WIT Press


A Duty-block Network Approach For An Integrated Driver Rostering Problem In Public Bus Transport

Price

Free (open access)

Volume

128

Pages

12

Page Range

295 - 306

Published

2012

Size

467 kb

Paper DOI

10.2495/UT120261

Copyright

WIT Press

Author(s)

L. Xie & L. Suhl

Abstract

The driver rostering problem (DRP), arising in public bus transport companies, aims to generate for each driver a roster while management considerations, labor laws, and desires of drivers have to be satisfied. Optimal rosters are characterized by maximal satisfaction of drivers, minimal difference of overtime among all drivers, and minimal number of unassigned duties. The DRP is mostly solved sequentially due to its high complexity, namely firstly the rota scheduling problem, and secondly the duty sequencing problem. However, this method may generate sub-optimal rosters. In order to avoid a sub-optimal solution, the paper discusses an integrated DRP, which is solved with a duty-block network approach for real-world instances and compared with a solution of a heuristicbased integrated DRP. Keywords: integrated driver rostering problem, network optimization model, public transport, rotation-based rostering. 1 Introduction The operational planning process of public bus transport companies is divided into four steps, which are carried out sequentially. Figure 1 shows the relationship between the four operational planning problems at a public bus transport company (from Freling [1]). Line routes and their frequencies are given as input for a timetabling problem. In addition, the travel times along the lines as well as any potential layover times at stations are assumed to be known. In timetabling timetables are determined and a set of timetabled trips with start and end locations and times are given for the next step. In vehicle scheduling vehicles are assigned to timetabled trips resulting in vehicle blocks. After

Keywords

integrated driver rostering problem, network optimization model, public transport, rotation-based rostering.