Solving airline crew scheduling problems by branch-and-cut
Management Science
An Efficient Implementation of Edmonds' Algorithm for Maximum Matching on Graphs
Journal of the ACM (JACM)
Discrete Optimization Algorithms with Pascal Programs
Discrete Optimization Algorithms with Pascal Programs
SchedSP: a Grid-based application service provider of scheduling solutions: Research Articles
Software—Practice & Experience
Set partitioning/covering-based approaches for the integrated vehicle and crew scheduling problem
Computers and Operations Research
Resolving scheduling issues of the London Underground using a multi-agent system
International Journal of Intelligent Systems Technologies and Applications
Multiresponse optimization of dispatch rules for public bus services
Computers and Industrial Engineering
Truck Driver Scheduling in the European Union
Transportation Science
SchedSP: providing GRID-enabled real-world scheduling solutions as application services
EuroWeb'02 Proceedings of the 2002 international conference on EuroWeb
| Hi-index | 0.01 |
The daily bus and driver scheduling, for all bus companies that operate a non-fixed daily schedule of legs, is a difficult combinatorial problem that must be solved every afternoon. The work of the next day is changing on a daily basis either due to different load requirements on the standard routes or due to additional services and trips that the busses need to perform and the bus companies do wait until late afternoon before solving the scheduling problem. In addition, there exist hard customer requirements on the time required for the solution of the problem. This paper firstly presents a quick heuristic scheduling procedure named QS for the solution of the problem. QS has worked very well in the production environment of several bus companies of Greece. The main algorithms used by QS are minimum cost matching, set partitioning and shortest path. In addition, a column generation procedure named CGQS that uses an LP-solver and the QS process as its integer solution finder is presented. CGQS starts from the solution point of a single QS run and then performs several iterations in which LP problems are solved and new promising shifts are created using the LP dual solution.