VesselLink AI: Revolutionizing Cost-Efficient Scheduling and Linkage for Eastern Indias Steel

Abstract

The steel industry relies significantly on efficient logistics to move raw materials like coking coal and limestone. The conventional method of planning and executing logistics, which involves fragmented SAP systems and Excel-based workflows, often results in inefficiencies and increases in costs and time. This paper proposes a new intelligent system called VesselLink AI, which optimizes vessel scheduling, port selection, and dispatch planning from ports to the plant. The proposed system incorporates machine learning models to forecast pre-berthing delays, make scheduling decisions, and minimize logistics costs. It takes into account many constraints like stock availability, railway logistics, and sequential discharge. The proposed system can significantly reduce the total logistics costs by 10 to 15 percent, which can be validated through experimental results. The proposed system can provide efficient and cost-effective solutions to the steel industry.

Introduction

The text discusses the importance of efficient logistics management in the steel industry, where uninterrupted supply of raw materials such as coking coal and limestone is essential for continuous production. Traditional logistics systems based on SAP and Excel are limited in handling real-time data, uncertainty, and coordination challenges like ship delays, port congestion, and railway constraints. These inefficiencies increase operational costs and reduce productivity. To address these issues, the paper proposes an AI-driven intelligent logistics system called VesselLink AI.

The proposed system integrates data from SAP, Excel, and external sources into a unified framework. It uses machine learning models for predicting vessel arrival times and delays, while optimization algorithms improve vessel scheduling, port allocation, and dispatch planning. The system also supports real-time monitoring, scenario analysis, and decision-making tools to handle disruptions efficiently.

The literature survey reviews existing AI and optimization-based logistics systems, including ETA prediction, berth allocation, multimodal transport optimization, and disruption handling. Although these systems improve specific logistics functions, they lack end-to-end integration, railway logistics consideration, real-time data integration, and support for bulk material transportation required in steel industries. These limitations highlight the need for a comprehensive AI-based solution like VesselLink AI.

The system architecture of VesselLink AI consists of four layers: presentation, application, intelligence, and data layers. The presentation layer provides dashboards and monitoring tools, the application layer manages system operations and data flow, the intelligence layer performs AI-based prediction and optimization, and the data layer handles storage and integration of logistics data. The workflow includes data collection, preprocessing, delay prediction, optimization, scheduling, dispatch planning, and scenario analysis.

The methodology involves input data processing, machine learning-based delay prediction, cost estimation, and optimization. The prediction model considers vessel parameters, weather conditions, port congestion, and historical data. The optimization engine minimizes total logistics costs, including freight, port handling, railway transport, and demurrage costs. The system also supports adaptive scheduling, dispatch planning, and “what-if” scenario analysis for better decision-making.

The results show significant improvements in logistics efficiency, scheduling, resource utilization, and cost optimization. The system achieves approximately 10–15% reduction in logistics costs while improving scheduling accuracy and reducing delays. Machine learning models accurately predict vessel delays, and the optimization engine efficiently utilizes resources such as ports, railways, and stock capacity. Scenario analysis further enhances decision-making by comparing different logistics conditions.

The discussion concludes that VesselLink AI has strong potential for transforming steel industry logistics through AI and optimization techniques. Although the system depends heavily on the availability and quality of real-time data, it provides scalable, reliable, and cost-effective logistics management. The proposed system can also be extended to more complex industrial and supply chain applications in the future.

Conclusion

The proposed system, VesselLink AI, is successful in the application of artificial intelligence techniques for the enhancement of logistics operations within the steel supply chain. The system effectively integrates machine learning-based techniques for the prediction of delays, cost optimization, as well as real-time data from various sources such as SAP and Excel. This improves the efficiency of the scheduling process as well as the cost associated with the logistics process. The system also ensures the proper coordination between ports, vessels, as well as the transport system, while also considering constraints such as stock capacity and railway availability. The results show that the system is efficient, scalable, and suitable for the support of logistics decision-making processes. Therefore, the system is suitable for the enhancement of logistics operations within the steel supply chain.The proposed system, VesselLink AI, is successful in the application of artificial intelligence techniques for the enhancement of logistics operations within the steel supply chain. The system effectively integrates machine learning-based techniques for the prediction of delays, cost optimization, as well as real-time data from various sources such as SAP and Excel. This improves the efficiency of the scheduling process as well as the cost associated with the logistics process. The system also ensures the proper coordination between ports, vessels, as well as the transport system, while also considering constraints such as stock capacity and railway availability. The results show that the system is efficient, scalable, and suitable for the support of logistics decision-making processes. Therefore, the system is suitable for the enhancement of logistics operations within the steel supply chain. In addition, the system is effective in the enhancement of logistics operations within the steel supply chain as well as the contribution towards the digital transformation of logistics systems. The system is also suitable for the support of the development of intelligent logistics systems.

