Case study: Supply Chain Demand Monitoring for a Leading Networking Equipment Company

About the Client

The client is an American multinational corporation that develops and sells networking products. Its products include routers, switches, network management software, network security products, and software-defined networking technology.

Business Context

The core objective of this engagement was to help the client develop a scientific method to capture the customer demand getting fulfilled by the distributors. At the same time the client wanted to develop the sell in forecast to improve the forecast accuracy for themselves and their distributors.

BRIDGEi2i Solution

BRIDGEi2i partnered with the client to create a demand monitoring system to keep a track on forecast accuracy and bias at different levels. The Machine Learning and AI lab experts at BRIDGEi2i developed a future forecast visibility system to give a better visibility of forecasts which eventually lead to the improvement of the demand planning process.

  • BRIDGEi2i’s data engineering team created a Data Mart to integrate sell through data, sell in data, drop shipments to customers and distributor’s inventory snapshots.
  • Features were determined based on static information like COGS, list price, product hierarchy and dynamic information like run rate, volatility and product lifecycle stage.

BRIDGEi2i’s AI Labs and Data Engineers created SKU segmentation to classify them into primary and secondary, intermittent, fast moving, transition etc. The team implemented a robust forecasting methodology based on segments and choice model.

The AI Lab experts and the consulting team at BRIDGEi2i deployed a demand planning optimizer solution for searching and fitting the best algorithm for each of the segment. The Optimizer solution was used to forecast sell through, sell in and drop shipments at SKU level. The solution also provided a front-end tool for tracking the forecast accuracy.

Business Impact

0
%
over Baseline Improvement in Sell-Through Forecast Accuracy
0
%
Improvement in Sell-in accuracy over Client Forecasting Alogrithm
0
M
$ Potential Saving Annually