Building a Machine Learning Model to Predict the Sales of Auto Parts


Maruti Techlabs' client, A20 Motors, manufactures over 10,000 premium-quality spare parts for Asian, American, and European vehicles. A20 Motors is one of the largest manufacturers and distributors of aftermarket car and truck parts in Central and South America, North America, Europe, Asia, Australia, and the Caribbean. (Disclaimer - The name A20 Motors is a placeholder as there is an NDA signed between both parties.)


For the smooth functioning of A20 Motors, correctly stocking up inventory to meet sales and demand of parts is as important as maintaining the highest manufacturing standards for those parts. The team of A20 Motors prepared a mathematical formula to predict the sales cycle of different parts. They used this formula to ensure accurate stocking up of inventory. As the mathematical formula helped predict the sales cycle, it formed the basis of the most crucial pillars of their business - inventory management, restocking, maintaining the warehouse, and optimizing the shipping process. Over time, they found that the mathematical formula was not accurately predicting the upcoming demands. It resulted in overstocking & understocking of different parts and, consequently, loss of potential sales and profit.


A20 Motors sought an advanced solution to accurately forecast sales of vehicle parts based on historical trends. They needed a statistically derived machine learning model to predict sales of both current and new parts. After extensive research, their VP of Product Development selected Maruti Techlabs as their machine learning solutions partner. Maruti Techlabs conducted a four-week feasibility study to analyze data correlations and preprocess the data. They developed a model using variables such as model/make, number of vehicles in operation, part type, and total units sold per vehicle. After experimenting with various algorithms, they finalized an LSTM model with Derived Sale Days, which accurately predicted sales numbers. Challenges included predicting sales for new parts with little historical data and handling skewed data from client APIs, which Maruti Techlabs addressed by building an API for extracting current data and retraining the model.

Image Gallery