城市道路非机动车骑行品质评价:一种自然骑行实验方法

Evaluation of bicycling environment on urban roads:A naturalistic bicycling-based method

中国的很多城市道路往往以机动车为主导进行规划设计,导致非机动车骑行环境安全性和舒适性较低,对路段非机动车骑行品质进行评估,找出问题路段进行改善至关重要。本文基于自然骑行实验,提出了一种包含设施属性和骑行行为特征参数等评价指标的城市道路非机动车骑行品质评估模型。首先设计实验车,通过自然骑行实验采集交通设施参数、骑行行为参数及骑行品质评分等数据,随后基于有序Logit模型识别关键指标,并建立随机森林评价模型;最后,以上海市杨浦区某小路网为例对模型进行应用,绘制了骑行品质地图,可直观评价路段骑行品质。本文所提出的评价模型可结合共享单车进一步推广应用,突破数据样本数量和质量的局限性,实现大规模路网骑行品质评估。

Many urban road segments in China are often designed with focus on vehicles,resulting in unsafe and uncomfortable cycling environments,thus,it is crucial to evaluate cycling quality on road segments to identify problematic segments for improvement.Based on naturalistic cycling experiments,this paper proposes a cycling quality evaluation model for urban roads that in-cludes facility and behavior feature indexes.First,an experimental bicycle is designed to collect data on traffic facility parameters,riding behavior parameters and riding quality scores through naturalistic cycling experiments.Then,critical indexes are identified based on ordered Logit model,and a random forest evaluation model is established.Finally,the proposed model is applied to a road network in Yangpu District,Shanghai,a cycling quality map is drawn,which can visually evaluate the cycling environment on road segments.The evaluation model can be further promoted and applied in combination with sharing bicycles,breaking the limitations of quantity and quality of data sample,to achieve cycling quality evaluation on large-scale road networks.