Waiting for signalized crossing or walking to footbridge/underpass? Examining the effect of weather using stated choice experiment with panel mixed random regret minimization approach
Introduction
Walkability is increasingly popular in urban and transport planning because walking can reduce traffic emissions, relieve traffic congestion, improve physical health, and support the well-being of society (Pucher and Buehler, 2010; Lawlor et al., 2003; Lo, 2009; Elias and Shiftan, 2012). This is particularly true for transit-oriented cities such as Hong Kong, where 89% of trips are made by public transportation and walking is the primary means of access (Hong Kong Transport Department, 2014; Li and Loo, 2016; Sze et al., 2019).
Safety is a key attribute of walkability. To resolve pedestrian-vehicle conflicts and improve traffic safety at intersections and crosswalks, different pedestrian facilities have been installed to separate pedestrian and vehicular traffic by either time (signalized crossings) or space (footbridges and underpasses) (World Health Organization, 2018; Rankavat and Tiwari, 2020; Zhu et al., 2021a). However, these approaches may also suffer from shortcomings. For example, traffic signals increase the waiting time of pedestrians and vehicles and worsen traffic congestion (Vandaele et al., 2000). In contrast, the construction and maintenance costs of footbridges and underpasses are much higher than those of signalized crossings (Yip et al., 2014). Additionally, pedestrians may struggle with the use of footbridges and underpasses because of extra walking distance, climbing up or down stairs, poor lighting and hygiene, and security concerns (Sharples and Fletcher, 2001; Sinclair and Zuidgeest, 2016; Zhu et al., 2021b). One study indicated that females and elderly individuals were hesitant to use footbridges and underpasses, especially at night (Rankavat and Tiwari, 2016). Hence, it is crucial to identify the factors that affect pedestrians’ perception and choice among different crossing facilities for the design and planning of transport infrastructure. In particular, the trade-off between additional waiting time at a traffic signal and effort to access footbridges and underpasses should be considered (Rankavat and Tiwari, 2020; Chowdhury and Van Wee, 2020).
Many developed societies are facing the problem of an ageing population. In the United Kingdom, the proportion of the population above 65 years of age doubled in 15 years (Hanson, 2004). In Hong Kong, the proportion of older inhabitants is expected to increase from 12% in 2015 to 25% in 2035 (Hong Kong Legislative Council, 2015). Hence, there has been increasing concern for universal access in building design and urban development in recent years. For instance, it is necessary to consider the needs of individuals with limited mobility for access to different public facilities, such as underground stations, footbridges and underpasses, and elevated walkways (Sze and Christensen, 2017; Xu et al., 2022; Yi and Ling, 2020). In Hong Kong, many footbridges and underpasses were built more than 40 years ago. To enhance the accessibility to footbridges and underpasses for all, facilities such as ramped walkways, elevators, and people-movers have been installed in recent years (Hong Kong Building Department, 2011; Hong Kong Highway Department, 2020). Furthermore, weather conditions can also affect pedestrians’ perception and walking experience. Summer in Hong Kong is hot and humid because of the subtropical climate, and the average annual precipitation is over 2300 mm. In 2020 and 2021, 55 “very hot weather warnings” and 78 “rainstorm warning signals” were issued (Hong Kong Observatory, 2022). Thus, the effects of extremely hot weather and rainy conditions on the choice decisions of pedestrians must be considered.
Data collection methods including self-report questionnaires (Cambon de Lavalette et al., 2009; Dommes et al., 2015), video observation surveys (Diependaele, 2019; Zhu et al., 2021b), computer vision and automated tracking (Papadimitriou, 2012), simulated experiments (Calvi et al., 2020), global positioning systems (Lassarre et al., 2012), and revealed preference surveys (Arellana et al., 2022) have been adopted to examine the walking behaviour of pedestrians. Moreover, factors including road design, walking environment, and traffic conditions that affect pedestrian route choices have been identified (Garrod et al., 2002; Perdomo et al., 2014; Beitel et al., 2018). A few studies have examined the factors that affect pedestrians’ preference and tendency to use different crossing facilities. In particular, one study has considered the effects of pedestrian gender and age on the choice of crossing facilities (Anciaes and Jones, 2018). On the other hand, two recent studies have explored the effect of pedestrian occupation on the choice of grade-separated crossing facilities. Results indicated that working professionals had higher tendency to use footbridge, compared to students (Chandrappa et al., 2021; Bhatia et al., 2022). However, it is rare that other socio-economic characteristics like personal income and education level are considered for the choice of crossing facilities. Furthermore, interaction effect of weather conditions on the relationship between personal characteristics and choice of pedestrians should be accounted.
