Each year, approximately 1 million guests are relocated due to overbooking. Walking guests costs the hotel industry an estimated $5 billion per year.
What it does
4C is a cloud-based application that leverages predictive analytics to get actionable insights for optimizing overbookings by hotel sales managers. It relies on historical occupancy data, current & future reservations, variable room costs and walk costs to forecast the optimal rate of overbooking.
How we built it
See architecture in gallery above
Challenges we ran into
- Setting up authorization flow with 3rd-party hotel API
- Getting access to data & sort through it
- Narrow down problem scope
- Develop algorithm & data science to get problem insights
Accomplishments that we're proud of
- Team-wide contributions where each member was able & willing to help in their field of expertise
- Fully-functional app that actually works
- Working data pipelines to ingest & analyze real data in real-time from 3rd party APIs.
- Rapid focus in defining problem so we can prioritize objectives, divide tasks, plan and execute in short timeframe
- Fantastic team chemistry
What we learned
- The power of working in a multi-functional team
- The invaluable insights one can get from existing data
What's next for 4C - Predictive Analytics for Overbooking Optimization
- Develop ability to input additional variables such as events and weather to algorithm to calculate cost and opportunity
- Give hotels ability to view data by day of week, reservation type, loyalty status and more
- Incorporate data from additional APIs to accurately forecast demand for similar properties
- Leverage more machine learning to optimize forecast for all variables.
Log in or sign up for Devpost to join the conversation.