Overview
Current rail transport infrastructure design is led by aesthetics and prescriptive regulations. This is a problem in passenger intensive operations, where performance and comfort are major issues. Better terminal design would provide increased passenger throughput and less overcrowding but it is very difficult to design for performance without accurate simulation tools. Existing tools are available to simulate passenger evacuation but none have been developed specifically for general circulation movements.
The goal of the AVATARS project was to develop a range of innovative technologies to allow the simulation of passenger movements in and around rail terminals in normal circulation behaviour. The tools have been created to enable designers to populate structures with large groups of realistic, interacting agents and follow their movements through a series of pre-defined, realistic scenarios. In this way AVATARS allows building and infrastructure design to be tested for passenger comfort and convenience at the design stage.
In order to achieve the project goal, AVATARS envisaged the development of three major software components.
- The first looked at agent interaction, specifically their interaction with each other and with their environment.
- The second component was a tool to rapidly generate realistic test scenarios in order to speed up and simplify simulation runs.
- The third was a tool to assist users in isolating the cause of any problem within the terminal design
The evaluation of the tools has been carried out under a series of case studies, developed throughout the project taking real-world information from operating terminals and trains and produce complex, multi-event scenarios capable of evaluating the abilities of the AVATARS software.
Funding
Results
The AVATARS project was completed and the prototype software was developed according to the requirements and specifications of the end users partners. The software was tested against real world data in realistic scenarios and the AVATARS partners achieved their quantifiable objectives for the software.