To examine individuals' preferences in hypothetical settings, the stated preference (SP) approach has been applied to study transport mode choice (Jin et al., 2020), perception of community severance (Anciaes et al., 2018), walking path choice (Anciaes and Jones, 2020), electric vehicle ownership (Jia and Chen, 2021), and traffic enforcement (Li et al., 2014; Steinbakk et al., 2019). With the use of SP approach, it is possible to measure the pedestrians’ willingness to “pay” for the use of specified crossing facilities (Anciaes and Jones, 2018). However, difference in the perceived values between walking time and waiting time was rarely considered. Furthermore, moderating effect of environmental conditions on the association between waiting time, walking time, and pedestrian choice behaviour is yet to be explored.
For the analysis method, random utility maximization (RUM) approach is usually adopted in conventional discrete outcome models for pedestrian choice behaviour (Anciaes et al., 2018; Anciaes and Jones, 2018; Beitel et al., 2018). The RUM approach assumes that individuals favour the choice that provides the highest level of satisfaction (Train, 2009). However, bias in parameter estimation is possible since the RUM approach allows for self-compensation between underperforming and outperforming attributes (Chorus et al., 2008). Therefore, an alternate approach, random regret minimization (RRM), should be considered. The advantages of RRM are threefold. First, RRM considers the attributes of both selected and unselected options in the estimation: a choice that has less regret is preferred. In addition, RRM can improve the model fit. Furthermore, the constant assumption for willingness-to-pay estimates (i.e., trade-offs between attributes) can be relaxed when RRM is adopted (Zhu et al., 2021a; Iraganaboina et al., 2021).
In this study, the roles of facility design and planning, environmental conditions, and personal characteristics in pedestrian choice among signalized crossings, footbridges, and underpasses are examined using a stated choice experiment. For example, hypothetical scenarios for different combinations of geometric design, crowdedness, accessible design (i.e., elevators, stairs, ramped walkways), traffic flow, and weather conditions are visualized (; Mukherjee and Mitra, 2020; Zhu and Sze, 2021). In addition, personal characteristics, including socio-demographics, walking habits, and the self-rated importance of the design attributes of crossing facilities are considered. Furthermore, the trade-off between waiting time, walking time, and other factors of pedestrians are estimated using RRM in a panel mixed multinomial logit model. Moreover, the effects of unobserved heterogeneity and panel data are accounted for (Mannering et al., 2016). The findings shed light on the design and planning of crossing facilities to promote walkability and improve pedestrian safety.
Section snippets
Stated preference experiment
To gauge the trade-off between waiting time at traffic signals and additional walking time of pedestrians to use footbridges and underpasses, a stated preference (SP) experiment is adopted. The SP choice sets are presented based on a hypothetical scenario of walking travel. The pedestrian is assumed to be facing three crossing alternatives (i.e., signalized crossing, footbridge and underpass) and makes crossing decisions based on the given information. Important attributes and relevant levels
Data and sample description
A total of 500 participants completed the questionnaire survey. Table 3 summarizes the characteristics of the participants. Overall, ratio of male to female is 689 to 1000. It is consistent to that of Hong Kong population (ratio of male to female equal to 838 to 1000) (Census and Statistic Department, 2021a). For the age, participants of age between 18 and 25 years constitute more than one-third (37.8%). It is much higher than that of Hong Kong population (7.9%). On the other hand, participants
Descriptive analysis
Since there are four choice scenarios in each questionnaire, the total number of observations is 2000. Among the 2000 scenarios, signalized crossing, footbridge, and underpass are preferred in 641 (32.0%), 784 (39.2%), and 575 (28.8%) scenarios, respectively. Fig. 1 shows the distribution of choice frequency. As shown in Fig. 1, the frequency of choosing signalized crossing increases as the waiting time decreases (Fig. 1(a)), traffic volume decreases (Fig. 1(c)), and when the weather is 28 °C
Summary of influencing factors
Table 6 summarizes the factors that affect pedestrians’ choice of crossing facility identified in this study. The findings are generally consistent with those of previous studies, particularly for waiting time, walking time, traffic conditions, crowdedness and pedestrian characteristics.
