Telangana Road Accident Hotspot Prediction Using Quantum Machine Learning with a Variational Quantum Neural Network

Road accidents remain a significant public safety concern, and a small number of road segments account for a large share of the harm. This paper proposes SafeRoute Telangana, a risk-aware navigation framework that identifies accident-prone road segments and recommends alternative routes across Telangana, India. The framework integrates historical accident analysis, Quantum Machine Learning (QML) and a Variational Quantum Neural Network (VQNN) implemented in Qiskit to estimate accident risk from accident and road-network features. It covers data preprocessing, spatial hotspot identification, risk estimation and route optimisation. A routing engine generates feasible alternatives, which are compared on distance, estimated travel time and accident-risk exposure, so that the shortest route and a potentially longer but lower-risk route can be shown side by side. A React Native application with a FastAPI backend visualises routes and hotspots on a digital map and gives visual and voice alerts near risk zones. The evaluation plan compares the VQNN with classical baselines (logistic regression, random forest, support vector machine) using accuracy, precision, recall, F1-score and AUC-ROC, together with route-level measures. Telangana-specific accident records are essential for regional validity, and the compatibility of supplementary datasets must be verified before training. Measured results are to be reported once training on verified regional data is complete, this version contains no measured performance figures. The system is intended to support informed travel decisions, not replace driver judgement.

© 2026 — Authors retain the copyright of this article. This is an open access article distributed under the Creative Commons Attribution License (CC BY 4.0) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.