• 한국컴퓨터정보학회논문지
• /
• 제26권9호
• /
• pp.143-153
• /
• 2021

본 논문의 목적은 빠른 배송을 위한 배송 차량의 유연 회차와 크라우드쉬핑(Crowd Shipping) 배달원인 플렉스를 활용한 "유연 회차와 플렉스" 방식을 제안하고, 풀이 알고리즘을 제안하는 것이다. 알고리즘은 선행 연구에서 성능이 검증된 2-opt를 사용하여 차량 경로를 계산하고, 절약 휴리스틱(Saving heuristic)을 변형한 방식으로 플렉스에 할당할 고객을 계산한다. 알고리즘은 차량과 플렉스를 분리하는 것이 아니라, 상호 보완적으로 운용될 수 있게 한다. 이를 위해 차량 정보가 플렉스 할당 과정에 고려되도록 하였다. 본 연구의 실험은 다양한 도심지의 상황을 상정한 Random, Mixed, Cluster 인스턴스로 구성되었다. 실험 결과 모든 인스턴스에서는 플렉스의 사용이 차량 유연 회차의 효율성을 높여주는 것을 확인하였다. 또 플렉스와 배송 지연 시간 비용에 대한 민감도 분석에서는 "유연 회차와 플렉스"가 차량만을 사용하는 방식의 차량 대수에 따라 값에 차이를 보였지만, 플렉스 비용이나 배송 지연 시간 비용이 200% 증가하는 경우에도 우위를 유지하는 것을 확인하였다. 본 연구의 결과는 플렉스가 차량과 연계되어 활용되는 경우, 플렉스에 의해 서비스되는 고객뿐만 아니라 차량으로 서비스되는 고객들의 서비스 품질을 높일 수 있음을 보여준다.

• 경영과학
• /
• 제24권2호
• /
• pp.43-56
• /
• 2007

VRP(Vehicle Routing Problem) is studied in this paper, where two different kinds of missions are to be completed. The objective is to minimize the total vehicle operating distance. A mixed integer programming formulation and a heuristic algorithm for a practical use are suggested. A heuristic algorithm consists of three phases such as clustering, constructing routes, and adjustment. In the first phase, customers are clustered so that the supply nodes are grouped with demand nodes to be served by the same vehicle. Vehicle routes are generated within the cluster in the second phase. Clusters and routes are adjusted in the third phase using the UF (unfitness) rule designed to determine the customers and the routes to be moved properly. It is shown that the suggested heuristic algorithm yields good performances within a relatively short computational time through computational experiment.

• Journal of Electrical Engineering and Technology
• /
• 제13권6호
• /
• pp.2468-2478
• /
• 2018

We propose the Sparse Feature Convolutional Neural Network (SFCNN) to reduce the volume of convolutional neural networks (CNNs). Despite the superior classification performance of CNNs, their enormous network volume requires high computational cost and long processing time, making real-time applications such as online-training difficult. We propose an advanced network that reduces the volume of conventional CNNs by producing a region-based sparse feature map. To produce the sparse feature map, two complementary region-based value extraction methods, cluster max extraction and local value extraction, are proposed. Cluster max is selected as the main function based on experimental results. To evaluate SFCNN, we conduct an experiment with two conventional CNNs. The network trains 59 times faster and tests 81 times faster than the VGG network, with a 1.2% loss of accuracy in multi-class classification using the Caltech101 dataset. In vehicle classification using the GTI Vehicle Image Database, the network trains 88 times faster and tests 94 times faster than the conventional CNNs, with a 0.1% loss of accuracy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권3호
• /
• pp.1213-1237
• /
• 2019

Vehicular Ad-hoc Network (VANET) facilities envision future Intelligent Transporting Systems (ITSs) by providing inter-vehicle communication for metrics such as road surveillance, traffic information, and road condition. In recent years, vehicle manufacturers, researchers and academicians have devoted significant attention to vehicular communication technology because of its highly dynamic connectivity and self-organized, decentralized networking characteristics. However, due to VANET's high mobility, dynamic network topology and low communication coverage, dissemination of large data packets (e.g. multimedia content) is challenging. Clustering enhances network performance by maintaining communication link stability, sharing network resources and efficiently using bandwidth among nodes. This paper proposes a mobility-based, multi-hop clustering algorithm, (MBCA) for multimedia content broadcasting over an IEEE 802.11p-LTE-enabled hybrid VANET architecture. The OMNeT++ network simulator and a SUMO traffic generator are used to simulate a network scenario. The simulation results indicate that the proposed clustering algorithm over a hybrid VANET architecture improves the overall network stability and performance, resulting in an overall 20% increased cluster head duration, 20% increased cluster member duration, lower cluster overhead, 15% improved data packet delivery ratio and lower network delay from the referenced schemes [46], [47] and [50] during multimedia content dissemination over VANET.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2009년도 춘계 종합학술대회 논문집
• /
• pp.884-888
• /
• 2009

Vehicular Adhoc Networks(VANET)은 차량과 노변장치 또는 차량간 통신 기술로 MANET의 보다 구체적인 활용 분야이면서 MANET의 Quality of Service(QoS) 기술이 Video on Demand(VoD), Video streaming, Voice over IP(VoIP) 등 멀티미디어 데이터의 품질을 위하여 연구되고 있다면, VANET에서는 사고나 기타 예상될 수 있는 응급한 상황에 대하여, 즉시 주변 차량에 전달하여 사고를 미연에 방지하는 등의 부분에 초점이 맞추어져있다. 본 논문에서는 Link State Routing(LSR)을 이용하여 실제 네트워크를 구성하고, Common Open Policy Service(COPS)를 이용한 QoS Provisioning 모델을 구현하였다. 그리고 VANET의 특성에 맞는 보다 효율적인 k-hop Cluster, inter-domain policy negotiation에 대하여 제안한다.

