Predicting Density in Weaving Areas by Developing Trajectory Data-Based Models Using Artificial Neural Network and Particle Swarm Optimization

Density in weaving areas is a corresponding value which represents the level of operation in these areas. Therefore, in this paper, it was attempted to predict the density of weaving areas by simulating 10368 different types of weaving area with different geometry and different traffic characteristics. Density was obtained for each weaving area by analyzing the trajectory data. A database containing geometric and traffic characteristics as variables and density as a function was generated by determining density in all instances. By using this database, two models were developed to predict the density of these areas using Artificial Neural Network and Particle Swarm Optimization algorithm. The models were tested, validated and their errors were checked. The results showed a good accuracy of similarity between the results of models in predicting the density of weaving areas and that of simulations. Six weaving areas as case studies were surveyed to verify the models and statistical analysis indicated an acceptable accuracy and validity of the models.