Data Lake for Manufacturing

Client profile

Our client is a pharmaceuticals major into manufacturing in India.

The Problem

Application silos has lead to data fragmentation over the years and there is a huge cost related to  datawarehousing to get a simple reports, either historical or forecasting, for a specific business process. New technologies like machine learning can not be adopted to identify process improvement area and there is a huge data quality issues depending on the data/report consumer.

Goals

Create a consistent 360 degree view of the business data, use machine learning to improve the internal manufacturing processes and provided a single consistent reporting tools/methods and experience.

Kanerika was instrumental in creating data lake of 15+ application data including SAP supply chain, sales, inventory, marketing, finance & HR.   Architected, designed and implemented the entire Hadoop stack with Hortonworks, Hive, Sqoop, Apache wifi data pipes and then land the data in the lake. Supplement this process with right model, data cleansing and use the AngularJS visualization to achieve project goals.

Impact

20X

20x improved data response time

Consistent

Consistent Business Data View and Policy Enforcement Enabler

Self-serviced

Self serviced Reports.

Key Technologies

Hortonworks, HDFS, Hive, NiFi, Squoop, Kafka,  Kibana, AngularJS, Apache Spark

Client Testimonial

 

Never thought getting reports could be so easy. No longer waiting for getting the right insights. Big time saver!