Bus Signal Priority (BSP), which has been deployed in many cities around the world, is
a traffic signal enhancement strategy that facilitates efficient movement of buses
through signalized intersections. Most BSP systems do not work well in transit
networks with nearside bus stop because of the uncertainty in dwell time. Unfortunately,
most bus stops on arterial roadways are of this type in the U.S.
This dissertation showed that dwell time at nearside bus stops could be modeled
using weighted least squares regression. More importantly, the prediction intervals
associated with the estimate dwell time were calculated. These prediction intervals were
subsequently used in the improved BSP algorithm that attempted to reduce the negative
effects of nearside bus stops on BSP operations.
The improved BSP algorithm was tested on urban arterial section of Bellaire
Boulevard in Houston, Texas. VISSIM, a micro simulation model was used to evaluate
the performance of the BSP operations. Prior to evaluating the algorithm, the
parameters of the micro simulation model were calibrated using an automated Genetic
Algorithm based methodology in order to make the model accurately represent the
traffic conditions observed in the field.
It was shown that the improved BSP algorithm significantly improved the bus
operations in terms of bus delay. In addition, it was found that the delay to other
vehicles on the network was not statistically different from other BSP algorithms
currently being deployed. It is hypothesized that the new approach would be
particularly useful in North America where there are many transit systems that utilize
nearside bus stops in their networks.
| Identifer | oai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/1460 |
| Date | 17 February 2005 |
| Creators | Kim, Wonho |
| Contributors | Martin, Amy Epps, Rilett, Laurence R. |
| Publisher | Texas A&M University |
| Source Sets | Texas A and M University |
| Language | en_US |
| Detected Language | English |
| Type | Book, Thesis, Electronic Dissertation, text |
| Format | 1514687 bytes, electronic, application/pdf, born digital |
Page generated in 0.002 seconds