Intelligent Safety

Schöneburg, Rodolfo; Buschbeck, F.; Gärtner, Eric; Struck, Matthias; Bogenrieder, Ralf; Bullinger, Wilfried; Richert, Julien

doi:10.1007/978-3-658-18443-8_9

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since most of these relationships seem to be probabilistic and there fore cannot be expressed as deterministic rules, market predictions are among the most well suited and promising applications of artificial neural networks. Several proposals have been made to use neural network models for prediction and forecasting problems in business-related tasks, such as locating sources of forecast uncertainty in a recurrent gas market model [25], corporate bond rating [4], currency exchange rate analysis [8, 181, economical modelling [13], mortgage delinquency prediction [7], chaotic timeseries prediction [24], prediction of IBM daily stock prices 26 , prediction of three selected German stock prices [22], prediction of the FAZ-Index [6] and rediction of the weekly Standard & Poor 500 index pl7J. In some of these proposals, the neural networks performed better than regression techni ues [4,22] or as good as the Box-Jenkins techni ue [21, while in others the results were disappointing 77, 261.…”

Section: Introductionmentioning

confidence: 99%

Controlling airline seat allocations with neural networks

Freisleben

Gleichmann²

[1993] Proceedings of the Twenty-Sixth Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

In this paper we present a neural network that is intended t o support airline marketing specialists in controlling seat allocations on flight departures. The focus of our investigation is the prediction of overbooking rates in order t o avoid that an aircrafl departs with empty seats when passengers who have booked seats do not participate in the flight. The neural network proposed t o solve the problem is an extension of the forward-only counterpropagation model. The network learns to approximate the mapping between the input data (the number of booked seats f o r each reservation class at distinct time periods prior to departure) and the desired output (the number of no-shows). The trained network is then used t o make the gredictions f o r the future. The feasibility of our approach is demonstrated by an efficient implementation. Ezperimental results obtained on real-life booking data f o r a particular flight indicate that the proposed neural network model is superior t o the standard forward-only counterpropagation model and quite competitive t o traditional, non-neural methods applied to the overbooking prediction problem.

show abstract

Section: Introductionmentioning

confidence: 99%

Controlling airline seat allocations with neural networks

Freisleben

Gleichmann²

[1993] Proceedings of the Twenty-Sixth Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

show abstract

Intelligent Safety

Cited by 1 publication

References 0 publications

Controlling airline seat allocations with neural networks

Controlling airline seat allocations with neural networks

Contact Info

Product

Resources

About