The Optimization of Intelligent Transport Systems: Planning, Energy Efficiency and Environmental Responsibility

1. Introduction

Modern urban development processes have created new demands on mobility systems, shaped by growing spatial constraints, evolving land-use patterns, and a marked increase in individual travel behaviors [1]. These dynamics have been accompanied by a surge in private vehicle ownership, often at the expense of public and active transport modes [2]. The consequences, including congestion, environmental degradation, rising energy consumption, and deteriorating urban health indicators, have reinforced the urgent need to rethink how transport systems are planned and managed.

In this Special Issue, the focus was placed on the optimization of intelligent transport systems (ITS) within the broader context of energy efficiency and environmental responsibility. Contributors addressed the inherent tensions between limited urban space and the growing demand for mobility, offering multidimensional perspectives on how ITS and smart planning tools can support more sustainable urban futures [3]. The accepted articles reflect the increasing relevance of approaches that combine technological innovation with spatial and social awareness [4].

Key themes covered in this issue include the optimization of public transport operations, spatial and infrastructural planning for electromobility, the integration of information technologies into transport systems, and the development of transit-oriented strategies aligned with the principles of sustainable mobility. Several contributions also explored the role of autonomous vehicles, MaaS (Mobility-as-a-Service) platforms, and novel driving systems in enhancing system-wide efficiency and reducing environmental impact.

Collectively, the articles in this Special Issue demonstrate the critical importance of holistic and data-driven planning approaches in modern transport systems. They also underscore the potential of ITS as a catalyst for systemic change, enabling not only better service quality but also measurable improvements in energy use and environmental outcomes. The findings presented here are of interest not only to researchers but also to decision-makers tasked with shaping the future of urban mobility.

2. A Short Review of the Contributions in This Special Issue

Brief summaries of the six selected papers belonging to this Special Issue of Energies, titled “The Optimization of Intelligent Transport Systems: Planning, Energy Efficiency and Environmental Responsibility”, are included in the next subsections.

2.1. Manifold Learning in Electric Power System Transient Stability Analysis [Contribution 1]

Sarajcev and Lovrić propose an innovative application of manifold learning methods for the transient stability analysis (TSA) of electric power systems. Using data derived from IEEE 39-bus system simulations and wide-area PMU monitoring, they compare multiple dimensionality reduction techniques including PCA, kernel PCA, isomap, MDS, LLE, and t-SNE. Their findings show that kernel PCA, when combined with a support vector machine classifier, offers robust performance for identifying system disturbances, even in a fully unsupervised learning setting. The study contributes both methodological advancements and practical insights into the integration of machine learning with real-time power grid stability diagnostics.

2.2. The Influence of Stops on the Selected Route of the City ITS on the Energy Efficiency of the Public Bus [Contribution 2]

This study investigates how bus stops, both scheduled and traffic-induced, affect energy consumption in urban public transport, using data from Rzeszów, Poland. Based on GPS tracking and fuel usage records from a diesel-powered city bus operating on route 13, the analysis quantifies energy losses resulting from stopping, idling, and subsequent acceleration. The findings indicate that approximately 26.2% of total fuel consumption is attributable to stops at bus stops, while 42.5% is due to stops caused by traffic conditions. The study highlights that implementing on-demand stops and improving traffic flow management through intelligent transport systems (ITS) could significantly enhance the energy efficiency of urban bus operations.

2.3. Reliable Energy Optimization Strategy for Fuel Cell Hybrid Electric Vehicles Considering Fuel Cell and Battery Health [Contribution 3]

This article presents a novel hierarchical energy management strategy (HSEMS) for fuel cell hybrid electric vehicles (FCHEVs) aimed at reducing hydrogen consumption while enhancing energy efficiency and component durability. The upper-level controller employs fuzzy fault-tolerant mechanisms to ensure reliable operation and preserve the health of the battery and proton-exchange membrane fuel cell (PEMFC) under fault conditions. The lower-level controller uses dynamic programming and Pontryagin’s minimum principle for optimal real-time power distribution between the fuel cell and battery, integrating a proportional–integral controller for precise tracking. Simulations using TruckMaker/MATLAB demonstrate the proposed strategy’s superiority over equivalent consumption minimization and fuzzy logic methods, yielding up to 9.8% hydrogen savings while effectively managing faults. The strategy’s predictive capability is enhanced through adaptive fuzzy and neuro-fuzzy SOC estimators, contributing to extended component lifespan and robust performance under varying driving cycles.

2.4. The Attractiveness of Regional Transport as a Direction for Improving Transport Energy Efficiency [Contribution 4]

This article examines the role of regional public transport frequency in improving energy efficiency by encouraging a modal shift from private to public transport. Based on empirical observations from 27 regional bus lines across 19 Polish districts, the authors develop novel indicators (AttK and AttYK) to assess transport attractiveness for different demographic groups. Their statistical analysis confirms a strong positive correlation (Spearman’s Rs = 0.807, p = 0.001) between the number of daily connections and attractiveness for adults and seniors, suggesting that at least four daily pairs of connections are necessary to make regional transport a viable alternative to private cars. The study estimates that each additional connection can attract 1.5 new adult passengers and reduce energy consumption by 0.33–0.69 kWh per kilometer, contributing to the EU’s transport energy efficiency goals. The findings highlight the importance of service frequency in combating transport exclusion and achieving sustainability in non-urban areas.

