Introduction
RailSys uses a microscopic description of the real world in the model to reduce the risk of wrong decisions based on an inexact planning. The program has been developed by the University of Hannover and RMCon, Rail Management Consultants. It has been applied successfully in different projects such as the high speed lines Cologne-Rhein/Main and Sydney-Canberra, the city railways of Munich, Cologne, Sydney, Melbourne… or the rail network in Berlin and Copenhagen.
RailSys consists on four modules:
• Infrastructure manager £¨STED£©, its function is to model the existing infrastructure and to build up different infrastructure variants. This editor allows to model with an accuracy of one meter. The modelized elements of the infrastructure include switches, signals, stations, speed indicators, platforms, stop boards, routes… and they contain attributes about track parameters like length, gradient, max. speed… This module contains several alternatives to refer the signalling system, such as ATC £¨Automatic Train Control£© or M/P £¨main signal/presignal£©, or referring to the interlocking machine. The starting point is the usual partition of the rail network into stations, junctions, and sections connecting them. The infrastructure is built up as a succession of nodes and links in RailSys, where a link symbolises a track and a node is either a connecting point of several links or a location for signals and other elements.
Microscopic network display in STED £¨infrastructure manager of RailSys£©
• Timetable manager: the program is ready to get the exact running time calculation and they are transferred into the timetable construction phase done with SIMU; according to this, it is possible to see the effect of any track change or other changes immediately. In this module, exact routes and different alternatives are set for every train, apart from the definition of the characteristics of the rolling stock £¨acceleration, deceleration, train length, priorities, train patterns…£© and other dispatching rules. The main task of the program is the optimal allocation of locomotives in large lines or networks, so it indicates inconsistencies such as unfeasible connection times, not enough headways and conflicts with other trains.
Display of timetable manager:
1- Hierarchic train and pattern list
2- Timetable table
3- List of detected conflicts
4-Track layout
5- Velocity route diagram
6- Graphical timetable with block occupation and conflicts.
• Simulation manager: Nominal or perturbed simulations can be carried out. Nominal £¨or single£© simulation serves normally to check that everything is working fine £¨some conflicts may not have been detected previously£©, Nevertheless, further simulations have to be analyzed to emulate real train operations, because of the very complex relationship when a train is delayed £¨this delay can spread and disturb other trains in the system£©. These perturbed simulations can be edited according to several criteria £¨delays in dwell time, accumulated delay at the departure of some trains…£©. In the case of the perturbed simulations, the program will take into consideration the dispatching methods to reroute the trains. Results about delays, obstructions… are displayed to be considered.
Display of Simulation manager
1- Simulation time
2- Delays
3- Current data of a specific train
4- Animated simulated with display of all trains and their delays
• Evaluation manager £¨PE£©: it has been developed to evaluate the impact of the infrastructure or timetable alternatives. It evaluates the performance of the simulated operational program, by means of preparing and analysing the delay data. To identifying the most favourable option, comparisons should be made. This involves determining whether different options result in significant improvements to the overall system performance.
Examples of statistical evaluation by PE:
On the left hand: average arrival delay of commuter trains in every station
The results can be displayed as performance of the whole network, for lines only, in stations, for the total amount of trains or for different patterns… and different parameters can be analyzed: arrival and departure delays, additionally generated delays £¨which could identify critical sections£©, on time running performance at stations, number of delayed trains, number of operational manoeuvres £¨or rerouting£©, block occupation…
