Search results for: quick training algorithm

Authors: Hironori Karachi, Haruka Yamashita

Abstract:

Recently, the system of quick response (QR) code is getting popular. Many companies introduce new QR code payment services and the services are competing with each other to increase the number of users. For increasing the number of users, we should grasp the difference of feature of the demographic information, usage information, and value of users between services. In this study, we conduct an analysis of real-world data provided by Nomura Research Institute including the demographic data of users and information of users’ usages of two services; LINE Pay, and PayPay. For analyzing such data and interpret the feature of them, Nonnegative Matrix Factorization (NMF) is widely used; however, in case of the target data, there is a problem of the missing data. EM-algorithm NMF (EMNMF) to complete unknown values for understanding the feature of the given data presented by matrix shape. Moreover, for comparing the result of the NMF analysis of two matrices, there is Discriminant NMF (DNMF) shows the difference of users features between two matrices. In this study, we combine EMNMF and DNMF and also analyze the target data. As the interpretation, we show the difference of the features of users between LINE Pay and Paypay.

Keywords: Data science, non-negative matrix factorization, missing data, quality of services.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 390

Authors: Asha Rani, Christian Micheloni, Gian Luca Foresti

Abstract:

In this paper, a neural tree (NT) classifier having a simple perceptron at each node is considered. A new concept for making a balanced tree is applied in the learning algorithm of the tree. At each node, if the perceptron classification is not accurate and unbalanced, then it is replaced by a new perceptron. This separates the training set in such a way that almost the equal number of patterns fall into each of the classes. Moreover, each perceptron is trained only for the classes which are present at respective node and ignore other classes. Splitting nodes are employed into the neural tree architecture to divide the training set when the current perceptron node repeats the same classification of the parent node. A new error function based on the depth of the tree is introduced to reduce the computational time for the training of a perceptron. Experiments are performed to check the efficiency and encouraging results are obtained in terms of accuracy and computational costs.

Keywords: Neural Tree, Pattern Classification, Perceptron, Splitting Nodes.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1186

Authors: Branislav Vuksanović, Dragana Nikolić

Abstract:

This paper describes the implementation and testing of a multichannel active noise control system (ANCS) based on the filtered-inverse LMS (FILMS) algorithm. The FILMS algorithm is derived from the well-known filtered-x LMS (FXLMS) algorithm with the aim to improve the rate of convergence of the multichannel FXLMS algorithm and to reduce its computational load. Laboratory setup and techniques used to implement this system efficiently are described in this paper. Experiments performed in order to test the performance of the FILMS algorithm are discussed and the obtained results presented.

Keywords: Active noise control, adaptive filters, inverse filters, LMS algorithm, FILMS algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1587

Authors: B. Bakken, O. Boe

Abstract:

The Norwegian Military Academy (Army) has initiated a project with the main ambition to explore possible avenues to enhancing operational effectiveness through an increased use of simulation-based training and exercises. Within a cost/benefit framework, we discuss opportunities and limitations of vertical and horizontal integration of the existing tactical training system. Vertical integration implies expanding the existing training system to span the full range of training from tactical level (platoon, company) to command and staff level (battalion, brigade). Horizontal integration means including other domains than army tactics and staff procedures in the training, such as military ethics, foreign languages, leadership and decision making. We discuss each of the integration options with respect to purpose and content of training, "best practice" for organising and conducting simulation-based training, and suggest how to evaluate training procedures and measure learning outcomes. We conclude by giving guidelines towards further explorative work and possible implementation.

Keywords: Effectiveness, integration, simulation, training.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1394

Authors: Mahmud Hesain ALdwaik, Omar Hsiain Eldwaik

Abstract:

This paper deals with tracking and estimating time delay between two signals. The simulation of this algorithm accomplished by using Mathcad package is carried out. The algorithm we will present adaptively controls and tracking the delay, so as to minimize the mean square of this error. Thus the algorithm in this case has task not only of seeking the minimum point of error but also of tracking the change of position, leading to a significant improving of performance. The flowchart of the algorithm is presented as well as several tests of different cases are carried out.

