How can customer expectations such as reliability and in-time delivery processes be fulfilled? How can logistics stakeholders deal with unexpected situations in day-to-day business? There are many challenges logistics actors face in a highly competitive arena. Within the Transforming Transport research project, the Sustainable Connected Trucks pilot is dealing with these challenges by exploiting Big Data analytics.

Better understanding of planning processes with analysis of travel times and handling activities

One important aspect for this pilot is to enhance planning and optimisation systems for planners, fleet managers or technical systems. This requires assessing the transportation process and all its aspects, including the pick-up of orders at designated locations, routing, driving and delivery. Gaining better understanding of these processes involves investigating the traffic flow of truck journeys, as well as identifying and analysing logistics hotspots such as terminals, toll stations and ferry stations.

This requires large amounts of Big Data processing specifically related to truck fleets all over Europe. In this context, the pilot is concentrating on a defined truck corridor between Amsterdam and Frankfurt with alternative driving routes and different considerations such as travel times, traffic situations and airport operating times which must all be analysed. Understanding all these aspects can help improve logistics planning.

Improved accuracy of the estimated time of arrival for truck journeys

Planning the transportation process entails more accurate estimation of arrival times at delivery locations or interim spots, as well as good understanding of travel times and handling activities. Without this information, large buffer times at these locations must be included to account for travel time losses. To address this challenge, the pilot is developing a stochastic routing approach that improves the reliability of long-term travel time predictions and allows logistic companies to optimise the needed buffer time for in-time arrival. This is being achieved by analysing historical data and calculating delays on the pilot corridor between Frankfurt airport and Amsterdam airport.

To illustrate, the figure below displays possible alternative routes with a snapshot on a Monday morning. The colours on the route indicate the travel time loss probability for each segment of the route, showing how this number increases from low (blue, 0-20 %) to high (red, 80-100 %). Around morning rush hour peak, large parts of the routes are not affected. However, areas around Arnhem, Belgium and Düsseldorf show a high probability. With the help of the pilot’s newly developed network prediction model, different routes can now be compared in terms of reliability and travel time losses.

New insights for truck traffic and long-term changes based on satellite images

In addition to analyses based on truck data and software components for routing and travel times, satellite images play an important role in the pilot. The use of this additional data source provides further opportunities not only for actors in the transportation sector but also for technical end-users who can use the images to validate their own data. Satellite images can help analyse the current traffic situation or state of a location, as well as changes over time. Therefore, fields of application within and beyond the project include landslide forecasts, air quality mapping and the identification of vehicle types.

On another front, satellites can also help identify the speed of vehicles with the help of a single picture (figure below). This involves identifying different spectral ranges of the light to estimate speed.

Overall, the pilot opens a world of possibilities for using Big Data analytics to better understand transportation planning processes, estimate arrival times more accurately and gain new insights based on satellite images. It is indeed very interesting to note what’s possible with the help of Big Data and what opportunities it can provide in terms of future developments and fields of application.