Poor management of stock levels leading to poor inventory turnover and loss of revenue
Sequence Number
2
Industry
Retail
Banner
How 5G enabled
5G enables real-time stock management – AI (ML) predicts demand and analyse optimised stock levels on shelves
Data Flows
Title
Devices
Icon
Description
Sensors & cameras to monitor inventory levels
POS data can provide inventory insights
Data from other relevant sources: Weather; Temperature; etc. that impact customer behaviour
Title
Connectivity
Icon
Description
5G will provide high bandwidth low latency connections, especially for processing video streams
Title
Edge Compute
Icon
Description
MV at edge to process real-time visual inventory and customer behaviour data
Title
Cloud Compute & Storage
Icon
Description
All historical and real-time data stored in Enterprise owned storage
Title
Applications & Services
Icon
Description
AI: (MV) monitor current stock levels; (ML) to predict stock demand (based historical, weather etc. data) for re-ordering; and to optimize inventory Planogram [1] recommendations
Title
Inform Decision Makers
Icon
Description
Should be an autonomous process; Normally no need to ask decision makers
Real-time inventory Planogram change instructions can be sent directly to in-store staff smartphone, tablet, or smart glasses
Title
Support Decision Making
Icon
Description
End of process
Application Logic
Description
Smart Shelf Sensors (e.g., Weight Sensors and RFID antennas) to automatically scan and determine store inventory levels Fixed and mobile (e.g., in-store staff smart glasses, autonomous robots) cameras to monitor store inventory levels
Data collected will be used for stock management applications and inputs for AI
Relevant data will be stored either on-site or in a central enterprise storage from multiple stores
Description
Model will be stored and maintained by AI application
Consideration for a separate system (e.g., other sensors) for goods that cannot be tracked in the same manner.
Model will predict over-time, the optimized stock level forecasts for all SKUs to prevent out-of-stock (OOS) issues and to increase inventory turnover, based on customer purchase behaviour, seasonality, weather etc.
Description
Data collection, the execution of models and addressing the outcome will be fully automated
AI (MV) models will be improved over time and will be fully relied on in the future
SME and Data Scientists will collaborate in developing and deploying MV models and other AI solutions
Expected benefits
Ensure availability of goods to sell
Improved consistency in quality and freshness of goods through timely placement of orders and demand
Effectively Planogram management to increase inventory turnover and maximize sales