This study proposes a digital twin–based simulation environment that incorporates real-time weather variations to support future research on autonomous driving and mobility simulation. Conventional traffic simulation environments typically rely on fixed scenarios, which limits their ability to capture dynamic external conditions during runtime. To address this limitation, an integrated simulation environment that couples microscopic traffic simulation with a three-dimensional virtual environment is required. The proposed environment integrates the microscopic traffic simulation platform SUMO with the 3D visualization platform Unity through an API-based data exchange interface. The target road network was modeled using field-surveyed geometry, traffic volumes, and signal operation parameters. Time-series rainfall data were fed into the system as an external input, and the saturation headway parameter was dynamically updated as a function of rainfall intensity to represent traffic flow variations under changing weather conditions. To evaluate the model's applicability, a comparative analysis was conducted between a baseline SUMO-based simulation and the proposed real-time responsive model under identical traffic and operational conditions. The results show that the proposed model continuously captured rainfall variations over time, yielding dynamic fluctuations in control delay, whereas the baseline simulation retained fixed parameter values throughout the run. Unlike conventional fixed-scenario simulations, the proposed framework enables the continuous integration of external environmental changes during simulation execution. These results demonstrate that the proposed framework successfully implements a real-time data integration structure that supports timevarying traffic operation analysis. The proposed digital twin–based simulation environment is expected to be applicable not only to traffic operation analysis under diverse weather and road conditions but also as a foundational platform for future mobility and traffic operation research.

ABSTRACT
1. 서론
2. 문헌 고찰
3. 연구 방법
    3.1. 디지털 트윈 기반 교통 시뮬레이션 환경 구성
    3.2. 디지털 트윈 기반 모의실험 플랫폼 구축
    3.3. 디지털 트윈 기반 모의실험 연계 체계 구축
4. 연구 검증
    4.1. 검증 환경
    4.2. 검증 결과
5. 결론
REFERENCES

• 김영율((주)유에이치알앤디 대표이사) | Kim Young-Yul
• 김진태(국립한국교통대학교 교수) | Kim Jin-Tae (Professor Korea National Univ. of Transportation, Graduate Shcool of Transportation, Dept. of Transportation Systems Eng., 157 Cheoldobangmulgwan-ro, Uiwang, Gyeonggi-do 16106, Republic of Korea) Corresponding author
• 오상태(국립한국교통대학교 교통시스템공학과 박사후연구원) | Oh Sang-Tae