Modeling urban resident travel satisfaction during the morning and the evening peak hours: A case study in Beijing

Abstract

Understanding how various factors influence travel satisfaction can assist in traffic policy-making. In the study, we aim to develop innovative urban resident travel satisfaction evaluation models by building a comprehensive travel satisfaction evaluation index system and considering the asymmetric traffic flow and difference in travel time urgency during the morning and evening peak hours. We consider both the internal factors reflecting resident-related characteristics including socio-economic attributes and travel characteristics, and the external factors reflecting road-related characteristics including traffic facilities, road traffic conditions, traffic environments, and service levels. Then, for the morning and the evening peak hours, we respectively build a structural equation model (SEM) to capture the intrinsic interactions between latent factors, and an ordered logit model to describe the direct influencing factors of travel satisfaction considering its ordered nature. Finally, we examine our proposed models with the travel survey data collected in the Yizhuang district of Beijing, China. The numerical results show that both the internal and the external factors have significant impacts on travel satisfaction. The SEM models captures the interactions between the latent variables such as the positive relationship between traffic facilities and traffic environments. The ordered logit modeling results show that most external factors except the satisfaction of the road obstacles have positive influences on travel satisfaction. Our research findings provide a better understanding of the intrinsic interactions between latent variables and direct influencing factors of travel satisfaction and offer the government valuable guidance on how to improve travel satisfaction.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 3 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.