What is a turning movement count? TMC basics for capacity work
A turning movement count, usually just called a TMC, is a tally of vehicles at an intersection broken out by the move they make: left, through, right, on every approach, for every hour the study covers. It's the input that feeds intersection capacity analysis, signal retiming, and most Highway Capacity Manual Level of Service work. If you've ever opened a Synchro file and seen a 15-minute bin table for each approach leg, that table came from a TMC.
The format is standard enough that most DOTs and consultants use the same layout: approach direction down the side, movement (left/through/right) across the top, 15-minute bins down the page, usually for one AM peak period and one PM peak period, sometimes a midday count if the corridor has retail or school traffic that shifts the peak. Trucks and peds get their own columns on most count sheets too, since heavy vehicle percentage changes the capacity math.
How a TMC gets collected
Three ways, mostly. A crew sits at the corner with clickers or a tally sheet, which still happens for small studies or when a signal needs retiming next week and there's no time to wait on a vendor. A traffic camera or portable video unit records the intersection and someone, or some software, counts the footage afterward, frame by frame if it's a contested movement. Or a firm with Miovision-style units or similar drops a pole-mounted camera for 24 to 48 hours and runs automated turning classification on the video.
A tube count catches axles crossing a line and gives you mainline link volume. It has no way to record which way a vehicle turned, so it can't produce the approach-by-approach breakdown a signal study needs. That's the distinction that trips up people new to traffic studies: a link count and a turning movement count answer different questions, and a capacity study usually needs both, just not from the same device.
When you need the full TMC versus just a link number
A lot of scoping time gets spent answering the wrong question first. If you need to know whether an intersection needs a new signal phase, or what the delay is on a left-turn lane at peak, you need the full TMC. Before that, though, many agencies are really asking which of six intersections on a list is busy enough to justify sending a crew at all. That screening question needs a directional volume on the link, not a turn-by-turn breakdown.
That's a cheaper problem to solve. A single VHR pass over the corridor gives you a vehicle count by link and direction without a crew, without a lane closure, and without a video unit mounted for two days. It won't replace the TMC once you've confirmed an intersection is worth studying in depth, but it answers the screening question: which of the links on your list are carrying enough volume to warrant the full effort, and in which direction. For agencies working through a long list of uninstrumented links before a capacity or signal-timing study, that's one pass over a corridor turned into a number you can act on instead of a week of tube deployment just to find out a segment isn't worth the study.
A few things that trip people up
A TMC counted on a Tuesday in a school-in-session week looks nothing like one counted during summer break on the same corridor, so the count date matters as much as the count method. Peak hour in the TMC data isn't always the same clock hour as peak hour on the link upstream, since turning traffic redistributes depending on what's downstream. And a TMC gives you movement volumes, not movement delay. You still need a field visit or a signal timing sheet to know whether that left-turn volume is causing a queue.
If your list of corridors is long and your crew time isn't, start by finding out which links are worth a closer look before you schedule the counters.