Algorithms Based on Artificial Agents to optimize the capacity of Railway Networks (2016-Present)

In a context of liberalization of the Spanish rail sector, the objective of the project is to investigate and develop new methodologies for the allocation of sections and hours of use of roads between the operating companies of the system. In particular, the suitability of using combinatorial auctions is studied along with agent based modelling methodologies.

The main deliverable is a pilot model of combinatorial auction applied to a sample of lines of the Spanish railway network, and a set of recommendations, derived from the simulation of different scenarios, of how to design the allocation process.

The project lasts 3 years with extension. The achievement of the objectives of the project involves substantive and objective innovations in the area of ​​capacity allocation in the railway sector, as there are not in reality sufficiently satisfactory models that integrate the complexity of real constraints and conditioning of the railway system. The challenge of the project is to show that by means of agent based modelling, this complexity can be incorporated.

The project team consist of 9 people who integrate experience and youth. It has extensive research experience in agent based modelling and combinatorial auctions, having participated in projects to assign capacities in airports and in resource allocation in multiproject environments.

The project has as interested companies ADIF (the manager of the railway infrastructures in Spain) and 2it Ingeniería (consulting company in the field of railway infrastructures).


Grupo de Investigación Reconocido de la Universidad de Valladolid
Paseo de Belén, 11. Edificio UVainnova. Campus Miguel Delibes. 47011 Valladolid
983 18 47 03

Este sitio web utiliza cookies para que usted tenga la mejor experiencia de usuario. Si continúa navegando está dando su consentimiento para la aceptación de las mencionadas cookies y la aceptación de nuestra política de cookies, pinche el enlace para mayor información.plugin cookies

Aviso de cookies