Construction of a predictive model of pedestrian crossing traffic in a mountainous region taking into account weather and time factors
Year: 2024
Type: article
Abstract: Pedestrian traffic in mountainous regions is significantly affected by weather conditions and time factors. Optimization of urban infrastructure and ensuring pedestrian safety require forecasting traffic changes depending on climate and time of day. The purpose of this study is to develop a predictive model that takes into account key factors affecting pedestrian traffic intensity. The study collected data on pedestrian traffic in the mountainous region, including parameters: air temperature, precipitation, day of the week, time of day, proximity to public transport stops and large objects. Machine learning and statistical analysis methods were used to process the information. Data visualization and model construction were performed using pandas DataFrame. A multivariable regression model validated using the coefficient of determination (R2) and root mean square error (RMSE) was used as a forecasting tool. The study showed that pedestrian traffic has a clear dependence on weather conditions. The optimal temperature for pedestrian activity is 15–25°C, while at temperatures below 0°C and above 30°C, a significant decrease in traffic is observed. Precipitation reduces the number of pedestrians by 10–30% during rain and by 20–40% during snowfall. The highest activity is recorded on weekdays in the morning (8:00–10:00) and in the evening (17:00–19:00), while traffic decreases on weekends. The accuracy of the model is confirmed by the value R2 = 0.89 and RMSE = 54.32. The obtained results demonstrate the importance of taking into account seasonal and temporal factors when organizing the urban environment. The use of a predictive model allows for the infrastructure to be adapted in advance to changes in weather conditions, minimizing the impact of unfavorable factors on pedestrian traffic. The stability of temporal patterns indicates the possibility of effective planning of urban resources. The developed model for predicting pedestrian traffic in mountainous regions is highly accurate and allows predicting traffic changes depending on weather conditions and time factors. Practical application of the model helps to improve pedestrian safety, optimize the transport network and improve the quality of the urban environment. The predictive model of pedestrian crossing traffic takes into account key climatic and time parameters that affect traffic intensity. Data analysis revealed patterns of pedestrian behavior, which allows using the model for planning urban infrastructure and improving road safety. The developed model can be used by city services to adapt road infrastructure to seasonal changes, improve travel conditions and ensure pedestrian safety.
Language: English
Subfield: Transportation
Field: Social Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4408831383