This study aims to develop a regression model using data from the Ammunition Stockpile Reliability Program (ASRP) to predict the shelf life of 81mm mortar high-explosive shells. Ammunition is a single-use item that is discarded after use, and its quality is managed through sampling inspections. In particular, shelf life is closely related to the performance of the propellant. This research seeks to predict the shelf life of ammunition using a regression model. The experiment was conducted using 107 ASRP data points. The dependent variable was 'Storage Period', while the independent variables were 'Mean Ammunition Velocity,' 'Standard Deviation of Mean Ammunition Velocity,' and 'Stabilizer'. The explanatory power of the regression model was an R-squared value of 0.662. The results indicated that it takes approximately 55 years for the storage grade to change from A to C and about 62 years to change from C to D. The proposed model enhances the reliability of ammunition management, prevents unnecessary disposal, and contributes to the efficient use of defense resources. However, the model's explanatory power is somewhat limited due to the small dataset. Future research is expected to improve the model with additional data collection. Expanding the research to other types of ammunition may further aid in improving the military's ammunition management system.

In order to prevent accidents via defective ammunition, this paper analyzes recent research on ammunition life prediction methodology. This workanalyzes current shelf-life prediction approaches by comparing the pros and cons of physical modeling, accelerated testing, and statistical analysis-based prediction techniques. Physical modeling-based prediction demonstrates its usefulness in understanding the physical properties and interactions of ammunition. Accelerated testing-based prediction is useful in quickly verifying the reliability and safety of ammunition. Additionally, statistical analysis-based prediction is emphasized for its ability to make decisions based on data. This paper aims to contribute to the early detection of defective ammunition by analyzing ammunition life prediction methodology hereby reducing defective ammunition accidents. In order to prepare not only Korean domestic war situation but also the international affairs from Eastern Europe and Mid East countries, it is very important to enhance the stability of organizations using ammunition and reduce costs of potential accidents.

이 연구는 1세대 스마트 온실의 재배환경 데이터와 장미 절 화의 품질 특성 데이터를 수집하고 그 요인들 간의 상관 관계 를 분석하여 절화수명 예측 및 최적 환경 조성의 기초 자료를 얻고자 수행되었다. 이를 위해, 토경재배(SC) 및 암면배지경 양액재배(RWH) 하우스 각 1개소를 선정하여 1년간 기온, 상 대습도(RH) 및 수증기압차(VPD), 일적산광량(DLI), 근권온도 등의 환경 데이터와 매월 말 수확된 장미 ‘Miss Holland’ 절 화의 품질 특성 데이터를 수집하였으며, 이 데이터와 절화수 명과의 상관관계를 분석하였다. 절화수명은 10월과 11월을 제외하고는 SC 하우스에서 RWH 하우스보다 더 길었다. 절 화수명과 환경 및 생육 특성 간의 상관관계 분석에서 SC 하우 스의 상관계수는 RWH 하우스보다 조금 더 높았으며, 절화수 명 예측을 위한 요소들도 두 하우스 간에 차이가 있었다. SC 하우스의 절화수명 Y=0.848X1+0.366X2-0.591X3+2.224X4- 0.171X5+0.47X6+0.321X7+9.836X8-110.219(X1-X8: 최고 RH, RH 일교차, DLI, pH, Hunter’s b value, EC, 절화 장, 잎 두께; R2=0.544)로 예측되었고, RWH 하우스의 절화수명 Y=-1.291X1+52.026X2-0.094X3+0.448X4-3.84X5+0.624X6 - 8.528X7+28.45(X1-X7: 경경, 야간 VPD, 최고 근권온도, 최 저 근권온도, 기온 일교차, RH 일교차, 최고 VPD; R2=0.5243) 로 예측되었다. 이 두 모델식으로부터 SC 하우스에서는 RH, EC 및 pH가, 그리고 RWH 하우스에서는 근권 온도가 절화수명에 더 큰 영향을 미친다는 것을 추론할 수 있다. 따라서 각 재배 방법에 따라 장미의 절화수명에 더 큰 영향을 미치는 환경적 요인을 효율적으로 관리할 필요가 있다.

Predicting remaining useful life (RUL) becomes significant to implement prognostics and health management of industrial systems. The relevant studies have contributed to creating RUL prediction models and validating their acceptable performance; however, they are confined to drive reasonable preventive maintenance strategies derived from and connected with such predictive models. This paper proposes a data-driven preventive maintenance method that predicts RUL of industrial systems and determines the optimal replacement time intervals to lead to cost minimization in preventive maintenance. The proposed method comprises: (1) generating RUL prediction models through learning historical process data by using machine learning techniques including random forest and extreme gradient boosting, and (2) applying the system failure time derived from the RUL prediction models to the Weibull distribution-based minimum-repair block replacement model for finding the cost-optimal block replacement time. The paper includes a case study to demonstrate the feasibility of the proposed method using an open dataset, wherein sensor data are generated and recorded from turbofan engine systems.

