
It is 9 p.m., the driver has forty minutes left on the clock, and the next two rest areas are full. That is the moment a truck parking availability system is built for. It answers one question before the driver commits to an exit: how many spaces are open at the next lot.
This guide explains what a truck parking information management system, or TPIMS, actually is, how the states that have built one did it, what it costs, and the detection decision that decides whether the number on the sign can be trusted.
FHWA's most recent Jason's Law survey, completed in 2019, found that 98 percent of drivers report problems finding safe parking, and 75 percent hit the problem at least once a week. The country has roughly 313,000 truck parking spaces, about 40,000 of them at public rest areas and 273,000 at private truck stops. When those fill up, drivers end up where enforcement agencies see them most: on exit ramps and shoulders, overwhelmingly between 7 p.m. and 9 a.m.
The cost lands on drivers first. ATRI's parking diary study put the search at 56 minutes of revenue driving time sacrificed per day, about 9,300 revenue miles a year, or roughly $4,600 in lost wages per driver. In ATRI's 2025 Top Industry Issues survey, drivers ranked parking as their second biggest concern, behind only pay.
Then it lands on everyone else. A tractor-trailer on a ramp shoulder at 2 a.m. is a fixed object at highway speed, and a driver who keeps rolling past the hours-of-service limit because there is nowhere to stop is a fatigue crash waiting for a reason.
A truck parking information management system has three layers. Agencies tend to spend their planning time on the third one and their maintenance budget on the first.
The 2025 NCHRP Guide for Truck Parking Information Management Systems notes that roadside signs have consistently been the dissemination method drivers prefer, and recommends placing them three to five miles upstream of the facility and no more than 60 miles out. Iowa went the other way: it skipped signs entirely, published availability only through 511, apps, and in-cab systems, and used the savings to instrument more sites. Its evaluation found the system performed about the same as neighboring states that installed signs, because most drivers plan overnight parking more than an hour ahead on a phone.
MAASTO. Eight Midwestern states (Iowa, Indiana, Kansas, Kentucky, Michigan, Minnesota, Ohio, and Wisconsin) built a shared regional TPIMS that went live in January 2019. Iowa alone instrumented 33 sites along I-80 plus 10 more on connecting interstates, with capacities from 5 to 850 spaces. By April 2023, 35 users from 23 companies were pulling Iowa's data feed into their own apps and logistics tools.
The I-10 Corridor Coalition. California, Arizona, New Mexico, and Texas are deploying a joint system covering roughly 550 truck spaces at 37 public rest areas, funded by a $6.85 million federal ATCMTD grant matched one-to-one by the states. New Mexico's piece, a $2.7 million system at six I-10 rest areas from Lordsburg to Anthony, pushes availability to electronic signs and NM 511. Omnisight FusionSensors do the counting at those New Mexico sites.
The outcomes are real. Iowa's evaluators found parking spread more evenly along I-80 after deployment, system downtime under 5 percent at most sites, and the share of roadside inspection violations that were hours-of-service violations fell from 1.82 percent to 1.08 percent after the system went live, a 40.7 percent reduction.
Every TPIMS comes down to one of two detection methods, and the choice drives cost, accuracy, and how much operator time the system eats for the next decade.
Space-by-space occupancy. A sensor in or above every space reports whether it is occupied. The NCHRP guide calls this the most accurate approach because it is self-correcting: a miss in one space does not corrupt the lot total. The downside is that cost scales with the number of spaces. In-pavement magnetometer pucks are the common choice, and the guide is blunt about their weak point: battery life "has been the most significant challenge identified." They also have to be replaced whenever the lot is resurfaced, and freezing climates can affect them.
Entrance and exit counting. Sensors at the driveways count trucks in and trucks out, and the system subtracts. It is cheaper to install and scales to an 850-space lot as easily as a 20-space one. The catch is that a missed count does not go away. The guide says that for most systems "accuracy checks and count resets will be required once or twice a day," and Washington State's 2026 evaluation documented in/out counts drifting steadily away from ground truth until someone reset them. We cover why those counts drift, and what Iowa and Washington learned, in the next post.
Underneath both methods is the question that decides everything: can the sensor tell a tractor-trailer from a pickup, at night, in the rain? A rest area entrance sees cars, RVs, and bobtails too. A counter that cannot classify is adding cars to the truck count.
Published figures are scarce, but two anchors exist. Michigan's planning estimate on FHWA's National Coalition on Truck Parking page was about $115,000 per public rest area and $65,000 per private site. New Mexico's six-site system came in at $2.7 million including signs. Iowa's contract to implement, test, and maintain its system for three years was $3.9 million. The number to watch is not the install. It is the maintenance and operator hours for recalibration, puck replacement, and manual verification over the life of the system.
Our specification checklist expands each of these into language you can put in an RFP.
Omnisight started in a truck parking pilot in Colorado, where every available sensor technology was benchmarked through a mountain winter and only HD3D radar fused with HD video kept counting through whiteout snow and freezing rain. That became FusionSensor. One unit mounted at a lot entrance counts every vehicle in and out, classifies it (tractor-trailer, bobtail, pickup, car, RV) at 98.7 percent accuracy, and can also watch spot-specific occupancy, all processed on the device by TrueEdge. It installs on an existing pole in under an hour with no pavement cut, runs from minus 30 to plus 75 Celsius in an IP66 housing, and publishes to signs, 511, and your TMC in NTCIP, SDLC, MS2, JSON, or XML. That is what runs the New Mexico I-10 corridor today.
See how it fits a corridor on our truck parking management page, or request a demo and we will walk through your sites with you.
Subscribe to The Edge Brief for monthly field notes, deployment stories, and product updates.
Subscribe →