An in depth analysis of the Preferential Bidding System for SAS Crew Planning

by Isabelle Kinnunen

Institution: Linköping University
Year: 2015
Keywords: PBS; Preferential Bidding System; Crew Planning; SAS; Scandinavian Airlines; Optimization; Engineering and Technology; Civil Engineering; Transport Systems and Logistics; Teknik och teknologier; SamhÀllsbyggnadsteknik; Transportteknik och logistik; Technology; teknik; Logistik; Logistics Management
Record ID: 1352661
Full text PDF: http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-118570


Crew planning is a complex optimization problem that usually is divided into two separate parts, pairing and rostering. In the rostering process crewmembers are assigned a personal roster, a schedule, for the next coming time period. In order for the crew to have an influence on their roster, SAS have implemented a Preferential Bidding System (PBS) where the crew has the ability to express their wishes regarding their roster, formulated as bids. Generally there is an outspoken dissatisfaction among the crew about PBS and how it works and therefore the questions were raised of how the system could be more transparent and how the bid fulfillment could be communicated both internally and externally. The performed analysis was made with regard to bidding pattern and bid fulfillment and was both quantitative and qualitative. The quantitative part consisted of a thorough review of historic data (delimited to only four months) and a questionnaire, targeting the whole crew collective, was set up as a complementary qualitative analysis. In this thesis the work and result is presented. The result of the analysis shows that satisfaction felt by the crew cannot be directly compared to bid fulfillment as it is measured today. The major aspects of improving the PBS should include a simplification regarding mainly bid types, but also actions in order to increase the transparency of the system and awareness. The recommendation to SAS is to present statistics to crew for each bid type regarding number of placed bids, number of possible fulfillments and how much that is fulfilled thus visualizing the possibility to get a bid fulfilled. A revised range of bid types have also been recommended along with a few minor changes.