Keywords: Tracking time delay, Algorithm simulation, Mathcad, MSE

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2002

Authors: Moth'd Belal. Al-Daoud

Abstract:

Clustering is a very well known technique in data mining. One of the most widely used clustering techniques is the k-means algorithm. Solutions obtained from this technique are dependent on the initialization of cluster centers. In this article we propose a new algorithm to initialize the clusters. The proposed algorithm is based on finding a set of medians extracted from a dimension with maximum variance. The algorithm has been applied to different data sets and good results are obtained.

Keywords: clustering, k-means, data mining.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2059

Authors: C. Rogerson, E. Scott

Abstract:

The purpose of this paper is to consider the introduction of online courses to replace the current classroom-based staff training. The current training is practical, and must be completed before access to the financial computer system is authorized. The long term objective is to measure the efficacy, effectiveness and efficiency of the training, and to establish whether a transfer of knowledge back to the workplace has occurred. This paper begins with an overview explaining the importance of staff training in an evolving, competitive business environment and defines the problem facing this particular organization. A summary of the literature review is followed by a brief discussion of the research methodology and objective. The implementation of the alpha version of the online course is then described. This paper may be of interest to those seeking insights into, or new theory regarding, practical interventions of online learning in the real world.

Keywords: Computer-based courses, e-learning, online training, workplace training.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1646

Authors: B.Bahmani Firouzi, T. Niknam, M. Nayeripour

Abstract:

Clustering is a very well known technique in data mining. One of the most widely used clustering techniques is the kmeans algorithm. Solutions obtained from this technique depend on the initialization of cluster centers and the final solution converges to local minima. In order to overcome K-means algorithm shortcomings, this paper proposes a hybrid evolutionary algorithm based on the combination of PSO, SA and K-means algorithms, called PSO-SA-K, which can find better cluster partition. The performance is evaluated through several benchmark data sets. The simulation results show that the proposed algorithm outperforms previous approaches, such as PSO, SA and K-means for partitional clustering problem.

Keywords: Data clustering, Hybrid evolutionary optimization algorithm, K-means algorithm, Simulated Annealing (SA), Particle Swarm Optimization (PSO).

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2232

Authors: Chaiwat Waree, Kalanyoo Petcharaporn

Abstract:

This research aims to develop and evaluate a training course to promote learning activities of 2nd year, Suan Sunandha Rajabhat University, faculty of education students using multiple intelligences theory. The process is divided into two phases: Phase 1 development of training course to promote learning activities consisting of principles, objectives of the course, structure, training duration, content, training materials, training activities, media training, monitoring, measurement and evaluation quality of the course. Phase 2 evaluation efficiency of training course was to use the improved curriculum with experimental group which is 2nd year, Suan Sunandha Rajabhat University, faculty of education students was drawn randomly 152 students. The experimental pattern was randomized Control Group Pre-Test Post-Test Design, Analysis Data by t-Test with the software SPFSS for Windows. Research has shown that: 1). the ability of teaching and learning according to the theory of multiple intelligences after training is higher than before training significantly in statistic at .01 level, 2). The satisfaction of students to the training courses was overall at the highest level.

Keywords: A training course, learning activities, multiple intelligences.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1361

Authors: B. Arun Kumar, T. Ravichandran

Abstract:

Workflow scheduling is an important part of cloud computing and based on different criteria it decides cost, execution time, and performances. A cloud workflow system is a platform service facilitating automation of distributed applications based on new cloud infrastructure. An aspect which differentiates cloud workflow system from others is market-oriented business model, an innovation which challenges conventional workflow scheduling strategies. Time and Cost optimization algorithm for scheduling Hybrid Clouds (TCHC) algorithm decides which resource should be chartered from public providers is combined with a new De-De algorithm considering that every instance of single and multiple workflows work without deadlocks. To offset this, two new concepts - De-De Dodging Algorithm and Priority Based Decisive Algorithm - combine with conventional deadlock avoidance issues by proposing one algorithm that maximizes active (not just allocated) resource use and reduces Makespan.