As the main findings in this research, the variables that represent weather conditions, income and interaction effects (Female x Raining) have a significant impact on the signalized crossing, footbridge, and
Conclusion
In this study, the influences of waiting time, walking time, traffic and weather conditions, and personal characteristics on the choice of crossing facility of pedestrians are examined using a stated choice experiment. The contributions of this study are twofold. First, trade-offs between the waiting time and walking time of different pedestrian groups are estimated, with which the moderation effects of weather conditions (i.e., rain and extreme hot weather) and crowdedness are considered.
Author statement
Dianchen Zhu: Methodology, Data collection, Data curation, Writing- Original draft preparation; N.N. Sze: Conceptualization, Supervision, Writing - Reviewing and Editing; Zhongxiang Feng: Methodology, Writing - Reviewing and Editing; Ho-Yin Chan: Writing – Reviewing and Editing.
Funding
The work that is described in this paper was supported by the grants from the Research Grants Council of Hong Kong (15209818), Research Committee of the Hong Kong Polytechnic University (H-ZJNG), the Fundamental Research Funds for the Central Universities (JZ2022HGQA0123) and the National Natural Science Foundation of China (52272345, 71971073).
References (111)
- et al.
Estimating preferences for different types of pedestrian crossing facilities
Transport. Res. F Traffic Psychol. Behav.
(2018) - et al.
A comprehensive approach for the appraisal of the barrier effect of roads on pedestrians
Transport. Res. Pol. Pract.
(2020) - et al.
A stated preference model to value reductions in community severance caused by roads
Transport Pol.
(2018) - et al.
Analyzing pedestrian behavior when crossing urban roads by combining RP and SP data
Transport. Res. F Traffic Psychol. Behav.
(2022) - et al.
Parametric action decision trees: Incorporating continuous attribute variables into rule-based models of discrete choice
Transport. Res. Part B: Method.
(2007) - et al.
A systematic review of the factors associated with pedestrian route choice
Transport Rev.
(2022) - et al.
Assessing safety of shared space using cyclist-pedestrian interactions and automated video conflict analysis
Transport. Res. Transport Environ.
(2018) Quasi-random maximum simulated likelihood estimation of the mixed multinomial logit model
Transp. Res. Part B Methodol.
(2001)Simulation estimation of mixed discrete choice models using randomized and scrambled Halton sequences
Transp. Res. Part B Methodol.
(2003)- et al.
The impact of stop-making and travel time reliability on commute mode choice
Transp. Res. Part B Methodol.
(2006)
Skywalk facility utilization and factors influencing its use: a case study of Mumbai Metropolitan Region, India
Case Studies on Transport Policy
Weather, transport mode choices and emotional travel experiences
Transport. Res. Pol. Pract.
The impact of waiting time and other factors on dangerous pedestrian crossings and violations at signalized intersections: a case study in Montreal
Transport. Res. F Traffic Psychol. Behav.
Effectiveness of augmented reality warnings on driving behaviour whilst approaching pedestrian crossings: a driving simulator study
Accid. Anal. Prev.
Pedestrian crossing decision-making: a situational and behavioral approach
Saf. Sci.
Impacts of the walking environment on mode and departure time shifts in response to travel time change: case study in the multi-layered Hong Kong metropolis
Travel Behaviour and Society
Pedestrian route choice with respect to new lift-only entrances to underground space: case study of a metro station area in hilly terrain in Hong Kong
Tunn. Undergr. Space Technol.
