基于模糊推理的公共交通分担率预测研究

研究基于模糊推理建立公共交通分担率预测模型的方法。通过分析公共交通出行的影响因素,选择线网密度、国内生产总值、平均车速等十项指标作为影响因素,建立模糊层次结构模型。确定各因素的模糊规则,采用模糊推理预测公交分担率。并以黑龙江省某市交通数据进行预测,最后分析了影响因素与分担率水平的影响关系,应用得到了预期的效果。     

The New Regionalism: Key Characteristics of an Emerging Movement

This article analyzes the emergence of a “new regionalism” and situates this movement within the historical evolution of regional planning. Key characteristics include (1) a focus on specific territories and spatial planning; (2) a response to the particular problems of the postmodern metropolitan region; (3) a holistic perspective that integrates planning specialties as well as environmental, equity, and economic goals; (4) a renewed emphasis on physical planning, urban design, and sense of place; and (5) a more activist or normative stance on the part of planners. The implementation of new regionalist concepts is likely to come about not through top-down regional government, but through incremental development of social capital, institutions, ad hoc partnerships, and frameworks of incentives and mandates between existing levels of government.     

What Is a Forecast for?

Abstract

Problem, research strategy, and findings: The forecasts transit agencies submit in support of applications for federal New Starts funding have historically overestimated ridership, as have ridership forecasts for rail projects in several countries and contexts. Forecast accuracy for New Starts projects has improved over time. Understanding the motivations of forecasters to produce accurate or biased forecasts can help forecast users determine whether to trust new forecasts. For this study I interviewed 13 transit professionals who have helped prepare or evaluate applications for federal New Starts funds. This sample includes interviewees who have had varying levels of involvement in all 82 New Starts projects that opened between 1976 and 2016. I recruited interviewees through a snowball sampling method; my interviews focus on the interviewees’ perspectives on how New Starts project evaluation and ridership forecasting has changed over time. Interview results suggest that ridership forecasters’ motivations to produce accurate forecasts may have increased with increased transparency, increased influence on local decision making, and decreased influence on external (federal) funding.

Takeaway for practice: Planners can evaluate the likely trustworthiness of forecasts based on transparency, internal influence, and external influence. If forecast users cannot easily determine a forecast’s key inputs and assumptions, if the forecaster has been tasked with producing a forecast to justify a predetermined action, and if an unfavorable forecast would circumvent decisions by the forecaster’s immediate client, forecasts should viewed with skepticism. Planners should seek to alter conditions that may create these conflicts of interest. Forecasters seem to be willing and able to improve forecast accuracy when the demand for accurate forecasts increases.     

Complementary Pricing and Land Use Policies: Does It Lead to Higher Transit Use?

Problem, research strategy, and findings: A shift toward more sustainable transportation requires both adequate pricing of externalities from driving and supportive land use policies. However, proponents of each approach often under-estimate the complementarity and potential synergy between them. This study investigates the interaction effects between gasoline prices and land use (policy) variables using a panel dataset of transit ridership in 67 urbanized areas between 2002 and 2010. We found that while doubling the average gasoline price would increase transit ridership by 8.4% in an urbanized area with mean density and no regional containment policy, in areas with slightly higher density and a regional containment policy, the impact of higher gasoline prices would rise to 21%. In communities that had adopted a package of smart growth land use options, the impact of higher gasoline prices on transit use is even greater.

Takeaway for practice: Pricing schemes will be more effective where alternatives to automobility and supportive land use policies exist. The impacts of urban form on travel behavior are also strengthened when driving externalities are correctly priced. Planners and policymakers should take advantage of the complementarity between pricing and land use planning approaches by implementing policies in combined and well-coordinated ways.     

基于视频稳像和视角变换的公交客流计数方法

已有的公交客流计数方法大多没有考虑车体振动引起的视频抖动问题和摄像头角度引起的图像梯形失真问题,因此提出一种基于视频稳像和视角变换的公交客流计数方法。首先采用基于块匹配的视频稳像技术,减小车体振动引起的图像序列间的偏移;然后通过对视频图像进行视角变换,校正摄像头角度引起的图像梯形失真;最后采用基于头肩部特征的人体检测及跟踪方法,统计各个时段的上下客流量和乘客总人数。经实验测试,稳像处理后的视频与稳像前相比,峰值信噪比（PSNR）提高了约5.5dB,稳像速率达到25帧/s。视角变换后的视频与原始视频相比,人体识别率提高了约10%,大大提高了客流计数精度     

基于客流数据的区域出行特征聚类

区域功能发现对完善城市规划有着重要的指导意义.区域居民的出行特征提取与发掘可以作为建立模型分析区域功能的数据支撑.随着智能交通技术在轨道交通系统的应用,大量蕴含行人移动性和出行目的地信息的客流数据被采集得到,发现客流数据与地铁站相关区域功能有紧密联系.从地铁客流数据中提取出乘客出行模式和地铁站客流模式,并以此为基础建立概率图模型,实现了区域出行特征聚类.首先,以地铁客流数据为基础提取了乘客出行模式和地铁站客流模式,发现地铁站客流集中性和潮汐性的特性,能在一定程度上反映地铁的区域功能.然后,采用了文本分析领域经典的概率图模型,建立基于潜在狄利克雷分配(latent Dirichlet allocation, LDA)主题模型的地铁客流出行特征聚类模型,将具有出行规律相似性的地铁站聚类在一起.最后,通过分析聚类实验结果,发现在不同客流峰段内的区域功能和相互客流关系.