Liping GAO defends his doctoral thesis on Monday, December 18, 2023 at 10:00 a.m. The defense takes place in room 334 of the IBGBI site. It is also possible to attend the defense via Zoom: https://univ-evry-fr.zoom.us/j/93676270202?pwd=UU9FaytzbWpIWlZ2cHB2dmNYUlY0QT09
Abstract:
Traveler’s preference-driven route planning is one of the most complicated and important tourism preparing activities. In reality, travelers may have multi-preferences such as travel time, beautiful scenery, safety, and low carbon, simultaneously. Moreover, the preference and travel time are both time-dependent. Considering that real-world road networks are typically large-scale and include hundreds of thousands of road segments and intersections, the routing planning process can be extremely time-consuming. Developing an efficient decision support platform that can be able to propose the best route planning to satisfy various traveler’s preferences in such diverse contexts is necessary. In this thesis, three new time-dependent route planning problems with traveler’s preference (TRPPs-TP) are investigated.
Firstly, a single-objective TRPP-TP is investigated in that the preference score on road segments is assumed to be time-dependent. The objective is to maximize the total preference score. For the problem, an integer linear programming model is proposed, and the NP-hard complexity of the problem is analyzed. To address the problem efficiently, a novel two-phase method is developed. Numerical experiments on randomly generated road networks and real-world road networks demonstrate the superiority of the developed method.
Secondly, a bi-objective TRPP-TP with the time-dependent preference score is studied. The first objective is to maximize the total preference score, and the second one is to minimize the total travel time. For the problem, an integer linear programming model is formulated. For the problem, an exact epsilon-constraint method is applied to find the Pareto front on small-sized instances. To handle large-sized instances, an efficient problem-specific non-dominated sorting genetic algorithm-II (NSGA-II) is developed. Especially, a new region-based coding is designed and a feasible route condition is provided to find near-optimal solutions in a reasonable computation time. Experiments on randomly generated road networks and real-world road networks demonstrate the performance of the proposed NSGA-II.
Finally, a bi-objective eco-friendly group-oriented TRPP-TP is addressed. The first objective is to maximize the total traveler preference score and the second one is to minimize the total carbon emissions. For this problem, a new integer linear programming model is proposed, and an epsilon-constraint method is used. Numerical experiments on randomly generated road networks are conducted to find the best balancing solutions.
Keywords: route planning, time-dependent preference, time-dependent travel time, mathematical model, epsilon-constraint method, NSGA-II
Composition of the doctoral thesis jury
| Membre du jury | Titre | Lieu d’exercice | Fonction dans le jury |
|---|---|---|---|
| Éric ANGEL | Professor | University of Paris-Saclay (Univ. Évry) | Examiner |
| Lyes BENYOUCEF | Professor | Aix-Marseille University | Reviewer |
| Chao CHEN | Professor | Chongqing University | Invited member |
| Feng CHU | Professor | Université Paris-Saclay (Univ. Évry) | Thesis Director |
| David DUVIVIER | Professor | Polytechnic University of Hauts-de-France | Reviewer |
| Issam NOUAOURI | Associate Professor | University of Artois | Examiner |
| Lydie NOUVELIÈRE | Associate Professor | Université Paris-Saclay (Univ. Évry) | Examiner |
- Date : lundi 18/12/2023, 10h
- Lieu : Salle 334 du site IBGBI [Plan d’accès au format PDF] et soutenance via Zoom: https://univ-evry-fr.zoom.us/j/96342040470?pwd=UUp0ZFpreDlZeXhRd3FpcENvNndldz09
- Doctorant : Liping GAO, Université d’Évry, Université Paris Saclay, IBISC équipe AROB@S
- Directrice de thèse : Feng CHU (PR IUT d’Évry, IBISC équipe AROB@S)
