Case Studies

About the Client

A grocery store or big mart chain seeking to optimize shelf management and improve customer experience.

The client faced two primary issues that were causing revenue loss:

Product out of stock: Products were unavailable on shelves despite being in store inventory. The manual process of checking stock was labor-intensive and time-consuming.
Product misplacement: Disarranged or misplaced products negatively impact the customer shopping experience, particularly in high-end fashion outlets where presentation is crucial.

Improved real-time stock monitoring, reducing out-of-stock instances by 15%

Enhanced product placement accuracy, increasing customer satisfaction

Reduced manual labor costs for inventory checks

JashDS developed a deep learning-based model to detect out-of-stock or misplaced products in real-time:

Implemented continuous video stream capture using existing CCTV infrastructure
Created a custom deep learning model to analyze the live video feed
Developed a system to predict and identify out-of-stock and misplaced items in real-time
Integrated automated alerts (SMS/email) to notify relevant personnel for immediate resolution
Utilized state-of-the-art learning-based models and custom datasets from local marts for training
Employed gradual resizing techniques during training to achieve higher accuracy and better model generalization