Athens University Economics and Business (AUEB)
Mathematical models and optimization methods for vehicle routing problems under uncertainty
Abstract
dc:descriptionThis doctoral dissertation examines mathematical models and optimization methods for routing and service systems under uncertainty, with particular emphasis on the endogenous assignment of service time windows. In urban transportation, last-mile delivery, and appointment-based services, travel and service times are affected by unforeseen factors, creating uncertainty in arrival times and service start times. Within this context, the thesis studies the Time Window Assignment Problem under stochastic travel or service times and introduces the Stochastic Time Window Assignment Problem (STWAP), in which time windows are treated as decision variables and used as a mechanism for absorbing operational uncertainty. The thesis comprises two complementary research directions. In the first research direction, the STWAP is integrated into the Vehicle Routing Problem and treated as a recourse function within a hierarchical two-stage solution framework. An Adaptive Large Neighborhood Search (ALNS) metaheuristic is developed to generate routing solutions, while the time-window assignment problem is solved using exact linear and mixed-integer programming methods. Uncertainty in travel times is represented through continuous and discrete random variables, while service-reliability requirements are explicitly incorporated into the proposed models. In the second research direction, the STWAP is studied as a standalone optimization problem over predetermined sequences of jobs or service requests with stochastic service times. The accumulation of uncertainty throughout the sequence and stochastic waiting are explicitly taken into account, while the expected customer inconvenience caused by time-window deviations is evaluated using quadratic penalty functions, leading to a mixed-integer quadratic programming formulation. Extensive computational experiments and simulation studies investigate different probability distributions, service-level requirements, and penalty structures. The results highlight the interactions among time-window width, service reliability, expected inconvenience, and operational cost, demonstrating the value of endogenous time-window assignment for the design of reliable transportation and service systems operating under uncertainty.
Degree
thesis:*- Grantor dc:publisher
- Athens University Economics and Business (AUEB)
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Vareias, Anastasios
- Βαρειάς, Αναστάσιος
Subjects
dc:subject × 18- Επιχειρησιακή έρευνα
- Στοχαστική βελτιστοποίηση
- Πρόβλημα δρομολόγησης στόλου οχημάτων
- Χρονικά παράθυρα
- Στοχαστικό πρόβλημα ανάθεσης χρονικών παραθύρων
- Αξιοπιστία εξυπηρέτησης
- Operations research
- Stochastic optimization
- Vehicle routing problem
- Time windows
- Stochastic time window assignment problem
- Service reliability
- Οικονομικά και Επιχειρήσεις
- Κοινωνικές Επιστήμες
- Διοικητική Επιστήμη και Επιχειρησιακή Έρευνα
- Economics and Business
- Social Sciences
- Management Science and Operations Research
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/62570