FHR-NSGA-III: A hybrid many-objective optimizer for intercity multimodal timetable optimization considering travel mode choice

•We propose a many-objective multimodal timetable optimization model under travel mode choice.•We develop a Fast Hypervolume indicator-based MOEA to improve efficiency for solution seeking.•We provide a hybrid MOEA that combines FH-MOEA with R-NSGA-III to improve solutions’ quality.•We validate our...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Information sciences Jg. 649; S. 119654
Hauptverfasser: Zhao, Jiandong, Feng, Yingzi, Wu, Jianjun, Gao, Ziyou
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Inc 01.11.2023
Schlagworte:
Hybrid many-objective evolutionary algorithm
Hypervolume (HV)
Integrated intercity transportation
Multimodal timetable optimization problem (TOP)
Passenger travel mode choice
Multimodal timetable optimization problem (TOP)
Hybrid many-objective evolutionary algorithm
Passenger travel mode choice
Integrated intercity transportation
Hypervolume (HV)
ISSN:0020-0255, 1872-6291
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•We propose a many-objective multimodal timetable optimization model under travel mode choice.•We develop a Fast Hypervolume indicator-based MOEA to improve efficiency for solution seeking.•We provide a hybrid MOEA that combines FH-MOEA with R-NSGA-III to improve solutions’ quality.•We validate our proposed algorithm using intercity networks from three urban agglomerations. Multiple transport modalities often have low frequency and uneven service, which puts customers through an uncomfortable and costly experience due to disorganized transportation. This paper presents a passenger-centric approach for multimodal timetable optimization. We suggest a many-objective mixed programming model that takes the travel mode choice into account. By simulating the utility function of passengers’ perceived travel cost, scheduling measures are developed that also take the time window, capacity, and overtaking into account. Afterwards, by minimizing the number of related solutions, we present a Fast Hypervolume indicator-based Many-objective Evolutionary Algorithm (FH-MOEA), which can swiftly determine the hypervolume contribution of each solution. Furthermore, by merging FH-MOEA with Reference Point Based Multi-Objective Optimization Using Evolutionary Algorithms (R-NSGA-III), we propose the hybrid algorithm FHR-NSGA-III for finding high-quality solutions. Intercity networks between Guangzhou and Qingyuan, Chengdu and Chongqing, and Beijing and Zhangjiakou are the subjects of computational experiments. The results demonstrate the performance of the algorithm in finding high-quality solutions in a computationally efficient manner.
ISSN:0020-0255
1872-6291
DOI:10.1016/j.ins.2023.119654