http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
다변량 시계열 이상 탐지에서의 센서 간 관계 유형을 반영하는 그래프 구조 학습
박민재,김명호 한국정보과학회 2024 정보과학회논문지 Vol.51 No.3
Sensors are used to monitor systems in various fields, such as water treatment systems and smart factories. Anomalies in the system can be detected by analyzing multivariate time series consisting of sensor data. To efficiently detect anomalies, information about the relationships between sensors is required, but this information is generally difficult to obtain. To solve this problem, the previous work used sensor data to identify relationships between sensors, which were then represented using a graph structure. However, in this process, the graph structure only reflects the presence of relationships between sensors, not the types of relationships between sensors. In this pap er, we considered the types of relationships between sensors in graph structure learning and analyzed multivariate time series to detect anomalies in the system. Experiments show that improving detection accuracy in graph structure learning for multivariate time series anomaly detection involves taking into account the different kinds of relationships among sensors.
박민재,박동호 한국신뢰성학회 2018 신뢰성응용연구 Vol.18 No.4
Purpose: The purpose of this research is to develop a maintenance cost model which is applicable for Korean Lemon Law and to optimize relevant decision variables such as length of warranty period and imposed warranty premium price for the vehicles. Methods: We propose a two-dimensional warranty policy based on age and usage from the manufacturer’s perspective which can be implemented to reduce the warranty cost of a repairable product incurred during the warranty period. We consider both minimal repair service and refund offered under the two-dimensional warranty services. The numerical application is implemented using the proposed approach. Results: This paper shows that when the threshold for the critical component failure increases, the expected cost rate increases while the optimal length of warranty period decreases. On the contrary, when the threshold for the noncritical component failure increases, the expected cost rate decreases while the optimal length of warranty period decreases. Conclusion: We find the optimal length of warranty period which could be very helpful for the business managers to make an appropriate decision for the manufacturer to minimize the total warranty cost and to maximize the customer’s satisfaction.
Two-dimensional warranty for items with preventive maintenance based on Korean Lemon Law
박민재,정기문,박동호 한국신뢰성학회 2018 한국신뢰성학회 학술대회논문집 Vol.2018 No.10
In this paper, we develop a cost model under two-dimensional warranty with age and usage based on Korean Lemon Law and determine decision variables such as optimal preventive maintenance cycle for age and usage perspectives and optimal length of warranty period. We propose a two-dimensional maintenance strategy, under which the item is preventively maintained according to a specified age interval or usage interval, whichever occurs first. A periodic preventive maintenance service is considered which can be implemented to reduce the repair cost of a repairable product under warranty. The optimization of imperfect preventive maintenance for repairable items is investigated from the manufacturer’s perspective by taking into account the moments of customers purchasing two-dimensional warranty. During the warranty period, preventive maintenance service and refund service is conducted to reduce the degradation rate which the product is in operation and to increase customers’ satisfaction. We consider minimal repair service, refund service, preventive maintenance service for the warranty services. The numerical application is implemented using the proposed approach and examples are discussed to exemplify the applicability of the methodologies derived in this paper.