• 정보처리학회논문지C
• /
• 제9C권6호
• /
• pp.865-874
• /
• 2002

HTTP(HyperText Transfer Protocol)는 월드 와이드 웹 분산 시스템이 객체를 인출하기 위해 사용하는 전송 프로토콜이다. HTTP는 연결 지향 프로토콜이므로 트랜스포트계층에서 TCP(Transmission control Protocol)를 사용한다. 그러나 HTTP는 TCP와의 상호운용이 좋은 편이 아닌 것으로 알려져 있다. 본 연구에서는 TCP의 성능에 영향을 주는 요인들을 살펴보고, HTTP 접근을 위해 TCP의 slow-start 오버헤드 및 연결에 소요되는 트랜잭션 시간과 TCP의 성능 향상 대안중의 하나인 T-TCP(Transaction TCP)의 트랜잭션 시간을 검토한다. 평균 대기 시간과 평균 반환 시간은 사용자의 서비스 품질을 만족시키기 위한 중요한 파라메터들이다. TCP와 T-TCP 트랜잭션 시간이 주어지는 경우 그러한 파라메터들의 계산 공식이 유도되었다. 실험 및 계산 경험을 통해 제안된 공식이 잘 작동됨을 확인하였고, 대역폭의 확장이 필요한 환경에 적용될 수 있으며 T-TCP의 시간 특성이 TCP 보다 우수함을 확인하였다. 아울러, 평균 대기 시간과 평균 반환 시간을 줄이기 위해 대역폭을 조합하여 서버를 분산하는 방법이 제시되었다.

• 한국컴퓨터정보학회논문지
• /
• 제23권8호
• /
• pp.133-142
• /
• 2018

Subway metro scheduling is one of the most important problems impacting passenger convenience today. To operate efficiently, the Seoul metro uses regular, periodic schedules for all lanes, both north and southbound. However, many past studies suggest that non-periodic scheduling would better optimize costs. Since the Seoul metro is continuously facing a deficit, adopting a non-periodic schedule may be necessary. Two objectives are presented; the first, to minimize the average passengers' waiting time, and the second, to minimize total costs, the sum of the passenger waiting time, and the operational costs. In this paper, we use passenger smart card data and a precise estimation of transfer times. To find the optimal time-table, a genetic algorithm is used to find the best solution for both objectives. Using Python 3.5 for the analysis, for the first objective, we are able to reduce the average waiting time, even when there are fewer trains. For the second objective, we are able to save about 4.5 thousand USD with six fewer trains.

• Industrial Engineering and Management Systems
• /
• 제15권1호
• /
• pp.19-31
• /
• 2016

A design method of the network for automated transporters mounted on rails is addressed for automated container terminals. In the network design, the flow directions of some path segments as well as routes of transporters for each flow requirement must be determined, while the total transportation and waiting times are minimized. This study considers, for the design of the network, the waiting times of the transporters during the travel on path segments, intersections, transfer points below the quay crane (QC), and transfer points at the storage yard. An algorithm, which is the combination of a modified Dijkstra's algorithm for finding the shortest time path and a queuing theory for calculating the waiting times during the travel, is proposed. The proposed algorithm can solve the problem in a short time, which can be used in practice. Numerical experiments showed that the proposed algorithm gives solutions better than several simple rules. It was also shown that the proposed algorithm provides satisfactory solutions in a reasonable time with only average 7.22% gap in its travel time from those by a genetic algorithm which needs too long computational time. The performance of the algorithm is tested and analyzed for various parameters.

