Car park yield management

K-Yield

Parking yield management

From business intelligence to dynamic pricing

The interest paid to dynamic pricing techniques for parking lots (or car park yield management) has considerably increased in the very recent years, due to the remarkable shift of the sector, becoming more and more digital. 

The development of smart city concepts (pricing policies becoming a recognized component of a city eco-mobility policy), the extension of new services like parking bookings, concerning sometimes inventories that were not even digitalized before (hotel parking spots around an airport, for example), etc.: all these moves have led to a completely new parking competition context, with an absolute necessity for traditional actors (airports, urban parking operators, etc.) to become price agile.

Kowee’s car park yield management solution is a true dynamic pricing engine: it does not consist in enabling a parking manager to blindly push on-line rates according to his feeling (or, in the best case, his expertise). Based on forecasts and unique machine learning algorithms, it finds the optimal price to set for a given booking request, considering the alternatives and their degree of certainty (other vehicles, likely to occupy the same inventory at the same time and leading to other revenue expectancies).

 

K-Yield 

*To understand what a “true” car park dynamic pricing solution is, please have a look at this: Parking dynamic pricing: how to not get it wrong

K-Yield is “a true”* car park Dynamic Pricing solution that helps you increase your revenues immediatly.

Its classic version is backed by a pre-booking system and each time a request is received, the K-Yield engine responds with an optimal price.

Another version of K-Yield exists which allows to change prices also optimally but without the need for a pre-booking system: prices evolve dynamically at the parking entrance, displayed on digital panels. This is the so-called K-Yield version for drive-ups (“drive-ups” are vehicles without reservation).

Benefits

K- Yield is a dynamic pricing (or yield management) software solution specifically designed for the parking sector.

  • By defining on-line pricing tactics (also called “yield strategies“), the manager can have a direct impact on the on-line parking revenues
  • A minimum of 5 to 10% additional revenues are usually observed after a 6-month transient period

 

K-Yield is a sophisticated dynamic pricing solution giving full control on the parking revenues. It is based on the most advanced revenue optimization techniques for drive-ups and pre-bookings at the same time.

Benefits

K- Yield is a dynamic pricing (or yield management) software solution specifically designed for the parking sector.

  • By defining on-line pricing tactics (also called “yield strategies“), the manager can have a direct impact on the on-line parking revenues
  • A minimum of 5 to 10% additional revenues are usually observed after a 6-month transient period

 

K-Yield is a sophisticated dynamic pricing solution giving full control on the parking revenues. It is based on the most advanced revenue optimization techniques for drive-ups and pre-bookings at the same time.

Needed data
The sources of data are equivalent to that of K-Analytics and K-Pricing.

A minimum of 1 to 2 years of past invidual transactions (tickets) is necessary. This holds for the gate system data as well as, ideally, for the booking system (but it is well possible to launch a brand new e-commerce platform, including parking space pre-booking features, together the implementation of K-Yield).

The APIs to build with the third party systems are the following:

  • A real-time connection with the gate system to make sure the latest information about the car parks are “known” by the system: any new entry or new exit may well have an impact on the price proposed for a later booking request
  • A real-time connection with the e-commerce platform (or “parking pre-booking system”), to ensure an instantaneous answer to any booking request
Features

K-Yield has been designed with the most innovative algorithms to address the question of the optimal trade-off to find between the various lengths of stay expected to enter the parking.

  • K-Yield uses innovative machine learning methods combined with cutting-edge optimization algorithms to create superior demand predictions and inventory management.
  • The comprehensive predictive models combine detailed historical demand along with current booking and drive-up demands (K-Yield does not only forecast the bookings but also the drive-ups)
  • K-Yield does not rely on occupancy forecasts to produce optimal prices (because this is not a driver of parking revenue optimization)
  • The seasonal models are easily customizable to properly address special events, car park maintenance periods, holidays, etc.
  • K-Yield is all but a black box: it offers an extremely simplified user interface making it immediate to control / fine tune the proposed dynamic prices in the various so-called Yield situations.
  • The applied Yield policies are therefore always 100% aligned with the sales and marketing strategy of the parking operator
Implementation

The implementation of K-Yield follows follows a specific path to ensure 100% efficiency as for the Yield strategies that will be put in place:

  • An understanding phase: comprehension of the  local parking demand drivers, market habits, marketing positioning of the various products, past pricing policies, occupancy constraints, etc. ; all this is made possible by a preliminary one-off extract of 2 years of past data (bookings and tickets)
  • A forecast fine tuning phase: Kowee data scientists adjuts the Kowee machine learning models, together with managing of the calendar matching rules (past vs future) and Yield expert parameters
  • In parallel, the APIs are put in place between the various third party systems and Kowee’s
  • A phase dedicated to the Yield strategy definition, including customer segmentation, dynamic pricing simulation over a future demand period (and related gain estimates), etc.

All this is ran over a period of 10 to 12 weeks, where Kowee project team will propose recommendations while also ensuring knowldege transfer.

Close all items