Powered by Crisp Analytics

DATABASE ASSIGNMENT
Problem Statement 1

I created an insurance company data model using dummy data.

Policy Data Table :-

This data includes a unique customer ID, a policy number, the type of insurance policy, the coverage
amount, the premium amount, and the start and end dates of the policy. This data can be used for various
purposes such as testing and development, training and analysis, or as a template for generating more
complex data.

CONFIDENTIAL 2
Primary Key - Customer ID

Watermark Column – None

Customer Data Table :-

Primary Key – Customer ID

Payment Data Table :-

Primary Key – Payment ID

Claim Data Table :-

CONFIDENTIAL 3
Primary Key – Claim ID

Data Model Design:

CONFIDENTIAL 4
  Customer: Contains customer data such as name, address, phone number, email, and any other
relevant information.

 Policy: Contains policy data such as policy number, policy type, coverage amount, premium
amount, and start/end dates.

 Claim: Contains information about customer claims, including claim number, policy number, date
of claim, description of claim, and status of the claim.

 Payment: Contains information about customer payments, including payment ID, policy number,
payment amount, payment date, and payment method.

Data Rules:

 Each customer can have multiple policies, but each policy can only be associated with one
customer.
 A policy can have multiple claims associated with it, but each claim can only be associated with one
policy.
 A policy can have multiple payments associated with it, but each payment can only be associated
with one policy.
 Claims can only be filed for policies that are currently active.
 Payments can only be made for policies that are currently active and have outstanding premium
amounts.
 The coverage amount and premium amount of a policy must be within acceptable limits and
calculated based on the policy type and other factors.
 The start and end dates of a policy must be valid and not overlap with any other policies for the
same customer.
 Claims and payments must be processed and recorded accurately and in a timely manner.
 These data rules are just a few examples of what could be included, and they can be expanded or
modified depending on the specific needs of the insurance company.

Database - MySQL

Relational databases such as MySQL, PostgreSQL, and are widely used in the insurance industry due to
their ability to handle structured data and enforce data integrity through the use of primary keys, foreign
keys, and other constraints. They also offer robust querying capabilities and support for transactions, which
can be important for processing claims and payments accurately and efficiently.

Alternatively, NoSQL databases such as MongoDB or Cassandra may be suitable if the data is less
structured or if the system needs to handle a high volume of data with varying schema. NoSQL databases
can be more flexible and scalable than relational databases, but may sacrifice some of the data integrity and
consistency guarantees that relational databases provide.

Ultimately, the choice of database will depend on the specific needs of the insurance company, such as the
volume and complexity of the data, the performance and scalability requirements, and the budget and
resources available for managing the database. A thorough analysis of these factors should be conducted
before selecting a database solution.

CONFIDENTIAL 5
P ROBLEM S TATEMENT -2

USE-CASE
A database for an airline service now the database would comprise of 6 entities that would help the airlines
track the bookings and transactions and the user details to handle their business needs and make strategies
for profit.

CONFIDENTIAL 6
Entities in the data model are:

1. Aircrafts
2. Airports
3. Flights
4. Users
5. Bookings
6. Transactions

The ER diagram for the same is built as:

Q U E S T I O NS :
There can be locking system implemented which means whenever a person starts booking a seat the seat
gets locked and no other person can book the seat until the person cancels their booking or the transaction is
ended.

Sharding can be done based on countries in which the airlines is operating i.e. the database can be
partitioned based on airport location without changing the schemas of the table and the data can be stored at

CONFIDENTIAL 7
multiple servers as shards of the original data which would help efficiently maintaining the database. Such
that all the domestic flights and airports would be in one server and can be retrieved with that server only
but for international flights multiple server will be used for the booking and transactions

To have an ACID complaint database, the company can use RDBMS which supports ACID transactions.
Examples of such databases are Oracle, MySQL, PostgreSQL etc.

To have a dynamic data company can use in-memory Database (IMDB) in which the data is entirely stored
in memory rather than disk used for fast transactions. Examples are Redis and Memcached. Also, memory-
optimized databases can also be used which offer memory and disk combination storage and are used for
fast data access and durability. Examples are Apache Ignite etc.

To have analytics on the time-based user behavior, the company can use time-series databases which offers
storing and retrieving time-stamped data which in this case would be time-stamped transactions and would
give faster querying and analytics some examples are Prometheus, InfluxDB, Graphite etc.

CONFIDENTIAL 8