Demand Forecasting For A Leading Car Manufacturer

Screen Shot 2020-05-20 at 10.29.33 am


Demand forecasting presents itself as one of the foundational challenges in retail inventory management and budgeting. Traditional heuristic methods are resource intensive when scaled to large sets of products and often underperforms in accuracy by failing to incorporate the effects of substitutes and complementary goods.


The Problem

A leading Australian car manufacturer wanted to be able to produce high-quality forecasts for their products. This included knowing how much and what type of stock each dealer should hold to maximise sales but minimise excess capital being held up in inventory. Additionally, they wanted to know how competitors and their own sales promotions would affect the sales of their products.


Our Approach

Our approach effectively provided stable and accurate forecasts at the colour, model and dealer level by integrating the effects of promotional forecasting, marketing expenditure, delivery lags and supply constraints in addition to yearly, monthly, weekly, and cyclical seasonalities. Our implementation also factored in the effects of exogenous variables such as holidays, weather, and local demographics. Moreover, promotional forecasting identified the effect of promotions on substitute and complementary goods such that all affected inventories were optimised to meet demands and reduce surplus stock.

We developed a balance between the statistical and machine learning disciplines – constructing a hybrid system that benefits from both the flexibility of nonparametric models and the stability of statistical methods. The initial stage of preprocessing augments existing data by imputing outliers and restructuring the dataset for various predictive models. Additional features were also engineered through the underlying time series in reflection from our past experiences and expertise. Subsequently, predictions from statistical models, in conjunction with nonparametric and deep learning based recurrent networks, were stacked together to mitigate overfitting of any singular model.



Our demand forecasting implementation provided a dynamic, large-scale and systematic way of optimising retail inventory management and increased the accuracy of budgeting. As a result, the client was able to better focus resources and expertise in their core objective of providing a better experience and service for their customers.