Every city has the corridor. The one where the signals seem to turn red as you arrive, where the left-turn arrow runs for nobody at 9 p.m., where the timing plan was built from a count taken before the shopping center opened. The Federal Highway Administration puts a number on it: outdated signal timing accounts for more than 10 percent of all traffic delay on major routes.
Adaptive signal control is the fix for that corridor. But when it comes time to fund it, the question from the council or the commission is always the same: what do we actually save? This post pulls together what agencies have reported, where the savings come from, and the one variable that decides whether your deployment lands anywhere near those numbers.
Fixed-time signals run a schedule. Actuated signals respond to a detector call at the stop bar. Adaptive signal control goes a step further and continuously re-plans cycle lengths, splits, and offsets along a corridor based on what the detectors are seeing right now. If you want the fundamentals first, start with our explainer on how adaptive signal control works and then come back.
The figures below are from public evaluations, not vendor brochures. Results vary by corridor, and the biggest gains come where timing was most out of date.
It helps to separate the ROI into four buckets, because they land in different budgets and different people care about each one.
Driver time. This is the biggest number and the hardest one to put in a city budget, because it accrues to the public rather than the agency. Travel-time savings multiplied by daily volume and a value of time is how FHWA and most benefit-cost studies calculate it, and it is why the ratios run so high.
Fuel and emissions. Idling is where fuel goes to die. Every stop you remove from a corridor is fuel not burned and emissions not produced. For cities with climate action plans, this is often the line item that unlocks a funding source.
Retiming labor. This is the bucket that lands squarely in your budget. FHWA pegs conventional signal retiming at roughly $2,500 to $3,100 per signal per update, and the standing guidance is to retime every three to five years. Most agencies go far longer because the staff time is not there. An adaptive system retimes itself every cycle. The engineers you have can spend their time on the corridors that need judgment instead of the ones that need maintenance.
Safety. Fewer stops means fewer rear-end conflicts. Better detection of the dilemma zone means fewer drivers caught between stopping hard and running the red. Pair adaptive control with a dynamic all-red extension and the intersection can hold the all-red for the vehicle it sees about to run the light.
The cost side of the ledger is dominated by what is already at the intersection. An intersection with a modern controller, working communications, and reliable detection on every approach is a software upgrade. An intersection with a 20-year-old cabinet, no fiber, and half its loops dead is a construction project before it is an adaptive project.
Detection is usually the biggest variable. FHWA's Traffic Detector Handbook notes that the detection element is only about 4 to 10 percent of the cost of implementing intersection control, yet accounts for a significant share of the maintenance dollars. Budget for it up front rather than discovering it in year two.
Here is the part that gets left out of the sales deck. Adaptive logic is only as good as the detection feeding it. A missed call is a green that never comes. A false call is a green nobody uses. A dead loop set to recall is a phase that runs to max every cycle regardless of demand. An adaptive system fed by detectors that fail at night, in rain, or quietly in the pavement will optimize beautifully for traffic that is not there.
The corridors that hit the FHWA numbers are the ones where detection was treated as the foundation rather than an afterthought. We go deeper on the specific failure modes in why adaptive signals underperform, and why it is usually the detection.
This is the problem FusionSensor was built to solve at the intersection. One device per approach fuses HD3D radar with AI video, so it sees presence, speed, and range in any weather while classifying vehicles, pedestrians, and cyclists at 98.7 percent accuracy. Radar does not care that it is 2 a.m. and raining. Video knows the difference between a bus and a pickup. Together they give the adaptive controller stop-bar presence, advance detection, per-lane counts, and pedestrian calls from a single unit mounted on the existing mast arm, with nothing cut into the pavement.
The same data does double duty. The counts and classifications that drive signal timing also feed your traffic data collection program and your performance measures, which means the detection you buy for adaptive control is also the turning-movement study you would otherwise pay for separately. That is how the ROI on adaptive signal timing gets better over time instead of worse.
If you are writing the spec, our checklist for specifying intersection detection for adaptive control covers what to require and what to ask every vendor. Or talk to us about what FusionSensor sees at your worst intersection.
Subscribe to The Edge Brief for monthly field notes, deployment stories, and product updates.
Subscribe →