PURPOSES : The purpose of this study is to compare applicability, explanation power, and flexibility of traffic accident models between estimating model using the statistical method and the machine learning method. METHODS: In order to compare and analyze traffic accident models between model estimated using the statistical method and machine learning method, data acquisition was conducted, and traffic accident models were estimated using statistical methods such as negative binomial regression model, and machine learning methods such as a generalized regression neural network (GRNN). Then, the fitness of model as R2, root mean square error (RMSE), mean absolute percentage error (MAPE), accuracy, etc., were determined to compare the traffic accident models. RESULTS: The results showed that the annual average daily traffic (AADT), speed limits, number of lanes, land usage, exclusive right turn lanes, and front signals were significant for both traffic accident models. The GRNN model of total traffic accidents had been better statistical significant with R2: 0.829, RMSE: 2.495, MAPE: 32.158, and Accuracy: 66.761 compared with the negative binomial regression model with R2: 0.363, RMSE: 9.033, MAPE: 68.987, and Accuracy: 8.807. The GRNN model of injury traffic accidents also showed similar results of model’s statistical significance. CONCLUSIONS: Traffic accident models estimated with GRNN had better statistical significance compared with models estimated with statistical methods such as negative binomial regression model.

PURPOSES : The purpose of this study is to compare noise reduction quantities between before/after two-layer low noise pavement implementation using equivalent noise level analysis and to analyze the noise reduction effects of the two layer low noise pavement with a statistical method such as the Anderson-Darling Test.METHODS: In order to compare and to analyze noise reduction effects between before/after two-layer low noise pavement implementation, data acquisition as noise levels on a roadside and an apartment rooftop was conducted in the study area. The equivalent noise level was estimated in order to compare noise reduction quantities and the Anderson-Darling Test was carried out for estimating noise reduction effects of the two-layer low noise pavement.RESULTS: The equivalent noise levels of before/after two-layer low noise pavement implementation for the roadside during the daytime are 65.355 dB and 63.520 dB and during the nighttime are 62.463 dB and 59.088 dB. The equivalent noise levels for the apartment rooftop during daytime are 57.301 dB and 59.088 dB and during the nighttime are 54.616 dB and 52.464 dB. Also two-layer low noise pavement decreased the noise reduction effects estimated with the statistical method as the Anderson-Darling test for the roadside during the daytime by around 66.68% and decreased noise reduction effects on the roadside during the nighttime by 0.70%. Moreover it reduced noise reduction effects in the apartment rooftop during the daytime and nighttime by 0% and 96.32%, respectively.CONCLUSIONS : Based on the result of this study, two-layer low noise pavement can positively affect noise reduction during both the daytime and nighttime according to the results of estimating the equivalent noise levels and the Anderson-Darling test.

In this study, two Fe-30Mn-0.2C-(1.5Al) high-manganese steels with different surface conditions were hydrogencharged under high temperature and pressure; then, tensile testing was performed at room temperature in air. The yield strength of the 30Mn-0.2C specimen increased with decreasing surface roughness(achieved via polishing), but that of the 30Mn-0.2C- 1.5Al specimen was hardly affected by the surface conditions. On the other hand, the tendency of hydrogen embrittlement of the two high-manganese steels was not sensitive to hydrogen charging or surface conditions from the standpoints of elongation and fracture behavior. Based on the EBSD analysis results, the small decrease in elongation of the charged specimens for the Fe-30Mn-0.2C-(1.5Al) high-manganese steels was attributed to the enhanced dislocation pile-up around grain boundaries, caused by hydrogen

PURPOSES : The purpose of this study was to develop safety performance functions (SPFs) that use zero-inflated negative binomial regression models for urban intersections in central business districts (CBDs), and to compare the statistical significance of developed models against that of regular negative binomial regression models. METHODS : To develop and analyze the SPFs of intersections in CBDs, data acquisition was conducted for dependent and independent variables in areas of study. We analyzed the SPFs using zero-inflated negative binomial regression model as well as regular negative binomial regression model. We then compared the results by analyzing the statistical significance of the models. RESULTS : SPFs were estimated for all accidents and injury accidents at intersections in CBDs in terms of variables such as AADT, Number of Lanes at Major Roads, Median Barriers, Right Turn with an Exclusive Turn Lane, Turning Guideline, and Front Signal. We also estimated the log-likelihood at convergence and the likelihood ratio of SPFs for comparing the zero-inflated model with the regular model. In he SPFs, estimated log-likelihood at convergence and the likelihood ratio of the zero-inflated model were at -836.736, 0.193 and -836.415, 0.195. Also estimated the log-likelihood at convergence and likelihood ratio of the regular model were at -843.547, 0.187 and -842.631, 0.189, respectively. These figures demonstrate that zero-inflated negative binomial regression models can better explain traffic accidents at intersections in CBDs. CONCLUSIONS : SPFs that use a zero-inflated negative binomial regression model demonstrate better statistical significance compared with those that use a regular negative binomial regression model.

