Agent-based Microsimulation of Land Use, Technology Adoption, and Transportation within an Integrated Urban Modelling Framework

点击翻译按钮获取中文摘要

Canadian cities are experiencing increased traffic congestion, greenhouse gas (GHG) emissions, and energy consumption due to private vehicle dependency and urban sprawl. For example, the transportation sector accounts for around 25% of the GHG emissions in Canada. To address these sustainability challenges, Canada will invest $180 billion over the next 10 years for sustainable city and transportation planning. Advanced simulation models are required to analyze land use and transportation policies within an integrated modelling framework. Travel and location choices are complex inter-dependent decisions, which closely interact with the adoption of technologies such as vehicle technology (e.g. autonomous vehicles), shared mobility (e.g. Uber), and ICT (e.g. smartphones). To analyze such decision interactions, fundamental research on developing innovative simulation techniques for improved prediction, as well as advanced econometric models for better behavioural representation, are required. The goal of my research program is to develop an agent-based urban microsimulation tool-kit that predicts the evolution of population technology adoption, land use, transportation, and environment of an urban region over time and space. The short-term objectives are to: 1) develop an innovative agent-based microsimulation model for predicting technology adoption, housing, and mobility choices, 2) develop novel econometric methods for understanding technology adoption, housing, and mobility behaviour, and 3) develop an advanced activity-based traffic microsimulation and environment model. A hybrid of event-based and discrete-time simulation techniques will be adopted to allow two-way feedback among technology use, land use, transportation, and environment components. Hybrid econometric methods will be developed to accommodate decision interactions and capture multi-layer heterogeneity among agents. The key outputs of this model will be traffic flows, vehicular emissions and dispersion, and residential and transportation energy use. During my PhD and post-doctoral training, I developed agent-based simulation and econometric models. I confirmed one-directional decision interactions, such as the effects of demographic events on location choices and travel activities. I will expand this knowledge to explore the two-way relationships among technology adoption, transportation, and land use decisions, and their environmental implications. This research will have a significant impact as the simulation tool will assist engineers and planners from industries and government agencies to develop sustainable policies, and will support future infrastructure investment decision-making. The development of these new technologies will benefit Canada by facilitating high-quality training (e.g. programming languages such as C#DotNET), creating new jobs (e.g. simulation specialists), and fostering sustained economic growth through ensuring an efficient transportation network.