Former Los Angeles mayor Eric Garcetti’s recent move into an advisory role at an AI company is one more indication that cities are taking intelligent infrastructure more seriously. The public conversation often focuses on AI itself, but for transportation agencies, the harder question is more practical: can a city measure what is actually happening on its streets with enough historical depth, spatial precision, and engineering reliability to improve mobility?
That is where traffic monitoring becomes more than a dashboard. Mobility improvement depends on knowing how traffic behaves across seasons, weekdays, hours, intersections, and individual road segments. A single count, a short survey, or a generic traffic layer cannot reveal whether congestion is caused by demand growth, poor signal timing, seasonal fluctuation, turning movement imbalance, capacity limits, or non-recurring disruption. Ticon’s approach is built around that distinction.
In AADT Estimation by Various Methods: Accuracy and Reliability, Gregory Brodski and Alex Chaihorsky describe Ticon as a traffic information consolidator, combining traffic engineering methodology, multivariate analysis, and large-scale mobility datasets. The purpose is not only to estimate Annual Average Daily Traffic, or AADT, but to produce traffic flow volumes at hourly resolution and, where appropriate, 15-minute intervals, with spatial resolution down to an exact address.
This matters because transportation systems do not fail at average conditions. They fail at specific times and locations. A corridor may look acceptable in annual averages while breaking down every Tuesday afternoon near one intersection. A retail access point may appear adequate in a daily total while producing a short but severe left-turn queue during the evening peak. A signal retiming project may reduce delay on one approach while increasing it downstream. Historical traffic data allows engineers to see these patterns before deciding what to change.
Ticon’s methodology is designed around ampleness of observation. Its high-resolution dataset supports 100% temporal and spatial coverage, with traffic speeds, volumes, and derived metrics available for short road segments, as short as 10 meters and averaging about 40 meters in one Ticon mobility improvement dataset. That level of resolution is important because congestion is often spatially narrow. A bottleneck may begin at a merge, driveway, turn bay, signalized approach, or short weaving area, then propagate across a much wider network.
The limitations of conventional traffic data collection are well known. Portable counts may observe only 48 hours, roughly 0.5% of the year. A one-week count covers about 1.9% of annual time. Manual counting during selected peak intervals may capture about 1.35% of time. These methods can still be useful, especially for calibration and field verification, but they are vulnerable to seasonal effects, weather, incidents, school calendars, construction, and atypical daily demand. Ticon’s reports note that low time coverage is a common source of biased results in before-and-after analysis, especially when traffic demand varies by week, month, or season.
Ticon’s traffic data collection approach addresses this by consolidating multiple independent sources and cross-verifying them. The platform incorporates geospatial information, road geometry, intersection geometry, road segment connectivity, speed distributions, expected vehicle composition, driver behavior under different conditions, time of day, congestion state, and external influences such as weather. Its analytical stack includes speed-volume analysis, saturation analysis, capacity analysis, before-and-after analysis, retrospective analysis, and network bandwidth utilization.
The accuracy work behind this methodology is central to its value. In the AADT evaluation described by Brodski and Chaihorsky, Ticon compared estimates with publicly available DOT count data across Georgia, Nevada, and California. The study selected 695 counting points. After removing 28 points due to detector-counting errors and 30 due to GPS data gaps, 637 points remained, most of them bidirectional. That produced more than 1,200 Ticon AADT estimations. The reported median average percentage error was 4.78%, with a Relative Root Mean Square Error of 11.97%. Ticon reports that this level of accuracy keeps expected AADT error within 20% boundaries at 90% confidence.
For planners and traffic engineers, the confidence interval is not an academic detail. It determines whether a traffic estimate can support a capital project, signal timing decision, safety study, traffic impact analysis, or commercial site evaluation. If uncertainty is unknown, the analysis may look precise while being unreliable. Ticon’s methodology explicitly reports expected error, MAPE, RRMSE, and confidence boundaries, which allows the user to judge whether the data is suitable for the task.
Historical traffic monitoring also changes how cities evaluate mobility improvement projects. Ticon’s TrafficScope workflow supports analysis for any period within the past 10 years, allowing agencies to compare conditions before and after a signal retiming, lane reconfiguration, access change, new development opening, construction project, or ITS deployment. Instead of asking only whether average speed improved, the platform can identify time slots and locations where a measure helped, where it had no effect, and where it caused degradation.
That distinction is essential for public accountability. A new traffic control system may reduce delay during the morning peak but worsen midday operations. A corridor project may improve one intersection while shifting queues to the next. A restriction may reduce volumes but fail to restore free-flow speed. In Ticon’s COVID-era traffic management analysis, traffic restrictions produced delay reductions of up to 60%, but average speeds reached free-flow conditions only after average daily traffic fell by almost 10 times relative to standard levels. The engineering implication was clear: local traffic control quality, including signal timing, seasonal timing plans, and adaptive ITS measures, can be more useful for mobility improvement than demand suppression alone.
This is where historical traffic data becomes a practical tool for continuous operations. Mobility is not improved once and then left alone. Vehicle fleets change, delivery patterns change, retail and employment patterns shift, school schedules vary, roadworks alter capacity, and community expectations around safety and emissions continue to rise. Ticon’s high-resolution monitoring allows agencies to rank problematic zones, study level of service, measure saturation, evaluate network bandwidth utilization, and verify whether implemented measures meet the intended performance target.
The same logic applies to private-sector location decisions, including the recent opening of QuikTrip’s first Utah store near Interstate 15 and the sale of a fully leased retail strip center in Fontana, California. These are not transportation projects in the public-agency sense, but they depend on the same traffic reality. A corridor’s daily volume is only the beginning. Retail and fuel sites depend on directional flow, heavy-vehicle share, commuter peaks, pass-by traffic, access quality, turning opportunities, and seasonal demand. Ticon’s research on site selection emphasizes that standard AADT alone is insufficient when intra-day, intra-week, and monthly variation can alter business performance.
For mobility improvement, the broader lesson is the same: traffic monitoring must be both historical and granular. Engineers need to know not only how many vehicles use a road, but when they arrive, how speeds deteriorate, how queues form, how volumes relate to capacity, and whether a change improved the system or merely moved the problem. Ticon’s combination of year-round observation, cross-verified multisource traffic data, high spatial resolution, and engineering analytics gives planners a more reliable basis for these decisions.
The future of traffic management will involve more AI, more automation, and more connected infrastructure. But the value of those tools will depend on the quality of the historical traffic data beneath them. Without accurate monitoring, cities risk optimizing noise. With detailed and verified traffic intelligence, mobility improvement becomes measurable, repeatable, and easier to defend.
Join the network to access Ticon case studies and to receive Ticon newsletters
@2023 Ticon
