Budget Rent a Car
Client: Budget Rent a Car System, Inc. is one of the world's best-known car rental brand with more than 2,700 locations in more than 120 countries.
Challenge: Budget faced a critical logistics optimization problem on a daily basis which was to price and deploy cars maximizing fleet utilization while maintaining a high service level with minimum costs. Two key challenges they were facing:
Rental Cost Determination: In order to rent a car, consumers pay a rental fee to rent a car for a short period of time, ranging from a few hours to several weeks. Budget offers different categories of cars, from premium to full-size luxury. Rental fee is usually determined by two factors; duration and category of the rented car. The client was in need of a statistical mathematical model which would plan rental fee for coming days in the most optimal way.
Fleet Management: Budget’s typical policy for car rentals was to accept reservations based on a macro analysis of fleet availability. They were not operating by pre-assignments of specific cars to forecasted customers but by planning based on a cursory matching of reservations with the pool of available cars for different groups. To achieve a high service level, providing all customers that hold a reservation with the car of the requested group (or an upgrade) was extremely important, which resulted in issues like overbooking etc.
The client was looking for a predictive model which would forecast reservations they should take for a specific car group on a given day, to optimize resources and reservation allocation accordingly.
Solution: With extensive experience in revenue and yield management solutions, SSI’s team of data engineers developed an advanced statistical forecasting platform. By carefully analyzing data of current reservation and past reservations, the tool runs various scenarios based on the variation of capacity, variation of customer choice, costs of overbooking, duration of booking a car and overbooking.
Through intelligent algorithm processing of various scenarios and historical data, the platform forecasts the optimal number of reservations the client should accept for each car plan. Moreover, the solution also makes pricing recommendations based on historical data and demand forecasts.
Result: The advanced statistical forecast platform for revenue and yield management helped the client in optimizing and maximizing the total revenue and yield in capacity. Importantly, it reduced operating headaches for the company by lowering customer facing and marketing issues like overbooking, losing customers due to pricing, pre-assignment of correct car category etc. The client also experienced better utilization of resources leading to cost savings.