Insurance Company Case Study - Hadoop /Bigdatasolution:
We have this basic use case to implement using Bigdata solutions:
An insurance company with regional offices in each state is in business of Home and Auto Insurance.
**Each regional office send the details of monthly transaction ( enrollments , updates to policy , cancellation) to Central System
**The feed is form of Raw txt file converted by Mainframe system and sent to central processing server (Linux ) using FTP.
** The data is in FIXED line format .
1 single line contains :
- Transaction ID,
- SSN ,
-InsuranceID ,
-RegionalOfficeID,
-Insurance Type
- Insurance Duration
-Amount
-Notes
-Timestamp
Business Need:
The Company wants to launch a new product for Home and Auto insurance users and the management would like to give some real time facts based on the data they have like:
- No of user using the insurance national wide
- Pace with which the users are joining
- Which type of insurance package ( Home , Auto , Dual ) is more lucrative to customers as per their location geographies.
.
Some Hard Numbers:
No of users: 20 Million
Data File feed size: 50 MB / Cycle
No of Data cycles: 2 / week.
Time expected for output: Monthly Basis
Please analyse this business problem and provide your inputs for design and technology choice
Please join this forum and add your inputs. If you need more clarifications , please let me know the same. Wherever possible make suitable assumptions based on your exp and the type of work you have done.
I am looking for multiple options to design this solution.
Linkedin group:
