ABOUT THE CLIENT
Client is a manufacturer, distributor and retailer of window treatments and components in Australia. Client operates a number of manufacturing plants including textile weaving and coating, plastic components and the fabrication of timber, fabric and aluminum venetians, curtains and vertical blinds.
CHALLENGE
The client offers a category of products to its customers. The demand on a given product depends on many factors including a product’s own properties such as price or brand, prices of competing products in the category, sales events, and even the weather. The goal is to build a demand model that incorporates these factors and allows one to perform what-if analysis to forecast response on price changes, assortment extensions and reductions, compute optimal stock levels, and allocate shelf-space units.
DURATION

1 – 2 Months

APPROACH

Demand prediction models are generally useful in marketing campaign design because they explain the impact of regressors on demand. For instance, a demand prediction model can reveal that the price sensitivity (the measure of how much the demand changes when the price changes) on a given product strongly correlates with the package size and demographic properties of the neighborhood, suggesting the use of different prices at different stores and setting different per-unit margins for different package sizes.

We will use the demand prediction models dedicated to price optimization and assortment planning.

SOLUTION

Demand prediction can be considered a relatively straightforward data mining problem that boils down to building a regression model and evaluating it over historical data. However, the design of the regression model is not so straightforward because the demand is influenced by many factors with complex dependencies. We studied a regression model suggested and evaluated for a supermarket chain in the Netherlands. This model is based on earlier marketing studies for fashion retailers and consumer durables providers who also reported usage of similar models in practice. However, it is important to understand that different optimization problems require different demand prediction models and it is hardly possible to build a universal demand model that incorporates a wide variety of factors that influence demand.

TECHNOLOGIES USED

R

Tableau

SQL Database

ALGORITHMS USED

Time Series Forecasting

KEY BENEFITS

√ Clients can reduce revenue loss associated with excess inventories and avoid the profit loss associated with the expiry of products.

√ Client can maintain enough stocks that fulfills customer demands

√ Help various departments (sales, operations, planning, manufacturing, finance, etc.) to ensure proper time for planning and coordination.

RESULTS

Our solution enabled the client to ensure efficient manufacture based on seasonal demand and, thereby increased profit by 6%