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

@article{203369,
        author = {Koppisetti Jyothi Priyanka and D Kanakasatya},
        title = {TRAFFIC CRASH COUNTERMEASURE RECOMMENDATIONS USING A DEEP NEURAL NETWORK.A DECISION SUPPORT TOOL FOR TRAFFIC SAFETY ENGINEERS},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {11498-11507},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=203369},
        abstract = {Traffic crashes cause major social and economic losses. Existing studies identify crash factors and suggest scenario-specific countermeasures. However, few works integrate causes with preventive actions into a single decision-support framework. This study proposes a data-driven approach to recommend countermeasures for crash types using machine learning and deep learning. We develop a Deep Neural Network (DNN) multilabel classifier to map crash characteristics to interventions automatically. The model learns from historical crash records paired with documented preventive measures. Inputs comprise six significant features selected from the dataset. Training and validation employ K-fold cross-validation to ensure robustness and generalisation. Experimental results show the DNN achieves 90.2% accuracy. Precision is 91.6%, recall 90.2%, and F1-score 90.1%. We compare the DNN with XGBoost, K-Nearest Neighbours, Support Vector Machine, and Random Forest. Most models show competitive performance near 90.2% accuracy. Support Vector Machine records slightly lower performance at 87.4% accuracy. The system effectively recommends tailored countermeasures by crash type. It consolidates fragmented reports into a unified, data-driven framework. Outcomes support policymakers and transport authorities in evidence-based decision-making. The approach improves traffic safety management and optimises resource allocation. It provides a reproducible tool for implementing targeted road-safety strategies. Future work will expand feature sets and evaluate deployment in real-world settings.},
        keywords = {Traffic Safety, Countermeasure Recommendation, Deep Neural Network, Multilabel Classification, Crash Analysis, Machine Learning, K-Fold Cross-Validation, Road-Safety Management, Feature Selection, Decision Support System.},
        month = {May},
        }