Key Challenges

Initially, each food package design required time-consuming manual inspection to meet FSSAI standards, adding a significant burden to the process. The specificity and additional information required by FSSAI rules further complicated manual verification. Incorrect designs had to be corrected and resubmitted, resulting in repeated human verification and extended approval timelines. Additionally, the existing manual process was not scalable, making it difficult to efficiently handle a large volume of packaging designs.

Key Results

The verification process for food packaging designs was drastically reduced from days or weeks to a maximum of 120 seconds. Automated checks significantly enhanced accuracy by ensuring all FSSAI regulatory labels were correctly identified, with any discrepancies or missing labels highlighted. This led to a substantial improvement in operational efficiency, faster product launches, and reduced costs. The seamless integration and deployment on Google Cloud Functions provided Bizongo with a scalable and reliable automated verification system.

Overview

Bizongo, a leading player in the packaging industry, faced significant challenges in ensuring that each food package design met the stringent Food Safety and Standards Authority of India (FSSAI) regulations. The manual verification process was time-consuming and prone to errors, often taking days or weeks for a single design approval. This inefficiency not only delayed product launches but also increased operational costs.

Challenges

  • Manual Inspection: Each design had to be manually checked for compliance with FSSAI standards, which was labor-intensive and time-consuming.
  • Complex Regulations: FSSAI standards are highly specific, and certain food products require additional information, further complicating the manual review process.
  • Delays in Approval: Incorrect designs needed to be corrected and resubmitted, leading to additional delays as the human verification process repeated.

Solution

  1. GCP Project Setup: We helped Bizongo set up a Google Cloud Platform (GCP) project to streamline their processes.
  2. Annotation of Designs: Food packaging designs were annotated to facilitate automated checks.
  3. AutoML Object Detection: We trained an AutoML Object Detection model to identify regulatory labels on the packaging and return bounding boxes for further verification.
  4. Automated Script Development: A script was developed to perform checks against FSSAI norms. This script was deployed using the Flask framework on Google Cloud Functions, enabling efficient, automated verification.

Business Outcome

  • Accelerated Approval Process: The automated system significantly reduced the time required to check each food packaging design to a maximum of 120 seconds, compared to the days or weeks needed for manual inspection.
  • Improved Accuracy: The system ensured that all regulatory labels were present and correct, highlighting any missing labels or necessary changes in the design.
  • Operational Efficiency: Bizongo experienced a substantial increase in operational efficiency, reducing costs and speeding up product launches by minimizing delays in the approval process.
Share this post

Related Case Studies

Migration from AWS to GCP for an Ed Tech

GCP, Cloud Migration, AI/ML
Read Case Study

Migration a Saas platform from On-Prem to GCP

GCP, Cloud, Saas
Read Case Study

Bitech AG DevOps Migration from on-prem to AWS for German ISV

AWS, DevOps, SaaS
Read Case Study

The Ankercloud Team loves to listen