Keywords: Workflow Scheduling, cloud workflow, TCHC algorithm, De-De Dodging Algorithm, Priority Based Decisive Algorithm (PBD), Makespan.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2751

Authors: Xiangbin Zhu

Abstract:

EDF (Early Deadline First) algorithm is a very important scheduling algorithm for real- time systems . The EDF algorithm assigns priorities to each job according to their absolute deadlines and has good performance when the real-time system is not overloaded. When the real-time system is overloaded, many misdeadlines will be produced. But these misdeadlines are not uniformly distributed, which usually focus on some tasks. In this paper, we present an adaptive fuzzy control scheduling based on EDF algorithm. The improved algorithm can have a rectangular distribution of misdeadline ratios among all real-time tasks when the system is overloaded. To evaluate the effectiveness of the improved algorithm, we have done extensive simulation studies. The simulation results show that the new algorithm is superior to the old algorithm.

Keywords: Fuzzy control, real-time systems, EDF, misdeadline ratio.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1448

Authors: Safa Adi

Abstract:

This paper will consider the problem of sequential mining patterns embedded in a database by handling the time constraints as defined in the GSP algorithm (level wise algorithms). We will compare two previous approaches GTC and PSP, that resumes the general principles of GSP. Furthermore this paper will discuss PG-hybrid algorithm, that using PSP and GTC. The results show that PSP and GTC are more efficient than GSP. On the other hand, the GTC algorithm performs better than PSP. The PG-hybrid algorithm use PSP algorithm for the two first passes on the database, and GTC approach for the following scans. Experiments show that the hybrid approach is very efficient for short, frequent sequences.

Keywords: Database, GTC algorithm, PSP algorithm, sequential patterns, time constraints.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 648

Authors: K. M. Faraoun, A. Boukelif

Abstract:

In the present work, we propose a new technique to enhance the learning capabilities and reduce the computation intensity of a competitive learning multi-layered neural network using the K-means clustering algorithm. The proposed model use multi-layered network architecture with a back propagation learning mechanism. The K-means algorithm is first applied to the training dataset to reduce the amount of samples to be presented to the neural network, by automatically selecting an optimal set of samples. The obtained results demonstrate that the proposed technique performs exceptionally in terms of both accuracy and computation time when applied to the KDD99 dataset compared to a standard learning schema that use the full dataset.

Keywords: Neural networks, Intrusion detection, learningenhancement, K-means clustering

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3555

Authors: Isaias Ramirez, Jose Luis Silva, Carlos Gonzalez, Gustavo Candelaria, Jose Pepe Rasgado, Carlos Carrillo

Abstract:

This paper presents the experience in an eLearning training project that is being implemented for electrical planning engineers from the national Mexican utility Comision Federal de Electricidad (CFE) Distribution. This modality is implemented and will be used in the utility for training purposes to help personnel in their daily technical activities. One important advantage of this training project is that once it is implemented and applied, financial resources will be saved by CFE Distribution Company because online training will be used in all the country; the infrastructure for the eLearning training will be uploaded in computational servers installed in the National CFE Distribution Training Department, in Ciudad de Mexico, and can be used in workplaces of 16 Distribution Divisions and 150 Zones of CFE Distribution. In this way, workers will not need to travel to the National Training Department, saving enormous efforts, financial, and human resources.

Keywords: Moodle, eLearning, corporate training, electrical planning engineer.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 384

Authors: O. Starineca, I. Voronchuk

Abstract:

Human Resources (HR) selection and training have various implementation possibilities depending on an organization’s abilities and peculiarities. We propose to base HR selection and training decisions about on a competence-based approach. HR selection and training of employees are topical as there is room for improvement in this field; therefore, the aim of the research is to propose rational decision-making approaches for an organization HR selection and training choice. Our proposals are based on the training development and competence-based selection approaches created within previous researches i.e. Analytic-Hierarchy Process (AHP) and Linear Programming. Literature review on non-formal education, competence-based selection, AHP form our theoretical background. Some educational service providers in Latvia offer employees training, e.g. motivation, computer skills, accounting, law, ethics, stress management, etc. that are topical for Public Administration. Competence-based approach is a rational base for rational decision-making in both HR selection and considering HR training.

