Scheduling Of Rail Driver Duties

Abstract

Operator costs account for a very high amount of any large conventional transport organisation's expenditure. A scheduling system that could save even a small proportion of this would clearly be very useful. An exhaustive literature search has revealed that although there have been some feasibility studies, there appears to be no practical driver scheduling system for the rail industry yet. Among other transport modes, bus is by far the closest to rail operation, airlines being concerned with rotating rosters covering many days, while other surface modes tend to use ad hoc methods. Bus driver scheduling systems are well developed with proven successes over the past fifteen years. It is therefore sensible to try to adapt a bus driver scheduling system for rail operation. During 1990-91, a model to estimate the numbers of driver duties required to cover any combination of train work and labour conditions was developed for British Rail in an exercise to appraise alternative working practices. This was based on the widely used bus driver scheduling system, IMP ACS, which employs heuristics and integer linear programming techniques. This paper describes the model and its results. The IMPACS-based model has been demonstrated to be capable of being adapted to a full scheduling system for British Rail, and IMP ACS was used to develop reasonably realistic rail schedules in some simplified test cases. There are, however, substantial differences between the modes which make it impossible to use a bus-related system directly for rail operation, and further work is being undertaken to develop a full rail driver scheduling method. The paper also discusses some rail conditions that complicate the scheduling problem. Transactions on the Built Environment vol 7, © 1994 WIT Press, www.witpress.com, ISSN 1743-3509 82 Railway Operations