Modeling Mode and Departure Time Responses to Congestion Pricing: A Spatial and Behavioral Analysis Using Cross-Nested Logit Model
Mohammad Amin Ashena, Adam Weiss, Jason Hawkins, Lina Kattan · 2026
Read Modeling Mode and Departure Time Responses to Congestion Pricing: A Spatial and Behavioral Analysis Using Cross-Nested Logit Model by Mohammad Amin Ashena, Adam Weiss, Jason Hawkins, Lina Kattan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-29756.txt.
Introduction
Ashena, Weiss, Hawkins, and Kattan model how people respond to congestion pricing, using stated preference survey data from Calgary, Canada. Effective pricing needs close knowledge of how people shift mode and departure time. The authors build and compare three discrete choice models: Multinomial Logit, Nested Logit, and Cross-Nested Logit.
The Cross-Nested Logit model provides superior realism, capturing simultaneous substitutions across both modes and leaving times. Spatial analysis shows substantial geographic variation in price sensitivity, strongest for central, peak-period trips. Policy analysis favours targeted, dynamic tolling, especially cordon-based pricing paired with time-specific toll adjustments. The authors stress that complementary measures, improved transit and focused discounts, are needed for equitable outcomes.
Catalog record
| CREATOR | Mohammad Amin Ashena, Adam Weiss, Jason Hawkins, Lina Kattan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 113 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.29756
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Modeling Mode and Departure Time Responses to Congestion Pricing: A Spatial and Behavioral Analysis Using Cross-Nested Logit Model Mohammad Amin Ashena, Adam Weiss, Jason Hawkins, Lina Kattan Abstract Effective congestion management strategies require a detailed understanding of how travellers respond to different pricing interventions. This paper presents an in-depth analysis of traveller behaviour under congestion pricing scenarios, focusing specifically on mode and departure time decisions. Utilizing stated preference survey data from commuters in Calgary, Canada, three discrete choice models including Multinomial Logit, Nested Logit, and Cross-Nested Logit are developed and compared. Results indicate that the Cross-Nested Logit model provides superior behavioural realism and flexibility by capturing simultaneous substitutions across modes and departure times. Spatial analysis and elasticity assessments reveal substantial geographic variation in traveller sensitivity to pricing, particularly highlighting stronger responses among commuters travelling to high-demand central locations and during peak travel periods. Further elasticity analyses clarify behavioural patterns, identifying traveller groups with varying degrees of flexibility. Policy analyses underscore the effectiveness of targeted, dynamic tolling, particularly cordon-based pricing combined with time-specific toll adjustments, in reducing congestion levels. Additionally, the findings highlight the necessity of complementary measures, including improved transit services and targeted discounts, to ensure equitable outcomes. The findings offer targeted insights into how specific pricing strategies such as cordon, distance, and travel time-based tolls can be used to influence travel behaviour, reduce peak-period congestion, and guide equitable policy design in urban transportation planning. keywords Stated Preference Survey, Cross-Nested Logit Model, Congestion Pricing, Elasticity Analysis, Mode Choice, Departure Time Choice 1 Introduction Traffic congestion represents one of the most pressing challenges facing urban areas globally, significantly impacting economic productivity, environmental sustainability, and overall quality of life. As cities expand and urban populations grow, the strain on existing transportation infrastructures intensifies, necessitating innovative and effective solutions to mitigate congestion ( lindsney2001 ) . Among various policy measures, congestion pricing has gained considerable attention as an economic instrument designed to manage travel demand, reduce traffic volumes during peak periods, and encourage shifts towards more sustainable transportation modes ( de2011 ) . Congestion pricing schemes vary in implementation, often categorized into cordon-based, distance-based, and time-based pricing. Cordon-based pricing typically involves charging vehicles entering a predefined congested urban area, whereas distance-based pricing applies charges according to the length of travel. Time-based pricing schemes levy fees based on the duration of travel within designated zones or specific peak periods. While these mechanisms show promise in influencing travel behaviours, accurately forecasting their effectiveness depends critically on understanding traveller behavioural responses, which involve complex choices related to both travel mode and departure time ( gu2018 ; simoni2019 ) . Traditional modelling frameworks, such as Multinomial Logit (MNL) and Nested Logit (NL) models, have historically been employed to predict these behavioural responses. However, these models often fall short in capturing realistic decision-making processes, particularly because travellers simultaneously evaluate both mode and timing of their trips ( wen2001generalized ) . To address