PURPOSES: This study aimed to evaluate the performance of pavement management works and to develop a function for estimating the level of service (LOS) and cost of service (COS) for the systematic and quantitative management of pavement performance in the public sector. METHODS: The International Roughness Index (IRI) was used as the performance index for pavement management. Long-term pavement performance data for a period of 7 years (2007-2014) collected by the National Highway Pavement Management System and historical maintenance budget data published by the South Korean government were used to develop the LOS-COS function. Based on the function, a model for estimating the appropriate budget as well as the network conditions was suggested. RESULTS : There was high degree of correlation between pavement performance and the investment level (R = - 0.74). The developed LOS-COS function suggested that the unit cost to improve the network IRI to 1 m/km was 32.6 billion KRW. Further, the maintenance costs normalized with respect to the LOS levels were LOS-A = 88.2 billion KRW, LOS-B = 55.6 billion KRW, and LOS-C = 23.0 billion KRW. CONCLUSIONS: This study proposes a simple way of developing a LOS-COS function. It also shows how to develop a network budget demand and condition estimation model using the LOS-COS function. In addition, it is the first attempt to evaluate the road maintenance budget in South Korea. It is expected that these results will help in the negotiations between the road managers and budget makers.

PURPOSES: The purpose of this study is to estimate the impact of variable message signage (VMS) on traffic safety as a function of road curve radius using statistical methods. METHODS: In order to analyze the impact of VMS installations on traffic safety, travel speed, lateral distance, and geometric data relating to road curvature in each study area was acquired and analyzed for the impact of providing VMS information on driver performance and traffic safety using statistical methods including student t-test, Mann-Whitney test, and the Anderson-Darling test for estimating traffic safety hazard zone in each lane. RESULTS: As a result of analyzing driver performance characteristics before and after providing VMS information, it was determined that by providing VMS information, mean travel speed is deceased and vehicles are driven with increased precision, following the centerline in the first and second lanes. Also the results of analyzing traffic safety impacts of VMS indicate that traffic safety performance factors in the first lane of the Gapyeong section can, on average, increase in the left and right side of the lane by 19.22% and 68.98%, respectively, and in the case of the second lane, safety impacts, on average, can increase in both sides by 100%. For the Hongcheon section, traffic safety impacts in the first lane, on average, can increase along the left and right sides of the lane by 32.31% and 47.18%, and within the second lane, traffic safety can be increased along the left and right side of the lane by 10.97% and -0.01%, respectively. CONCLUSIONS: Based on the results of this study, the impact on traffic safety obtained by providing VMS information for road sections with smaller curve radii is greater than can be obtained for road sections with larger curve radii.

PURPOSES : The purpose of this study is to assess removal efficiency of non-point pollutants and applicability for non-point pollutant reduction facilities by conducting the demonstration project operation. METHODS : In order to analyze removal efficiency of non-point pollutants for facilities such as a grassed swale, a small constructed wetland, a free water surface wetland, a horizontal sub-surface flow wetland, and a sand filtration, the field data including specifications of facilities, rainfall, inflow and runoff rainfall effluent etc. was acquired after occurring rainfall events, and the acquired data was analyzed for removal efficiency rate to assess road non-point pollutants facilities using event mean concentration (EMC) and summation of load (SOL) methods. RESULTS: The results of analyzing rainfall effluent, non-point pollutant sources showed that total suspended solid (TSS), chemical oxygen demand (COD), biochemical oxygen demand (BOD), total nitrogen (TN), total phosphorus (TP), chrome (Cr), zinc (Zn), and lead (Pb) can be removed through non-point pollutant reduction facilities by 60.3% ~ 100%. Especially removal efficiency of TSS, COD and BOD is relatively higher than removal efficiency of other non-point pollutant sources in all kind of non-point pollutant facilities. CONCLUSIONS : Based on the result of this study, even though natural type of non-point pollutant reduction facilities for roads occupy small areas comparing with drainage basin areas, most of non-point pollutant sources would be removed through the facilities.

