Conference Papers 2025

Seminar
Day 2 (18 Sep 2025), Session 7, Advancing Rail Freight, 15:30 - 17:30

Status
Accepted, documents submitted

Submitted by / Abstract owner
Sofia Cerqueira

Authors
Sofia Cerqueira, National Laboratory for Civil Engineering (presenter)
Elisabete Arsenio, National Laboratory for Civil Engineering
Rui Henriques, INESC ID and IST, University of Lisbon

Short abstract
Existing systems face limitations in capturing short- and long-term fluctuations in service quality while addressing sustainability challenges. To bridge this gap, this research develops an Advanced Analytical Decision Support System for the i

Abstract
Despite the increasing digitalization of logistics, there remains a research gap remains on the comprehensive monitoring and assessment of rail service performance levels across end-to-end journeys. Existing systems struggle to capture both short- and long-term fluctuations in service quality while addressing sustainability challenges [1,2]. Recent advancements in Intelligent Rail Transportation Systems (IRTS) leverage big data analytics to enhance railway operations and decision-making. These applications span multiple domains, including predictive analytics for delays, demand forecasting, optimization, and simulation models for operational improvements [2,3]. While these innovations have significantly improved railway efficiency, there is still a lack of support systems to assess the capability of these advancements in improving efficiency, sustainability, and overall service performance, integratively. To this end, key challenges need to be tackled: (i) the integration of diverse data sources, encompassing both transport views at ports and corridors, for holistic analysis, (ii) defining key metric (e.g. capacity shortages, delays, suboptimal loads) that capture service performance variability across time and space, and (iii) ensuring robust machine learning and statistical analysis to enhance observability of spatiotemporal pattern dynamics in the scope of service quality [4,5].
Considering the above challenges, this research focus on the development of an advanced analytical decision support system (ADS2) for the comprehensive identification and assessment of key performance indicators in rail freight systems such as capacity shortages, travel time, delays, suboptimal load factors [7,8]. The system is designed to provide actionable insights for each journey by offering: (i) a tailored, multi-dimensional metric system that evaluates service (e.g., average delay times at entry and exit points, capacity, travel time, empty trips, and suboptimal transfer time); (ii) context-sensitive analytics that captures temporal trends and seasonal patterns across various terminal locations; (iii) capabilities to assess service suitability for different types of goods (e.g., hazardous and non-hazardous goods); and (iv) a ranking system that classifies the spatiotemporal features of end-to-end journeys and compares their quality against similar journeys with similar mobility dynamics within the rail service.
The methodological framework of ADS2 includes: (i) integrating available data sources to ensure completeness and alignment; (ii) structuring measurable service quality indicators as time series to represent end-to-end rail service performance along time (e.g., capacity shortages, delays, suboptimal loads, among others); (iii) applying machine learning models for feature representation of multivariate time series to detect anomalies and temporal patterns while clustering similar dynamics; (iv) identifying trends, seasonal patterns, and key characteristics within clusters; (v) ranking services within similar dynamics using a voting system based on service quality metrics; and (vi) implementing a dashboard for route selection, visualization of key statistics, and journey ranking comparisons to support data-driven decisions on service quality metrics. The ADS2 is designed for an easy adaptation to other contexts and transport networks, such as public, private or shared multimodal systems. To ensure broad applicability and support continuous improvements, the code and dashboard for the underlying system is open access. This open-source approach enables operators to modify and extend the tool to suit different service models, incorporate new metrics, and adapt to changing operational requirements.
The case study in this research focuses on the evaluation of services in the national freight system that connects the multimodal Port of Sines with key cargo destinations, including the Sines-Madrid corridor. The proposed tool aims to support an effective transport planning – operational, tactical and strategic – to foster the connection between rail services and the logistics operations covering the entire supply chain.

[1] Rong, C., Ding, J., & Li, Y. (2024). An interdisciplinary survey on origin-destination flows modeling: Theory and techniques. ACM Computing Surveys, 57(1), 1-49.
[2] Wang, Y., & Sarkis, J. (2021). Emerging digitalisation technologies in freight transport and logistics: Current trends and future directions. Transportation Research Part E: Logistics and Transportation Review, 148, 102291.
[3] Tang, R., De Donato, L., Besinović, N., Flammini, F., Goverde, R. M., Lin, Z., ... & Wang, Z. (2022). A literature review of Artificial Intelligence applications in railway systems. Transportation Research Part C: Emerging Technologies, 140, 103679.
[4] Marinagi, C., Reklitis, P., Trivellas, P., & Sakas, D. (2023). The impact of industry 4.0 technologies on key performance indicators for a resilient supply chain 4.0. Sustainability, 15(6), 5185.
[5] Atluri, G., Karpatne, A., & Kumar, V. (2018). Spatio-temporal data mining: A survey of problems and methods. ACM Computing Surveys (CSUR), 51(4), 1-41.
[6] Zhang, R., Li, L., & Jian, W. (2019). Reliability analysis on railway transport chain. International Journal of Transportation Science and Technology, 8(2), 192-201.
[7] Zaid, F., Gazder, U., & Barbieri, D. M. (2024). Multi-criteria analysis for freight transport decision-making with fuzzy analytic hierarchy process: A top management’s perspective for Bahrain. Transportation Research Interdisciplinary Perspectives, 23, 101017.

Programme committee
Rail Policy and Planning

Topic
Utilising technology, Artificial Intelligence and digital twins to improve transport systems and transport research