This paper describes research in modelling train timetable and route generation as a constraintsatisfaction problem (CSP). The key objective of this research is to design a constraint-based scheduling algorithm that can be used to generate a train timetable given headway requirements at different times of the day. The key constraint is to avoid track circuit or route contentions while maximising train utilisation. The objective of the scheduling algorithm is to determine how service levels can be increased without jeopardising passenger safety. This research investigated traffic at a train terminus where two types of trains are competing for the use of the same tracks; trains that are reversing and trains that are being dispatched from the depot. The contention problem is particularly serious during the rush hour train build-up. The current timetable and train routing are generated using two separate rule-based systems. However, due to the complexity of constraints involved, the current systems cannot generate a plan that can meet the desired service levels.
[1]
David A. McAllester,et al.
Nondeterministic Lisp as a Substrate for Constraint Logic Programming
,
1993,
AAAI.
[2]
Pascal Van Hentenryck.
Constraint satisfaction in logic programming
,
1989,
Logic programming.
[3]
Vipin Kumar,et al.
Algorithms for Constraint-Satisfaction Problems: A Survey
,
1992,
AI Mag..
[4]
J.-F. Puget.
Object oriented constraint programming for transportation problems
,
1993
.
[5]
Jean-francois Puget,et al.
A C++ implementation of CLP
,
1997
.
[6]
David L. Waltz,et al.
Understanding Line drawings of Scenes with Shadows
,
1975
.
[7]
Alain Colmerauer,et al.
An introduction to Prolog III
,
1989,
CACM.
[8]
Alan K. Mackworth.
Consistency in Networks of Relations
,
1977,
Artif. Intell..
[9]
Guy L. Steele,et al.
The definition and implementation of a computer programming language based on constraints
,
1980
.