Global Beverage Company Intelligent Supply Chain

Background

A Global Beverage Company with over 29,000 stores spread across the globe is in a continuous innovation drive to enhance customer experiences. Implementing an augmented intelligence platform to transform supply chain planning is one such initiative. The goal was to optimize the delivery of products to various stores across the globe and balance supply and demand at a store level granularity. This was to be achieved through insights gained from millions of point of sale transactions.

Challenge

Data preparation is an important and critical step in data driven decision automation. Data from multiple operational systems must be ingested, cleansed and transformed into a standardized data model. The result is consumed by AI algorithms to produce actionable insights. Source systems include:

• ERP systems for order management, in transit, on order, receipts.
• Point of Sale systems for sales transactions.
• Supply chain network
• Delivery schedules
• Inventory management systems for on-hand quantities.

This data includes a vast amount of semi-structured data in the form of JSON documents. The volume, variety and velocity of data from these systems made it imperative to implement a greenfield big data processing platform to enable the flow of transactions, orders, master data into an advanced analytical platform to perform sales, plan and consumption forecasts involving 100s of millions of transactions at scale in the cloud.

Solution

pSolv partnered with a leading Augmented Intelligence Platform vendor to architect a scalable data lake platform and implemented the solution to ingest and process data from multiple on-prem data sources including batch and realtime on Azure. The platform enables orchestration of data ingestion, transformation and delivery jobs in a scalable and cost-effective manner.

Outcome

• Solution Architecture integrating state of the art technologies supported with best practices
• Scalable, automated, and in-time processing of huge volumes of data
• Data pipelines with built-in enforcement of merge contract to guarantee data integrity and quality
• Optimized use of cloud resources – compute, storage and network