• 한국항만학회지
• /
• 제5권2호
• /
• pp.19-37
• /
• 1991

This work aims to : establish a model of the container physical distribution system of Pusan port comprising 4 sub-systems of a navigational system, on-dock cargo handling/transfer/storage system, off-dock CY system and an in-land transport system : examine the system regarding the cargo handling capability of the port and analyse the cost of the physical distribution system. The overall findings are as follows : Firstly in the navigational system, average tonnage of the ships visiting the Busan container terminal was 33,055 GRT in 1990. The distribution of the arrival intervals of the ships' arriving at BCTOC was exponential distribution of $Y=e^{-x/5.52}$ with 95% confidence, whereas that of the ships service time was Erlangian distribution(K=4) with 95% confidence, Ships' arrival and service pattern at the terminal, therefore, was Poisson Input Erlangian Service, and ships' average waiting times was 28.55 hours In this case 8berths were required for the arriving ships to wait less than one hour. Secondly an annual container through put that can be handled by the 9cranes at the terminal was found to be 683,000 TEU in case ships waiting time is one hour and 806,000 TEU in case ships waiting is 2 hours in-port transfer capability was 913,000 TEU when berth occupancy rate(9) was 0.5. This means that there was heavy congestion in the port when considering the fact that a total amount of 1,300,000 TEU was handled in the terminal in 1990. Thirdly when the cost of port congestion was not considered optimum cargo volume to be handled by a ship at a time was 235.7 VAN. When the ships' waiting time was set at 1 hour, optimum annual cargo handling capacity at the terminal was calculated to be 386,070 VAN(609,990 TEU), whereas when the ships' waiting time was set at 2 hours, it was calculated to be 467,738 VAN(739,027 TEU). Fourthly, when the cost of port congestion was considered optimum cargo volume to be handled by a ship at a time was 314.5 VAN. When the ships' waiting time was set at I hour optimum annual cargo handling capacity at the terminal was calculated to be 388.416(613.697 TEU), whereas when the ships' waiting time was set 2 hours, it was calculated to be 462,381 VAN(730,562 TEU).

• 대한교통학회지
• /
• 제13권4호
• /
• pp.105-115
• /
• 1995

This paper deals with an operation concept of "time-transfer of buses", in which buses arrive a transfer center at the same time and allow passengers to transfer to other bus lines, so that passengerscan go anywhere all the timed-transfere buses operate with only one transfer. This timed-transfer bus system is known as an efficient operating technique which can be adopted in an area with sparsely distrubuted bus demand. A model of timed-transfer is estabilished in terms of vaious factors such as the expected(or average) arrival time, distribution of arrival time, timed-transfer cycle, shceduled departure time, etc. It is assumed that the objective of timed-transfer bus system is to minimize the total transfer waiting time. The optimal schedualed arrival time or buffer time(time required to arrive ealy in consideration of delay) is analyzed for a general case and various speicial cases. It was found that the optiaml buffer time is an inverse funcion of the timed-transfer cycl and increases with the cycle time, assumin g that there is a fixed scheduled departure time at the transfer cetner regardless of whether one or more buses fail to arrive before the scheduled departure time. If buses are to wait uhtil all the buses arrive at the transfer center, that is, the transfer departure time is vairable, the optimal scheduled arrival times can be obtained by a mathmatical programming.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2011년도 정기총회 및 추계학술대회 논문집
• /
• pp.1722-1729
• /
• 2011

This paper presents an optimization model for revising train timetable based on an existing timetable to improve transfer time at each station. The transfer time consists of walking and waiting time. The model is formulated as a mixed integer programming. The objective function is to minimize the transfer time from one train to another train at each station. To reflect real situations, range of revising departure time is considered as major condition in the model. To validate the effectiveness of the model, rudimentary computational results are included, and the results are analyzed in terms of transfer time.

