Constraint sensitivity analysis of airline crew rostering using preferential bidding

Date

2018-10-08

Authors

Palav, Vaibhav

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Airline crew rostering is an important part of airline operations that can save costs by optimized assignments of crew members to existing sequence of flights (pairings). Crew rostering with preferential bidding considers the bids by crew members that determine their preferences on specific pairings so that they obtain their preferred overall schedule (roster). This problem is traditionally solved as an extension of the classical set covering problem. In this report, we present a preferential bidding-based set covering model with all the preferences are treated as hard constraints and perform a sensitivity analysis of the model by varying all the controllable parameters to uncover the robustness of the model’s solutions under these changes. These parameters include the block hour hard limits, block hour soft limits, number of off-days in the planning horizon (a month) among others. The parameters are varied with respect to the base parameter values and the changes in the optimal solutions are studied with respect to the objective function value, computational time and other performance metrics (total block hours assigned, standard deviation of block hours across crew members, etc.) We observe which of these parameters have a large impact on the model and perform this sensitivity analysis to check the robustness of the base model. This report’s analysis is a first step towards creating a full-fledged model with additional extensions such as seniority-based bidding.

Description

LCSH Subject Headings

Citation