tru-traffic.com

I am using Tru-Traffic to replicate existing timing for the "before" conditions travel time runs. Most of my corridors are not on a common cycle length before I retime them, so many of the intersections get changed to a multiple of the Network Cycle Length. Although I still get correct delay and stop information, it decreases the effectiveness of the plots overlayed with the time-space diagram if a large percentage of the intersections do not have they correct cycle length. My first choice would be the elimination of this limitation altogether, or at least where you could specify cycle length by sections. But if that is not possible, could we get some additional options such as 1/4 increments (1/4, 3/4, 5/4)? A common instance is a combination of 90" and 120" cycle lengths, which currently cannot be shown together on a diagram.

Thank you for the fine suggestions.

You'll be pleased to hear that the developmental version of the software already adds more adjusted cycle length options: 3/4, 4/3, 1/4, and 4x.

To work around the problem In the meantime, or even thereafter if the actual cycle length is not related to the network-wide cycle length by any of these close harmonic ratios,

1. Choose whichever adjusted cycle length is closest to the actual cycle length,
2. Set the splits (in seconds) to their correct values,
3. Ignore the subsequent warnings about the splits' sum not equaling the cycle length, and
4. Set the Vertical Scale on the diagram to show sufficient cycles so even the special-cycled intersections are fully represented in terms of how they relate to the signals with different cycle lengths.

Since ver. 9.0 will include the additional adjusted cycle length options listed above, it will allow mixing 90" and 120" cycles without complaint. All the same, it will obviously not accommodate all possible mixtures of cycle lengths without this workaround. For example, any cycle ratios involving 5 or above (e.g., 1/5, 2/5, 3/5, 4/5, 5/4, 5/3, 5/2, 5x, or anything with yet higher integers) still require this workaround. I'm reluctant to add options involving 5 or greater, as it seems to me that the notion of "coordination" is mostly abandoned when the cycle lengths are related this remotely, and at some point the warnings about the splits' sum not equaling the cycle length can be a useful reminder that this intersection is not in coordination.

Your further comments most welcome.