Activity-Based Models (ABMs) have traditionally been the domain of mega regions because they are resource-intensive to develop, require significant investments in data collection, software development, and operational costs. The challenge lies particularly in obtaining statistically robust household travel surveys, which must capture diverse behaviors, preferences, and communication arrangements across populations. Efforts to enhance the transferability of ABMs between regions have proven instrumental in addressing these challenges. By adapting foundational behavioral parameters from existing models, ABMs can be efficiently transferred, calibrated, and fine-tuned to reflect local conditions, significantly reducing development costs. This approach not only democratizes access to advanced travel forecasting tools but also provides critical insights into demographic behaviors and preferences, furthering the understanding and assessment of equitable planning.
The Joint REgional TranSportation Simulator (JESS) ABM, developed for the Joint Regional Transportation Agency and the City of Halifax was built on the above principles of transferability and adaptability. In doing so, it has proven that a new age in transportation analytics and data-driven decision-making in planning has arrived, regardless of the size of the urban area.
Activity-based travel demand models (ABM) often use econometric models or heuristic rules to study activity participation and duration, focusing on independent variable effects. However, joint effects and the influence of emerging trends like telecommuting and online activities remain underexplored. This study investigates how individual attributes influence activity participation and duration, relationships between activity engagement and duration, and interdependencies among activities. Using data from Metro Vancouver, the study examines these dynamics across four groups: commuters, telecommuters, non-workers, and hybrid workers, based on 24-hour activity logs from the British Columbia Activity Time Use Survey. Findings from the Bayesian Belief Network (BBN) revealed significant direct and causal relationships between in-home and out-of-home activity participation and their durations. Furthermore, in-home activities influence each other, as do out-of-home activities. For example, out-of-home recreational activities and in-home recreational durations directly influence in-home social activity participation, while in-home personal business durations have an indirect effect for commuters. The number of children in a household directly influences participation in pick-up/drop-off activities. Similar activity relationships are observed among telecommuters, non-workers, and hybrid workers. The developed model will be implemented within a 24-hour ABM framework, currently under development at UBC Okanagan.
Assistant Professor, University of British Columbia - Okanagan
Dr. Mahmudur Fatmi is a transportation professor at UBCO. He has started as an assistant professor at UBCO since July 2018. Dr. Fatmi contributes by developing advanced econometric modelling methods and agent-based microsimulation modelling techniques to assist in making effective... Read More →
Passenger demand models are used to understand the movement of people across the multimodal transportation systems, which usually includes a combination of the modes ranging between car, bus, train, ferry, etc., and often include other access modes such as walking, taxis, etc. Traditional models are often complex and require substantial amount of data for their development, validation, and calibration. Consequently, this complexity contributes to increased project costs and extended timelines.
This study presents a simplified approach towards modeling passenger flows across a regional transportation network, focused on ferry transport. The model considers travel times, intermodal transfer times, boarding and alighting times, and the corresponding fares associated with the movement of passengers. These costs and times are categorized by passenger type and are represented as a link- or node-based generalized cost. The generalized cost function build-up also accounts for bottlenecks, inefficiencies, congestion, boarding issues, and delays, etc.
The model has been developed in collaboration with BC Ferries to understand demand on the ferry network and to test shifts in demand with potential new ferry routes. Future applications include analyzing demographic shifts, modeling future capacity needs, and assessing the impact of mode shift policies, providing a robust framework for strategic transportation planning.
Job mobility is a critical factor in understanding long-term employment transitions and their interplay with transportation systems, residential choices, and urban planning. This study examines job mobility using data from the British Columbia Time Use Survey (BC ATUS), focusing on survey design, data accuracy, and the reliability of self-reported employment histories. The study evaluates the structure of job mobility-related survey questions and the challenges associated with recall bias and data consistency in longitudinal data collection. To analyze job mobility dynamics, this study employs a continuous hazard-based duration model, which explicitly accounts for individual and household life-cycle stages, including residential relocation, partner relocation, and retirement decisions. The model integrates socioeconomic and demographic attributes, such as age, marital status, education level, and homeownership, to examine their influence on job duration and transition likelihoods. The results indicate that job relocation is most frequent in the early years of employment and is significantly influenced by individual and household factors, particularly residential/partner relocation and retirement decisions.
Assistant Professor, University of British Columbia - Okanagan
Dr. Mahmudur Fatmi is a transportation professor at UBCO. He has started as an assistant professor at UBCO since July 2018. Dr. Fatmi contributes by developing advanced econometric modelling methods and agent-based microsimulation modelling techniques to assist in making effective... Read More →
Effective pedestrian facilities are essential for promoting walking and supporting urban mobility objectives for municipalities. Conventional pedestrian level-of-service (PLOS) assessments rely on density as an indicator but fail to account for movement restriction, such as “shoulder brushing” or “bumping others.” This study introduces an innovative approach to real-time PLOS estimation that captures density and accounts for restricted movement. The system development consists of three phases: Detection, PLOS Modeling, and Application. Initially, a virtual camera within a game engine (Unity) replicates a real-world detection system to capture the trajectory of pedestrians on a second-by-second basis. Next, a traffic micro-simulation (Vissim) models pedestrian movement, deriving mathematical indicators to quantify density and restricted movement. These indicators are computed in real-time, offering enhanced PLOS measurement. Finally, the model is tested through a case study with countermeasures proposed to improve PLOS. The pilot study conducted on York Lane at York University, Toronto, Ontario, evaluates one base scenario and five alternatives. Under similar conditions, removing static obstacles improved restricted movement by 10%, while eliminating dynamic obstacles yielded a 15% improvement in comparison to the base scenario. The real-time virtual PLOS estimation system offers urban planners an effective tool to minimize restricted movement and enhance pedestrian infrastructure.
This study examines the factors influencing work duration, shopping, and recreation time, along with their variability, with a particular focus on demographic characteristics, employment status, household composition, and the built environment. The study utilizes data from the 2022 Halifax Travel Activity Survey, which collected detailed 24-hour activity logs from the respondents. Ordinary Least Squares (OLS) regression models were developed using the survey data to analyze the relationships between various independent variables and activity durations. The results reveal that individuals aged 25-65 generally work longer hours, while part-time workers and students report significantly fewer hours. Retired individuals exhibit greater variability in work duration, suggesting more flexible schedules, while students and part-time workers show less variability due to fixed routines. In terms of shopping and recreation, retirees and the unemployed spend more time on leisure activities, while students and households with children spend less time. License ownership and proximity to mixed land-use areas are significant factors influencing the duration and variability of activities. The findings underscore the importance of considering both intrapersonal and interpersonal variability when modeling travel behavior and emphasize the need for models that account for increased flexibility in work and leisure activities post-COVID.
