http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
의사결정 단계와 시간지연을 고려한 이산시간 전투 모형 연구
윤봉규 한국국방경영분석학회 2023 한국국방경영분석학회지 Vol.49 No.2
Recently, there has been a growing interest in combat models due to the increasing need for the development of tactics in 'algorithmic warfare', particularly with the rise in the utilization of autonomous weapons on the battlefield including drones. Two primary approaches have emerged for the development of combat models: equation-based analytic models and simulation-based models. While both approaches possess their own strengths, simulation methods have gained popularity due to their ease of use. Moreover, NetLogo, a dis- crete-time simulation language, has become increasingly prevalent in various applications, owing to its low barrier to entry and its ability to simulate real-world phenomena. Nevertheless, NetLogo models are limited by their reliance on discrete-time simulation languages, thereby hindering their utilization of numerous analytic results based on continuous-time methodologies. Moreover, the challenge lies in the complexity of customizing a model to various scenarios. This paper presents two discrete-time combat models employing both equation-based analytic approach and simulation approach. The equation-based analytic model utilizes Phase-type distribution to describe the decision stages of combatants, while the simulation model is constructed using NetLogo. By con- ducting a comparative analysis of these two models for verification purposes, this paper aims to provide a vali- dated discrete-time combat model that exhibits strong adaptability to various battlefield situations.
ABC를 활용한 통신 설비 원가 산정 방법론 및 활용 방안
윤봉규,양원석 대한산업공학회 2007 산업공학 Vol.20 No.3
In the telecommunication industry, estimation of the cost of network facilities is very important since depreciation cost of the facilities accounts for a large portion of the product cost. Moreover, cost estimation in the industry becomes more difficult because of increasing indirect cost upon digital convergence, expanding multi-purpose facilities, complexity of service product, etc. Nevertheless, not much seem to have been done in improving estimation methodology of the cost of network facilities. As a result, the quality of cost information on network facilities has deteriorated, and now even decision-makers in the industry dismiss the information. Recently, two major telecommunication companies adopted a new network cost estimation method to deal with the issue. In this paper, we study the concept of new cost estimation method and the procedure to develop and apply it. We also suggest the method to carry out the cost allocation using Matlab which is more efficient and time-saving than other commercial cost calculation packages.
이산시간 단계형 분포를 활용한 도착 오차가 있는 예약 도착시스템 분석
윤봉규 ( Bongkyoo Yoon ),최경환 ( Kyunghwan Choi ) 한국로지스틱스학회 2021 로지스틱스연구 Vol.29 No.5
Since Scheduled Arrival phenomena are encountered in various area, they have been studied by many researchers using Queueing Models. Previous works, however, could not deal with the exact nature of the arrival and service process due to the tricky property of the system which has the non-Markovian arrival and service. In this regard, previous Analyses under the Markovian assumption with Poisson Arrival have the possibility to overestimate performance measures such as the number of customers and waiting time, which could lead to the excessive investment on resources to prepare for the variability of the system. We propose a novel way to capture the non-Markovian nature of arrival process while avoiding the complexity of analysis by using Phase-type random variables. With the model suggested in this paper, we investigate the impact of non-punctual arrivals in scheduled system under non-Markovian assumption and show how serious the distortion of performance measures in Markovian models.
Text Mining을 활용한 북한의 도발 수준 및 형태 예측
차선교,윤봉규 대한산업공학회 2023 대한산업공학회 춘계학술대회논문집 Vol.2023 No.5
Text Mining을 활용하여 특정 이벤트의 발생 가능성을 파악하기 위한 연구는 Machine Learning의 발달과 더불어 활발히 진행되어 왔으며, 북한의 도발 또한 Text Mining을 활용하여 예측하는 것이 가능할 것으로 보인다. 그러나 양질의 학습데이터 확보 어려움과 이벤트 분류의 난이도로 인해 이에 대한 연구는 다른 분야에 비해 미흡한 수준이다. 이 과정에서 본 연구는 pre-trained BERT 모델을 활용하여 북한의 도발행태를 기존의 연구와 같은 이진분류(도발징후 또는 평시)가 아닌 다차원으로 분류할 수 있는 기준을 정립하고 학습데이터를 북한의 ‘조선중앙통신’ 원문과 ‘국내언론지’ 기사로 구분하여 비교 분석했다. 이를통해 국내언론지 기사를 활용하는 것보다 ‘조선중앙통신’ 원문 자료를 활용하는 것이 예측의 정확도를 높일 수 있다는 점을 확인했다. 본 연구는 북한의 도발에 대한 과학적 예측을 통해 정보로서의 가치를 높이고 정성적인 전문가 판단의 신뢰도를 향상시킬수 있는 방안을 제시했다는 점에서 의의가 있다.
Text Mining을 활용한 북한의 도발 수준 및 형태 예측
차선교,윤봉규 대한산업공학회 2023 대한산업공학회지 Vol.49 No.5
Research into the feasibility of predicting specific events using Text Mining techniques has been actively pursued in conjunction with the advancement of Machine Learning. Consequently, the potential for predicting North Korea's provocations utilizing Text Mining methods has emerged. However, the field lags behind other domains due to challenges in acquiring high-quality training data and the complexity associated with event classification. This study addresses these limitations by leveraging a Pre-trained BERT model to establish a comprehensive classification framework for North Korea's provocative behavior, moving beyond binary classifications (provocation or peace) used in previous research. Original data from the Korean Central News Agency (KCNA) and domestic media sources were gathered and analyzed as training data. Notably, the findings demonstrated that employing original data from the KCNA increased prediction accuracy compared to utilizing data from domestic media. This study offers a way to enhance the informational value of North Korea's provocations through scientific predictions, ultimately bolstering the reliability of qualitative expert judgments.
흡수 마코프체인을 활용한 함정 근접무기체계 효과성 분석
김성우,윤봉규,Kim, Seong-Woo,Yoon, Bong-Kyoo 한국군사과학기술학회 2014 한국군사과학기술학회지 Vol.17 No.6
Since the technology of anti-ship missiles has advanced rapidly, defending battleships from the threat of anti-ship missiles is a crucial factor for the survival of warships. In this paper, we analyze the performance of an anti-ship missile defense system whose name is Close-In-Weapon-System. We show the survival probability of a warship equipped with the Close-In-Weapon-System as the number of anti-ship missiles attacking the warship varies. Because of the complex and dynamic operational characteristics surrounding the Close-In-Weapon-System such as speed of missiles, different range of an individual weapon in the weapon system, and the continuous change of the kill probability of the missiles corresponding to the distance of missiles from ships, few work has been done for the performance of Close-In-Weapon-System. We present a model to incorporate all the dynamic characteristics of the system using absorbing Markov Chain. With our results, we expect commanders of warships equipped with Close-In-Weapon-System to be provided with more helpful information on how to deal with the anti-ship missiles.