MLOps-enabled bottle manufacturing platform

Optimizing operations for beverage bottle manufacturing units with model training workflows, using Machine Learning Operations (MLOps) strategies and AI

the story

Our client is a prominent player in crafting innovative solutions for the liquid food and beverage industry. The client offers an IIoT platform to optimize production through KPI optimization, operators' interaction, and smart, condition-based monitoring. An edge device collects raw data from the machines on the processing belt. This data snapshot is used to train a reinforcement learning model which calculates the most optimized KPIs for a manufacturing unit.

The client wanted to implement a productionized, scalable, and repeatable AI pipeline on AWS by adopting MLOps best practices. This automated AI pipeline was required to ingest the raw data from the edge device and the developed AI solution was to be deployed back to the edge device. Also, the overall cost of infrastructure and ownership for such a shared platform was to be taken into consideration. Nagarro's experts created a custom AI workflow to enable a serverless MLOps implementation, that helped in crafting better business strategies for the platform.

challenge

The changing needs of the market demand upgrading existing systems. The client's existing AI implementation was disconnected from the data ingested from the edge device into the AWS storage bucket; a partial model training pipeline was on the Azure cloud platform and the remaining was on-premise. The client needed to streamline the workflows and perform troubleshooting for edge scenarios. Additionally, a commercial model training platform license had to be attached to the ML solution, which was a manual process. Lastly, the pipeline had to be built as a multi-tenant platform to be used by the client’s customers.

solution

Nagarro partnered with the client to build a custom, MLOps-enabled AI workflow for deployment on AWS. The solution was designed to be serverless, with the computing cost generated on a pay-as-you-go model. Data pre-processing was done on Spark EMR cluster and multiple model training was performed on AWS SageMaker. Currently, deployment on edge device (using SageMaker Neo, Edge Manager and IoT Greengrass) is in progress. In the last phase, data ingestion and deployment will be performed on the cloud.

outcome

Nagarro’s serverless MLOps implementation of the client’s AI model has provided the client full control of the various AI steps in the end-to-end pipeline. The custom-tailored pipeline components are configurable for the solution requirements. The defined multi-tenant application comprising of MLOps and DevOps best practices and infra-as-code schemes have allowed the client to evaluate the total cost of ownership of the platform and thereby, define the business strategies for the solution to be shared with end customers.

Thinking Breakthroughs

Filter by Goal

Trending technologies

Industry

Events and Webinars

Our Event Series

Featured Event

Automate and Accelerate your Risk Management Process

06 Dec

14:00 - 15:00 CET

Our Latest Talk

Innovation powered by world class research

By Kanchan Ray, Dr. Sudipta Seal

60 mins

Fluidic Enterprise

Africa & Asia-Pacific

Central & South America

Europe

Middle East

North America

success story

MLOps-enabled bottle manufacturing platform

the story

challenge

solution

outcome

Interesting? Spread the word

What can we help you achieve?

Stay up to date with insights from Nagarro!

©2023

Fluidic
Enterprise

Stay up to date with insights
from Nagarro!