PURPOSES: The purpose of this study is to evaluate moisture susceptibility of a dense graded and an open graded asphalt mixtures by the method of AASHTO T-283. METHODS: To simulate moisture damage of asphalt pavements with continuously rainfall during summer rainy season, the dense graded and the open graded asphalt mixtures were immersed in water for 15 days and were measured the weight and the change of strength. Also, the mixtures were performed five freeze-thaw cycles to simulate moisture damage of the mixtures by freeze-thaw during winter and were measured the change of strength. The degradation characteristics model was used to analyze the relationship between strength and moisture damage. RESULTS: According to the results, the dense graded and the open graded asphalt mixtures were shown in the similar trends of the strength changes by immersion time and freeze-thaw cycle. However, the moisture damage reduction of open graded asphalt mixture was more sensitive in early phase than that of dense graded asphalt mixture.

일반국도유지보수 의사결정지원시스템에는 DB 이외에 1)포장상태지표의 공용성 모형, 2)포장상태지표 기반의 차량주행속도모형, 3)경제성 분석 모형과 4)의사결정지원 모형으로 구성된다. 특히 국토교통부의 「교통시설 투자지침서」에서 제시하는 도로이용자 및 사회·환경비용은 모두 차량운행속도의 함수로 구성되어있기 때문에 신뢰성 있는 차량운행속도모형의 개발이 요구된다. 즉, 도로의 상태기반 주행속도모형은 도로 부문의 편익 비용 항목 중 1)통행시간비용, 2)차량운행비용, 3)대기오염 비용, 4)온실가스비용, 5)소음비용 항목 등 이용자 비용 산정에 활용된다. 본 연구에서는 포장상태가 차량주행속도의 변수로 포함되도록 모형화 하는 것이 핵심이다. 먼저 유지보수 전·후의 『포장상태-차량속도』자료를 확보하기 위해 NC-200을 통해 현장조사를 실시하였다. 순창, 제천, 단 양, 증평의 네 구간을 선정하여 각 구간의 지형 분석, 교통량 측정 및 특성 분석을 실시하였다. 필터링 된 속도 자료를 활용하여 유지보수 전·후의 속도 차이를 AD(Anderson-Darling) 검정, t 검정, Mann-Whitney U 검정을 통해 통계적 유의성을 검정하였다. 모형개발을 위해 통계적 유의성이 검증된 구간 및 아스팔트 포장구간의 속도자료를 활용하여 표 1과 같이 모형을 개발하였다. 연구결과, 승용차를 대상으로 제한속도가 60km/h, 80km/h인 경우의 평지를 대상으로 한 모형구축 과정을 통해 세 가지 도로 상태(균열, 소성변형, IRI)별 속도는 통계적으로 유의한 차이가 있음을 확인하였다. 개발된 속도모형을 통해 도로 이용자비용의 산출이 가능하며 향후 일반국도유지보수 의사결정지원시스템의 경제성 분석 및 유지보수 대상구간의 우선순위 결정 등 다양하게 활용가능하리라 기대된다.

