NEMT routing is the process of deciding which vehicle runs which trips in what order: assigning each trip to a driver, sequencing pickups and drop-offs so every stop lands inside its pickup window, and multi-loading compatible riders traveling the same direction. Good routing turns a day’s trip list into workable driver manifests — and keeps reworking them as the day changes.
Every weekday morning, an NEMT operation faces the same puzzle: a list of trips — dialysis runs, clinic appointments, discharges, standing orders — and a smaller list of vehicles and drivers to run them with. Routing is the discipline of matching one list to the other, and it is where a medical transport fleet either earns its margin or burns it. Two providers with identical vans and identical trip volume can finish the day with very different fuel bills, very different on-time scores, and very different broker standings, purely on how the trips were routed.
This guide explains the three moving parts — trip assignment, multi-loading, and pickup windows — and the part most explanations skip: why the morning’s beautiful route plan rarely survives contact with the actual day.
What does NEMT routing actually decide?
Three questions, answered for every trip on the board. Which vehicle: the trip’s level of service — ambulatory, wheelchair, stretcher — rules out every vehicle that cannot legally and safely carry the rider, and capability has to match before geography even matters. What order: each vehicle’s trips get sequenced so the driver flows from stop to stop without doubling back, without idle gaps, and without one late appointment cascading into the next. And what stays flexible: which trips are locked to hard times, and which have slack the desk can use when the day wobbles.
Routing is often confused with dispatching, and the two overlap on the same desk, but they are distinct jobs: routing is the planning — building the manifests; dispatching is the live execution — working the phones, tracking the vehicles, and rebuilding the plan when reality edits it. A fleet needs both, and the second is what rescues the first.
How does trip assignment work?
Assignment starts from constraints, not preferences. The stretcher trip can only go on the stretcher vehicle, so it anchors that vehicle’s day; the wheelchair trips distribute across the lift-equipped vans; ambulatory trips fill in around them. Time-critical trips — a dialysis chair, an infusion slot — anchor next, because they cannot move. Flexible trips get placed last, in the gaps, where they can absorb disruption.
Geography comes after capability and time. A well-assigned board keeps each vehicle working a coherent area of the map instead of criss-crossing the county, because empty miles between stops — deadhead — are cost with no revenue attached. When an assignment forces forty minutes of deadhead to reach a single trip, a good router asks whether the trip belongs on a different vehicle, a different slot, or — under some broker contracts — whether it is a trip worth declining at all.
What is multi-loading in NEMT?
Multi-loading is carrying more than one rider at a time when their trips are compatible: same direction, overlapping timing, and a vehicle with capacity for both. The classic example is a dialysis run — three riders from the same part of town, all due at the same clinic for morning chair times, picked up in sequence and dropped together. One vehicle, one driver, three completed trips.
Done well, multi-loading is the biggest efficiency lever in NEMT routing, which is exactly why it is also the easiest to overdo. Every added pickup extends the first rider’s time on board, and a route built purely for vehicle efficiency can leave a frail passenger riding for an hour on what should be a fifteen-minute trip. Broker contracts and common sense both put limits on ride time, so the router’s real skill is knowing when to stop loading: the question is never just “do these trips fit?” but “do these riders fit — this vehicle, this route, this much time on board?”
Real-time driver coordination and routing around the clock — overnight, weekends, holidays, and peak surges covered.
What are pickup windows, and why do they rule everything?
A pickup window is the agreed band of time — commonly a stretch around the scheduled pickup — inside which the vehicle must arrive. Windows are the contract term that turns routing from a preference into a constraint: brokers timestamp arrivals against the window automatically, the resulting on-time percentage ranks you against other providers, and a route that looks efficient on the map but misses windows is a failed route.
Windows are why NEMT routing cannot simply borrow its playbook from parcel delivery. A late package waits on a porch; a late pickup is a missed chair time, a stranded rider, and a recorded failure on the scorecard your next quarter’s trip volume depends on. Every sequencing decision — which stop first, how much slack between stops, how many riders to load — is really a bet on hitting windows, and a sane router builds the route around the tightest windows first, then lets the flexible stops fill in.
Why does the morning route fall apart by 10 a.m.?
Because the trip list will not sit still. An appointment runs long and the will-call return everyone expected at 11 fires at 12:40. The broker pushes two same-day adds. A rider cancels after the van is en route; a driver hits a closed road; a discharge that was “ready in twenty minutes” takes two hours. None of this is unusual — it is the normal texture of a medical transport day, and it means routing is not a task that finishes at 7 a.m. It is a loop that runs all day.
This is where routing and dispatching fuse. The plan gets the fleet out the door; the desk keeps the plan honest — reslotting the late return, working the adds into whichever manifest has slack, and protecting the anchored trips while the flexible ones shuffle. A provider whose routing exists only as a morning printout is really running unrouted by lunchtime.
Does routing software handle all this automatically?
Modern NEMT platforms — the RouteGenie and TripMaster class of software — genuinely help: they hold the trip list, know each vehicle’s capability, flag window conflicts, and propose optimized routes and multi-loads far faster than a whiteboard ever did. A fleet running more than a handful of vehicles should be using one.
What the software does not do is absorb the day. An optimizer proposes; someone still has to accept the 12:40 will-call, judge whether the frail rider should really be the third pickup on that load, decide which trip slips when a van dies, and answer the facility calling to move a time. Routing quality, in practice, is the software’s math plus the desk’s judgment plus the coverage to apply both at 5 a.m. and 7 p.m. — which is why the same platform produces very different results in different providers’ hands.
A fair test when you evaluate any routing setup — software, in-house desk, or outsourced — is to walk one messy scenario end to end: a driver calls out at 4:30 a.m. with a full dialysis manifest. Who sees it first, who rebuilds the board, how do the riders and the facility find out, and what does the broker portal show at 6 a.m.? The answer tells you more about your routing operation than any feature list, because routing is judged on the day it goes wrong, not the day it goes to plan.
Common questions
Where this guide fits: it is part of our patient transport (pts) answering and dispatch coverage. Next step: try the desk free for your first week.
Daniel Okoro is TransportBPO's Content Editor and Dispatcher — a former dispatcher who moved into editorial after years running live boards for taxi, NEMT and trucking accounts. He writes from dispatch-floor experience, not the marketing department, on what actually keeps a 24/7 desk running.
Connect on LinkedIn