Geoscience Reference
In-Depth Information
1 Introduction
The train platforming problem is a subproblem of the generation of a timetable for a
railway company. The generation of a timetable is a hierarchical process. At the
rst
stage, a preliminary timetable for the whole network is proposed. In this phase, a
macroscopic viewpoint at the railway network is applied. Stations are considered as
black boxes. Capacity limits of particular stations and the movement of trains inside
the stations are not taken into account. Then, at the second stage, a microscopic
viewpoint related to stations is applied. At every station, the network timetable is
checked whether it is feasible with respect to capacity, safety and train operators
'
preferences. This process results in a track occupancy plan which speci
es for each
arriving or departing train the platform track along with the time slot during which
the track will be occupied by the train. Cargo trains do not affect the plan since they
travel mostly in night, when there are fewer passenger trains, they use different
tracks in the station, and in case of con
fl
icting movements they can wait at the entry
signal.
In the Czech and Slovak Republic, planning train movements through the station
is done by hand, using planner
is experience and a set of rules determined by a
railway company. The main goal of this research is to design a more sophisticated
approach which would serve as a planner
'
s decision supporting tool and result in a
better track occupancy plan. Improvement in the plan quality results in:
'
1. better management of train operation in the station, namely:
(a) shorter times of routes occupation by arriving and departing trains,
(b) uniform workload of the infrastructure elements, such as tracks, switches,
and platforms, which leads to a more robust plan resistant to random
disturbances;
2. higher service quality perceived by passengers, namely:
(a) shorter distances needed for changing trains,
(b) more appropriate platforms (platforms near to ticket sales points and to the
station entrance, platforms equipped by station shops or catering etc.),
(c)
less probability of changing the planned platform when the train delays;
3. meeting train operators
'
requirements on arrival and departure times and plat-
forms assigned to trains.
Routing and scheduling trains at a station has been studied by researchers in
countries, where large, busy stations with capacity constraints can be found.
Billionet [ 1 ] addresses only the routing problem. The problem is modelled using a
graph theory and the integer programming formulation of the resulting graph col-
ouring problem is solved. However, the k colouring problem is not indeed an
optimisation problem, it means any feasible solution is acceptable and the problem
formulation does not re
ect the solution quality, such as route lengths or platform
preferences for individual trains. Zwaneveld [ 2 ] and Zwaneveld et al. [ 3 ] formulate
fl
Search WWH ::




Custom Search