• 한국구조물진단유지관리공학회 논문집
• /
• 제8권4호
• /
• pp.185-193
• /
• 2004

현재의 교량의 유지관리 시에 적절한 보수 시기나 최적화된 유지관리 정책을 결정하기 위하여 교량의 성능저하를 정확히 예측하는 것은 가장 중요한 일이다 이률 위해 제안된 방법은 정량적 평가, 마르코프체인, 베이시안 추정법 등으로 구성되었다. 제안된 방법에 따라 국내의 콘크리트 슬래브 교량을 예로서 예측을 하여는데, 기존의 전문가 의견조사 빛 외관조사에 의한 예측보다 좀 더 합리적인 결과를 보여주었다.

• 한국품질경영학회:학술대회논문집
• /
• 한국품질경영학회 2004년도 품질경영모델을 통한 가치 창출
• /
• pp.413-418
• /
• 2004

The bounded adjustment is known to be more efficient than repeated adjustment when the cost is incurred for engineering process control. The procedure of the bounded adjustment is to adjust the process when the one-step predicted deviation exceeds the adjustment limit by the amount of the prediction. In this paper, two run rules are proposed and studied in order to improve the efficiency of the traditional bounded adjustment procedure. The efficiency is studied in terms of the standardized cost through Monte Carlo simulation when the procedure is operated with and without the run rules. The adjustment procedure operated with run rules turns out to be more robust for changes in the process and cost parameters. The Markov chain approach for calculating the properties of the run rules is also studied.

• 한국품질경영학회:학술대회논문집
• /
• 한국품질경영학회 2004년도 품질경영모델을 통한 가치 창출
• /
• pp.499-502
• /
• 2004

Multi-server Markovian retrial model with heterogeneous servers is analyzed. Arriving customers constitute the MMAP(Marked Markovian Arrival Process). Distribution of the primary customers among the servers is performed randomly depending on the type of a customer and the number of the server. Customers from the orbit have the exponential service time distribution with a parameter depending on the server only. The choice of the server for a retrial is made randomly as well. Multidimensional continuous time Markov chain describing operation of the model is investigated by means of reducing to asymptotically quasi-toeplitz Markov chaius.

• Journal of the Korean Statistical Society
• /
• 제4권1호
• /
• pp.79-87
• /
• 1975

However, the mechanism for producing revised estimate is the principal means of varying the resulting precisions of estimate. Therefore, a scheduling system including physician's revision should be checked by a computer simulation to evaluate possible gains to admissions scheduling accruing from the use of these estimates. The ability to accurately predict bed occupancy has long been an objective of hospital management. If the one were able to anticipate bed accupancy, then the one could more accurately plan for bed needs, schedule personnel, allocate service and supply. We may conclude that even though the Markov chain analysis would not lead to ready-made answers for the scheduling system of elective patients, it can provide the more detailed and systematic knowledgy for the solutions on the weekly base as well as the foundations for long run planning in relative sense.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제18권4호
• /
• pp.295-304
• /
• 2014

We analyze theMMPP/G/1/K queue with a modified state-dependent service rate. The service time of customers upon service initiation is changed if the number of customers in the system reaches a threshold. Then, the changed service time is continued until the system becomes empty completely, and this process is repeated. We analyze this system using an embedded Markov chain and a supplementary variable method, and present the queue length distributions at a customer's departure epochs and then at an arbitrary time.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제17권2호
• /
• pp.103-113
• /
• 2013

We analyze the queueing model for leaky bucket (LB) scheme with dynamic arrivals and token rates. In other words, in our LB scheme the arrivals and token rates are changed according to the buffer occupancy. In telecommunication networks, the LB scheme has been used as a policing function to prevent congestion. By considering bursty and correlated properties of input traffic, the arrivals are assumed to follow a Markov-modulated Poisson process (MMPP). We derive the distribution of system state, and obtain the loss probability and the mean waiting time. The analysis is done by using the embedded Markov chain and supplementary variable method. We also present some numerical examples to show the effect of our proposed model.

