Car Counting Traffic Survey: Manual and Automated Methods

Counting cars sounds simple until you have to do it for an 8-hour intersection survey. This guide covers the methods traffic engineers and students actually use to count vehicles, from a multi-counter app on a phone to pneumatic road tubes and video analytics, and what the resulting data is used for in road planning, signal timing and city design.

Why traffic counts are done

Traffic volume drives nearly every road design decision: lane counts, signal timing, intersection geometry, roundabout-versus-traffic-light choices, and capacity expansion priority. Average Daily Traffic (ADT) is the headline number — total vehicles per day in both directions — but Peak Hour Volume is more useful for capacity planning. Turning movement counts at intersections (left, through, right by approach) are essential for signal optimisation and modern transport planning.

Method 1: Manual counts with a tally counter

A counter sits at the intersection and clicks once per vehicle by movement type. For a four-leg intersection, that is twelve movements (left, through and right per approach). A multi-counter app lets one observer track three movements per approach on a single phone. SimpleCounter supports unlimited named counters — perfect for a single observer covering one approach with three named counters and a synchronised clock.

Method 2: Pneumatic road tubes

Rubber tubes stretched across a lane register an axle pulse per vehicle pass. They are cheap to deploy ($500–2,000 per setup) and good for 24-hour or multi-day counts on rural and suburban roads. The catch: they cannot distinguish turning movements at intersections, so they are best suited to through-volume measurement on road segments rather than intersection studies.

Method 3: Inductive loops in pavement

Wire loops embedded in the road detect metal vehicles passing over them. They are used permanently at signalised intersections to drive signal logic. Inductive loops provide continuous data 24/7 once installed but cost $5,000+ per loop and require pavement cuts. They are standard equipment at most major signals in developed cities and are invisible to drivers.

Method 4: Video and AI counting

Modern systems mount a camera on a pole, stream video to a server with computer vision software, and output counts by movement, vehicle class and direction. Install costs run $5,000–20,000, but the per-day cost is much lower over time. Used for both temporary surveys and permanent monitoring. Accuracy is 95%+ in good conditions and degrades in heavy rain or low light without infrared support.

How to run a manual count survey

Plan the intersection diagram and movement codes. Print a sheet for each 15-minute period or, better, use a multi-counter app with one named counter per movement. Position observers safely off the road — one per approach for high-volume sites, one per intersection for low-volume. Synchronise counters with the same phone clock and use end-of-period beeps to prevent missed transitions between intervals.

Common count survey errors

The most common mistake is missing the actual peak. Always include the real rush hour (7–9 AM, 4–6 PM in most cities) — two hours is the minimum for capacity analysis. Other common errors: skipping bicycle and pedestrian counts (modern complete-streets analysis requires them), failing to classify heavy vehicles separately, and counting in atypical weather and treating it as normal.

Choosing intervals and durations

Standard practice is 15-minute bins so peak hours can be precisely identified. For most studies, an 8–12 hour count covering both AM and PM peaks gives a defensible volume figure. For capacity studies at signalised intersections, multiple weekdays improve confidence and average out random variation.

Free starter setup for a student survey

All you need: a smartphone with a multi-counter app, a clipboard for the intersection diagram, and a watch or timer for 15-minute intervals. Two observers can cover a four-leg intersection by splitting approaches. Total cost: zero if you already have a phone, with accuracy comparable to professional manual counts.

From count to decision

Raw counts feed into signal timing software, capacity analysis spreadsheets, and long-term transport planning models. Even a single accurate intersection count can change a signal's green-time split, justify a new turning lane, or influence a planning decision. Good counting genuinely changes how a road works for everyone using it.

Frequently Asked Questions