Optimal Pricing of Configurable Products using Machine Learning

In this paper, we proposed a Machine Learning (ML) based methodology to optimally price configurable products. The methodology uses an ensemble of ML models to forecast the demand; along with an optimization framework to find the prices and upgrades that achieve highest business goals. The novelty of proposed methodology is providing an intuitive upgrade pricing experience for customers while targeting maximum objective(s) attainment for business. It was demonstrated that an ensemble of ML models can achieve errors lower than 30-40 percent allowing effective pricing of a portfolio of configurable products. The prices were close to the current manually generated prices by business experts.