VeRoLog is excited to launch its new initiative: the VeRoLog Webinar. It takes place every last Thursday of each month, from 4 p.m. to 5.30 p.m. (CEST) and it will feature plenary talks, paper presentations from academia and/or industry as well as interviews with famous researchers from the VeRoLog community.
Next Webinar: July 29, 2021, 4 pm (CEST)
Plenary talk: “Shortest Path Problems with Resource Constraints”
Stefan Irnich (Johannes Gutenberg Universität Mainz)
Abstract: In most vehicle routing and crew scheduling problems solved by column generation-based methods, the subproblem is a variant of the shortest path problem with resource constraints (SPPRC). The SPPRC has contributed to the success of these methods in at least two ways: First, through its resource constraints, it constitutes a flexible tool for modeling complex cost structures as well as a wide variety of rules that define the feasibility of a route or a schedule. Second, there exist efficient algorithms at least for some important variants of the SPPRC. The talk presents selected examples of modeling with SPPRC and an overview of state-of-the-art methods for solving SPPRC.
Followed by an interview with
Angel Corberan (Universitat de València)
Meeting ID: 940 0638 5217
August 26, 2021, 4 pm (CEST)
Research talk: “Routing a fleet of gliders“
Maria Battarra (University of Bath)
Abstract: This talk will discuss routing algorithms for a fleet of gliders. Gliders are unmanned flying vehicles without a propulsion system, and they are most commonly deployed to take aerial pictures or collect information via sensors. The advantages and disadvantages of gliders will be discussed, and compared and contrasted with drones. Applications in surveillance and humanitarian settings will be presented. The underlying routing problem is modelled by explicitly accounting for flying dynamics, and alternative linearization techniques are tested to make the problem tractable. We propose a metaheuristic based on a novel sequential trajectory optimisation, which computes flyable trajectories for a given route, and a routing matheuristic, combining iterated local search and a set-partitioning-based integer programming formulation. Our computational results will showcase the strengths and weaknesses of this approach, as well as lessons learned while integrating complex system dynamics into a routing algorithm.
Research talk: “The multi-depot vehicle routing problem with profit fairness“
Margaretha Gansterer (University of Klagenfurt)
Abstract: One of the main obstacles in horizontal collaborative transportation is the proper distribution of gains among partners. In practice, a posteriori gain sharing mechanisms rarely guarantee that all partners feel fairly treated. We present the multi-depot vehicle routing problem with profit fairness, a bi-objective optimization problem that adds a fairness objective function to the classical cost minimization function. We explore the effects of integrating fairness in the optimization process. To that end, we approximate the Pareto front of any problem instance using an adaptive large neighbourhood search algorithm, embedded within an ε-constraint scheme. This problem is solved for different instance types and time horizons. We show that the economic cost of fairness is rather low and tends to decrease when fairness is considered for longer time horizons.
September 30, 2021, 4 pm (CEST)
June 24, 2021, 4 pm (CEST)
Research Talk: “Machine Learning for Time Slot Management“
Niels Agatz (Erasmus University Rotterdam)
Abstract: The COVID-19 pandemic provided a huge boost to online grocery shopping. Online grocers typically let customers choose a delivery time slot to receive their goods. To ensure a reliable service, the retailer may want to close time slots as capacity fills up. The number of customers that can be served per slot largely depends on the specific order sizes and delivery locations. Conceptually, checking whether it is possible to serve a certain customer in a certain time slot given a set of already accepted customers involves solving a vehicle routing problem with time windows. This is challenging in practice as there is little time available and not all relevant information is known in advance. We explore the use of machine learning to support time slot decisions in this context.
Industry talk: “Time Slot Management in Practice“
Thomas Visser (ORTEC)
Abstract: Attended home delivery is growing, even more so during the COVID pandemic. While consumers get more and more accustomed to booking time slots online for a variety of (delivery) services, it becomes more demanding for retailers to manage the availability/incentives of time slots on their website. ORTEC offers time slot management software services which utilize dynamic vehicle routing heuristics. In this talk, we discuss some of the challenges (and opportunities for further research) of implementing such time slot management services in practice. Furthermore, we highlight some of our own research findings including recent work on dynamic time slot pricing and machine learning.
May 27, 2021, 4 pm (CET)
Plenary talk: “Split-demand vehicle routing problem: state of the art“
Juan José Salazar González (Universidad de La Laguna)
Abstract: The talk addresses several vehicle routing problems where customers are allowed to be visited several times to serve their demand from a fleet of vehicles. It will focus mainly on the single-vehicle one-commodity pickup-and-delivery problem which illustrates clearly the difficulties when the split-demand features must be mathematically formulated with a Mixed Integer Programming Model to be later solved with a row-generation mechanism. We will also survey early approaches for the classical Capacitated Vehicle Routing Problem and other recent variants with additional side constraints.
Followed by an interview with
Grazia Speranza (Università degli Studi di Brescia)
April 29, 2021, 4 pm (CET)
Daniele Vigo (Università di Bologna) on “Integrating Machine Learning into state-of-the-art Vehicle Routing Heuristics“
Abstract: The Vehicle Routing Problem (VRP) is one of the most studied combinatorial optimization problems for which hundreds of innovative heuristic and exact algorithms have been proposed in more than fifty years of research. Recently, some attempts were performed towards the integration of Machine Learning into heuristics to enhance their performance and guide their design. We report some initial efforts we performed in this direction and highlight the difficulties and promising research directions which merit further investigation.
Contact us if you have comments or suggestions!
To stay up to date, bookmark the webpage https://www.euro-online.org/websites/verolog/webinar/, subscribe to the VeRoLog Newsletter, or follow us on Twitter (@EWGVeRoLog)