Every battery on your fleet ships with a management system that measures voltage, current and temperature thousands of times an hour. Almost all of it is discarded. Choir LITE reads that feed and turns it into one number per pack: how much usable life is left, and how fast it is being spent.
The full platform assumes you will instrument the asset. Choir LITE assumes you will not. It reads the battery management system that is already there, runs the prediction on a small box at the asset, and sends out over private cellular.
No new sensors. No install on the pack. Nothing written back to the battery.
How the full platform differs →On a battery-electric asset, the pack is the largest single line item and the first thing to reach end of life. It is also the one component nobody can see inside.
State of charge tells you how full the pack is right now. State of health tells you how much capacity it has permanently lost. Fleets run on the first and are surprised by the second.
Two buses bought on the same day, running the same route, can be years apart in usable life depending on how they were charged, how hot they got, and how deep they were cycled. Fleet-average planning replaces good packs early and misses the bad ones.
Pack warranties are written against capacity retention. If you cannot show which packs degraded and when, the window closes and the cost moves from the manufacturer to you.
Nothing is added to the battery and nothing touches your corporate network.
Every pack ships with a battery management system. Choir LITE reads its output — voltage, current, temperature, cycle history — and never writes to it. On fixed assets that is usually Modbus and readable on day one. On vehicles it is a CAN map, which varies by manufacturer and takes a decode step per platform.
The prediction runs on the box itself. For mobile fleets there is one box per vehicle, because the box has to be where the pack is. For depot chargers and depot storage there is one box per site. If the connection drops the box keeps predicting, stores what it produced, and backfills when the link returns.
Our connectivity partner LinkWorx supplies a multi-carrier SIM that fails over between networks, on a private carrier APN. Traffic does not cross the public internet, and your network never touches ours — which removes the IT review that stalls most fleet pilots before they start.
Choir LITE is pointed at battery-electric assets and the depot infrastructure around them. That is a deliberate boundary, not a starting position.
Diesel and internal-combustion fleets. That market is well served by established fleet telematics and heavy-equipment vendors, and we have no edge there. We are not going to invent one to widen a slide.
You are being asked to consider a forecast from a company you have not worked with. Here is the honest position.
Not simulations. The evidence page carries the evaluations, including the cases where a simpler model beat ours and the cell where our forecast never called the crossing. If you want to read the losses before the wins, they are published.
Choir LITE has not run on a customer fleet. We are looking for a first site and will run that pilot at cost, because the reference matters more to us than the invoice. Anyone telling you otherwise about a company this size is selling you something.
Charger power electronics is a different problem. The battery work does not transfer to it automatically, and we will not pretend it does — that needs its own validation before we sell it. And the system abstains. On stale or incomplete data it returns no forecast and tells you why, rather than a confident number it cannot support.
A school bus fleet does not run in July. Charging a full monitoring rate through the months an asset is parked is how a good idea gets cancelled at renewal.
Full prediction while an asset is working, a low watch rate while it is parked, flipping automatically. You keep visibility on a stored pack — which is when some of the worst degradation happens — without paying working-season rates for it.
Or a single annual figure across the fleet, so the finance conversation happens once a year instead of every month per asset.
Hardware is a small one-time cost per box and is close to a pass-through — we are not in the box business. The recurring prediction subscription is the product. Connectivity is priced by LinkWorx.
Fixed depot storage on Modbus is readable on day one, which makes it the fastest place to find out whether this is worth anything to you. One site, at cost, and you keep the output either way.
