PURPOSES : In this study, we aim to broaden the understanding of the factors influencing the accuracy of WIM systems for overload enforcement. Particularly, we explored the proportions and causes of secondary influencing factors (driving path, vehicle class, and acceleration), which have been relatively less studied and reduced the accuracy of the WIM system. METHODS : Overload enforcement data were recorded by the WIM system, and enforcement officers were gathered. The ratios of each data point, which are the relative errors, are used to estimate the accuracy of the WIM system. These relative errors were classified into four driving-path groups, four vehicle-class groups, and three acceleration groups. The change in the accuracy of the WIM system caused by each influencing factor was analyzed by comparing the difference in the average relative error between the classified groups. Analysis of variance (ANOVA) and Welch's ANOVA were used to determine significant differences between groups. RESULTS : Vehicles departing from a normal driving path make it difficult for the GVW compensation algorithm of the WIM system to operate properly. For these abnormal paths, the standard deviation of the average GVW relative error was 22%. There was no specific trend in the difference in accuracy by vehicle class. However, we found that the rear axle and retractable axle were the main causes of the reduced GVW accuracy in each vehicle class. The average GVW relative error remained the same regardless of the acceleration, but the average FAW relative error of the accelerated vehicle was approximately 2.5% lower than that of the unaccelerated vehicle. CONCLUSIONS : An abnormal driving path, lifting of a retractable axle, and rapid acceleration (or deceleration) reduce the accuracy of WIM systems. Intelligent transportation systems, such as traffic signals, telematics devices, and applications that induce desirable driving are required for effective overload enforcement. Additionally, it is necessary to smoothen the road pavement to minimize the dynamic effects on the rear axle.

1. 서론
2. 연구방법
    2.1. 연구대상 선정 및 데이터 수집
    2.2. WIM 시스템의 정확도 표기
    2.3. 검측영향인자 조사 및 구분
    2.4. 검측영향의 통계분석
3. 연구결과 및 검토
    3.1. 주행경로별 검측영향 분석결과
    3.2. 차종별 검측영향 분석결과
    3.3. 가속도별 검측영향 분석결과
4. 과적단속 시스템의 운용방안 논의
    4.1. 회피 주행 통제
    4.2. WIM 시스템의 설치장소 관리
5. 결론
감사의 글

• 이동해(정회원 · 한국건설생활환경시험연구원 내진센터 연구원) | Lee Donghae (Department Seismic Safety Center, Korea Conformity Laboratories, 73, Yangceong 3-gil, Ochang-eup, Cheongju-si, Chungbuk 28115, Korea)
• 이종석(한국건설생활환경시험연구원 내진센터 선임연구원) | Lee Jongseok (Department Seismic Safety Center, Korea Conformity Laboratories, 73, Yangceong 3-gil, Ochang-eup, Cheongju-si, Chungbuk 28115, Korea)
• 조재우(한국건설생활환경시험연구원 방재화재본부 본부장) | Cho Jaewoo
• 김종우((주)유디엔에스 기술연구소 CTO) | Kim Jongwoo
• 김태상(정회원 · 한국건설생활환경시험연구원 내진센터 센터장) | Kim Taesang (Department Seismic Safety Center, Korea Conformity Laboratories, 73, Yangceong 3-gil, Ochang-eup, Cheongju-si, Chungbuk 28115, Korea) Corresponding Author