By: Eugene Kirpichov, Sasha Luccioni & David Rolnick, Conference Chairs, Predictive Analytics World for Climate
In anticipation of his upcoming presentation at Predictive Analytics World for Climate Livestream, May 24-28, 2021, we asked Daniel Rohr, Senior Data Scientist at Tracks GmbH, a few questions about their deployment of predictive analytics. Catch a glimpse of his presentation, Tracks For Trucks: Turning Data Science into CO2 Savings, and see what’s in store at the PAW Climate conference.
Q: In your work with predictive analytics, what behavior or outcome do your models predict?
A: Tracks’ mission is to reduce the CO2 emissions in the road freight industry, eventually decarbonizing road freight. To this end we are developing an artificial intelligence that will predict fuel consumption of trucks.
Q: How does predictive analytics deliver value at your organization – what is one specific way in which it actively drives decisions or operations?
A: We use our technology to help our clients to do efficiency dispatch. We predict the fuel consumption on different routes, with different trucks and different drivers. Our clients can then choose the most efficient combination for dispatch.
Q: What surprising discovery or insight have you unearthed in your data?
A: We currently focus on two types of clients: Carriers, that is, companies that own their own trucks. These carriers can benefit from our services by reducing their fuel consumption. Shippers are companies that use the services of carriers to send shipments from A to B. Shippers benefit from our accurate measurement of CO2 emissions. This has a direct result on their CO2 reporting, accurate measurements can be used to reduce CO2 emissions and consequently reduce the cost associated with offsetting these emissions.
Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.
A: One of the areas we focused on was truck/driver combinations. We invented an algorithm that quantifies the efficiency of truck/driver combinations. With this algorithm we were then able to identify good and bad pairings, and these results even surprised veterans in the industry.
Q: Additional Questions: What adjacencies do you see to machine learning for making predictions useful in your domain?
A: Probably the most important take-away and message from my talk is to use natural intelligence rather than artificial intelligence. It’s more important to thoroughly think about your problem, instead of throwing the most sophisticated Machine-Learning model at it.
—————————–
Don’t miss Daniel’s presentation, Tracks For Trucks: Turning Data Science into CO2 Savings at PAW Climate on Tuesday, May 25, 2021 from 10:20 AM to 11:05 AM. Click here to register for attendance.
By: Eugene Kirpichov, Sasha Luccioni & David Rolnick, Conference Chairs, Predictive Analytics World for Climate
 |
 |
 |
