Roadrunner: A Tool for Real-World Solutions
Four Points to Consider
In the real world, people don’t usually travel in a straight line: On occasion they will make a left or right turn. Sometimes they even do both. Yet current radar, video, loop, and other common systems provide arterial estimates. The only thing known for certain is the radar, loop, or video measurement itself. Travel times must be made by assumptions. Assumptions which, as everyone knows, will make a fool out us more often than not.
True, 802.11 sampling systems are not as accurate radar and loop counts, but radar, video, and loop counts are disconnected from time, they are discrete, individual samples. 802.11 sampling, however, connects the dots and enables us to construct a vision of the entire matrix and see it for what it is.
Traditional systems like road tubes measure points: a singular value useful for counts. More sophisticated systems like radar measure counts and speeds–the point has direction, it has a speed, but little else is known. Roadrunner measures paths and velocity: after two measurements we know not only where the vehicle has been, but how fast it took to get to the next segment of its journey. Roadrunner tells us not just what the rate of change is, but the rate at which the rate of change changes.
Yes, Roadrunner can measure a single vehicle–a probe–as it travels through a municipality. But that is missing the forest for the trees. In fact, Roadrunner is a matrix sampler–it divides of a set of measurements into discrete segments within an overall system boundary. Matrixed designs, like other large-form statistical methodologies, have both advantages and disadvantages. For example, instead of a sample time taking the entire transit time of an individual vehicle (as with a floating car test), Roadrunner assembles the segment measurements into a snapshot that is only minutes in duration. In other words, instead of taking 30 minutes to complete a transit, numerous transits are broken apart and a new pathway chain is assembled. Instead of some vague average we have a more precise measure.
- The public gets accurate travel and transit times and optimal routes for trip planning and daily commutes.
- Maintenance gets a simple, easy to use system that is inexpensive and easy to maintain.
- Engineering gets the data it needs for progressive signalization and incident management and detection.
- Planning gets the longitudinal data it needs for not only infrastructure improvement, but optimization.
The outcomes here are subtle but powerful. With Roadrunner the system can not only run logically, but optimally and with a degree of precision and efficiency executed on multiple levels and not currently enjoyed.
- Reduce arterial delay by 80%
- Reduce arterial travel time by 70%
- Reduce total system delay by 60%
- Reduce intersection delay by 70%
In order for these systems to succeed, though, requires accurate and timely traffic profiles. Roadrunner not only provides approach segment travel times, but segment travel times that can be weighted by time of day, day of week, and week of month. The system can help you build a robust database of transit characteristics that won’t be thrown off by outliers, as well as the flagging of outliers that signal a change in traffic characteristics due to accidents, special events, and road closures–both anticipated and unanticipated.