On Shelf Availability: 3-Wk Pilot Implementation

Tredence Inc

Automated ML based predictive alerts around stock-outs, low inventory, off scan anomalies etc. to reduce lost sales opportunities

Objective: Develop a solution using ML based diagnostic and predictive analytics that identifies root-causes of stock-outs and off-sales behavior and generates custom alerts to take preemptive action against lost opportunity.

Key Challenges Addressed:

  1. Current systems can track and capture Out-of-Stock (OOS) but lack accurate forward looking forecasts
  2. End to end inventory visibility
  3. Lack of a system that proposes preemptive alerts and strategic changes"

How do we address your challenges:

  1. ML driven predictive models based solution is build to identify the OSA levels in the stores and thus identify underlying issues.
  2. Provide competence in handling issues of inventory and shelf mismanagement by monitoring phantom inventory and safety stock.
  3. An accelerator to actively generate prioritized alerts at channel-Store-SKU-Daily level to minimize lost opportunity.

Pilot Outcome: Scope: 1 country; 1 retailer; ~500 stores; 2-3 SKUs

One time alert list generated based on sales behavior by SKUs

Implementation Plan The break-up of the implementation plan is as below: Week 1 - Data discovery and data ingestion Week 2 - OSA Solution building - Exploratory data analysis followed by ML modelling to identify OSA levels in the stores Week 3 - Alerts generated with required granularity and prioritization.

This implementation uses the following native Azure components: ADF pipelines ADLS Gen 2 Azure SQL database Azure ML Power BI

https://store-images.s-microsoft.com/image/apps.17517.90336073-d895-42ae-a836-77f1f9cf7894.2f2bd233-d7d5-46dc-9daf-1b8ed5bea150.c971ce66-a012-4657-8bb0-ef31f72297aa
https://store-images.s-microsoft.com/image/apps.17517.90336073-d895-42ae-a836-77f1f9cf7894.2f2bd233-d7d5-46dc-9daf-1b8ed5bea150.c971ce66-a012-4657-8bb0-ef31f72297aa