Gaussian processes for unconstraining demand
From MaRDI portal
Abstract: One of the key challenges in revenue management is unconstraining demand data. Existing state of the art single-class unconstraining methods make restrictive assumptions about the form of the underlying demand and can perform poorly when applied to data which breaks these assumptions. In this paper, we propose a novel unconstraining method that uses Gaussian process (GP) regression. We develop a novel GP model by constructing and implementing a new non-stationary covariance function for the GP which enables it to learn and extrapolate the underlying demand trend. We show that this method can cope with important features of realistic demand data, including nonlinear demand trends, variations in total demand, lengthy periods of constraining, non-exponential inter-arrival times, and discontinuities/changepoints in demand data. In all such circumstances, our results indicate that GPs outperform existing single-class unconstraining methods.
Recommendations
- Unconstraining methods in revenue management systems: research overview and prospects
- Gaussian processes with linear operator inequality constraints
- Multi-class unconstrained estimation models based on multi-distribution of demand in revenue management systems
- Advanced Lectures on Machine Learning
- Estimating shape constrained functions using Gaussian processes
Cites work
- 10.1162/15324430260185646
- `Horses for courses' in demand forecasting
- Complete characterization of Hadamard powers preserving Loewner positivity, monotonicity, and convexity
- Detection of abrupt changes: theory and application
- Forecasting and Stock Control for Intermittent Demands
- Gaussian processes for machine learning (GPML) toolbox
- Gaussian processes for machine learning.
- scientific article; zbMATH DE number 3860238 (Why is no real title available?)
- scientific article; zbMATH DE number 3583145 (Why is no real title available?)
- scientific article; zbMATH DE number 3054885 (Why is no real title available?)
- Inference about the change-point in a sequence of random variables
- Model Evaluation and Spatial Interpolation by Bayesian Combination of Observations with Outputs from Numerical Models
- Models of the Spiral-Down Effect in Revenue Management
- On fractional Hadamard powers of positive definite matrices
- Revenue management without forecasting or optimization: an adaptive algorithm for determining airline seat protection levels
- Revenue management: Research overview and prospects
- The theory and practice of revenue management
- Theory and use of the EM algorithm
- Unconstraining methods in revenue management systems: research overview and prospects
Cited in
(5)- Unconstraining methods in revenue management systems: research overview and prospects
- Multi-class unconstrained estimation models based on multi-distribution of demand in revenue management systems
- Explainability through uncertainty: trustworthy decision-making with neural networks
- Bayesian variable selection in Kriging metamodeling for quality design
- Optimal control by policy iterations and constrained Gaussian process regressions
This page was built for publication: Gaussian processes for unconstraining demand
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1713757)