카쉐어링 시스템은 여러 나라에서 활성화 되어있고 이용하는 사람들이 지속적으로 증가하고 있다. 국내에서는 현재 서울을 중심으로 사업이 시행되고 있다. 국내의 카쉐어링 서비스 활성화를 위해선 우선 많은 사람들이 손쉽게 이용할 수 있는 공간에 서비스를 받을 수 있도록 위치를 선정하는 것이 중요하다. 따라서 본 연구에서는 대전시를 대상으로 카쉐어링 서비스를 제공하기 위한 위치 선정방안을 제시하는 것을 목적으로 한다. 카쉐어링의 입지 분석을 위해 GIS Arc/Info 10 소프트웨어를 활용하였고 GIS그리드 분석을 수행하여 주요 지점의 입지선호도를 순위별로 선정하였다. 그리고 주요지점을 바탕으로 GIS Arc/Info의 네트워크 분석 (Network Analysis)모듈에 내장되어있는 입지배분모델(Location-allocation Model)을 이용하여 주요지점의 입지를 선정하였다. 먼저 카쉐어링 입지분석을 위한 특성으로 교통적 요인, 인구요인, 잠재 수요력, 외부접근 성, 내부접근성으로 나누었고 각각의 특성을 구현할 수 있도록 공간데이터와 속성데이터를 구축하였다. GIS공간데이터를 얻기 위해 우선 KTDB의 전국 행정구역 및 도로망 자료를 바탕으로 카쉐어링 대상지역인 대전지역의 행정구역과 도로망 GIS공간데이터를 얻었다. 그리고 위성지도를 이용하여 분석에 필요한 대전지역 대학교, 기차역, 버스터미널, 지하철 위치를 각각 추가하였고 이를 바탕으로 주요지점 93곳을 선정하였다(주요 교차로 50개, 대학교 13개, 기차역 및 버스터미널8개, 지하철역 22개) 그리고 공간데이터에 들어갈 속성데이터로써 교통적 요인으로는 대전광역시 교통조사 및 분석보고서(2012)의 첨두시간 교통량데이터(대/시)를 이용하였고, 인구요인으론 행정동별 인구밀도(명)와 대학교 재학생(명) 수를 이용하였다. 잠재 수요력으로는 전국 여객 기종점 통행량 조사(2011) 자료의 동 별 자동차 보유대수를 이용하였다. 다음으로 각각의 레이어를 5x5셀 크기의 래스터 자료로 변환 한 후 서로 다른 요인들의 측정치를 통일시키기 위해 각 요인들의 측정단위 들을 퍼지논리에 의해 0과 1사이 의 실수로 분류(S형 증감형태)하여 표준화 하였다. 그리고 각각의 레이어에 가 중치를 부여한 후 각각의 레이어 중첩을 통한 셀 별 연산을 수행하여 주요지점의 우선순위를 결정하였다. 마지 막으로 주요지점 93곳을 공급 후보지로 정하고 각 동 별 중심지를 수요지점으로 선정하여 GIS프로그램내의 입 지배분 모델 중 min-distance model을 적용하여 주요지점의 입지를 선정하였다. 본 연구를 통해 향후 카쉐어링 서비스 도입 시 위치선정에 있어 주요지점의 우선순위는 물론 수요를 최대화 할 수 있는 서비스지역을 선정하는데 기초자료가 될 것으로 기대된다.

현재, 국도 포장관리시스템(PMS; Pavement Management System)에서 HDM(Highway Development and Management)-4를 이용하여 예산 분석 및 보수우선순위 검토, 경제성 분석을 시행하고 있다. 하지만 분석에 필요한 입력 자료가 우리나라 현실에 맞지 않고, 공용성 모형의 보정이 어려우며 또한 분석결과의 신뢰성 등에 의문이 제기되면서 국내현실에 적합한 의사결정시스템의 개발이 필요하게 되었다. 본 연구에서 개발된 국도 포장관리시스템(PMS)의 의사결정지원시스템의 구성요소 크게 1)도로포장상태의 공용성 모형, 2) 도로포장상태에 따른 차량속도변화 추이모형, 3)경제성 평가시스템으로 구성되어져 있다. 먼저, 공용성 모형이란 포장의 파손과정에 대한 해석을 다루는 내용으로 유지보수가 요구되기까지의 기대수명과 그 기대수명 내에서의 파손특성의 변화과정을 표현하는 것으로 정의할 수 있으며, 본 연구에 서 공용성 모형은 균열, 소성변형, 종단평탄성에 대한 년간의 파손량을 함수로 모형화하였다. 또한, 도로포장상태에 따른 차량속도변화 추이모형은 국도의 유지보수 전·후 “포장상태-차량속도”자료 를 모형화하였다. 마지막으로 경제성 평가 시스템은 관리자 및 이용자 비용의 산정과 우선순위 선정기법으로 구성되었는데, 본 시스템에서의 관리자 비용은 유지보수 비용으로 한정하여 산정하였다. 이용자비용은 국토교통부의 ‘교통시설 투자 평가지침’을 기준으로 도로편익비용 항목 중 1)통행시간비용, 2)차량운행비용, 3) 대기오염 비용, 4)온실가스비용, 5)소음비용을 기준으로 산정하였다. 우선순위선정법으로는 연등가법(EUAC; Equivalent Uniform Annual Cost), NPV(Net Present Value)법과 비용효율성법(Cost-effectiveness Method)을 활용하여 우선순위를 산정하였다. 본 연구에서 개발한 의사결정시스템은 마이크로 소프트 엑셀(Microsoft Excel)의 VBA(Visual Basic Application)를 기반으로 이용자의 보다 쉬운 이해와 활용, 확인 및 보정/수정이 가능하도록 설계되었다. 분석시스템은 활용목적에 따라 “프로젝트 레벨”과 “네트워크 레벨”로 구성(그림 1. 참조)하였으며, 주요 기능으로는 1) 유지관리비 산정 및 예측, 2) 유지보수 대상구간의 우선순위 결정, 3) 유지보수시기(지연) 에 따른 도로관리자 및 이용자비용의 변화추세 예측, 4) 경제성지표 도출 등이 있다.

