Personas and responsibilities

Introduction

This manual outlines the operational steps for managing product data. It

assigns specific responsibilities to designated personas and details their

roles in new product introductions, design updates, and data maintenance

1. Product Master Data

1.1. New Product Check

● Product Master Data Persona: Determine if there is a new product.

● If Yes: Import or ingest data via API or manual registration. Backoffice manual Page 11

● If No: Skip to the relevant section based on the product's lifecycle status.

1.2. Data Import and Registration

● Product Master Data Persona: For new products, ensure that all relevant data is

accurately imported into the system.

2. Product Design

2.1. Design Update Check

● Product Design Persona: Verify if there's a new design.

● If Yes: Proceed to update the product image collection.Backoffice manual Page 18

● If No: Continue to the Planogram Maintenance section.

2.2. Image Collection Update

● Product Design Persona: Update the image collection to reflect the new product

design.

3. Planogram Maintenance

3.1. Product and Image Collection Review

● Image Supervisor: Check for new products or image collections.

● If Yes: Update the planogram. Backoffice manual Page 24

● If No: Proceed with Data Maintenance procedures.Backoffice manual Page 18

2

Operations Manual: Product Data Management

3.2. Planogram Update

● Image Supervisor: Incorporate new products or image collections into the

planogram as necessary. Backoffice manual Page 41

4. Data Maintenance

4.1. Annotation Quality Check

● Image Supervisor: Ensure that annotation approvals are accurate, ensuring

high-quality data for machine learning model training and recognition. Backoffice

manual Page 41

4.2. Image Selection for Annotation

● Image Supervisor: Select and annotate images, maintaining a high level of accuracy

in product annotations. Backoffice manual Page 41

4.3. Dashboard Review for Accuracy

● Image Supervisor: Regularly review the dashboard to identify and correct accuracy

issues in product images. Backoffice manual Page 41

4.4. Annotation Diversity

● Image Supervisor: Generate diverse annotations across various images to facilitate

optimal machine learning training. Backoffice manual Page 41

4.5. Enhancing Model Accuracy

● Image Supervisor: Continuously update annotations to maintain and improve the

accuracy of the machine learning models. Backoffice manual Page 41

4.6. Machine Learning Model Training

● Image Supervisor: Train the machine learning system regularly, especially after any

updates or changes, to enhance model accuracy.Backoffice manual Page 31

Conclusion

The successful implementation of this manual relies on the diligent execution of

responsibilities by the Product Master Data Persona, Product Design Persona, and the

Image Supervisor. Consistent adherence to these guidelines will ensure the integrity and

accuracy of our product data and the efficacy of our machine learning models.