Keywords: Competence-based selection, human resource, training, decision-making.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1057

Authors: Yongzhong Wu, Ping Ji

Abstract:

This paper proposes a genetic algorithm based on a new replacement strategy to solve the quadratic assignment problems, which are NP-hard. The new replacement strategy aims to improve the performance of the genetic algorithm through well balancing the convergence of the searching process and the diversity of the population. In order to test the performance of the algorithm, the instances in QAPLIB, a quadratic assignment problem library, are tried and the results are compared with those reported in the literature. The performance of the genetic algorithm is promising. The significance is that this genetic algorithm is generic. It does not rely on problem-specific genetic operators, and may be easily applied to various types of combinatorial problems.

Keywords: Quadratic assignment problem, Genetic algorithm, Replacement strategy, QAPLIB.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2692

Authors: Luciano Nieddu, Giuseppe Manfredi, Salvatore D'Acunto, Katia La Regina

Abstract:

In this paper a non-parametric statistical pattern recognition algorithm for the problem of credit scoring will be presented. The proposed algorithm is based on a clustering k- means algorithm and allows for the determination of subclasses of homogenous elements in the data. The algorithm will be tested on two benchmark datasets and its performance compared with other well known pattern recognition algorithm for credit scoring.

Keywords: Constrained clustering, Credit scoring, Statistical pattern recognition, Supervised classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2005

Authors: Bi Kun, Gu Nai-jie, Tu Kun, Liu Xiao-hu, Liu Gang

Abstract:

Traditional parallel single string matching algorithms are always based on PRAM computation model. Those algorithms concentrate on the cost optimal design and the theoretical speed. Based on the distributed string matching algorithm proposed by CHEN, a practical distributed string matching algorithm architecture is proposed in this paper. And also an improved single string matching algorithm based on a variant Boyer-Moore algorithm is presented. We implement our algorithm on the above architecture and the experiments prove that it is really practical and efficient on distributed memory machine. Its computation complexity is O(n/p + m), where n is the length of the text, and m is the length of the pattern, and p is the number of the processors.

Keywords: Boyer-Moore algorithm, distributed algorithm, parallel string matching, string matching.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2147

Authors: Pavel K. Lopatin, Artyom S. Yegorov

Abstract:

The Algorithm 2 for a n-link manipulator movement amidst arbitrary unknown static obstacles for a case when a sensor system supplies information about local neighborhoods of different points in the configuration space is presented. The Algorithm 2 guarantees the reaching of a target position in a finite number of steps. The Algorithm 2 is reduced to a finite number of calls of a subroutine for planning a trajectory in the presence of known forbidden states. The polynomial approximation algorithm which is used as the subroutine is presented. The results of the Algorithm2 implementation are given.

Keywords: Manipulator, trajectory planning, unknown obstacles.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1200

Authors: M. Kalamani, S. Valarmathy, M. Krishnamoorthi

Abstract:

In this paper, Fuzzy C-Means clustering with Expectation Maximization-Gaussian Mixture Model based hybrid modeling algorithm is proposed for Continuous Tamil Speech Recognition. The speech sentences from various speakers are used for training and testing phase and objective measures are between the proposed and existing Continuous Speech Recognition algorithms. From the simulated results, it is observed that the proposed algorithm improves the recognition accuracy and F-measure up to 3% as compared to that of the existing algorithms for the speech signal from various speakers. In addition, it reduces the Word Error Rate, Error Rate and Error up to 4% as compared to that of the existing algorithms. In all aspects, the proposed hybrid modeling for Tamil speech recognition provides the significant improvements for speechto- text conversion in various applications.

Keywords: Speech Segmentation, Feature Extraction, Clustering, HMM, EM-GMM, CSR.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2096

Authors: Cemil Turan, Mohammad Shukri Salman

Abstract:

The convergence rate of the least-mean-square (LMS) algorithm deteriorates if the input signal to the filter is correlated. In a system identification problem, this convergence rate can be improved if the signal is white and/or if the system is sparse. We recently proposed a sparse transform domain LMS-type algorithm that uses a variable step-size for a sparse system identification. The proposed algorithm provided high performance even if the input signal is highly correlated. In this work, we investigate the performance of the proposed TD-LMS algorithm for a large number of filter tap which is also a critical issue for standard LMS algorithm. Additionally, the optimum value of the most important parameter is calculated for all experiments. Moreover, the convergence analysis of the proposed algorithm is provided. The performance of the proposed algorithm has been compared to different algorithms in a sparse system identification setting of different sparsity levels and different number of filter taps. Simulations have shown that the proposed algorithm has prominent performance compared to the other algorithms.

