Date: Apr 27, 2012 Source: DOT Success Story
Through a Small Business Innovation Research (SBIR) award from the U.S. Department of Transportation’s Federal Highway Administration, Traffax Inc. developed and deployed a system of 48 real-time BluFax BluetoothTMTraffic Monitoring (BTM) sensors on major freeways and signalized arterials in Northern Virginia and suburban Maryland in 2011. The B™technology was developed at the University of Maryland and licensed for commercialization in 2008 to Traffax Inc. Initially developed as a portable, easily deployable stand alone system, BluFax B™has provided significant utility for traffic studies and origin--destination studies in the US and abroad. The focus of the SBIR project was to bring high-resolution B™data to real-time applications for use by operations, traffic, and other purposes. The 48 unit deployment has been in full operation for over six months, providing high-resolution travel time data in real-time travel, sampling approximately one in twenty vehicles from these heavily traveled corridors. A portion of the network is publically viewable at http://traffax1.blufaxweb.comusing a login of sbir and password of sbirpw. The portion of the network in suburban Maryland is shown in Figure 1, and a sample installation pictured in Figure 2 (see attachments).
With the network fully deployed, attention has turned to applications of the data delivered by the system. Anticipated applications include: travel time data with means and distributions accurately reflecting real-world flows, validation of other traffic data systems, incident detection and characterization, observation of traffic diversion onto arterials during congestion, and performance measure calculations. Specifically for interrupted-flow facilities (signalized arterials), real-time monitoring, performance measures, quality of progression and signal timing assessment have been targeted.
Arterial monitoring is highlighted in this brief summary because tools used prior to the development of B™sensors were inadequate in characterizing complex flows. A single measure of central tendency (such as the mean, median, inter-quartile mean, etc.) is often inadequate to convey the quality of flow. Describing the quality of traffic flow requires a detailed analysis of the distribution of speeds and travel time in a corridor. To demonstrate the capabilities of B™data to support this kind of analysis, the project includes two major arterial corridors. In Maryland, MD 355 parallels I-270 in Montgomery County, providing access to services, and alternative routing in the event of freeway incidents. Likewise, in Virginia, Route 1 parallels I-95 to the south of the I-495 beltway. The B™deployment covered both corridors.
In partnership with Purdue University, a guidebook for the use of B™data on signalized arterials is being constructed (currently in draft) and is targeted at multi-lane signalized arterials. The signal coordination along such corridors has a significant impact on reliable corridor travel time. Through the combination of B™data and high resolution controller data, the guide illustrates an effective methodology to identify traffic signal timing problems with B™data, develop solutions with high-resolution controller data, and then verify the solution and quantify user benefits with B™data. The method has substantial cost and resource benefits over existing methods that rely on floating car runs.
Bluetooth based methodology has potential not only to improve traffic signal operations, but is efficient enough to begin to systematically assess when and how signal timing should be addressed, in a way that tracks the aging and adequacy of all signal timing in a region. The tutorial in the guidebook provides an easy to follow example that identifies a Saturday timing problem on a major arterial, quantifies the impact in terms of user cost, verifies the solution, and then describes the benefit in easy to use terms for decision makers and the general public.
The graphic in Figure 3 is a before and after cumulative frequency diagram (CFD) analysis clearly depicting the benefits of the methodology for one movement of the case study presented in the tutorial. [Note the CFD is a plot of the percentiles within a distribution, and is by definition a continuously increasing plot from 0 to 100%.] The sample in Figure 3 illustrates the before case (labeled Base) in which the median travel time was approximately 4.5 minutes, and several measurements made after signal timing improvements were performed.
