Stretch-by-stretch fog and heavy-rain warnings for night bus and truck fleets
TL;DR Fogwarn scores fog and heavy-rain risk for each stretch of highway, hour by hour, and pushes warnings into the telematics apps that bus and truck fleets already run. Fleet owners and telematics vendors pay a per-vehicle fee. It's a narrow, seasonal product and it should be priced and sold that way.
Weather is not the main cause of Indian road crashes, and any pitch that says otherwise will fall apart. A 2025 study of Uttar Pradesh notes that most two-wheeler crashes there happen in clear weather[1]. The same study shows how local the weather effect is - Kanpur district sees more crashes in fog, Agra and Lucknow more in rain, and the fog season runs from November to February[1].
Fog also forms in patches. Work on Chinese expressways traced dense fog to specific segments, driven by terrain and surface temperature, mostly in early mornings in autumn and winter[2]. A driver at 4 a.m. has no way to know the next 10 km is the bad one.
The seed pitched a safety platform for driverless cars in bad weather. Who would you sell that to? India has no driverless cars on public roads, and we found no rule that would allow them. Even in the UK, a two-year test by the National Physical Laboratory and the Met Office measured how rain and fog degrade radar, LiDAR and camera sensing[3]. So Fogwarn keeps the goal and serves human drivers.
The product builds a fog and heavy-rain risk score for each 2–5 km stretch of chosen highways, refreshed hourly. It draws on public forecasts, hyperlocal station data and a handful of visibility sensors put up at known trouble spots for calibration. Fleets get it inside their existing tracking app, with a warning to the driver 15–20 minutes before a high-risk stretch, a suggested speed and a log the fleet manager can check. Start with one corridor where night running and winter fog overlap, such as the routes through western and central Uttar Pradesh.
Drivers use it but fleet owners pay. Intercity sleeper-bus operators and truck fleets on night routes are the first buyers, at a per-vehicle monthly fee for the four winter months, plus the monsoon for ghat sections. The faster route to scale is selling the feed to telematics companies as a data layer. I found no public price for a weather-risk feed in India, and the best-known global weather-intelligence firm keeps its route-weather product on an enterprise plan with no listed price[4].
Hyperlocal weather data has become cheap. One food-delivery company's weather network runs 750+ automated stations across 60 cities with a one-minute refresh and says it will keep the data free for public use[5]. Fleet telematics is now common in Indian trucks, and one Indian telematics firm lists support for the major Indian truck makers[6]. The 2019 amendment to the Motor Vehicles Act also set up a National Road Safety Board to advise on road safety[7].
There's no published figure for India's road-weather risk market. The base is large and the slice is small. India recorded 467,044 crashes and 151,417 deaths in 2018 by the MoRTH figures cited in the UP study[1], and only a minority of those happen in bad weather. The firmer signal is willingness to pay for fleet safety in general, and one AI dashcam maker says its cameras have analysed over 30 billion driving miles[8]. That's adjacent, not your market.
AI dashcam makers sell cameras that coach drivers, and Indian telematics firms sell Indian telematics with driver scoring[8][6]. Neither shows a weather or fog feature on its main pages. Global weather-intelligence firms sell route-based weather risk globally, but its own guide derives risk from rainfall intensity alone, with no visibility input[4]. The free station network covers cities, not highways. The gap is fog risk by highway stretch, delivered through tools fleets already pay for.
| Difficulty | Moderate (39/100) |
| Starting capital | ₹60 L–1.2 Cr |
| Time to launch | 6–9 months |
The modelling is doable. The hard part is proving that a warning changed anything, because fog crashes are rare per vehicle and one winter of pilot data may show fewer harsh-braking events but say very little about crashes. Highway coverage is thin, so you'll need your own visibility sensors on the worst stretches. Sales are seasonal too, which makes cash flow lumpy.
No sector licence applies to an advisory risk feed, as far as we found. Avoid any claim that it prevents crashes.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let AI claim it prevents crashes. Fog crashes are rare per vehicle, so a model can show safer driving but not crash reduction from one winter of data.
Want to work out the AI layer for Fogwarn? Talk to Yash →
Here's what can go south -
I'd try this as a founder with weather-data or telematics experience who can sell to two or three telematics firms and not to fleets one by one. Do not raise money on a crash-reduction story. First prove that drivers slow down on flagged stretches, using speed data from one corridor over one winter.
Expect roughly ₹60 L–1.2 Cr, covering the risk model, visibility sensors on the worst stretches and pilots through telematics firms.
No sector licence for an advisory feed. You need road owner permission for roadside sensors and DPDP-compliant handling of driver data.
Sales are seasonal and proof is slow, so it works best sold through two or three telematics firms. Prove that drivers slow down on flagged stretches before you raise money.
The free Fogwarn validation kit has the line-by-line capital breakdown, the effort scoring, the data and privacy checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for fleet owners and telematics vendors, and 5 assumptions to test. Download the kit →
Thinking about building Fogwarn or something close to it? The make-or-break work is showing measurable driver slowdowns on flagged stretches within a single fog season, and I'm happy to help you design that test. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.