Study On Headway Reliability Of Bus Rapid Transit

Increasingly global concerns about traffic congestion, air pollution, energy shortage as well as regional economic development have placed great traffic demand on the public transit systems. Bus rapid transit (BRT), serving as a green quick and relatively inexpensive mass public transit mode, is growing in popularity throughout the world nowadays. Over the past decades, BRT has developed very quickly in China with many cities have provided BRT systems. In additiona, with over 140 cities of more than 1 million in population, the BRT market in china is very large. On the other hand, given that unreliable service has been cited as one of the major deterrents to existing and potential passengers, transit agencies have faced more and more with the immediate problem of providing credible transit service to ensure their attractiveness. Lack of reliability in public transit results in excess waiting time and discomfort for passengers and it also increases operating cost for operators due to the lost mileage and lower fleet utilization. It is thus of great practical significance to invesgate the service reliability of bus rapid transit. As a high frequency public transit, service reliability of BRT is mainly reflected by its headway reliability, but research focus on the headway reliability of BRT is absent. Our research on the headway reliability of BRT has theoretical significance.Firstly, a microscopic measure index of headway reliability named headway adherence ratio (HAR) is defined. HAR is calculated as the ratio of difference between actual headway and scheduled headway with respect to the scheduled headway. As uneven scheduled headway is common in practice, and which is closely related to performance of bus headway on a bus route. HAR could measure the headway reliability in a relative way by taking scheduled headway into consideration. It also makes performance of the bus headway reliability comparable for those bus routes with different scheduled headways.This paper addresses the interesting and practical bus headway adherence issue for a given BRT route with a number of bus stops. The uncertain parameters affecting bus headway makes us to investigate the bus headway adherence issue from a probabilistic perspective. This study proceeds to customize a four-step procedure to estimate a unified probability distribution that can describe the random HAR at each BRT stop of the BRT route by using the automatic vehicle location (AVL) data. Our real case studies with 44,025 HAR data show that the 19 existing probability distributions including Lognormal, Gamma, Beta and Wakeby are unable to well fit these HAR data. A tailored Wakeby-type distribution with five parameters is proposed to fit HAR of BRT stop. After deriving two fundamental propositions for the tailored Wakeby-type distribution, a tangible L-moment based method to estimate those parameters involved in the tailored Wakeby distribution is presented. The tailored Wakeby-type distributions can meet our expectation via our real case studies.Methods for headway reliability analysis are proposed at the level of stop and route respectively. At a stop level, based on the distribution of HAR, some probability-related indexes of headway reliability are proposed. And then, this paper presents an in-depth analysis on the existing indexes of headway reliability performance, and groups them as headway adherence, headway regularity, and regularity of headway deviation. At a route level, the Spearman’s rank correlation coefficient is introduced to analysis the headway adherence correlation. Finally, the proposed mythologies are applied in a case study to show the different headway reliability performance among a main BRT route and two branch BRT route. The contrast analysis could expand the understanding of BRT characteristics and the range of possible reliablility performance and benefits。Factors effecting headway reliability of BRT are studied. They are qualitatively analyzed from the perspective of running way, design of BRT stop, priority strategies at signial intersections and BRT vehicles. To research these effecting factors in a quantitative way, this paper studies headway reliability via dispatch interval reliability and travel time reliability. To explore the effect of dispatch intervel reliability to headway reliability, a simulation model for BRT arrival time is proposed.Finally, this paper addresses factors affecting headway reliability of BRT form the perspective of travel time reliability, which handles the two components of the BRT travel time:driving time and dwell time, respectively. With Automatic Vehicle Location (AVL) data collected from one real BRT route, impact of these two components on BRT travel time variance is quantitatively analyzed. Four models for mean and variance in BRT driving time and dwell time are built by the linear regression method. Characteristics of BRT for decision making are involved in these models. To represent the effect of signalized intersection on the BRT driving time more reasonable, a variable named weighted signal density (WSD) is proposed, both signal timing and density of signalized traffic intersection could be reflected by this variable. In the BRT dwell time analysis models, BRT station location (near-side and far-side) is considered as an effecting factor.This paper is useful to make a depth understanding on the distribution rule and influence mechanism of the headway reliability of BRT. From the perspective of operator, this paper is helpful to provid a set of method for BRT service measure and management. From the view of researches, this paper is useful to headway reliability based reseach.