Advancing quantitative assessment and planning of Canadian cycling infrastructure

本研究提出了自行车基础设施规划与评价的定量方法。这项研究有两个主要目标。首先，开发一种基于图像的机器学习算法，以支持将交通压力水平（LTS）评估和累积可达性方法快速扩展到具有可用空中和街道摄影的新城市（例如，所有加拿大城市）。其次，将开发一种新的数学优化模型，以方便快速识别增加自行车基础设施的街道和十字路口，这些街道和十字路口将提供最大的整体效益。该项目将产生一个易于使用的框架，以帮助规划者评估和沟通自行车基础设施干预的定量影响。

This research advances quantitative methods for cycling infrastructure planning and evaluation. The research has two main goals. First, to develop an image-based machine learning algorithm to support rapid expansion of the Level of Traffic Stress (LTS) assessment and cumulative accessibility methods to new municipalities with available aerial and street level photography (e.g., all Canadian cities). Second, a novel mathematical optimization model will be developed to facilitate rapid identification of the streets and intersections where adding cycling infrastructure would provide the most overall benefit. The project will yield an easy to use framework to help planners evaluate and communicate the quantitative impact of cycling infrastructure interventions.