References

[1] N. Evmides, S. Aslam, T. T. Ramez, M. P. Michaelides, and H. Herodotou, “Enhancing prediction accuracy of vessel arrival times using machine learning,” Journal of Marine Science and Engineering, vol. 12, no. 1362, pp. 1–17, 2024. doi:10.3390/jmse12081362. [2] C. Archetti, L. Peirano, and M. G. Speranza, “Optimization in multimodal freight transportation problems: A survey,” European Journal of Operational Research, vol. 299, pp. 1–20, 2022. doi:10.1016/j.ejor.2021.07.031. [3] Z. Chu, R. Yan, and S. Wang, “Vessel turnaround time prediction: A machine learning approach,” Ocean and Coastal Management, vol. 249, 107021, 2024. doi:10.1016/j.ocecoaman.2024.107021. [4] J. Yang, D. Liang, Z. Zhang, H. Wang, and H. Bin, “Path optimization of container multimodal transportation considering differences in cargo time sensitivity,” Transportation Research Record, vol. 2678, no. 11, pp. 1279–1292, 2024. doi:10.1177/03611981241243077. [5] T. Binsfeld, S. Hamdan, O. Jouini, and J. Gast, “On the optimization of green multimodal transportation: A case study of the West German canal system,” Annals of Operations Research, vol. 351, pp. 667–726, 2025. doi:10.1007/s10479-024-06075-5. [6] T. Cui, Y. Shi, J. Wang, R. Ding, J. Li, and K. Li, “Practice of an improved many-objective route optimization algorithm in a multimodal transportation case under uncertain demand,” Complex & Intelligent Systems, vol. 11, p. 136, 2025. doi:10.1007/s40747-024-01725-4. [7] L. Kolley, N. Rückert, M. Kastner, C. Jahn, and K. Fischer, “Robust berth scheduling using machine learning for vessel arrival time prediction,” Flexible Services and Manufacturing Journal, vol. 35, pp. 29–69, 2023. doi:10.1007/s10696-022-09462-x. [8] H. Zhang, J. Guo, and S. Guo, “An optimization study on dynamic berth allocation based on vessel arrival time prediction,” Ocean Engineering, vol. 332, 121321, 2025. doi:10.1016/j.oceaneng.2025.121321. [9] S. Md. Saber, K. Z. Thowai, M. A. Rahman, M. M. Hassan, A. B. M. M. Bari, and A. Raihan, “High-accuracy prediction of vessels’ estimated time of arrival in seaports: A hybrid machine learning approach,” Maritime Transport Research, vol. 8, 100133, 2025. doi:10.1016/j.martra.2025.100133. [10] S. Prakash, U. Mehta, and B. Sharma, “Artificial intelligence-driven multimodal route planning: Addressing dynamic unavailability and disruptions,” IEEE Access, vol. 12, pp. 172088–172109, 2024. doi:10.1109/ACCESS.2024.3498863. [11] B. Lin, M. Zheng, X. Chu, W. Mao, M. Zhang, and D. Wu, “A novel method for the evaluation of ship berthing risk using AIS data,” Ocean Engineering, vol. 293, 116595, 2024. [12] Z. Xiao, X. Fu, L. Zhang, and R. S. M. Goh, “Traffic pattern mining and forecasting technologies in maritime traffic service networks: A comprehensive survey,” IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 5, pp. 1796–1825, May 2020.

Copyright

Copyright © 2026 Mir Wajid Ali, Zain Abdul Hai, Moahmmad Aamir, Mr. Sagar Babu Jeldi. This 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.