Abstarct: بومی ارزیابی ایمنی عابرین پیاده است که می‌تواند در تحلیل ریسک، بازطراحی هندسی تقاطع‌ها و بهینه‌سازی زمان‌بندی چراغ‌ها برای ارتقای ایمنی عابران پیاده در شهرهای ایران مورد استفاده قرار گیرد. کلمات کلیدی: ایمنی عابر پیاده، تداخل عابر–وسیله نقلیه، شبیه‌سازی خردنگر، SSAM، TTC، PET، شبیه‌ساز رانندگی، تقاطع چراغ‌دار Abstract Pedestrians, as the most vulnerable group of road users, account for a considerable proportion of fatalities and injuries resulting from traffic crashes. A significant portion of these accidents occurs at signalized intersections, where the complexity of movement patterns and the diverse behaviors of drivers and pedestrians increase the likelihood of conflicts and collisions. The pedestrian safety situation in Iranian cities, including Qazvin, is unsatisfactory, highlighting the need for analytical and simulation-baxsed models to accurately assess and enhance pedestrian safety. The aim of this research is to develop a comprehensive frxamework for evaluating pedestrian safety at signalized intersections through the integration of field data, microscopic traffic simulation, and driving simulator experiments, in order to reproduce and analyze real traffic behavior and pedestrian–vehicle interactions with high precision. In the first phase, field data were collected from seven signalized intersections in Qazvin during peak daytime and nighttime periods. Simulation models were then developed in VISSIM and calibrated and validated using the observed data. Pedestrian–vehicle conflicts were analyzed using SSAM, and the surrogate safety indicators Time to Collision (TTC) and Post-Encroachment Time (PET) were extracted and compared with field observations. The results indicated that at thresholds of TTC = 2.7 seconds and PET = 8 seconds , the Mean Absolute Percentage Error (MAPE) reached its minimum, reflecting the highest level of agreement between simulated and observed data. Although the VISSIM–SSAM model demonstrated statistically significant correlations with the real-world data, the simulated number of conflicts was generally lower than field observations, suggesting a tendency of the model to underestimate pedestrian–vehicle conflict risk. Subsequently, driving simulator experiments were designed to provide a deeper understanding of driver and pedestrian behavior in hazardous situations. The effects of driver individual factors (age and gender) and environmental variables (time of day, road type, crosswalk marking condition, pedestrian clothing color, vehicle movement, and pedestrian movement) on key parameters— including maximum deceleration rate, braking reaction time, TTC, and PET were analyzed using mixed modeling and )ANOVA( . The findings revealed that age, gender, pedestrian clothing color, and the movements of both pedestrians and vehicles significantly affected drivers’ deceleration and reaction behavior. Younger and male drivers, under good visibility conditions and when encountering pedestrians wearing bright clothing, reacted faster and exhibited higher deceleration rates. Moreover, roadway type (two-lane or three-lane) and the presence of crosswalk markings had a direct influence on the intensity of driver reactions and avoidance behavior. It was also found that drivers over 40 years old and pedestrians wearing dark clothing exhibited the longest brake-release times, increasing the likelihood of collision. Finally, by integrating field observations, traffic simulation results, and driving simulator outcomes, a predictive model for pedestrian–vehicle conflicts was developed. This model estimates the probability of conflict occurrence baxsed on behavioral and geometric characteristics of intersections. The main innovation of this study lies in the integration of three levels of data—empirical, simulated, and behavioral—and the development of a localized frxamework for pedestrian safety evaluation. This frxamework can be applied to risk analysis, geometric redesign of intersections, and optimization of signal timing to enhance pedestrian safety in Iranian cities.