Measures for improving pedestrian crossing facilities based on perceptions of urban commuters: an experience in Kolkata
Case Studies on Transport Policy
Evaluation of penalty and enforcement strategies to combat speeding offences among professional drivers: a Hong Kong stated preference experiment
Accid. Anal. Prev.
Examining the spatially heterogeneous effects of the built environment on walking among older adults
Transport Pol.
A random regret-minimization model of travel choice
Transp. Res. Part B Methodol.
Examining women's perception of safety during waiting times at public transport terminals
Transport Pol.
More than walking and cycling: what is ‘active travel’?
Transport Pol.
SWOT analysis and development strategies for underground pedestrian systems
Tunn. Undergr. Space Technol.
Non-compliance with pedestrian traffic lights in Belgian cities
Transport. Res. F Traffic Psychol. Behav.
Investigation of combined stairs elevators evacuation strategies for high rise buildings based on simulation
Simulat. Model. Pract. Theor.
Red light violations by adult pedestrians and other safety-related behaviors at signalized crosswalks
Accid. Anal. Prev.
The influence of individual's risk perception and attitudes on travel behavior
Transport. Res. Pol. Pract.
The built environment, walking, and physical activity: is the environment more important to some people than others?
Transport. Res. Transport Environ.
Pedestrian environment and route choice: evidence from New York City and Hong Kong
J. Transport Geogr.
Classical estimation methods for LDV models using simulation
Handb. Econom.
An intercept study of footbridge users and non-users in Malaysia
Transport. Res. F Traffic Psychol. Behav.
Walking and jobs: a comparative analysis to explore factors influencing flexible and fixed schedule workers, a case study of Rasht, Iran
Sustain. Cities Soc.
How gender differences and perceptions of safety shape urban mobility in Southeast Asia
Transport. Res. F Traffic Psychol. Behav.
Stated response to increased enforcement density and penalty size for speeding and driving unbelted
Accid. Anal. Prev.
Evaluating the influence of information provision (when and how) on route choice preferences of road users in Greater Orlando: application of a regret minimization approach
Transport. Res. C Emerg. Technol.
Pedestrian at-fault crashes on rural and urban roadways in Alabama
Accid. Anal. Prev.
Are Individuals' stated preferences for electric vehicles (EVs) consistent with real-world EV ownership patterns?
Transport. Res. Transport Environ.
Mode choice analysis in urban transport with shared battery electric vehicles: a stated-preference case study in Beijing, China
Transport. Res. Pol. Pract.
Maintaining balance on a moving bus: the importance of three-peak steps whilst climbing stairs
Transport. Res. Pol. Pract.
A GIS-based methodology for identifying pedestrians' crossing patterns
Comput. Environ. Urban Syst.
Towards people-centered integrated transport: a case study of Shanghai Hongqiao Comprehensive Transport Hub
Cities
Walking towards a happy city
J. Transport Geogr.
Unobserved heterogeneity and the statistical analysis of highway accident data
Analytic Methods in Accident Research
A comprehensive study on factors influencing pedestrian signal violation behaviour: experience from Kolkata City, India
Saf. Sci.
Elevator availability and its impact on stair use in a workplace
J. Environ. Psychol.
Theory and models of pedestrian crossing behaviour along urban trips
Transport. Res. F Traffic Psychol. Behav.
Modelling the effects of risk factor and time savings on pedestrians' choice of crossing facilities at signalised intersections
Case studies on transport policy
Sensitivity of footbridge vibrations to stochastic walking parameters
J. Sound Vib.
Pedestrian preferences with respect to roundabouts—a video-based stated preference survey
Accid. Anal. Prev.
Cited by (4)
Safety or efficiency? Estimating crossing motivations of intoxicated pedestrians by leveraging the inverse reinforcement learning
2024, Travel Behaviour and SocietyAged and wheeled mobility in transit-oriented development: The capabilities approach
2024, Transportation Research Part D: Transport and EnvironmentChoice and equity: A critical analysis of multi-modal public transport services
2023, Transport Policy