• 한국멀티미디어학회논문지
• /
• 제23권7호
• /
• pp.830-839
• /
• 2020

As a popular mathematical framework for modeling decision making, Markov decision process (MDP) has been widely used to solve problem in many engineering fields. MDP consists of a set of discrete states, a finite set of actions, and rewards received after reaching a new state by taking action from the previous state. The objective of MDP is to find an optimal policy, that is, to find the best action to be taken in each state to maximize the expected discounted reward of policy (EDR). In practice, MDP is typically unknown, so simulation-based policy improvement (SBPI), which improves a given base policy sequentially by selecting the best action in each state depending on rewards observed via simulation, can be a practical way to find the optimal policy. However, the efficiency of SBPI is still a concern since many simulation samples are required to precisely estimate EDR for each action in each state. In this paper, we propose a method to select the best action accurately in each state using a small number of simulation samples, thereby improving the efficiency of SBPI. The proposed method accumulates the simulation samples observed in the previous states, so it is possible to precisely estimate EDR even with a small number of samples in the current state. The results of comparative experiments on the existing method demonstrate that the proposed method can improve the efficiency of SBPI.

• 한국국방경영분석학회지
• /
• 제28권2호
• /
• pp.1-19
• /
• 2002

This study shows the effect of inventory policy change from supplier-based to customer-based. We focus on the service level, cost, and information distortion of the Military Supply Chain(MSC) with System Dynamics. We design MSC model according to field practician interviews by using Vensim. The simulation makes a comparison between supply-based inventory policy performances and order-based inventory policy performances. In order to evaluate the MSC performances, we measure the accumulation of backlog(service level), supply chain cost, and order percentage overshoot(information distortion). The results show that 1) changing inventory policy from supplier-based to end customer order-based gets a good customer service, reduces MSC cost, and prevents information distortion, 2) changing inventory policy from supplier-based to immediate customer order-based reduces a small amount of MSC cost and deteriorates customer service, and 3) supply level is main factor for MSC performances improvement. This study implicates the policy change makes a improvement of MSC performance without introducing information system.

• 디지털융복합연구
• /
• 제12권8호
• /
• pp.377-387
• /
• 2014

본 연구는 시뮬레이션을 이용하여 외래프로세스를 개선하여 기관 운영의 효율성을 높이고자 수행되었다. 3가지의 시나리오를 설정하여 시뮬레이션 분석을 수행하였으며 외래환자 전체 체류시간, 대기시간, 이동시간, 진료시간, 직원 활용도 지표를 비교하여 시나리오에 따른 외래프로세스의 효율성을 평가하였다. 병원의 진료자료를 수집하여 통계도구와 프로세스 마이닝 도구를 이용하여 분석하였다. 그리고 시뮬레이션 툴인 PIOS를 이용하여 모형의 타당성은 t-test로 검증하였다. 시뮬레이션 분석 결과, 센터제로 운영하는 경우의 외래프로세스가 가장 효율성이 높은 것으로 나타났다. 이를 볼 때 외래환자에 대해서는 센터제 형태로 운영되는 것이 기관의 효율성을 높이는 방안이라는 것을 확인할 수 있었다. 본 연구를 통하여 시뮬레이션이 최적의 외래프로세스를 선정하는데 활용될 수 있는 방법이라는 것을 확인할 수 있었다. 시뮬레이션을 이용하면 과거 경험, 감정, 직관에 의존하는 기존의 보건의료 관리 기법에 비해 효율적인 의사 결정을 지원하는 방법이라는 것을 알 수 있다. 따라서 본 연구에서 제시한 연구 모델은 보건 의료 시스템 상에 다양한 활용이 가능할 것으로 보인다.

• 대한산업공학회지
• /
• 제41권6호
• /
• pp.588-598
• /
• 2015

In this paper, a simulation model for assessing the impact of alternative field service engineer replenishment policies is introduced. The end-to-end supply chain simulation model is created using a discrete-event and agent-based simulation model, which enables accurate description of key individual entities in the investigated supply chain, such as field service engineers. Once the model is validated with the historical data, it is used to assess the impacts of field service engineer replenishment policies for a major printing equipment manufacturing firm.In the case study, newly proposed replenishment policies for post-sale distribution supply chain are assessed for the level of service improvement to end customers.

• 한국경영과학회지
• /
• 제24권3호
• /
• pp.39-47
• /
• 1999

It has been concluded that the performance of class-based storage policy is better than the performance of random storage policy in the literature. However, the rack shortage problem assigned to the 1st class items makes the decision hard to apply the class-based storage policy in practice. In this paper, a new common zone concept is introduced between two classes to resolve the problem with class-based storage policy. The common zone is the area to accept items from both classes. An AS/RS model is developed for computer simulation study and the effect of common area sizes with various AS/RS operation conditions is analyzed.

