Signal retiming is one of the highest-anxiety tasks for any traffic engineer. Is it going to work? Did we enter the timing plans correctly? Are we going to get complaints? The usual way to manage that uncertainty is to go all-in after deployment: send people out to monitor the corridor, then send them out again. Or just wait anxiously to see what comes back.

Optimus is built to answer those questions before deployment instead of after, and to keep answering them once the plan is live. This is the first part of a three-part blog post series that will help engineers understand how they can use Optimus to reduce the anxiety of timing plan deployments.

Ask an engineer how confident they are in a timing plan coming out of a traditional tool like Synchro and the answer is usually some version of “Not very, but it’s standard practice.”

That lack of confidence at the front end has a cost. When you can’t trust the plan before it ships, you compensate afterwards by spending more time in the field, more days spent monitoring, more waiting to see what the plan really did. The uncertainty doesn’t go away. It just moves downstream, where it’s more expensive to resolve.

Performance predictions are how Optimus moves that certainty back to where the decisions get made.

An accurate forecast, before anything reaches a controller

Every timing plan generated by Optimus includes an accurate forecast of how it will perform in the field across a wide range of metrics (including control delay, arrivals on green, queue length, V/C ratio, and others), providing them at the network level as well as more granular forecasts at the intersection, phase, and movement levels.

In empirical studies, Optimus’ delay forecasts carry 75% lower error than HCM-based methods. With high levels of accuracy you can be confident that you have a clear understanding of what’s going to happen in the field without having to go there.

High-level predictions: decide whether to retime at all

The first thing an accurate forecast lets you do is decide whether the timing plan is worth deploying at all. Bad performance can’t always be solved by a new timing plan. For example, an oversaturated corridor, where demand already exceeds capacity, has very little room to give back no matter how you change the splits. The constraint is capacity, not timing. Retiming it produces marginal gains at best, and those gains cost real engineering hours and effort to capture.

Network-level predictions let you see the total benefit before you commit to any of that. If the expected improvement is marginal, the most valuable decision you can make is not to deploy, and to spend those hours on a corridor where they’ll move the needle. If the benefit is real, you move forward knowing the project was worth starting. The decision is made on expected impact rather than on a fixed retiming cycle or the loudest complaint.

Granular predictions: find the movements that will get worse

Even a timing plan with substantial performance improvements at the network level will still have different impacts in different parts of the network. Most intersections get better but a few movements get worse. And some complaints have nothing to do with performance getting worse at all: they come from performance being different. A driver used to sailing through the mainline and stopping at one particular intersection now stops at an earlier one instead. Nothing about that is worse on paper, but it isn’t what they’re used to, and the phone rings anyway.

Optimus predicts performance down to the network, intersection, phase, and movement level, so you can see exactly where the plan degrades before it ever reaches the field. That granularity is what lets you get ahead of the complaint instead of reacting to it. You can flag the movements that will change, prepare a response for the locations that will feel worse, and tell the public ahead of time what they’ll experience at those spots and what they’re gaining everywhere else on the corridor. A complaint you predicted and pre-empted is a very different conversation than one that catches you off guard.

The payoff

Accurate, granular predictions change three things about a retiming. You avoid projects that were never going to pay off. You walk into stakeholder and public conversations with the numbers already in hand. And you know what the plan does before it does it, which is the certainty the whole process was missing.

Next in the series: Timing Plan Validation — using Split Analysis and the time-space diagram to validate and fine-tune a plan before it ships.

Are you retiming a corridor soon? Do it with confidence — contact our team.