•  
  •  
 

Article Type

Original Study

Abstract

In developing metropolises, there is a risk to public safety in addition to the usual traffic jams. Traffic delays can have life-altering effects every second that fire trucks and ambulances are delayed. Large cities like Baghdad, which suffer from haphazard road sealing and incompetent traffic management, are especially affected, creating daily emergency crisis problems. We create a new intelligent routing system for these top priority vehicles in order to solve this life-or-death problem. Our method uses a trained model with real-time spatial information instead of outdated static maps. A Long Short-Term Memory (LSTM) neural network at its core constantly forecasts the flow of traffic in the near future. Time-Dependent A* (TDA*), a dynamic path planning algorithm, uses this predictive power to identify the actually fastest path at each decision point. We conducted a number of experiments and used the SUMO simulation to create a comprehensive digital structure of a region of Baghdad in order to verify our proposal. The outcomes were striking. Our AI-based solution reduced response times by an impressive 38.8% on average during peak travel hours, and the overall distance traveled was shortened by 25.4% compared to shortest-path methods in use today. The results we have shown provide strong evidence for our initial claim: an intelligent, future-aware routing system can make urban emergency-response substantially more effective and reliable. We therefore deduce that there is no longer anything theoretical about this model's efficacy, and that it should now be seen as a practicable way to address one of Iraq's most urgent urban problems, in what could become a cost-effective phased approach to smart city infrastructure development for the country.

Keywords

Intelligent Transportation Systems (ITS), High-priority vehicle routing, Artificial intelligence, Spatial data analytics, Smart cities, Iraq

Download

Share

COinS