Rail pods are an emerging concept of modular self-propelled rail vehicles which can interchangeably move freight and passengers to provide a more customer-oriented rail service. Pods are envisaged to operate on demand with the possibility of forming platoons either physically (e.g. mechanical or digital couplers) or virtually (e.g. by radio communication) coupling at stations. This study addresses the need to understand how rail pod platoons affect rail corridor capacity by analysing the actual infrastructure occupation under different platoon compositions, taking into account train movement dynamics and signalling constraints. First, this study extends the consolidated rail capacity assessment method (UIC Code 406) by applying the blocking time theory to assess the infrastructure occupation of the rail pod platoons. Based on this extension, a nonlinear optimization model is developed to determine coordinated speed profiles that are structurally consistent with the platoon configuration, aiming to minimize rail capacity utilization. The model is applied to a case study considering the ETCS Level 2 signalling system. The results obtained for the case study illustrate the ability of the proposed model to identify operational speeds and composition of rail pod’ platoons that lead to the effective capacity use of the existing infrastructure. This capacity assessment framework provides a theoretical foundation for flexible allocation of modular rail cars in dynamically structured platoons.

This paper proposes an agent-driven autonomous operation control architecture for maglev transportation system. To address the inherent contradiction between the uncertainty of intelligence and strict security of system, a hierarchical decisionmaking architecture is presented. It comprises three layers: organization, interface and action, where the decision-making domain of maglev agent is confined to the first two layers. Further, to validate both the architectures, a framework simulating agent decision-making process is constructed. It is implemented by the integration of three core components, including large language model (LLM), domain toolset and maglev train dynamics module. The results from testing autonomous operation planning task, demonstrate that maglev agent has established the cognition of operation task, and achieved appropriate decision-makings by leveraging domain-specific knowledge, tools and instructions, and information interaction. Ultimately, the proposed architectures transform the operation task expressed in natural language semantics into executable train operation strategy through multilayer decision conversion.

Rail Pods are an emerging concept of modular self-propelled rail vehicle which can interchangeably move freight and transport for a more customer-oriented rail service. Pods are envisaged to operate on-demand with the possibility of forming platoons by either physically or virtually coupling at stations. In such a context, it is essential to quantify the actual service capacity of rail pod platooning, taking into account heterogeneous convoy structures and infrastructure constraints such as signalling rules and block occupation. This study extends UIC Code 406/blocking time theory to applied to Pod platoons, proposing a novel optimization model that integrates traction, cruising, and braking speed profiles alongside safe separation constraints. This approach enables coordinated optimization of cruising speeds across various platoon structures to minimize track capacity consumption. The model is applied to a case study considering the ETCS Level 2 signalling system. The results obtained for such a case study illustrate the ability of the proposed model to identify operational speed and composition of rail pods’ platoons which lead to capacity effective use of the existing infrastructure. The proposed method provides potentials for a more flexible allocation of modular rail cars based on demand configuration.