Gaussian processes for unconstraining demand

Price, Ilan; Fowkes, Jaroslav; Hopman, Daniel

doi:10.1016/j.ejor.2018.11.065

Cited by 6 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The case study only considers knowing the target labels in time due to limitations in the data; studies also satisfying the requirement for input features would help validate the findings. Although this paper focuses on uncertainty for NN classifiers, uncertainty can also be quantified for NN regression models and other ML models such as Gaussian Processes (Price et al, 2019) and Random Forests (Shaker & Hüllermeier, 2020). Finally, uncertainty as XAI can be used in active learning, where limited labeled training data is available and the ML system can ask a human expert to label the most uncertain observations (Kadziński & Ciomek, 2021).…”

Section: Discussionmentioning

confidence: 99%

Explainability through uncertainty: Trustworthy decision-making with neural networks

Thuy,

Benoit

2024

European Journal of Operational Research

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Explainability through uncertainty: Trustworthy decision-making with neural networks

Thuy,

Benoit

2024

European Journal of Operational Research

View full text Add to dashboard Cite

“…The (unconstrained) demand distribution is given in Table 2. Unconstraining is done through the framework outlined in Price et al (2019). The fares are also shown.…”

Section: Simulation Setupmentioning

confidence: 99%

Single-dimensional leg-level dynamic programming with booking-time dependent cancellation probabilities for revenue management

Hopman

Koole

et al. 2022

IJRM

View full text Add to dashboard Cite

In this paper, an optimisation method is introduced that accounts for cancellations. We do so by estimating the opportunity cost of a booking between the time of booking and the expected time of cancellation. The formulation involves an estimate of the value of the state of the system at the time of cancellation (which is in the future), found through novel heuristics we introduce. The fare that is used to determine whether a product is available for sale, is adjusted by the risk the airline faces. We introduce an example which shows that there may be cases where it is optimal to reject a higher-priced product if the risk of cancellation is high, while accepting a lower-priced product. Simulations show increases in revenues against a traditional formulations that does not explicitly models cancellations. We show our method is robust against choice of heuristic, misjudgement of cancellation probability and forecasting errors.

show abstract

“…In addition, they analyze the impact of the reference price on the gained revenue. Price et al (2019) use a Gaussian Process methodology to track and estimate the dynamic changes in demand, taking into consideration the necessity to unconstrain the demand (estimating the true demand in case inventory is assumed unlimited from finite inventory data). The Gaussian Process is a machine learning/statistical approach that models data as a joint multivariate Gaussian (Atiya et al 2020).…”

Section: Finite Inventorymentioning

confidence: 99%

Novel pricing strategies for revenue maximization and demand learning using an exploration–exploitation framework

2021

View full text Add to dashboard Cite

The price demand relation is a fundamental concept that models how price affects the sale of a product. It is critical to have an accurate estimate of its parameters, as it will impact the company's revenue. The learning has to be performed very efficiently using a small window of a few test points, because of the rapid changes in price demand parameters due to seasonality and fluctuations. However, there are conflicting goals when seeking the two objectives of revenue maximization and demand learning, known as the learn/earn trade-off. This is akin to the exploration/exploitation trade-off that we encounter in machine learning and optimization algorithms. In this paper, we consider the problem of price demand function estimation, taking into account its exploration-exploitation characteristic. We design a new objective function that combines both aspects. This objective function is essentially the revenue minus a term that measures the error in parameter estimates. Recursive algorithms that optimize this objective function are derived. The proposed method outperforms other existing approaches.

show abstract

Gaussian processes for unconstraining demand

Cited by 6 publications

References 32 publications

Explainability through uncertainty: Trustworthy decision-making with neural networks

Explainability through uncertainty: Trustworthy decision-making with neural networks

Single-dimensional leg-level dynamic programming with booking-time dependent cancellation probabilities for revenue management

Novel pricing strategies for revenue maximization and demand learning using an exploration–exploitation framework

Contact Info

Product

Resources

About