Infoscience

Presentation / Talk

Integrated schedule planning with supply-demand interactions for a new generation of aircrafts

We present a model where we integrate schedule design, fleet assignment and demand models to maximize the profit of an airline. The objective of the study is to identify the challenges behind the integration of demand modeling into the optimization model and develop methodologies to overcome these challenges. The study is in the context of a collaborative work between EPFL and EPFL Middle East. A new air transportation concept, Clip-Air, is developed at EPFL which is a modular innovative aircraft with detachable load units that increase the flexibility. The decoupling of the load (capsules) and carrying units (wings) is believed to improve the airline operations. Clip-Air, being a flexible transportation concept, brings more interest into the integration of supply and demand models. Therefore models are adapted to the Clip-Air case and comparative analysis is done between standard aircrafts and Clip-Air to quantify the potential advantages. Considered supply model is an integrated schedule design and fleet assignment model. Integrated demand model is specified as a logit model where utility of the itineraries are explained by fare, departure time and number of stops which are shown to have explanatory power in the literature. Fare class segmentation is considered in the optimization model as different fare classes have different sensitivities suggested by the demand model. Furthermore we include supply and recapture effects in order to better represent the demand where recapture ratios are based on a logit model similar to the demand model. The resulting model is a mixed integer nonlinear problem and as a first step to deal with the high complexity of the problem, we present a heuristic method based on Lagrangian relaxation and sub-gradient optimization.

    Reference

    • EPFL-TALK-196037

    Record created on 2014-01-20, modified on 2016-09-16

Fulltext

  • There is no available fulltext. Please contact the lab or the authors.

Related material