Client Introduction & Business Requirement:
Our client is a leading Indian FMCG brand with presence across millions of retail outlets across India.
They wanted to improve the process efficiency and quality of their premium range of cookies. In their assembly line, the quality supervisors would sample x number of cookies every 1 hour and discard the batch in case of quality defects in order to maintain consistency. The observations are paper-based and are not available online or sent to the Head Office for audit. Some of the quality checks include:
Check for dimensions: Increase in dimensions by more than a few mms will lead to the packaging equipment rejecting the cookies.
Check for color: Color of each brand of cookie is currently compared with a standard sample and there are acceptable thresholds for the cookie being ‘Pale’ or ‘Dark’. Both these cases are candidates for rejection.
Check for foreign objects: On many occasions, there are small particles from the manufacturing equipment or packaging material that get embedded in the cookie. These have to be removed from the batch.
The short-term requirement was to detect all these defects using Computer Vision in order to reduce manual effort, and maintain quality standards on the basis of objective manufacturing parameters. Their long-term requirement is for the results from the Computer Vision System to be fed to their Edge devices close to real time, so that the operational parameters can be regulated for quality control.
- The passage of each cookie through the assembly line from the oven till the boxing stage is captured via Computer Vision, and analyzed against the optimal diameter, stack height, color gradient and also for the presence of foreign objects.
- The observations are tagged for context and stored in a centralized data store.
- If any deviations from the standard are observed, a real-time alert is sent in order to effect immediate change in the manufacturing process.
- In this manner, the quality check is done in real-time, in a controlled environment, based on specific parameters, thus eliminating subjective decision-making.
- Periodical quality-check was replaced by real-time quality check & improved process efficiency.
- Real-time capture of quality observations leading to a database that has trends of the cookie production output for each batch. This dataset opens up opportunities for predictive, prescriptive analytics use cases.
- Dynamic change in manufacturing parameters helped improve quality standards by 40%.
At Knowledge Lens, we constantly work towards improving our product technologies, so your business can do more for you. Visit us here to learn how you can grow your business operations through our Industry 4.0 solution today.