PILOT : SMART HIGHWAYS IN AUSOL

The initial pilot’s scenario is the 105-kilometre Ausol Highway, a toll highway with barriers that belongs to the AP-7 highway, which is part of the European route E15, Costa del Sol.

This highly-congested semi-urban corridor connects the cities of Málaga, Estepona and Guadiaro in the South of Spain.

The concession runs parallel to the N340/A7, a free alternative dual carriageway with two lanes per direction and grade or signalized intersections at some points along the road. Pilot will focus not only in the Ausol Highway, but also in the free alternative.

Ausol highway pilot aims to measure the technical and economic impacts of Big Data technologies on operation in highways concessions by using big data technologies will be applied to meet the following three main business goals:

• Understanding the road traffic.
• Optimize highway operation.
• Guarantee safer roads.

For reaching those purposes, the IoT infrastructure will integrate different data sources from Highway’s ITS and different Project stakeholders (public and private entities: DGT, TOMTOM, Waze, Yahoo OpenWeather, Twitter…), where information about traffic, road accidents, meteorological conditions and other variables will be collected – both historical and real-time.

Combining this information can provide a good knowledge of the traffic status in both spatial and temporal dimensions.

On top of that, new information will be collected through a handful of new technologies and connected in real time both to the ITS for operation, and the Big Data/IoT Platform for prediction.

The main expected results of this pilot are:

• Optimize highway operation with prediction models. A top priority goal to get at Ausol is the prediction of traffic in the short term, from 15 minutes to 2 hours. First models at work show a good accuracy up to 60 minutes predictions; and from 60 to 120 minutes, which will become essential when toll stations should support high traffic peaks – as reversible lanes might need to be opened to reduce drivers’ waiting times.
• Understanding the road traffic is provided by online dashboard for displaying all the information in a user-friendly dashboard, which is, stablished in a Cloud service, available for different platforms: Laptop, Tab/Pad and Mobile Phone. Several add-ons / gadgets enable users to add new features and personalize user experience; main functionality is the traffic curve, which shows historical information as well as predicted traffic in a segment of the road.
• Guarantee safer roads thanks to an index model based on a descriptive analysis has been carried using several variables in order to reflect the traffic situation like the average daily traffic calculated at every step with the predictive model, day type, climatology.

The main expected results of this pilot are:

• Optimize highway operation with prediction models. A top priority goal to get at Ausol is the prediction of traffic in the short term, from 15 minutes to 2 hours. First models at work show a good accuracy up to 60 minutes predictions; and from 60 to 120 minutes, which will become essential when toll stations should support high traffic peaks – as reversible lanes might need to be opened to reduce drivers’ waiting times.
• Understanding the road traffic is provided by online dashboard for displaying all the information in a user-friendly dashboard, which is, stablished in a Cloud service, available for different platforms: Laptop, Tab/Pad and Mobile Phone. Several add-ons / gadgets enable users to add new features and personalize user experience; main functionality is the traffic curve, which shows historical information as well as predicted traffic in a segment of the road.
• Guarantee safer roads thanks to an index model based on a descriptive analysis has been carried using several variables in order to reflect the traffic situation like the average daily traffic calculated at every step with the predictive model, day type, climatology.

The main KPIs of the pilot are:

• Understanding the road traffic, where the main objective is a better understanding of traffic flows and drivers’ behavior along the corridor (commuters, heavy users, migration flows,..) and its impact on infrastructure operation.

• Optimize highway operation. Based on real time traffic and other variables, we have two main objectives:
  • To predict incident probability of any specific segment of the road in the next 2 hours, so that maintenance vehicles can re-schedule their tasks and stay as near as possible to reduce reaction time.
  • To predict traffic peaks approaching toll stations (with barriers) in the next 2 hours, so that operator can re-configure lanes and its linked payment type to adequate capacity to demand.

• Guarantee safer roads, by reducing the number of accidents, improving response times to accidents and reducing its impact on traffic.