Aimsun Predict
Process real-time data to forecast upcoming traffic states: receive alerts for sudden anomalies or problems, and support proactive decisions.
Aimsun Predict
Process real-time data to forecast upcoming traffic states: receive alerts for sudden anomalies or problems, and support proactive decisions.
By their very nature, technologies that support intelligent transportation programs capture massive amounts of real-time data.
Aimsun Predict uses techniques such as data cleaning, clustering, prediction, and incident detection to make sense of this real-time data, provide situational awareness and forecast future traffic trends.
In a tight timeframe (typically four to 10 weeks), Aimsun Predict can help cities achieve major goals for their transportation programs.
Aimsun Predict – Top benefits
Real-time situational awareness of network performance: increase network management capabilities
Dashboard with customized metrics and KPIs: visualize the most relevant outputs in your preferred format
Alerts for unusual network performance in real-time: increase responsiveness to incidents, including breaches of air quality thresholds
Incident detection alerts: greater responsiveness to non-recurrent events or incidents
Data propagation using AI: support event planning to increase customer satisfaction and optimize resourcing and management
Forecast of evolution: predicts future traffic trends to identify performance, safety or environmental issues to support infrastructure improvements and operational planning
Quantative network performance: support operational decisions
Aimsun Predict – Top use cases
- Online data cleaning: homogenizing data as it comes in from permanent sensors and preparing it for real-time applications.
- Network state estimation: extending measures coming in from sensors at specific locations to the entire road network.
- Network state prediction: predicting the evolution of the traffic conditions over the next few hours.
- Network performance alerts: monitoring the performance of the transportation system and alerting traffic operators if it diverges from the usual pattern.
- Online incident detection: detecting sudden changes in traffic data that may be the symptom of an incident, e.g., a sharp drop in traffic flow outside peak hours might indicate congestion due to a crash or collision.
- Incident risk prediction: detecting and reporting situations that correspond to high risk of incidents, always having access to weather and traffic data in real time.
- Air quality prediction: predicting the evolution of air quality in the upcoming hours and days.
- Situational awareness and monitoring: providing a web-based dashboard updated in real-time to monitor the state of the transportation system.
- Parking occupancy prediction: predicting parking occupancy in different areas over the next hour.
- Bus travel time prediction: predicting what time buses will arrive at stops.
- Online data cleaning: homogenizing data as it comes in from permanent sensors and preparing it for real-time applications.
- Network state estimation: extending measures coming in from sensors at specific locations to the entire road network.
- Network state prediction: predicting the evolution of the traffic conditions over the next few hours.
- Network performance alerts: monitoring the performance of the transportation system and alerting traffic operators if it diverges from the usual pattern.
- Online incident detection: detecting sudden changes in traffic data that may be the symptom of an incident, e.g., a sharp drop in traffic flow outside peak hours might indicate congestion due to a crash or collision.
- Incident risk prediction: detecting and reporting situations that correspond to high risk of incidents, always having access to weather and traffic data in real time.
- Situational awareness and monitoring: providing a web-based dashboard updated in real-time to monitor the state of the transportation system.
- Parking occupancy prediction: predicting parking occupancy in different areas over the next hour.
- Bus travel time prediction: predicting what time buses will arrive at stops.
Case study
AI-based traffic predictive systems and advanced incident detection
C-32 highway in Spain
Client: Abertis
Aim: a real-time traffic monitoring and forecasting solution to preempt incidents and test new technologies on the C-32, a primary highway in Catalonia, Spain.
- The solution offers short- and mid-term predictions of flow and speed along each segment of the highway, fusing data from loop detectors, toll plazas and probe vehicles.
- Traffic operators see alerts whenever the upcoming traffic conditions are similar to those that in the past led to incidents. They can therefore disseminate warnings and prevent the incidents.
- The solution aids in assessing the collision risk on each highway segment by integrating critical information, such as the current flow and speed.
- Monitoring and forecasting pollution levels includes calculation and prediction of vehicle emissions and speed recommendation.
