
Research Article
An Integrated Internet of Things and Prediction-based Optimization approach for real-time surface waste mining in Indian coal mining industry
@ARTICLE{10.4108/dtip.13143, author={Vijaya Kumar Manupati}, title={An Integrated Internet of Things and Prediction-based Optimization approach for real-time surface waste mining in Indian coal mining industry}, journal={EAI Endorsed Transactions on Digital Transformation of Industrial Processes}, volume={2}, number={2}, publisher={EAI}, journal_a={DTIP}, year={2026}, month={10}, keywords={Surface waste mining, Internet of Things, Machine learning, Dynamic dispatching, Fleet management, Discrete-event simulation}, doi={10.4108/dtip.13143} }- Vijaya Kumar Manupati
Year: 2026
An Integrated Internet of Things and Prediction-based Optimization approach for real-time surface waste mining in Indian coal mining industry
DTIP
EAI
DOI: 10.4108/dtip.13143
Abstract
INTRODUCTION: Surface waste mining (SWM) involves non-value-adding activities that affect operational efficiency in the mining industry. Improving fleet utilisation through real-time monitoring and coordinated dispatching can help address these challenges. OBJECTIVES: This study develops an integrated framework combining Internet of Things (IoT)-based activity recognition with dynamic fleet dispatching to improve productivity and balance excavator utilisation and tipper waiting time. METHODS: The framework comprises two phases. Phase I processes GPS and RFID telemetry to classify tipper activity states using machine learning. Phase II uses the recognised states to trigger a condition-based dispatching heuristic. A multi-objective model incorporating first-come, first-served queue recursion is evaluated through telemetry classification experiments and discrete-event simulations calibrated using operational data from an opencast coal mine in central India. RESULTS: The gradient-boosting classifier achieves a macro-F1 score of 0.986. Increasing the fleet from 20 to 36 tippers reduces average excavator idle time from 2.78 to 0.70 hours per 8-hour shift. For the existing 30-tipper fleet, the proposed framework projects a 13.6% increase in productivity, from 4,100 to 4,659 tonnes per shift. Compared with fixed assignment, the dispatching heuristic increases production by 2.5% and reduces tipper waiting time by 66%. Under heterogeneous operating conditions, the heuristic maintains excavator output imbalance at a coefficient of variation of no more than 2.4%. CONCLUSION: Integrating IoT-based activity recognition with condition-based dispatching provides a data-driven approach to improving fleet productivity and equipment utilisation in surface mining. The framework supports fleet-sizing and dispatching decisions while offering scope for extension to heterogeneous fleets and changing mine layouts.
Copyright © 2026 Vijaya Kumar Manupati, licensed to EAI. This is an open access article distributed under the terms of the CC BYNC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.