• 대한인간공학회지
• /
• 제34권3호
• /
• pp.235-245
• /
• 2015

Objective: The purpose of this study is to develop visual complexity metrics based on theoretical bases. Background: With the development of IT technologies, drivers process a large amount of information caused by automotive human-machine interface (HMI), such as a cluster, a head-up display, and a center-fascia. In other words, these systems are becoming more complex and dynamic than traditional driving systems. Especially, these changes can lead to the increase of visual demands. Thus, a concept and tool is required to evaluate the complicated systems. Method: We reviewed prior studies in order to analyze the visual complexity. Based on complexity studies and human perceptual characteristics, the dimensions characterizing the visual complexity were determined and defined. Results: Based on a framework and complexity dimensions, a set of metrics for quantifying the visual complexity was developed. Conclusion: We suggest metrics in terms of perceived visual complexity that can evaluate the in-vehicle displays. Application: This study can provide the theoretical bases in order to evaluate complicated systems. In addition, it can quantitatively measure the visual complexity of In-vehicle information system and be helpful to design in terms of preventing risks, such as human error and distraction.

• 지능정보연구
• /
• 제15권1호
• /
• pp.1-14
• /
• 2009

본 연구에서는 한정 용량 차량 경로탐색 문제(CVRP, Capacitated Vehicle Routing Problem)에서 이분 시드 검출 방법(Bisection Seed Detection)을 이용한 휴리스틱 알고리즘을 제안하였다. 이 알고리즘은 3단계로 구성된다. 1단계에서는 improved sweep 알고리즘을 이용해서 초기 클러스터를 구성한다. 2단계에서는 1단계에서 얻은 각 클러스터에 대하여 이분 시드 검출 법을 이용해서 seed 노드를 선택하고, regret 값에 따라 각 경로에 고객 노드들을 삽입 함으로서 차량 이동 경로를 생성한다. 3단계에서는 tabu 탐색 방법과 노드 교환 알고리즘(node exchange algorithm)을 이용하여 2단계에서 얻어진 각 경로를 더욱 향상 시킨다. 본 논문의 실험에서는 제안된 휴리스틱이 비교적 빠른 시간 내에 최적 근사 값을 얻을 수 있음을 보였으며, 이는 빠른 실행 시간을 요구하는 실 업무에 유용하다.

• 대한전자공학회논문지SP
• /
• 제40권3호
• /
• pp.203-216
• /
• 2003

In this paper a new feature extraction and target classification method is proposed for the recognition part of FLIR(Forwar Looking Infrared)-image-based ATR system. Proposed feature extraction method is "cluster(=set of classes)-based"version of previous fisherfaces method that is known by its robustness to illumination changes in face recognition. Expecially introduced class clustering and cluster-based projection method maximizes the performance of fisherfaces method. Proposed target image classification method is based on the mixture of experts model which consists of RBF-type experts and MLP-type gating networks. Mixture of experts model is well-suited with ATR system because it should recognizee various targets in complexed feature space by variously mixed conditions. In proposed classification method, one expert takes charge of one cluster and the separated structure with experts reduces the complexity of feature space and achieves more accurate local discrimination between classes. Proposed feature extraction and classification method showed distinguished performances in recognition test with customized. FLIR-vehicle-image database. Expecially robustness to pixelwise sensor noise and un-wanted intensity variations was verified by simulation.

• 지식경영연구
• /
• 제15권4호
• /
• pp.147-170
• /
• 2014

This study investigates the main characteristics of Chinese automobile industry based on the technology learning, architecture theory and cluster. As a case study sample, we chose three most representative automobile firms from three main cities in China, FAW from northern part of China, SAIC from middle part of China, and BYD from southern part of China. According to the research findings, FAW has equipped self-production ability in virtue of political support but felled behind in future transportation due to lack of convergence with local cluster. In case of SAIS, similar phenomenon happened in spite of highest purchasing power of shanghai. BYD has achieved great quantum jump through the aggressive investment strategy in electric vehicle even though there are still many technological learning and experience to be cumulated. Overall, this research extends the current literature on key roles (technological learning, architecture, and cluster features) in the automobile industry growth by suggesting their crucial aspects in knowledge management and strategic planning to a newly emerging market, China, and sheds light on the relationship between regional characteristics and automobile growth.

• 대한임베디드공학회논문지
• /
• 제8권2호
• /
• pp.69-78
• /
• 2013

As vehicles become more intelligent for safety and convenience of drivers, in-vehicle networking systems such as controller are network (CAN) have been widely used due to increasing number of electronic control unit (ECU). Recently, FlexRay was developed to replace CAN protocol in chassis networking systems, to remedy the shortage of transmission capacity and unsatisfactory real-time transmission delay of conventional CAN. However, it is difficult for vehicle network designers to calculate platform configuration registers (PCR) and determine a base cycle or slot length of FlexRay. To assist vehicle network designers for designing FlexRay cluster, this paper presents automatic field bus exchange format (FIBEX) generation algorithm from CANdb information, which is de-facto standard database format for CAN. To design this program, structures of FIBEX, CANdb and relationship among PCR variables are analyzed.