research involves N jobs
and M processors. The objective is to find a schedule which assigns N jobs to M processors in such a way that the
performance measures are optimized. Dispatching rules are usually applied dynamically to schedule jobs in manufacturing
systems.
Workflow balancing on a shop floor helps to remove bottlenecks present in the manufacturing system .Workflow
refers to the total time during which the work centers are busy. Earlier researchers have not specified the method for jobs to
be executed in parallel in order to balance the workflow to each machine. In parallel machine scheduling there are m
machines to which n jobs are to be assigned based on different priority strategies. The procedure is based on the idea of
workload balancing and on balancing the workload among machines. Different priority strategies are followed for the
selection of jobs. 8 different strategies are considered, namely random (RANDOM), shortest processing time (SPT),
longest processing time (LPT), most work remaining (MWKR), least work remaining (LWKR), first come first serve
(FCFS) and last come first serve (LCFS) for the selection of jobs for workflow balancing. The relative percentage of
imbalance (RPI) is adopted among the parallel machines to evaluate the performance of these strategies in a standard
manufacturing environment using neural network.
This paper discusses the application of neural network to solve a identical parallel machine scheduling. The
developed neural network model predicts the optimal solutions for any set of problem instances. This paper uses the Back
propagation and Delta rule based approach for training the neural network. The neural network approach is quite effective
and efficient for selecting the best strategies and their optimal sequence for a given scheduling problem.
KEYWORDS: Neural Network, Fuzzy Logic, Machine Scheduling, Dispatching Rules
INTRODUCTION
Most combinatorial optimization problems such as scheduling problems are NP-hard in the strong sense, implying
that it is highly unlikely that a polynomial algorithm can be designed to solve this problems. Since neural network are most
effective in solving classification and prediction problems, the emerging technology of neural network may be well suited
to solve this problem.
In this paper, we propose to use neural network systems to solve an identical parallel machine scheduling problems
and to predict the optimal solution for any set of problem instances. Neural network have been applied extensively to a