2.5. High-Capacity Energy Storage Devices Designed for Use in Railway Applications [Contribution 5]

This article investigates the potential of high-capacity supercapacitors, particularly those using starch-derived carbon materials, for application in railway transport to recover energy from electrodynamic braking. Through experimental testing on a diesel–electric multiple unit, the authors demonstrate that approximately 28.5% of energy can be recuperated during braking, irrespective of the drive type. The paper highlights the synthesis and performance of starch-based carbon electrodes, which achieved capacitances of up to 130 F/g, offering a sustainable and cost-effective alternative to conventional carbon materials. It discusses the electrochemical behavior, challenges of energy density and temperature sensitivity, and the need for effective voltage balancing in railway systems. The study further estimates that storing 15 kWh of recovered energy would require about 245 kg of supercapacitors, emphasizing both the feasibility and the technical considerations for large-scale implementation. Ultimately, the paper contributes to advancing environmentally responsible energy solutions for modern rail transport through supercapacitor integration.

2.6. The Impact of Mechanical Failure of 18,650 Batteries on the Safety of Electric Transport Operations [Contribution 6]

This study investigates the impact of mechanical damage on the safety of 18,650 lithium-ion batteries used in electric transport systems. Through finite element method (FEM) simulations and crash tests at various speeds, the authors demonstrate that collisions can lead to structural deformation, internal short circuits, and increased internal resistance, significantly impairing battery performance and potentially causing thermal runaway. Despite severe damage at high speeds (e.g., 200 km/h), thermal imaging confirmed that no internal short circuits occurred during charging, though capacity and reusability were compromised. The study underscores the importance of robust battery design, effective thermal management, and real-time monitoring systems to mitigate risks and improve safety and reliability in electromobility applications.

2.7. Combined Rough Sets and Rule-Based Expert System to Support Environmentally Oriented Sandwich Pallet Loading Problem [Contribution 7]

This article presents an integrated methodology that addresses the environmentally responsible optimization of the Sandwich Pallet Loading Problem (SPLP) within the logistics of fast-moving consumer goods. The authors combine Dominance-Based Rough Set Theory (DRST) and a rule-based expert system to minimize the number of stacked load units (LU), thereby reducing transport-related energy consumption. Key attributes such as product weight, height, and fragility are used to classify LUs. DRST handles imprecise data, while the expert system applies decision rules to determine optimal LU arrangements. The methodology is validated using real-world data from a major distribution company, achieving a 40% reduction in energy consumption per customer order. The paper demonstrates how machine learning and optimization can jointly support sustainable logistics operations by minimizing shipments, reducing resource use, and enhancing decision-making consistency.

2.8. Deep Learning vs. Gradient Boosting: Optimizing Transport Energy Forecasts in Thailand Through LSTM and XGBoost [Contribution 8]

This article investigates the accuracy of transport energy consumption forecasting in Thailand by comparing two advanced machine learning methods: Long Short-Term Memory (LSTM) neural networks and the XGBoost algorithm. Using data from 1993 to 2022, including vehicle registrations by size, vehicle kilometers traveled (VKT), GDP, and population, the study evaluates each model’s predictive performance. Results show that XGBoost significantly outperforms LSTM across all statistical measures, achieving an R² of 0.9508 versus 0.2005. Feature importance analysis highlights that medium vehicle registrations and truck VKT are dominant predictors of energy use, together accounting for over 57% of model influence, while demographic and economic variables contribute about 15%. The research not only demonstrates the superior forecasting capability of XGBoost, but also offers practical insights for policymakers seeking to enhance transport energy efficiency through evidence-based fleet and infrastructure planning.

3. Conclusions and Future Works

This Special Issue provides a comprehensive overview of current advances in the optimization of intelligent transport systems (ITS) with a strong emphasis on energy efficiency, environmental responsibility, and transport safety. The contributions demonstrate the versatility of methods ranging from machine learning and rough set theory to empirical transport energy analyses and safety testing of vehicle components applied to both urban and regional contexts. Together, these studies illustrate how integrating intelligent technologies, operational optimization, and energy-aware planning can foster more sustainable and resilient mobility systems.

The results confirm that ITS optimization requires not only technological innovation but also a context-sensitive approach that considers the physical, environmental, and social constraints of transport networks. Whether in public transport, logistics, electromobility, or infrastructure forecasting, the alignment of digital solutions with broader sustainability goals is essential for realizing systemic improvements.

Looking ahead, several avenues for future research emerge. These include the following:

• Integrating ITS planning with real-time big data analytics to improve adaptive transport management;
• Investigating the long-term environmental and economic impacts of electromobility infrastructure and alternative energy storage technologies (e.g., supercapacitors, hydrogen cells);
• Exploring hybrid modeling frameworks that combine machine learning with rule-based and simulation models for transport system forecasting and risk assessment;
• Extending multimodal energy efficiency studies to include active transport and last-mile logistics;
• Assessing the social acceptability and behavioral implications of ITS and Mobility-as-a-Service (MaaS) platforms in diverse urban and regional settings.

These directions offer promising opportunities for advancing transport research and supporting policy frameworks aimed at decarbonization, accessibility, and resilience in the mobility sector.