Study Estimating hourly traffic flow using Artificial Neural Network: A M25 motorway case


  • Ahmed Ibrahim Turki University of Samarra
  • Saad Talib Hasson Information Networks Department, College of information Technology, University of Babylon



Regression, traffic flow estimation, machine learning, multi-layered neural networks, technical indicators.


This paper examines the challenge of accurately computing highway performance measures by estimating traffic-flow between traffic sensors that are geographically dispersed. Consequently, predicting flows vehicle values is one of the most difficult issues in the field of traffic flow prediction. Therefore, there has been a rise in interest in combining machine learning (ML) methods with indicators from technical analysis. In this paper, we suggest a hybrid strategy for generating traffic to help with this issue. Our proposed method utilizes a technical indicator and an ANN technique to predict future flows. That this method can be applied to other technical indicators while still maintaining its simplicity and effectiveness thanks to the hybrid rules is what makes it novel. The performance of the proposed artificial neural network (ANN) was evaluated with a number of other machine learning techniques to help us choose the optimal ML approach. Daily traffic data from the Motorway Incident Detection and Automatic Signalling (MIDAS) system on the M25 highway was used to test the proposed method. The achieved results demonstrate that the predictive power of ML models is augmented when ML techniques are applied to technical analysis indicators.




How to Cite

Turki, A. I. ., & Hasson, S. T. . (2023). Study Estimating hourly traffic flow using Artificial Neural Network: A M25 motorway case . Samarra Journal of Pure and Applied Science, 5(1), 47–59.