Keywords: Adaptive filtering, sparse system identification, VSSLMS algorithm, TD-LMS algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 945

Authors: Elham Serkani, Hossein Gharaee Garakani, Naser Mohammadzadeh, Elaheh Vaezpour

Abstract:

Intrusion detection systems (IDS) are the main components of network security. These systems analyze the network events for intrusion detection. The design of an IDS is through the training of normal traffic data or attack. The methods of machine learning are the best ways to design IDSs. In the method presented in this article, the pruning algorithm of C5.0 decision tree is being used to reduce the features of traffic data used and training IDS by the least square vector algorithm (LS-SVM). Then, the remaining features are arranged according to the predictor importance criterion. The least important features are eliminated in the order. The remaining features of this stage, which have created the highest level of accuracy in LS-SVM, are selected as the final features. The features obtained, compared to other similar articles which have examined the selected features in the least squared support vector machine model, are better in the accuracy, true positive rate, and false positive. The results are tested by the UNSW-NB15 dataset.

Keywords: Intrusion detection system, decision tree, support vector machine, feature selection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1181

Authors: Kyu Chul Lee, Sung Hyun Yoo, Choon Ki Ahn, Myo Taeg Lim

Abstract:

Recent experimental evidences have shown that because of a fast convergence and a nice accuracy, neural networks training via extended kalman filter (EKF) method is widely applied. However, as to an uncertainty of the system dynamics or modeling error, the performance of the method is unreliable. In order to overcome this problem in this paper, a new finite impulse response (FIR) filter based learning algorithm is proposed to train radial basis function neural networks (RBFN) for nonlinear function approximation. Compared to the EKF training method, the proposed FIR filter training method is more robust to those environmental conditions. Furthermore , the number of centers will be considered since it affects the performance of approximation.

Keywords: Extended kalmin filter (EKF), classification problem, radial basis function networks (RBFN), finite impulse response (FIR)filter.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2358

Authors: Shrawan Kumar, D. K. Lobiyal

Abstract:

In this paper, we propose improved versions of DVHop algorithm as QDV-Hop algorithm and UDV-Hop algorithm for better localization without the need for additional range measurement hardware. The proposed algorithm focuses on third step of DV-Hop, first error terms from estimated distances between unknown node and anchor nodes is separated and then minimized. In the QDV-Hop algorithm, quadratic programming is used to minimize the error to obtain better localization. However, quadratic programming requires a special optimization tool box that increases computational complexity. On the other hand, UDV-Hop algorithm achieves localization accuracy similar to that of QDV-Hop by solving unconstrained optimization problem that results in solving a system of linear equations without much increase in computational complexity. Simulation results show that the performance of our proposed schemes (QDV-Hop and UDV-Hop) is superior to DV-Hop and DV-Hop based algorithms in all considered scenarios.

Keywords: Wireless sensor networks, Error term, DV-Hop algorithm, Localization.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2198

Authors: A.Sankara Subramanian, G.Gurusamy, G.Selvakumar, P.Gnanasekar, A.Nagappan

Abstract:

This paper presents an algorithm based on the wavelet decomposition, for feature extraction from the ECG signal and recognition of three types of Ventricular Arrhythmias using neural networks. A set of Discrete Wavelet Transform (DWT) coefficients, which contain the maximum information about the arrhythmias, is selected from the wavelet decomposition. After that a novel clustering algorithm based on nature inspired algorithm (Ant Colony Optimization) is developed for classifying arrhythmia types. The algorithm is applied on the ECG registrations from the MIT-BIH arrhythmia and malignant ventricular arrhythmia databases. We applied Daubechies 4 wavelet in our algorithm. The wavelet decomposition enabled us to perform the task efficiently and produced reliable results.