Most of automobile steering parts are manufactured through the multi-stage cold forging process using round-bar drawn materials. The same process is applied to the ball stud parts of the outer ball joint, and various research activities are being carried out to reduce the extreme manufacturing cost in order to survive in the limitless competition. In this paper, we present a quantitative prediction method for the limiting life of the die as a method for cost reduction in the multi-stage cold forging process. The load on the die was minimized by distributing the forming load based on process optimization through finite element analysis. In addition, based on the quantitative prediction algorithm of the limiting life of the die, the application of the split die and the optimization of the phosphate treatment of the material surface are presented as a conclusion as a method to improve the limiting life of the die.

Recently, the importance of preventive maintenance has been emerging since failures in a complex system are automatically detected due to the development of artificial intelligence techniques and sensor technology. Therefore, prognostic and health management (PHM) is being actively studied, and prediction of the remaining useful life (RUL) of the system is being one of the most important tasks. A lot of researches has been conducted to predict the RUL. Deep learning models have been developed to improve prediction performance, but studies on identifying the importance of features are not carried out. It is very meaningful to extract and interpret features that affect failures while improving the predictive accuracy of RUL is important. In this paper, a total of six popular deep learning models were employed to predict the RUL, and identified important variables for each model through SHAP (Shapley Additive explanations) that one of the explainable artificial intelligence (XAI). Moreover, the fluctuations and trends of prediction performance according to the number of variables were identified. This paper can suggest the possibility of explainability of various deep learning models, and the application of XAI can be demonstrated. Also, through this proposed method, it is expected that the possibility of utilizing SHAP as a feature selection method.

Spot welding is a representative process in automotive welding and the application of intelligent systems is accelerating. In particular, in the case of welding electrode management, the timing of electrode wear and dressing was determined by continuous spot welding evaluation, however there is concerned that errors in welding equipment or processes may work in a complex manner. In this study, a dynamic resistance waveform sensing and image measurement system that greatly affects the nugget formation, which is important to the quality of spot welding, was fabricated and used. Based on the experimental data of the galvanized steel sheet, an electrode life prediction algorithm for electrode wear was derived through CNN(Convolutional Neural Network) model of machine learning training.

The global trend is the application of heat-treated omission materials to reduce the manufacturing cost of automobile steering parts. Attempts have been made to apply heat-treated omission materials in domestic, but they are delayed due to concerns over rising cold forging process costs. For quantitative prediction of cold forging process cost, fatigue properties of forging die materials were evaluated. Based on this, the die life and cost were predicted quantitatively, and the manufacturing cost reduction of automobile steering parts using heat-treated material was found to be about 11%. Also, various methods to improve die life were additionally presented.

Recently, a study of prognosis and health management (PHM) was conducted to diagnose failure and predict the life of air craft engine parts using sensor data. PHM is a framework that provides individualized solutions for managing system health. This study predicted the remaining useful life (RUL) of aeroengine using degradation data collected by sensors provided by the IEEE 2008 PHM Conference Challenge. There are 218 engine sensor data that has initial wear and production deviations. It was difficult to determine the characteristics of the engine parts since the system and domain-specific information was not provided. Each engine has a different cycle, making it difficult to use time series models. Therefore, this analysis was performed using machine learning algorithms rather than statistical time series models. The machine learning algorithms used were a random forest, gradient boost tree analysis and XG boost. A sliding window was applied to develop RUL predictions. We compared model performance before and after applying the sliding window, and proposed a data preprocessing method to develop RUL predictions. The model was evaluated by R-square scores and root mean squares error (RMSE). It was shown that the XG boost model of the random split method using the sliding window preprocessing approach has the best predictive performance.

일반적으로 피로수명은 예측모델과 외부하중 및 기타 환경조건에서의 불확실성으로 인해 정확한 예측이 어렵다. 정확하고 신뢰성 있는 피로균열 예측을 바탕으로 강구조물의 수명관리가 효율적으로 수행될 수 있는데, 이를 위해서는 피로균열 성장 예측과 관련된 정보의 효과적이고 지속적인 업데이트가 필수적이다. 피로균열 예측 업데이트는 기존 예측모델의 변수들을 가용한 현재의 정보와 통합하는 과정을 의미하는데, 업데이트되는 변수들에 따라 예측모델의 결과가 달라진다. 본 연구에서는 피로균열 예측 업데이트에 필요한 가장 적절한 변수들을 결정하는 기법을 제시한다. 이를 위해 피로균열 예측 모델의 모든 변수들의 조합을 고려하여, 발견된 균열의 크기와 업데이트된 변수들의 차이를 평가한다. 피로균열 예측모델의 변수 업데이트를 위해 Markov chain Monte Carlo 시뮬레이션 기반 Bayesian 업데이트 기법을 적용하며, 발견된 균열의 크기와 업데이트된 예측균열의 크기 비교를 위해 평균절대오차(Mean absolute error) 기법과 Kullback-Leibler(KL) divergency 기법을 적용한다.

