Traffic Control Plan Quantity Takeoffs: Turning a Plan Sheet Into a Device Count

An estimator hands a number to a project manager. The project manager hands a load list to the yard. The yard sends a truck. Somewhere in that chain, a plan sheet became a count — and if the arithmetic was skipped in favour of "that looks like a two-truck job," the difference shows up as a second trip, a rented arrow board, or a device shortage discovered at 6 a.m. on a road you cannot legally occupy after 9.

The takeoff is the step that makes that chain honest. It is also the step most often done from memory.

A caveat before the arithmetic: the criteria referenced here are from the national MUTCD, which is a floor. Your state manual and your permit conditions govern your actual job and may be more restrictive. Verify with the agency having jurisdiction.

Start by resolving the letters

Traffic control plans and MUTCD typical applications express distances as letters rather than numbers, because one drawing has to serve a 25 mph street and a 70 mph highway. The takeoff begins by turning those letters into feet.

Taper length is the big one. MUTCD Table 6B-4 gives L equals W times S squared divided by 60 at 40 mph or less, and L equals W times S at 45 mph or more, where W is the width of the offset in feet and S is the posted speed limit, the off-peak 85th-percentile speed before work starts, or the anticipated operating speed. Table 6B-3 then scales L by taper type: a merging taper is at least L, a shifting taper at least 0.5 L, a shoulder taper at least 0.33 L, and the one-lane two-way and downstream tapers are a flat 50 to 100 feet.

That last one is worth noting for takeoff purposes, because it breaks the pattern: a one-lane two-way operation does not scale its tapers with speed at all, so its device count is largely fixed while its labour count is not — two flaggers, or a pilot car, or portable signals, each with quite different hour and equipment lines.

Device spacing is the other. The spacing between cones, tubular markers, vertical panels, drums, and barricades should not exceed a distance in feet equal to 1 times the speed limit in mph for taper channelization, and 2 times the speed limit for tangent channelization.

Those two rules produce most of your count.

The arithmetic, worked

A single-lane closure on a 45 mph arterial, 12-foot lane, merging taper.

Taper length. 45 mph is in the upper band, so L equals W times S: 12 × 45 = 540 feet.

Taper devices. Maximum taper spacing is 1 × 45 = 45 feet. 540 ÷ 45 = 12 intervals, so 13 devices to close the taper.

Tangent run. Say the work space plus buffer runs 400 feet. Tangent spacing is 2 × 45 = 90 feet. 400 ÷ 90 ≈ 4.4, round up to 5 intervals, 5 more devices (the first is shared with the taper's last).

Downstream taper. 50 to 100 feet; at 100 feet with 45-foot spacing, 3 devices.

That is roughly 21 channelizing devices for one direction of one closure — before spares, before the opposing direction, before signs.

Now change one input. The same 12-foot offset at 30 mph gives L = 12 × 30² ÷ 60 = 180 feet, with 30-foot spacing: 7 devices. Same picture, one third of the taper. This is precisely why "it's a lane closure, send the usual" is not an estimate. Our taper length and buffer space explainer goes deeper into why the buffer is a separate quantity from the taper and should never be borrowed from it.

Signs are counted differently

Signs come off the advance warning area, and the driver is distance, not spacing. On urban streets the nearest warning sign should sit at 4 to 8 times the speed limit in feet; on rural highways, 8 to 12 times; on freeways and expressways, placement extends as far as half a mile or more, with the advance warning area running 1,500 feet or more for open highway conditions.

The count follows from how many signs the plan calls for in that area and how many approaches need them. The approaches are where takeoffs go wrong. A closure on a through street with three side streets feeding into the advance warning area is not three signs — it is three signs plus whatever each side street approach requires so that a driver turning in is warned. On a job with a detour plan, the detour route itself carries a full sign count of its own, frequently larger than the closure's, and it is routinely omitted from the takeoff because it is drawn on a different sheet.

Count stands and bases separately from sign faces. They fail, blow over, and get stolen at different rates, and a takeoff that assumes one base per face will be short by the end of a long job.

The quantities people miss

A list worth checking every takeoff against.

The opposing direction. Most typical applications show devices for only one direction. If your plan is a mirror, your count is roughly double, and if it is not a mirror, you need to read both.

The second phase. Plans for staged work show phases on separate sheets. Two phases are two setups and two teardowns, not one setup that gets nudged.

Spares. Devices are destroyed, buried, run over, and taken. A takeoff with zero spare allowance is a takeoff that guarantees a mid-shift shortage. Whatever percentage you use, use it consistently so that the variance means something — and read it against what you actually lose, which is the subject of charging for damaged and lost devices and of equipment theft and vandalism.

Night-work additions. A daytime plan and its night version are different bills of material — lighting, additional retroreflective devices, and warning lights the daytime setup never needed. Barricade and warning light types covers which type belongs where, and it is not interchangeable.

Crew hours as a quantity. Setup and teardown time scales with the device count you just calculated. A 540-foot taper is not the same labour as a 180-foot one. Feeding the device count into the hour estimate — rather than estimating hours independently from a gut feel about job size — is what makes the two numbers agree.

Standby. If the plan has a hold point, an inspection, or a utility crew you are waiting on, the standby hours are a quantity. Pricing them is covered in pricing standby and show-up time; forgetting them is how a profitable job becomes a break-even one.

Write down your inputs, not just your answer

The single highest-value habit in takeoffs: record the assumptions next to the count.

Speed basis used, and whether it was posted, 85th percentile, or anticipated operating speed. Offset width. Taper type. Spacing rule applied. Number of approaches signed. Spare percentage.

Two reasons. First, it makes takeoffs comparable — when two estimators disagree, the disagreement is almost always in the inputs, and without them written down the argument is unresolvable. Second, it makes the takeoff defensible later. If a change order turns on whether the original scope contemplated a 45 mph or a 55 mph basis, the answer needs to exist somewhere other than in the estimator's recollection. That is the same discipline that makes change orders on traffic control jobs collectible rather than arguable.

Close the loop against reality

The takeoff is a prediction. The value compounds only if you check it.

Compare the planned device count against what the truck actually loaded and what came back. A persistent gap in one direction means one of two things, and both are worth knowing: either your takeoff method is systematically light and every bid is underpriced, or devices are leaving and not returning. Those have completely different fixes, and you cannot tell them apart without the comparison.

This is where the takeoff stops being an estimating exercise and becomes an operations one. Traffic OS tracks devices as inventory across yards and against jobs, so the quantity that was bid, the quantity that was loaded, and the quantity that came back are the same record rather than three different ones. That is also what makes tracking devices by job site tractable at more than a handful of concurrent jobs — and what turns the next takeoff into something better than an educated guess. The features page covers how the inventory and job records connect.

The short version

Resolve the letters into feet using your real speed and offset. Divide the taper by the spacing rule to get devices. Count signs by approach, not by closure. Add the opposing direction, the second phase, the detour, the spares, and the standby. Write down the inputs. Then check the prediction against the truck, and let the difference tell you which of your two possible problems you actually have.