• 대한산업공학회지
• /
• 제15권1호
• /
• pp.1-15
• /
• 1989

In the past few years, increasing numbers of automatic storage/retrieval system (AS/RS) using computer controlled storage/retrieval machine have been installed. This paper introduces two modeling approaches to determine the best operating policy for AS/RS : an M/G/1 queueing model and a computer simulation model. The operating policy consists of three elements. : the operation command cycle, the storage location method, and the operation dispatching rule. The analysis based on M/G/1 model is suitable for a quick and approximate evaluation, due to its inherent strict assumptions. The computer simulation can be used to perform a more realistic analysis. It is shown through the study that a significant improvement in the throughput and/or the space requirement can be expected by determining the best operating policy to a particular system. Most important, the computer simulation demonstrates its powerful capability in evaluating dynamic stochatic systems with imperfect information.

• 정보처리학회논문지B
• /
• 제14B권2호
• /
• pp.127-134
• /
• 2007

기계 학습 등 인공 지능 기법의 발전에 힘입어 지능형 주식 거래 시스템에 관한 많은 연구가 이루어져 왔다. 그러나 현실 주식 거래에서 적절한 거래 정책의 수립이 거래의 결과에 커다란 영향을 미치는 중요 요소로 작용하고 있음에도 불구하고, 기존의 연구에서는 예측 모듈의 예측 성능 향상에 주력하였거나, 거래 정책을 다룬 경우라도 예측 모듈에 종속적인 단순한 정책만을 제시하였다. 본 논문에서는 이러한 문제를 개선하기 위한 방안의 하나로, 신경망 기반 주식 거래 시스템의 구축을 위한 통합 개발 도고인 NXShell에서 채택하고 있는 ‘통합 다중 시뮬레이션‘ 기법을 제안한다. 통합 다중 시뮬레이션 기법에서는 신경망의 출력 값과 거래 정책 인자들 간의 모든 주어진 예측기의 특성에 맞는 고유의 최적 거래 정책을 수립한다. 제안된 기법의 효용성을 검증하기 위해, 한국 거래소 시장 및 코스닥 시장에서 수집한 데이터를 사용하여 수행한 거래 성능 비교 실험 결과를 제시한다.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2015년도 춘계학술대회
• /
• pp.649-652
• /
• 2015

차등서비스 네트워크의 AF(Assured Forwarding) 서비스에서 TCP 트래픽을 위한 기존 marking policy 연구는 TCP 트래픽의 RTT(Round Trip Time), 목표 전송률(target rate) 영향 등에 대한 고려가 부족하였다. 본 논문에서는 TCP 트래픽의 RTT의 영향에 의한 낮은 공평성을 개선하기 위하여 평균 전송률 예측 기반에서 TCP flow의 상태 정보를 이용한 개선된 TSW3CDM_FS(Time Sliding Window Three Color Dynamic Marker) 알고리즘을 제안한다. 제안한 알고리즘은 목표 전송률에 비례한 대역분배를 하기위한 dynamic marking policy 알고리즘이다. 제안된 알고리즘의 성능평가를 위하여 네트워크 시뮬레이터(NS-2)를 이용하여 컴퓨터 시뮬레이션을 수행하였다. 시뮬레이션 결과 제안한 TSW3CDM 알고리즘의 공평성이 기존의 TSW3CM 방식에 비해 향상된 결과를 보였다.

• 정보처리학회논문지A
• /
• 제8A권2호
• /
• pp.133-146
• /
• 2001

In a dynamic load distribution policies, each node gathers the current system sates information before making a decision on load balancing. Load balancing policies based on this strategy can suffer from processor thrashing. In this paper, we propose a new algorithm which attempts to decrease the frequency of the processor thrashing, the algorithm is based on the integration of three components. The first, the algorithm of which determine the size of jobs be transferred. The second, negotiation protocol with obtains a mutual agreement between a sender and a receiver on the transferring job size. And the third, a symmetrically-initiated location policy. The algorithm proposed in this paper used Siman IV as simulation tool to prove the improvement of performance. I analyzed the result of simulation, and compared with related works. The mean response time shows that there are no difference with existing policy, but appear a outstanding improvement in high load. The thrashing coefficient that shows the average response time, CPU overhead and the thrashing ratio at both the receiving and sending node has been used in the analysis. A significant improvement in the average response time and the CPU overhead ratio was detected using our algorithm when an overhead occurred in the system over other algorithm. The thrashing coefficient differed in the sending node and the receiving node of the system. Using our algorithm, the thrashing coefficient at the sending node showed more improvement when there was an overhead in the system, proving to be more useful. Therefore, it can be concluded that the thrashing ratio can be reduce by properly setting the maximum and minimum value of the system’s threshold queue.

• 보건행정학회지
• /
• 제3권1호
• /
• pp.146-168
• /
• 1993

The study was conducted to project supply and demand of the nurses till year 2010 based on analysis of supply and demand of nurses up to year 1991. Results of the study will provide invaluable information for nurses manpower planning as well as overall health manpower planning for the 21th century. It is projected that nurses will be oversupplied based on the current prductivity which is undesirable situation if the quality of care is considered, and undersupplied based on the the medical law as well as optimal productivity. Thus, it is desirable to increase active supply of nurses. One of the ways of increasing active supply would be increasing the size of training and education. But, considering low employment rate of nurses which is about 59% better way of solving problems related to nurses shortage would be improvement in nurses' employment rate. According to simulation study done as part of this study, if nurses' employment rate goes up to 80%, there is no need for increasing the size of training to meet the demand at the level of medical law.