다양한 활용용수 생산을 위한 막 소재 및 막여과 공정개발이 고도화됨에 따라 점차적으로 유지관리 측면의 연구개발 필요성이 부각되고 있다. 그 중에서 유지관리 측면의 막 손상 검지와 수명예측은 제도적으로 정립이 미흡한 실정이며, 막여과 공정 운영에서 경제성과 안정성을 고려한 막 손상 검지 기술개발은 막 특성 및 시설규모에 따른 오차보상이 불가피하게 수반되어야 하는 연구이다. 따라서 분리막의 물리/화학적 노출강도에 따른 특성 변화를 관찰하여 내구년한 산정 및 수명예측 알고리즘을 구축하고자 한다. 또한 분리막의 특성과 모듈의 배열에 따라 최적화된 막 손상 검지를 위한 기체-액체 치환율을 기반으로 물리적 한계도출과 병행하여 이를 극복하기 위한 센싱융합 기술을 소개하고자 한다.

PURPOSES : The study aims to predict the service life of national highway asphalt pavements through deep learning methods by using maintenance history data of the National Highway Pavement Management System. METHODS: For the configuration of a deep learning network, this study used Tensorflow 1.5, an open source program which has excellent usability among deep learning frameworks. For the analysis, nine variables of cumulative annual average daily traffic, cumulative equivalent single axle loads, maintenance layer, surface, base, subbase, anti-frost layer, structural number of pavement, and region were selected as input data, while service life was chosen to construct the input layer and output layers as output data. Additionally, for scenario analysis, in this study, a model was formed with four different numbers of 1, 2, 4, and 8 hidden layers and a simulation analysis was performed according to the applicability of the over fitting resolution algorithm. RESULTS: The results of the analysis have shown that regardless of the number of hidden layers, when an over fitting resolution algorithm, such as dropout, is applied, the prediction capability is improved as the coefficient of determination (R2) of the test data increases. Furthermore, the result of the sensitivity analysis of the applicability of region variables demonstrates that estimating service life requires sufficient consideration of regional characteristics as R2 had a maximum of between 0.73 and 0.84, when regional variables where taken into consideration. CONCLUSIONS : As a result, this study proposes that it is possible to precisely predict the service life of national highway pavement sections with the consideration of traffic, pavement thickness, and regional factors and concludes that the use of the prediction of service life is fundamental data in decision making within pavement management systems.

베어링은 많은 회전체에서 사용되는 핵심부품으로, 예기치 않은 고장을 방지하기 위해 많은 연구가 집중되고 있다. 이때 중요한 것은 되도록 초기에 건전성 상태를 잘 나타내는 적절한 특징신호를 추출하는 것이다. 그러나 기존의 연구들은 주로 진단관점에서 특징신호를 추출하여 고장예지에는 적합하지 않은 측면이 있었다. 본 논문에서는 이러한 문제를 극복하기 위 해 베어링 고장 주파수의 에너지와 시간 사이의 상관계수 가중 합을 이용하여 베어링 수명 예측에 용이한 특징신호를 추출 하는 방법을 개발하였다. 그 결과 일반적으로 고장진단에서 많이 사용되고 있는 특징신호인 RMS에 비해서 결함 초기부터 단조로운 증가 경향의 특징신호를 추출함을 알 수 있었다. 이를 입증하기 위해서 NASA Ames에서 제공한 IMS bearing 진 동 데이터를 이용하였고 제시한 특징신호와 일반적인 RMS와 의 거동을 비교하여 유효성을 검증하였다.

This research studies on the demand forecasting for service parts considering parts life cycle, that gets relatively less attentions in the field of forecasting. Our goal is to develop forecasting method robust across many situations, not necessarily optimal for a limited number of specific situations. For this purpose, we first extensively analyze the drawbacks of the existing forecasting methods, then we propose the new demand forecasting method by using these findings and reinforcement leaning technique. Using simulation experiments, we proved that the proposed forecasting method is better than the existing methods under various experimental environments.

OBJECTIVES : The objective of this research is to determine the integrity of pavement structures for areas where voids exist. Furthermore, we conducted the study of voided-area analysis and remaining life prediction for pavement structures using finite element method. METHODS : To determine the remaining life of the existing voided areas under asphalt concrete pavements, field and falling weight deflectometer (FWD) tests were conducted. Comparison methods were used to have better accuracy in the finite element method (FEM) analysis compared to the measured surface displacements due to the loaded trucks. In addition, the modeled FEM used in this study was compared with well-known software programs. RESULTS : The results show that a good agreement on the analyzed and measured displacements can be obtained through comparisons of the surface displacement due to loaded trucks. Furthermore, the modeled FEM program was compared with the available pavement-structure software programs, resulting in the same values of tensile strains in terms of the thickness of asphalt concrete layers. CONCLUSIONS: The study, which is related to voided-area analysis and remaining life prediction using FEM for pavement structures, was successfully conducted based on the comparison between our methods and the sinkhole grade used in Japan.