Turn pocket blockage and spillback models: Applications for signal timing and capacity analysis

William L. Reynolds, Nagui M. Rouphail, Xuesong Zhou

Research output: Contribution to journalArticlepeer-review

6 Scopus citations


The development of a macroscopic simulation tool is presented and tested to quantify the effects of short turn pockets on the sustainable service rate of a signalized intersection. Unlike the theoretical signal capacity, the sustainable service rate includes queue interaction effects and is thus influenced by blockage and spillback at the entrance to a short turn pocket. Previous research on the topic has focused either on the probability of spillback from a short turn pocket or the operation of a system with a single approach lane. No macroscopic model currently available has the ability to analyze throughput reductions due to the effects of short turn pockets on a multilane approach. The model described here uses a series of flow and density restrictions on cells of varying sizes on the approach to the intersection. Results, which indicate sensitivity of the model to turn pocket spillback, blockage, saturation flow rate, pocket length, lane utilization, cycle length, phase sequence, phase overlap, permitted phasing, and time-dependent demand, compare favorably with microsimulation. A phase optimization procedure is described to help efficiently allocate green time for a given set of turn pocket lengths and turn movement percentages. Implications of using the model for signal timing applications are discussed, and recommendations are made on additional model enhancements and testing needs.

Original languageEnglish (US)
Pages (from-to)112-122
Number of pages11
JournalTransportation Research Record
Issue number2259
StatePublished - Dec 1 2011
Externally publishedYes

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering


Dive into the research topics of 'Turn pocket blockage and spillback models: Applications for signal timing and capacity analysis'. Together they form a unique fingerprint.

Cite this