이 연구의 목적은 백색 LED 조명의 광학적 특성을 분석하고 그 특성에 따라 반응하는 학생과 교사들의 감성 반응을 평가하는 데 있다. 이를 위해 이 연구에서는 백색 LED모듈 5개를 시제품으로 만들어 광학적 특성을 측정하고 감성 평가용 모형을 제작하여 감성 반응을 평가하여 분석하였다. 감성 평가에 이용된 평가 방법은 의미미분법이고 선정된 문항은 16문항이며, 이 문항의 신뢰성과 타당성을 검증하기 위하여 예비 실험을 통하여 신뢰도 분석과 타당도 분석을 하였다. 이 과정에서 4가지 요인을 추출하였는데 제 1요인은 활동성, 제 2요인은 안정감, 제 3요인은 역량성, 제 4요인은 감성이미지 요인이라고 명명하였다. 색온도에 따른 감성 평가의 결과는 활동성과 역량성 요인에서는 색온도가 높은 조명을 선호하였으며 , 안정감 요인에서는 색온도가 낮은 조명을 선호하였다. 감성이미지 요인에서는 색온도와 관련 없이 5800K인 청색 계통의 조명을 선호하였다. 연색지수에 따른 감성 평가의 결과는 활동성, 안정감, 감성이미지 요인에서는 고연색 조명을 선호하였으며, 역량성 요인에서는 중 연색 조명을 선호하였다.

Robots are used in various industrial sites, but traditional methods of operating a robot are limited at some kind of tasks. In order for a robot to accomplish a task, it is needed to find and solve accurate formula between a robot and environment and that is complicated work. Accordingly, reinforcement learning of robots is actively studied to overcome this difficulties. This study describes the process and results of learning and solving which applied reinforcement learning. The mission that the robot is going to learn is bottle flipping. Bottle flipping is an activity that involves throwing a plastic bottle in an attempt to land it upright on its bottom. Complexity of movement of liquid in the bottle when it thrown in the air, makes this task difficult to solve in traditional ways. Reinforcement learning process makes it easier. After 3-DOF robotic arm being instructed how to throwing the bottle, the robot find the better motion that make successful with the task. Two reward functions are designed and compared the result of learning. Finite difference method is used to obtain policy gradient. This paper focuses on the process of designing an efficient reward function to improve bottle flipping motion.

Background : The development of an antioxidant to prevent disease by ROS-induced oxidative stress is necessary. This study investigated the changes of antioxidant capacities of two medicinal crops extracts by lactic acid fermentation. Methods and Results : The changes of free-radical scavenging activity of medicinal crops extracts by lactic acid fermentation were evaluated by using DPPH free-radical scavenging assay and ABTS free-radical scavenging assay. The DPPH free-radical scavenging activity of extracts or lactic acid fermented extracts were estimated as followed. samples were thoroughly mixed with 1 ㎖ ethanol solution of 0.1 mM DPPH. After stand for 30 min in the dark, the absorbance was measured at 570 ㎚ by using a UV Spectrophotometer. ABTS scavenging activity of extracts or lactic acid fermented extracts were estimated as followed. The working solution was prepared by mixing 1 ㎖ of ABTS solution with 88 ㎖ of 50% ethanol. A total of 25 ㎕ of samples were mixed with 225 ㎕ of ABTS working solution and allowed to stand for 10 min. The absorbance was read at 732 ㎚ in a UV spectrophotometer. The data were showed that lactic acid fermented extracts were higher antioxidant ability than the extracts. Conclusion : This study was showed that the antioxidant capacities of two medicinal crops extracts were improved by lactic acid fermentation.