In this case study learn how BRIDGEi2i helped a Fortune 500 Technology company to build a predictive algorithm for product failures in the market and to develop an inventory policy for Spares based on expected ASER.
1. A Case Study in
Predictive
Maintenance
A Fortune 500 Technology Company
Quick Context
Objective
a. Over 400 Printer SKUs with highly complex functionalities & components
b. Increasing trend in ASER** and long Warranty turn-around-times
• 70% accuracy in
predicting
warranty return
volumes at Product
Family level
• 15% lower TAT on
Warranty Events
Impact
• BRIDGEi2i specializes in several Line
Operations analytics with an
appreciation for the fact that the fixes
must be very well evaluated before
being operationalized
Key Success Elements
Our Approach
6 Months
2 Years
Client
Project length
Length of relationship with client
• Data was securely handled within Client
environment
• Warranty Management System was
accessed for actual product failure data
• Line Operation System data was
accessed via RFID database that
collects Sensor data every 100 seconds
in the manufacturing line
• SKU attributes like functionalities and
size was acquired from Engineering
• Complexity Index for each SKU based
on product design features
• Merge the installed base data with
WMS data to determine SKU level
failure rates (ASER)
• Accelerated Life Time Models to
determine failure probabilities for a new
product
• relationship between Line Audit % and
the modeled failure rates
• % of pallet to be audited in the line is
derived as a function of parameters that
determine failure rates
• Algorithm is plugged into the
manufacturing partners’ line
management system to provide
monthly Line Audit % numbers by SKU
• Solution was tested for Laser printers
and monthly reports showed marked
drop in ASER
Data Management Algorithmic Play Operationalization
a. To build a predictive algorithm for product failures in the market
b. To develop an inventory policy for Spares based on expected ASER
* ASER – Annualized Service Event Rate