• ETRI Journal
• /
• 제18권3호
• /
• pp.127-145
• /
• 1996

A new analytic framework based on a linear algebra approach is proposed for examining the performance of a direct sequence spread spectrum (DS/SS) slotted ALOHA wireless communication network systems with delay capture. The discrete-time Markov chain model has been introduced to account for the effect of randomized time of arrival (TOA) at the central receiver and determine the evolution of the finite population network performance in a single-hop environment. The proposed linear algebra approach applied to the given Markov problem requires only computing the eigenvector ${\prod}$ of the state transition matrix and then normalizing it to have the sum of its entries equal to 1. MATLAB computation results show that systems employing discrete TOA randomization and delay capture significantly improves throughput-delay performance and the employed analysis approach is quite easily and staightforwardly applicable to the current analysis problem.

• 한국조명전기설비학회:학술대회논문집
• /
• 한국조명전기설비학회 2008년도 추계학술대회 논문집
• /
• pp.363-366
• /
• 2008

It was accomplished that failure analysis not only failure numbers but also power system components every years. and these informations help power system operation considerably. power system equipment were occurred a break down by natural phenomenon and aging but it was not able to predict this failure number. But many papers and technical repots study for each equipment failure rate and reliability evaluation methods. so this paper show a failure number prediction whole power system component using Markov theory not each component failure probability. the result present a next month system failure number prediction.

• Journal of Information Processing Systems
• /
• 제10권1호
• /
• pp.119-131
• /
• 2014

In recent literature on traffic scheduling, the combination of the two-dimensional discrete-time Markov chain (DTMC) and the Markov modulated Poisson process (MMPP) is used to analyze the capacity of VoIP traffic in the cognitive radio system. The performance of the cognitive radio system solely depends on the accuracy of spectrum sensing techniques, the minimization of false alarms, and the scheduling of traffic channels. In this paper, we only emphasize the scheduling of traffic channels (i.e., traffic handling techniques for the primary user [PU] and the secondary user [SU]). We consider the following three different traffic models: the cross-layer analytical model, M/G/1(m) traffic, and the IEEE 802.16e/m scheduling approach to evaluate the performance of the VoIP services of the cognitive radio system from the context of blocking probability and throughput.

• 한국경제지리학회지
• /
• 제19권4호
• /
• pp.694-711
• /
• 2016

본 연구는 경제전반의 저성장 기조에도 불구하고 성장세를 보이는 지식서비스산업의 입지적 특성과 하위 업종별 고용기회의 변화를 분석한다. 특히 우리나라 20-30대 청년층이 선호하는 전문직서비스업 중 고용분포가 높은 업종인 출판업, 정보서비스업, 교육서비스업을 중심으로 지난 10년간 입지분포의 공간적 특성을 파악하고 고용분포의 변화를 예측한다. 이를 위하여 수도권지역을 대상으로 지식서비스업과 하위 3개 업종별 입지계수(LQ)를 산출하고, LISA 분석을 통해 집중분포지역을 파악한다. 또한 지식서비스업과 하위 3개 업종별 입지분포에 영향을 미치는 사회경제적 요인을 파악하고, 마르코프 체인 모형(Markov Chain Model)의 확률과정을 적용하여 고용분포지역의 변화를 단기예측한다. 분석결과, 수도권의 일부 특정지역을 중심으로 집중분포를 보이고 있는 세 업종별 특성에 따라 분포에 영향을 미치는 유의미한 변수에 차이가 나타났고, 업종에 따라 지역별 고용분포의 재분산에도 공간적 차이가 나타날 것으로 추정되었다. 본 연구에서 살펴본 지식서비스업종은 청년층의 고용분포가 가장 높지만 동시에 인력부족도 높은 것으로 나타났다. 이들 업종별 구인-구직과정의 마찰을 감소시키기 위한 노력이 필요할 것이며, 이는 청년실업 문제를 완화하는 데에도 기여할 수 있을 것이다.