Copyright © 2025 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{180580,
        author = {Vaishnavi Gomashe and Sahil Dalal and Renuka Nival and Nikita Khawase},
        title = {Predictive Asset Stewardship using AI & Data Analytics},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {1},
        pages = {2186-2201},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=180580},
        abstract = {Road infrastructure deterioration is a great problem in safety and transportation efficiency in large cities. Reactive maintenance approaches that are based on tradition have led to increased costs and inefficient resource allocation and ultimately result in more road accidents. In this paper, the authors discuss the technical implementation of the hybrid predictive maintenance framework that uses both Long Short-Term Memory (LSTM) Networks and Gradient Boosting Machines (GBM) to predict early road wear. The system, which uses real-time traffic and weather data, produces predictive insights that help city planners and maintenance teams to make informed decisions;
A comprehensive hybrid LSTM-GBM model is the result of a fully-tuned hyper-parameter process. In this application, the results are verified with the help of RMSE and R-Squared (R²) metrics in order to ensure the highest precision of prediction. The deployment of the model as a web-based application is beneficial as its municipality infrastructure system integrates easily. The dashboard also includes geospatial mapping, real-time alerts, and the trend analysis of periodic work on the road when there is wear and tear signs inescapable to the drivers.
The experiment findings unambiguously establish that the hybrid model outperforms the conventional predictive models produced using LSTM, which still has a sequential dependency, and GBM, which is further processed into predictions. Such a product will enhance road safety and reduce long-term maintenance costs while becoming a robust, scalable, and data-driven asset stewardship framework. The use of reinforcement learning, satellite imaging, and autonomous maintenance scheduling are the future prospects for the full optimization of predictive capabilities.
This research says that the groundbreaking implications of AI-based infrastructure management and the necessity of real-time data analytics in smarter urban planning are underlined. This result is major headway toward a more intelligent, safer, and ecologically sound road network assuring the life and activity of assets in a growing metropolitan area.},
        keywords = {Predictive Maintenance, Road Wear Index, LSTM, Gradient Boosting Machine (GBM), Infrastructure Analytics, Real-Time Data Processing, Smart Cities, Asset Management, Flask API, Dashboard Visualization, Time-Series Forecasting, Traffic Analytics, Weather Impact Modeling, Hybrid AI Models, Data-Driven Decision Making.},
        month = {June},
        }