Artificial Neural Networks for Automatic Dispatching Rule Selection

1995 ◽  
Author(s):  
Huajie Liu ◽  
Jian (John) Dong
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Real Time ◽  
Job Shop ◽  
Scheduling Algorithm ◽  
Scheduling Problems ◽  
Dispatching Rule ◽  
Planning And Scheduling ◽  
Real Time Scheduling ◽  
Artificial Neural

Abstract To schedule a job shop, the first task is to select an appropriate scheduling algorithm or rules. Because of the complexity of scheduling problems, no general algorithmic approach sufficient for solving all scheduling problems has been developed yet. Most scheduling systems offer multiple alternative algorithms to various circumstances, and experienced human schedulers are needed to select the appropriate dispatching rules in these systems. To automate scheduling process, this paper discusses the use of neural networks for scheduling decision making. For a given workshop, various situations are simulated to identify the best dispatching rule. The inputs and outputs of the simulation model are used to train an artificial neural network. The trained neural network then will be used to automatically selecting an appropriate dispatching rule for a given situation. Research results have shown great potential in using a neural network to replace human schedulers in selecting an appropriate approach for real time scheduling. This research is a part of an ongoing project of developing a real-time planning and scheduling system.

scholarly journals Efficient Training of Artificial Neural Networks for Autonomous Navigation

1991 ◽  
Vol 3 (1) ◽  
pp. 88-97 ◽  
Author(s):  
Dean A. Pomerleau
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Real Time ◽  
Autonomous Navigation ◽  
Training Techniques ◽  
Human Driver ◽  
Backpropagation Network ◽  
Artificial Neural ◽  
Network Project

The ALVINN (Autonomous Land Vehicle In a Neural Network) project addresses the problem of training artificial neural networks in real time to perform difficult perception tasks. ALVINN is a backpropagation network designed to drive the CMU Navlab, a modified Chevy van. This paper describes the training techniques that allow ALVINN to learn in under 5 minutes to autonomously control the Navlab by watching the reactions of a human driver. Using these techniques, ALVINN has been trained to drive in a variety of circumstances including single-lane paved and unpaved roads, and multilane lined and unlined roads, at speeds of up to 20 miles per hour.

QCSP Based Schedulability Test of Real-Time System

2013 ◽  
Vol 313-314 ◽  
pp. 1214-1217 ◽  
Author(s):  
Ju Yang Zhang ◽  
Yi Xiang Chen
Keyword(s):  
Real Time ◽  
Scheduling Algorithm ◽  
Constraint Satisfaction Problems ◽  
Test Method ◽  
Scheduling Problems ◽  
Real Time System ◽  
Scheduling System ◽  
Real Time Scheduling ◽  
Time Scheduling ◽  
Schedulability Test

The problem of schedulability test of real-time task in scheduling system has been addressed. In such as real-time scheduling problems with uncertain interrupt operation, Quantified Constraint Satisfaction Problems (QCSP) were introduced to model and solve CSP involving uncertainty or uncontrollability on the value taken by some variables. In this paper, we propose to a novel approach to schedulability test based on the QCSP model. Based on the QCSP model of scheduler and interrupter in the real-time scheduling system, we transform the schedulability test into the satisfability problem. Finally, we design the algorithm QCSP-SchTest for judging the satisfability of a schedule S(α). This leads to a new schedulability test method without considering the specific scheduling algorithm or strategy.

scholarly journals Real-Time Structural Damage Assessment Using Artificial Neural Networks and Antiresonant Frequencies

2014 ◽  
Vol 2014 ◽  
pp. 1-14 ◽  
Author(s):  
V. Meruane ◽  
J. Mahu
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Real Time ◽  
Damage Assessment ◽  
Structural Damage ◽  
Model Updating ◽  
Assessment Process ◽  
Artificial Neural ◽  
Assessment Problems

The main problem in damage assessment is the determination of how to ascertain the presence, location, and severity of structural damage given the structure's dynamic characteristics. The most successful applications of vibration-based damage assessment are model updating methods based on global optimization algorithms. However, these algorithms run quite slowly, and the damage assessment process is achieved via a costly and time-consuming inverse process, which presents an obstacle for real-time health monitoring applications. Artificial neural networks (ANN) have recently been introduced as an alternative to model updating methods. Once a neural network has been properly trained, it can potentially detect, locate, and quantify structural damage in a short period of time and can therefore be applied for real-time damage assessment. The primary contribution of this research is the development of a real-time damage assessment algorithm using ANN and antiresonant frequencies. Antiresonant frequencies can be identified more easily and more accurately than mode shapes, and they provide the same information. This research addresses the setup of the neural network parameters and provides guidelines for the selection of these parameters in similar damage assessment problems. Two experimental cases validate this approach: an 8-DOF mass-spring system and a beam with multiple damage scenarios.

scholarly journals Real-Time Quality Control and Infilling of Precipitation Data Using Neural Networks

10.29007/t5k7 ◽  
2018 ◽  
Author(s):  
Mohammadreza Moslemi ◽  
Darko Joksimovic
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Quality Control ◽  
Real Time ◽  
Network Model ◽  
Neural Network Model ◽  
Superior Performance ◽  
Precipitation Data ◽  
Artificial Neural

Due to advancements in instrumentation and communication technologies, monitoring of water infrastructure is experiencing a significant growth worldwide and water managers are increasingly deploying monitoring equipment for decision-making purposes. Hydrological events and relevant datasets including rainfall data are of a complex nature and are potentially susceptible to errors from various sources. Hence, it is essential to develop efficient methods for the quality control of the acquired data. The present work introduces an artificial neural network-based approach for real-time quality control and infilling of rain gauge data. Available rainfall measurements from neighboring rain gauges are employed to train and develop the neural network model. Trained artificial neural network model was able to validate up to about 97% of the data using 95% confidence intervals. This finding suggests that artificial neural networks can be successfully implemented for erroneous data identification/correction and reconstruction of missing data points. Given its short processing time and reportedly superior performance to traditional quality control strategies, neural network methodology can be deployed as an efficient tool for the processing and control of large sets of timeseries with complex natures including precipitation data.

Sign in / Sign up

Export Citation Format

Share Document