Data fusion for agent-based freight demand and traffic simulation

建议的研究将开发一个基于代理的货运需求模型，该模型与交通仿真模型相结合。基于代理的需求模型将预测城市货物运输的行为要素，包括卡车车队、装运地点、停车顺序、停车时间和路线选择。此类模型具有广泛的数据需求，将通过开发数据融合方法来解决这一问题，该方法将整合私营部门卡车GPS数据、制造商和零售机构的调查数据、卡车停放清单和使用数据以及来自各种其他来源的信息。该研究将基于代理的货运需求模型与城市道路网络中货车和乘用车联合交通的逐车仿真相结合。研究区域最初将是多伦多的一个小型测试研究区域。利用并行计算资源，该研究领域将扩大到代表大多伦多和汉密尔顿地区内的区域货物流动。基于主体的综合建模方法对于评估加拿大各级政府正在考虑的各种运输政策和基础设施改进将是有价值的。这些政策包括卡车停车政策、卡车路线计划、道路收费、卡车限制以及与城市地区工业用地有关的货运相关细则。可评估的影响包括拥堵、排放、噪音、可靠性和经济影响。

The proposed research will develop an agent-based model of freight demand that is integrated with a traffic simulation model. The agent-based demand model will predict behavioural elements of urban goods movement, including truck tour formation, shipment location, stop sequence, stop duration, and route choice. Such models have extensive data requirements, which will be addressed through the development of data fusion methods that will integrate private sector truck GPS data, survey data from manufacturers and retail establishments, truck parking inventory and usage data, and information from a variety of other sources. The proposed research will integrate the agent-based freight demand model with a vehicle-by-vehicle simulation of combined truck and passenger vehicle traffic on the urban road network. The study area will initially be a small test study area in Toronto. This study area will be expanded to represent regional goods movement within the Greater Toronto and Hamilton Area, making use of parallel computing resources. The integrated agent-based modelling approach will be valuable for evaluating a variety of transportation policies and infrastructure improvements that are under consideration by various levels of government in Canada. Such policies include truck parking policy, truck route plans, road pricing, truck restrictions, and freight related bylaws related to industrial land-uses in urban areas. Impacts that can be assessed include congestion, emissions, noise, reliability and economic effects.