Skip to main content

Integrating Flexible Discrete Choice and Revenue Management Models

Article Category

Article available in the folowing languages:

A new take on revenue management models

New approaches in revenue management coupled with novel algorithms are promising to meet customers' needs while boosting income for firms.

Digital Economy

Revenue management, which analyses consumer behaviour to optimise product availability, strives to sell the right product to the right customer at the right time and for the right price. This challenging objective can be achieved through better mathematical models that integrate consumer choices more effectively, potentially making industries such as hospitality and travel much more effective. The EU-funded FLEXREV (Integrating flexible discrete choice and revenue management models) project sought to design new revenue management systems that integrate consumer choices for more effective mathematical models. One promising strategy in this respect is to exploit what is known as estimation of network generalised extreme value (NetGEV) models, which are effective in modelling demand shifts when products become unavailable and overall trends in consumer purchasing. In more technical terms, the project team worked on two different NetGEV models, namely multinomial logit and nested logit to predict consumer behaviour more accurately. It studied NetGEV models for choice-based samples, as well as the integration of choice-based models with capacity-based revenue management models. This was done for both single-product and multi-product environments. The team also made progress in updating classical Expected Marginal Seat Revenue methodology for a multi-product environment. They tested the new, emerging methods on data from two different major international hotel chains, demonstrating how best to allocate resources to customers in order to maximise revenue. Several related allocation algorithms for different estimation methods were also developed and tested to ensure the quality of estimation and robustness. Lastly, in addition to testing the data for applications in the hospitality industry, the project team conducted modelling for the telecommunications sector, exploiting actual data from a key firm in the field. The newly developed methods and algorithms could ultimately contribute to better revenues in the business field and render firms more competitive.

Keywords

Revenue management, consumer behaviour, consumer choices, NetGEV, logit

Discover other articles in the same domain of application