Keywords: Daubechies 4 Wavelet, ECG, Nature inspired algorithm, Ventricular Arrhythmias, Wavelet Decomposition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2253

Authors: Naiqin Feng, Fang Wang, Yuhui Qiu

Abstract:

A new approach to promote the generalization ability of neural networks is presented. It is based on the point of view of fuzzy theory. This approach is implemented through shrinking or magnifying the input vector, thereby reducing the difference between training set and testing set. It is called “shrinking-magnifying approach" (SMA). At the same time, a new algorithm; α-algorithm is presented to find out the appropriate shrinking-magnifying-factor (SMF) α and obtain better generalization ability of neural networks. Quite a few simulation experiments serve to study the effect of SMA and α-algorithm. The experiment results are discussed in detail, and the function principle of SMA is analyzed in theory. The results of experiments and analyses show that the new approach is not only simpler and easier, but also is very effective to many neural networks and many classification problems. In our experiments, the proportions promoting the generalization ability of neural networks have even reached 90%.

Keywords: Fuzzy theory, generalization, misclassification rate, neural network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1492

Authors: Vandana Sharma, Parvinder S. Sandhu, Satwinder Singh, Baljit Saini

Abstract:

In field of Computer Science and Mathematics, sorting algorithm is an algorithm that puts elements of a list in a certain order i.e. ascending or descending. Sorting is perhaps the most widely studied problem in computer science and is frequently used as a benchmark of a system-s performance. This paper presented the comparative performance study of four sorting algorithms on different platform. For each machine, it is found that the algorithm depends upon the number of elements to be sorted. In addition, as expected, results show that the relative performance of the algorithms differed on the various machines. So, algorithm performance is dependent on data size and there exists impact of hardware also.

Keywords: Algorithm, Analysis, Complexity, Sorting.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2369

Authors: Rudi Kurniawan Arief

Abstract:

Manufacturing cost and setup time are the hot topics to improve in Metal Stamping industry because material and components price are always rising up while costumer requires to cut down the component price year by year. The Single Minute Exchange of Die (SMED) is one of many methods to reduce waste in stamping industry. The Japanese Quick Die Change (QDC) dies system is one of SMED systems that could reduce both of setup time and manufacturing cost. However, this system is rarely used in stamping industries. This paper will analyze how deep the QDC dies system could reduce setup time and the manufacturing cost. The research is conducted by direct observation, simulating and comparing of QDC dies system with conventional dies system. In this research, we found that the QDC dies system could save up to 35% of manufacturing cost and reduce 70% of setup times. This simulation proved that the QDC die system is effective for cost reduction but must be applied in several parallel production processes.

Keywords: Press die, metal stamping, quick die change, QDC system, single minute exchange die, manufacturing cost saving, SMED.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1140

Authors: Yahya H. Zweiri

Abstract:

The back-propagation algorithm calculates the weight changes of an artificial neural network, and a two-term algorithm with a dynamically optimal learning rate and a momentum factor is commonly used. Recently the addition of an extra term, called a proportional factor (PF), to the two-term BP algorithm was proposed. The third term increases the speed of the BP algorithm. However, the PF term also reduces the convergence of the BP algorithm, and optimization approaches for evaluating the learning parameters are required to facilitate the application of the three terms BP algorithm. This paper considers the optimization of the new back-propagation algorithm by using derivative information. A family of approaches exploiting the derivatives with respect to the learning rate, momentum factor and proportional factor is presented. These autonomously compute the derivatives in the weight space, by using information gathered from the forward and backward procedures. The three-term BP algorithm and the optimization approaches are evaluated using the benchmark XOR problem.

Keywords: Neural Networks, Backpropagation, Optimization.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1501

Authors: Neda Shahidi, Hadi Esmaeilzadeh, Marziye Abdollahi, Caro Lucas

Abstract:

In this paper, the problem of finding the optimal collision free path for a mobile robot, the path planning problem, is solved using an advanced evolutionary algorithm called memetic algorithm. What is new in this work is a novel representation of solutions for evolutionary algorithms that is efficient, simple and also compatible with memetic algorithm. The new representation makes it possible to solve the problem with a small population and in a few generations. It also makes the genetic operator simple and allows using an efficient local search operator within the evolutionary algorithm. The proposed algorithm is applied to two instances of path planning problem and the results are available.

Keywords: Path planning problem, Memetic Algorithm, Representation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1697