The next Data2Move event is all about big data in transport. As our special guest we welcome Prof. Stefan Minner from the TU in Munich. Among other things, we will discuss how historical data can reduce failed deliveries and how a better understanding of travel times can lead to better routing decisions. For this latter topic, we would like your help. When you fill out the registration form please note the extra question we’ve included. We ask you to guess the average speed of a truck on the Dutch road during the day (from door to door). We will compare your values with data of realized trips we have received from a retail distribution partner of Data2Move. So be aware that the trips are not only on highways but also in inner cities.
Get your tickets here!
Save the date: 19-02-2019 from 12:00 – 18:00
Location: Evoluon, Noord Brabantlaan 1A, 5652 LA Eindhoven
13:00-13:15 Welcome by Luuk Veelenturf (TU/e)
13:15-14:15 Workshop Part I: Travel time data analysis
14:15-15:00 Keynote by Stefan Minner (TUM) on routing with uncertain travel and service times
15:30-16:00 Results of Workshop Part I
16:00-16:30 Workshop Part II: One last time to improve your estimates
16:30-17:15 Wrap up and next steps
Read an abstract on what prof. Stefan Minner is going to talk about here below!
A Data-Driven Approach for Routing of Delivery Services under Uncertain Travel and Service Times
The trend to introduce highly customer-oriented services that requires a delivery within a certain time frame to a certain customer poses a great challenge for logistics service providers. In logistics, this phenomena is better known as the Vehicle Routing Problem with Time Windows.
In today’s keynote, Stefan Minner addresses the question “How can Big Data be used to improve the routing decision given the uncertain traffic and service conditions?” The subject of this keynote is derived from research that Minner has conducted with Szymon Albinski.
Companies are not just dealing with, often tight, time frames to deliver their products to their customers but they also have to plan their routes in such a way that they become cost effective. One of the main hurdles to take is that travel and service times are never the same and that they are subject to random and changing conditions such as the weather, congested roads or road works. These uncertainties need to be factored in.
In their research, Minner and Albinski propose a model that represents these uncertainties in a non-parametric way based on historical data. By integrating the steps of prediction and optimisation and not assuming set travel times, it allows them to factor in uncertainties such as for instance the weather conditions on the travel and service times.