AI Hospitality Alliance
Back to Research
Academic ResearchJanuary 1, 2024International Journal of Hospitality Management

Decoding the future: Proposing an interpretable machine learning model for hotel occupancy forecasting using principal component analysis

Most AI forecasting models are 'black boxes' that revenue managers don't fully trust — and tools you don't trust don't get used consistently. This paper proposes a machine learning approach that groups historical booking patterns into recognizable clusters (early corporate, last-minute leisure, event-driven spike, etc.) before applying a forecasting model to each, making the prediction explainable: not just 'occupancy will be 87%' but 'because booking pace matches the early corporate pattern, which historically reaches 87%.' The result is both more accurate than standard methods and more operationally useful, because revenue managers can see the reasoning and make better decisions around it.

Authors

Daniele Contessi, Luciano Viverit, Luis Nobre Pereira, Cindy Yoonjoung Heo

Article content

What the paper studied

This paper addresses a core challenge in hotel revenue management: the need for accurate and explainable occupancy forecasts. While machine learning has improved prediction accuracy, most models function as 'black boxes,' making it difficult for revenue managers to justify or trust their outputs. The study proposes a machine learning approach that leverages Principal Component Analysis (PCA) to group historical booking patterns into recognizable clusters—such as early corporate bookings, last-minute leisure surges, or event-driven spikes—before applying forecasting models to each cluster. This structure aims to make predictions not only more accurate but also transparent and operationally useful, allowing revenue managers to understand and communicate the reasoning behind forecasts.

Key findings

  • PCA reduces the complexity of booking data by organizing it into a small number of interpretable clusters that reflect actual booking behaviors observed at the property, rather than relying on generic industry templates.
  • Applying a separate pickup forecasting model to each identified cluster results in forecasts that are both more accurate than traditional and standard machine learning methods and more explainable to end users.
  • The model can provide actionable explanations for its predictions, such as linking an 87% occupancy forecast to a current booking pace that matches the historical early corporate pattern, which typically reaches that level.
  • Revenue managers who understand and trust the model’s reasoning are more confident in their pricing and yield decisions, better able to withstand short-term booking pace pressures, and more effective at communicating strategies to stakeholders.

Why it matters for hospitality

Forecasts are only as valuable as the trust and understanding they inspire in those who use them. In hospitality, revenue managers must regularly justify their predictions to general managers, owners, and asset managers. Black-box AI models, even when accurate, often fail to gain consistent adoption because their reasoning is opaque. By making the forecasting process interpretable and tying predictions to familiar booking patterns, this approach increases operational confidence and supports better decision-making. For technology buyers, interpretability becomes a critical criterion alongside accuracy. For asset managers and owners, explainable AI models support governance by enabling auditability and accountability, reducing the risk of over-reliance on unexplainable systems.

Practical takeaways

  • Use PCA to automatically identify and group booking patterns from historical data into clusters that reflect real, recognizable behaviors at your property.
  • Apply separate forecasting models to each cluster to improve both the accuracy and the interpretability of occupancy predictions.
  • When evaluating AI forecasting solutions, prioritize vendors who can clearly explain the reasoning behind specific predictions, not just their overall accuracy metrics.
  • Encourage revenue managers to adopt interpretable models, as understanding the forecast logic leads to better pricing, yield management, and communication with stakeholders.

Tags

Revenue ManagementOperationsEthics

Related research

Academic ResearchFebruary 3, 2023

Artificial Intelligence in the Tourism Industry: An Overview of Reviews

This paper is helpful because it shows where AI already has strong evidence behind it and where more caution is still needed. The clearest benefits are in efficiency, personalization, forecasting, and customer communication. At the same time, the research highlights issues that operators cannot ignore, including privacy, guest trust, staff impact, and the challenge of adopting AI in smaller or less data-rich businesses. For hotel leaders, the practical message is to move ahead in areas where the value is well established, while taking a more deliberate approach to ethics, employee transition, and whether a tool genuinely fits the business.

Academic ResearchApril 2, 2026

Smart Hospitality Operations: A Systematic Review of Digital Transformation, Intelligent Systems and Operational Excellence in Contemporary Hospitality

This paper gives hotel operators a practical structure for thinking about smart transformation. Rather than treating AI, IoT, robotics, analytics, and automation as separate projects, it shows how they work together to improve operational efficiency, service consistency, personalization, and sustainability. The most useful takeaway is that smart hospitality is not just about adding technology to existing workflows. It is about redesigning operations so connected systems, better data, and stronger staff readiness all reinforce one another. For managers, the paper points to a phased approach: invest in infrastructure, build employee capability, strengthen governance, and tie each technology decision to measurable business outcomes.

For Professors

Submit article for consideration

If you are a professor or researcher and would like to suggest a publicly available article for inclusion in the Research Hub, you can submit it for review and possible inclusion through our dedicated submission form.