evidoria

← Back to browse

Good practice Imported

Project iRASTE — Telangana's AI Collision-Warning System Cuts Bus Crashes 40%

India · Hyderabad · See the India profile · See the Hyderabad profile

Evidence: Quasi-experimental Top 74% 47/100 · Ask Evidence Copilot about this practice

Telangana fitted 200 state transport buses on three highway corridors with AI-based collision-warning cameras under Project iRASTE. A year-long comparison (April 2023–March 2024) found ADAS-equipped buses had 40% fewer accidents than buses without the system on the same routes.

200 buses across 3 highway corridors
Buses fitted with ADAS (from July 2022)
40% fewer accidents
Accident reduction, ADAS vs non-ADAS buses (same routes) (April 2023 - March 2024)
up to 66% projected reduction
Projected statewide crash reduction (modelled, full rollout) (modelled projection)
up to 40% projected reduction
Projected statewide fatality reduction (modelled, full rollout) (modelled projection)

Details

Maturity
Scaling
Promoter
Telangana State Road Transport Corporation (TSRTC) / IIIT Hyderabad (INAI)
Period
July 2022 – ongoing (evaluation April 2023 – March 2024)
Keywords
transport, road safety, public transit, ADAS

Context

Launched in July 2022 by the Telangana government with IIIT Hyderabad's INAI research centre, CSIR's Central Road Research Institute and Intel, Project iRASTE (Intelligent Solutions for Road Safety through Technology and Engineering) fitted 200 Telangana State Road Transport Corporation buses on three highway corridors — Hyderabad-Kodad, Hyderabad-Pullur and Hyderabad-Adilabad — with camera-based advanced driver-assistance systems (ADAS) that warn drivers in real time of forward collisions, lane departures and pedestrians in the road.

Objectives

The project aims to reduce road crashes and fatalities on Telangana's state bus fleet by warning drivers in real time of collision risks, lane departures and pedestrians.

Activities

Crash statistics from ADAS-equipped and non-ADAS buses operating on the same three highway corridors were compared by the Telangana State Road Transport Corporation and project partners from April 2023 to March 2024.

Results

The project partners reported that ADAS-equipped buses had 40% fewer accidents than buses without the system operating on the same routes. IIIT Hyderabad's analysis further projects that applying the project's full set of recommendations across the state transport fleet could cut road crashes by up to 66% and fatalities by 40%; the project has since been cited in the International Road Federation's 2023 yearbook as a reference case.

Conclusions

The reported figures come from the implementing partners themselves rather than an independent third-party evaluation, and the 200-bus, three-corridor deployment remains a small fraction of Telangana's total bus fleet — the wider 66%/40% projections are modelled estimates of a full-state rollout that has not yet happened.

Implementation
Indicative cost
Medium (€50k–€500k) — No public cost figures were found; conservatively banded medium for camera-based ADAS hardware fitted across 200 buses plus a multi-partner research programme.
Time to results
Medium (1–3 years) — Launched July 2022 with the comparative evaluation covering April 2023-March 2024; banded medium given the roughly two-year build-out and evaluation period.
Staffing & skills
Telangana State Road Transport Corporation operational staff, IIIT Hyderabad INAI research team, CSIR Central Road Research Institute, Intel as technology partner

Conditions for success

  • Multi-stakeholder partnership combining state government, a research institute and an industry technology partner
  • Real-time in-cab alerts that keep the driver as the final decision-maker

Common failure modes

  • Reported figures come from the implementing partners themselves, not an independent third-party evaluation
  • Deployment (200 buses, 3 corridors) remains a small fraction of Telangana's total bus fleet, so the 66%/40% statewide projections are modelled, not achieved
  • Public reporting says little about camera-data governance or privacy safeguards

Where it fits

Governance type
state government transport corporation with academic and industry partners
Scale
200 buses across 3 of Telangana's highway corridors
Income level
lower-middle-income

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful