Akhil Shijo
Conference Paper · Version v1
Open Access
Deep Learning and Optimization Techniques for Smart Transportation Systems
Abstract
Modern transportation systems face critical challenges including inefficient demand management, manual scheduling, poor resource utilization, and lack of real-time intelligence. This paper presents YATRIK ERP — an AI-powered intelligent transportation management system that integrates deep learning models and optimization algorithms for real-time decision-making.
The system utilizes Long Short-Term Memory (LSTM) networks for passenger demand prediction, Genetic Algorithms for autonomous scheduling, and multi-factor analysis for crew fatigue monitoring. The architecture follows a microservices-based design, separating the ERP backend (Node.js) from machine learning inference (Python Flask), ensuring scalability and maintainability.
The system utilizes Long Short-Term Memory (LSTM) networks for passenger demand prediction, Genetic Algorithms for autonomous scheduling, and multi-factor analysis for crew fatigue monitoring. The architecture follows a microservices-based design, separating the ERP backend (Node.js) from machine learning inference (Python Flask), ensuring scalability and maintainability.
≈25%
Operational Efficiency Improvement
≈30%
Cost Reduction Achieved
v1
Published Version · Zenodo
Publication Details
Keywords
Intelligent Transportation Systems
Deep Learning
LSTM
Genetic Algorithm
ERP
Demand Prediction
Autonomous Scheduling
Optimization
Cite This Paper
APA: Shijo, A., & Maria Sebastian, S. (2026, April 12). Deep Learning and Optimization Techniques for Smart Transportation Systems. Proceedings of the National Conference on Emerging Computer Applications (NCECA)-2026. Amal Jyothi College of Engineering, Kanjirapally. https://doi.org/10.5281/zenodo.19531457
Licensed under Creative Commons Attribution 4.0 International
· © Amal Jyothi College of Engineering 2026