Price Prediction of Used

Case Study

Price Prediction of Used Japanese Car


Japanese Used Car Exporters


Have a similar project in mind?

Business Challenges

The client exports used cars from Japan to 40 countries across the world enabling dealers to choose cars from online auction. During auctions, client gives quote for a particular make/model/ chassis ID to dealers with the help of a Subject Matter Expert (SME), which is costly and not scalable. The client wants to scale the business without the involvement of experienced SME.

Our Solution

We proposed to build Machine Learning algorithm which will predict the price of used car based on
the make/model/chassis ID available at upcoming auction.

Salient Features

  • The Algorithm gives 80% of accuracy while making the price prediction
  • The Algorithm will be retrained on a regular interval and will result in improvement of accuracy
  • Auction data acquisition from external sources
  • Data curation and data wrangling
  • Light weight frontend web application.
  • Easy to search based on Make/Model/Chassis ID
  • Option for users to send mails to their Mailbox
  • Option to view search history

Product / Service Description

We built a model using machine learning by training the model with historical data. An application was developed in Python for predicting the price of car available at auction& deployed on AWS for scalability. The Algorithm gives 80% of accuracy while making the price prediction and it can be further improved over time by retraining the algorithm.


Highly accurate results of the used cars auction

Case Studies
Explore Our Project Case Studies