North Carolina Department of Transportation
Research & Innovation Summit – 2021

Microsimulation-based model and SVM-ARIMA hybrid model for traffic condition prediction

Authors: Benjamin Lartey, Dr. A. Homaifar, Dr. A. Karimoddini
Dept. of Electrical and Computer Engineering, North Carolina A&T State University
View or Download PDF

Objectives

  • To develop a microscopic simulation model of a specific area using Aimsun Next traffic simulation software.
  • To extract simulation data which comprises travel time, delay time and speed (collected every 15 minutes).
  • Develop a hybrid machine learning (ML) model for predicting the travel time, delay time and speed over some time interval.
  • Evaluate and compare the performance of the hybrid model with other ML models.