• Journal of the Korean Operations Research and Management Science Society
• /
• 제29권4호
• /
• pp.175-187
• /
• 2004

We consider the simultaneous selection of part routes for multiple part types in Flexible Manufacturing Systems (FMSs). Using an optimization framework we investigate two alternative route assignment policies. The one, called routing mix policy in the literature, specifies the optimal proportion of each part type to be produced along its alternative routes, assuming that the proportions can be kept during execution. The other one, which we propose and call pallet allocation policy, partitions the pallets assigned to each part type among the routes. The optimization framework used is a nonlinear programming superimposed on a closed queueing network model of an FMS which produces multiple part types with distinct repeated visits to certain workstations. The objective is to maximize the weighted throughput. Our study shows that the simultaneous use of multiple routes leads to reduced bottleneck utilization, improved workload balance, and a significant increase in the FMS's weighted throughput, without any additional capital investments. Based on numerical work, we also conjecture that pallet allocation policy is more robust than routing mix policy, operationally easier to implement, and may yield higher revenues.

• Journal of Korea Technology Innovation Society
• /
• 제4권3호
• /
• pp.342-354
• /
• 2001

The appearance of on-demand business model of multimedia contents is one of the biggest changes in the broadband Internet which aims at the high-speed Internet users. Currently the interest in the webcasting service, a new converged service of the broadcasting and the telecommunications, has been increased. However the quality of service is not guaranteed due to the bottleneck problem in a middle-mile as the fundamental limit of the Internet, and has not been improved greatly due to the small-scaled operators of the webcasting services. This study suggests a new business model of the multi-channel satellite webcasting service based on the satellite multicasting and content delivery network, which guarantees the quality of service and channelizes the contents by the type or by the special purpose.

• Phonetics and Speech Sciences
• /
• 제7권4호
• /
• pp.9-16
• /
• 2015

In this paper, several experiments are performed according to deep neural network (DNN) based features for the performance comparison of speaker verification (SV) systems. To this end, input features for a DNN, such as mel-frequency cepstral coefficient (MFCC), linear-frequency cepstral coefficient (LFCC), and perceptual linear prediction (PLP), are first compared in a view of the SV performance. After that, the effect of a DNN training method and a structure of hidden layers of DNNs on the SV performance is investigated depending on the type of features. The performance of an SV system is then evaluated on the basis of I-vector or probabilistic linear discriminant analysis (PLDA) scoring method. It is shown from SV experiments that a tandem feature of DNN bottleneck feature and MFCC feature gives the best performance when DNNs are configured using a rectangular type of hidden layers and trained with a supervised training method.

• Management Science and Financial Engineering
• /
• 제20권2호
• /
• pp.7-12
• /
• 2014

Content Delivery Network (CDN) has evolved to overcome a network bottleneck and improve user perceived Quality of Service (QoS). A CDN replicates contents from the origin server to replica servers to reduce the overload of the origin server. CDN providers would try to achieve an acceptable performance at the least cost including the storage space or processing power. In this paper, we introduce a new optimization model for the CDN design problem considering the user perceived QoS and single path (non-bifurcated) routing constraints and analyze the computational complexity for some special cases.

• Proceedings of the IEEK Conference
• /
• 대한전자공학회 2007년도 하계종합학술대회 논문집
• /
• pp.23-24
• /
• 2007

Ethernet passive optical network (EPON) is a technology to solve the access network bottleneck and to realize fiber to the home (FTTH). In EPON, the dynamic bandwidth allocation (DBA) is important because all users share common line in upstream. In this paper, we propose new DBA named as modified EPON dynamic scheduling algorithm (MEDSA) and simulate new DBA using Matlab.

• IE interfaces
• /
• 제13권4호
• /
• pp.709-716
• /
• 2000

In this paper, We analyze a capacity dimensioning and performance of the access line to the X.25 packet network when the AO/DI internet service is connected via D-channel. We show the forecasting methodology of the traffic demand of the D-channel internet services when the TDX-1OA switching system accommodates the target number of subscriber. Based on the traffic demand forecast, the required capacity of the access line is suggested to prevent bottleneck problem in the access line to the X.25 packet network.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 한국컴퓨터정보학회 2019년도 제59차 동계학술대회논문집 27권1호
• /
• pp.123-124
• /
• 2019

This paper proposes an improved deep learning method based on small data sets for animal image classification. Firstly, we use a CNN to build a training model for small data sets, and use data augmentation to expand the data samples of the training set. Secondly, using the pre-trained network on large-scale datasets, such as VGG16, the bottleneck features in the small dataset are extracted and to be stored in two NumPy files as new training datasets and test datasets. Finally, training a fully connected network with the new datasets. In this paper, we use Kaggle famous Dogs vs Cats dataset as the experimental dataset, which is a two-category classification dataset.

• Proceedings of the Korea Information Processing Society Conference
• /
• 한국정보처리학회 2022년도 춘계학술발표대회
• /
• pp.9-11
• /
• 2022

Community detection is a hot topic in social network analysis, and many existing studies use graph theory analysis methods to detect communities. This paper focuses on detecting absolute fair maximal balanced cliques in signed attributed social networks, which can satisfy ensuring the fairness of complex networks and break the bottleneck of the "information cocoon".

• The Journal of Korean Institute of Communications and Information Sciences
• /
• 제34권12B호
• /
• pp.1475-1482
• /
• 2009

In ZigBee network, using energy efficiently is necessary because ZigBee node works by battery. To use energy efficiently, it is one of the way to reduce unnecessary network traffic. In this paper, it presents efficient shortcut routing algorithm using depth of destination node in ZigBee network. In traditional tree routing, each node transfers data only to its own parent or child node, which is inefficient way. Efficient shortcut routing algorithm is also based on tree routing. However, we suggests the algorithm with using neighbor table and depth of destination that is able to transfer data to other neighbor node, not only to parent or child node. It minimizes coordinator bottleneck state and unnecessary intermediate routing path which happens in traditional tree routing.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권1호
• /
• pp.308-324
• /
• 2018

HTTP adaptive streaming (HAS) is a promising technology for delivering video content over the Internet. HAS-based video streaming solutions rely on bandwidth estimation to select the appropriate video bitrate. Video streaming solutions that consider network conditions provide users with seamless video playback. However, when multiple clients compete for a common bottleneck link, conventional bandwidth estimation schemes that consider only one client overestimate the network bandwidth due to the ON-OFF traffic pattern. The bandwidth overestimation can cause Quality of Experience (QoE) degradation, such as unnecessary changes in video quality, and unfairness of video quality. In this paper, we propose a client-side bandwidth estimation scheme to obtain a better QoE of HAS in the multiple-client environment. The proposed scheme differentiates the client buffer status according to the buffer occupancy, and then estimates the available network bandwidth based on the buffer status and segment throughput. We evaluate the performance of HAS implemented in the ns-3 network simulator. Simulation results show that compared with the conventional schemes, the proposed scheme can enhance the QoE.