• 대한교통학회지
• /
• 제24권5호
• /
• pp.77-88
• /
• 2006

본 연구는 복합교통망의 실시간적 운영전략에 활용 가능한 동적 K최소시간경로탐색알고리즘을 제안하는 것을 목적으로 한다. 이를 위해 정적 K경로탐색에 적용되었던 전체경로삭제방안을 동적 최적경로탐색알고리즘에 확장 적용함으로써 복합교통망에서 시간종속적으로 변화하는 수단-링크 통행시간과 수단-링크간의 환승비용에 기초하여 경로를 순차적으로 탐색하는 K경로알고리즘을 제시하였다. 특히, 링크기반동적표지를 적용함으로써 수단간 환승시 발생되는 환승이동, 환승대기 및 기타 환승행태를 용이하게 모사하면서 최적해를 도출하도록 하였다. 최적식과 함께 제시된 알고리즘은 복잡도계산을 통해 효율성을 살펴보았으며. 버스와 지하철로 구성된 소규모 가상네트워크에 대해 적용해 봄으로써 알고리즘의 검증 및 활용성을 검토해 보았다.

• 한국융합학회논문지
• /
• 제12권4호
• /
• pp.217-228
• /
• 2021

버스카드를 활용한 통합 요금제 시행 이 후 대중교통 통행량 및 환승 통행량은 증가하였다. 이에 따라 버스운영에 있어서 환승객을 고려하는 것은 보다 중요해졌다. 기존 연구들에서는 환승객을 고려할 때 결정적 수리모델을 제안하여 승객과 버스의 확률적인 움직임을 반영하지 못하는 한계가 있었으나 본 연구에서는 미시 시뮬레이션 모델을 바탕으로 하여 보다 실제적인 버스 정류장 모델을 제안하였다. 제안한 시뮬레이션 모델을 기반으로 버스 도착 간격과 승객 대기 시간의 관계를 회귀 모델로 표현하였으며 환승객을 고려할 때와 고려하지 않을 때의 차이를 실증적으로 검증하였다. 또한 승객 대기 시간을 비용으로 변환하는 방법을 제안하고 이를 바탕으로 하여 최적 버스 도착 간격을 제안하였다. 연구에서 제안한 방법을 활용하면 버스 운영에 있어서 보다 현실을 반영한 상향식 의사결정을 가능하게 해줄 것으로 기대된다.

• 한국시뮬레이션학회논문지
• /
• 제17권4호
• /
• pp.249-257
• /
• 2008

본 논문에서는 로봇 한 대를 중심으로 여러 설비들이 제어되는 Flexible Manufacturing Cell(FMC)에서 로봇의 대기위치 선정과 작업물 선택 의사결정을 위하여 Q-learning 알고리즘을 응용한 방법을 제시하였다. 이를 위해 하나의 로봇과 다수의 설비로 구성된 전형적인 가상의 FMC를 시뮬레이션으로 설계, 제안한 알고리즘을 적용하여 다른 대안과의 비교 분석을 실시하였다. 비교결과 적용된 알고리즘이 로봇가동률을 높여 평균 처리시간, 총 처리량 모두 향상시키는 결과를 가져왔다. 또한, 이 방법은 본 논문에서 제시한 정보뿐 아니라 그 우선순위와 가중치에 따라 다른 정보들을 손쉽게 추가하여 적용할 수 있으므로 FMC의 생산성 향상에도 크게 기여할 것으로 기대된다.

• 대한산업공학회지
• /
• 제27권2호
• /
• pp.140-149
• /
• 2001

The paper discusses the problem of finding the Origin-Destination(O-D) shortest paths in internodal transportation networks with time-schedule constraints. The shortest path problem on the internodal transportation network is concerned with finding a path with minimum distance, time, or cost from an origin to a destination using all possible transportation modalities. The time-schedule constraint requires that the departure time to travel from a transfer station to another node takes place only at one of pre-specified departure times. The scheduled departure times at the transfer station are the times when the passengers are allowed to leave the station to another node using the relative transportation modality. Therefore, the total time of a path in an internodal transportation network subject to time-schedule constraints includes traveling time and transfer waiting time. In this paper, a genetic algorithm (GA) approach is developed to deal with this problem. The effectiveness of the GA approach is evaluated using several test problems.