EURO 2025 Leeds
Abstract Submission

2151. The Dynamic Multi-Objective Driver Dispatching Problem in Full Truck Load Transport

Invited abstract in session MC-59: Freight Transportation and Logistics, stream Transportation.

Monday, 12:30-14:00
Room: Liberty 1.14

Authors (first author is the speaker)

1. Neslihan Cevik
Eindhoven University of Technology
2. Albert Schrotenboer
Department of Industrial Engineering and Innovation Sciences, Eindhoven University of Technology
3. Tom van Woensel
Technische Universiteit Eindhoven

Abstract

The logistics sector faces driver shortages and has significant pressure for lower emissions, demanding more advanced dispatch strategies that balance driver satisfaction and sustainability in fleet management. We propose a dynamic, multi-objective framework that processes real-time data. At each decision point, newly revealed requests, each with pickup and delivery windows, vehicle requirements, and driver needs are matched with available drivers. Requests are grouped into feasible sequences, distinguishing between urgent and flexible orders. At each decision point, we combine a Markov Decision Process with a Mixed-Integer Programming (MIP) model. The MIP optimizes operational cost, emissions, and driver well-being by pairing drivers to orders, guided by metrics such as distance from home base, time-window adherence, and regulated work-hour limits. Recognizing the complexity of large state and action spaces, we employ a Cost Function Approximation in the MDP to maintain an anticipation of future outcomes. This dynamic assignment avoids short-sighted decisions—such as overloading drivers or creating inefficient subtours—and balances distance, emissions, and driver satisfaction. Preliminary tests show improvements in cost-efficiency, emission reduction, and driver satisfaction, emphasizing the importance of integrating operational, environmental, and human factors in modern logistics.

Keywords

Status: accepted

Back to the list of papers