this limitation, the Cross-Nested Logit (CNL) model has emerged as an advanced discrete choice approach capable of capturing the interdependencies and substitution patterns among various travel alternatives more effectively. By allowing alternatives to simultaneously belong to multiple nests such as mode and departure time nests, the CNL model enhances behavioural realism, providing more reliable predictions for policy analysis ( papola2004some ) . This paper aims to evaluate commuter behavioural responses to different congestion pricing strategies using an advanced CNL model. By modelling joint mode and departure time choices, the study contributes to a more nuanced understanding of how travellers respond to pricing incentives. The analysis incorporates empirical data, location-based elasticity estimates, and spatial heatmaps to identify nuanced behavioural and geographic patterns. While Calgary, Canada is used as the case study, due to its clear monocentric structure with a downtown core and surrounding suburban and peripheral zones, the findings are intended to inform broader congestion pricing policies in urban areas considering such schemes. Calgary’s absence of an existing CP program allows this research to simulate potential impacts in a similar context. The analysis employs detailed empirical data, spatial location analysis, elasticity calculations, and geographic heatmap visualizations to identify nuanced spatial and behavioural patterns among travellers. Specifically, the study investigates the effectiveness of cordon-based, distance-based, and time-based pricing mechanisms in influencing mode shifts and departure time adjustments. By comparing the CNL model with traditional MNL and NL models, this study further assesses the predictive performance improvements offered by the more flexible CNL structure. Ultimately, the research seeks to provide policymakers with robust, evidence-based insights and tailored recommendations for designing congestion pricing policies that effectively address traffic congestion in urban settings similar to Calgary, ensuring both equitable outcomes and enhanced transportation efficiency. Recent work by Heimgartner and Axhausen (2025) highlights the growing need for behavioural models that capture complex substitution patterns and interdependent decision processes. Their multi-stage stated preference study on telework adoption demonstrates that individuals’ responses to policy or contextual changes emerge from layered constraints, preferences, and higher-order behavioural adjustments, rather than isolated choices. Although their focus differs from congestion pricing, their findings reinforce the methodological importance of flexible model structures and carefully designed SP experiments for accurately representing traveller behaviour ( heimgartner2025multimodality ) . This perspective aligns with the motivation for employing a CNL modelling framework in the present study to better capture the overlapping correlations across mode and departure-time decisions. This study offers primary contributions. Methodologically, it empirically validates the CNL model as a more behaviourally aligned approach for jointly modelling mode and departure time choices, comparing its performance with standard MNL and NL models. It also integrates stated preference data, designed using a Bayesian D-efficient framework and customized with real-time travel information, into advanced spatial analysis, generating location-specific elasticity estimates and heatmaps that reveal heterogeneous behavioural patterns across urban space. From a policy perspective, the study provides actionable suggestion for urban decision-makers by identifying where and how congestion pricing strategies are most likely to be effective. Although Calgary is used as the case study, the findings and modelling approach are broadly transferable to other metropolitan regions with similar spatial and institutional conditions. The remainder of this paper is organized as follows: Section 2 provides background on congestion pricing schemes and discrete choice modelling approaches; Section 3 describes the survey design and data collection process. It also outlines the modelling methodology; Section 4 presents the results and comparative analysis of MNL, NL, and CNL models. It discusses elasticity findings and spatial patterns and offers policy insights; and Section 5 concludes with key implications and future research directions. 2 Background Traffic congestion remains a persistent challenge in urban environments, contributing to substantial economic, environmental, and social costs. These costs include lost productivity, elevated emissions, increased energy consumption, and adverse health outcomes. Exposure to traffic-related air pollution continues to pose significant risks; for example, approximately 10 million Canadians are regularly exposed to harmful levels of transportation-related pollutants ( brauer2013ambient ) . Motor vehicles are major contributors to this burden: in the United States, road traffic accounts for nearly 60% of Nitrogen Oxides (NOx) and about 75% of Carbon Monoxide (CO) emissions ( gately2017urban ) . These emissions, together with additional congestion externalities such as delays, crash risk, and noise, are typically not internalized by individual travelers, thereby generating