2 Use Case
Dishant Munjal edited this page 2019-07-31 14:02:02 +05:30
This file contains ambiguous Unicode characters!

This file contains ambiguous Unicode characters that may be confused with others in your current locale. If your use case is intentional and legitimate, you can safely ignore this warning. Use the Escape button to highlight these characters.

Use Case

The use case for demo purposes used here is explained as follows. The system we are designing is an e-commerce discount calculation system.

Rules

The rules for the discount calculation are

  1. Give the user a discount of 10% over MRP if the following conditions are followed
    • The users country is India.
    • The users loyalty factor is less than or equal to 2.
    • All the orders purchased by the user so far should amount to more than 5,000.
    • User should have at least made more than two successful orders.
    • The user should have visited the site more than two times every month.
  2. Give the user a discount of 20% over MRP if the following conditions are followed
    • The users country is India.
    • The users loyalty factor is equal to 3.
    • All the orders purchased by the user so far should amount to more than 10,000.
    • User should have at least made more than two successful orders.
    • The user should have visited the site more than two times every month.
  3. Give the user a discount of 25% over MRP if the following conditions are followed
    • The users country is not India.
    • The users loyalty factor is greater than or equal to 2.
    • All the orders purchased by the user so far should amount to more than 10,000.
    • User should have at least made more than two successful orders.
    • The user should have visited the site more than five times every month.
  4. Give the user a discount of 30% over MRP if the following conditions are followed
    • The users loyalty factor is greater than 3.
    • All the orders purchased by the user so far should amount to more than 50,000 but less than 100,000.
    • User should have at least made more than five successful orders.
    • The user should have visited the site more than fifteen times every month.
  5. Give the user a discount of 30% over MRP if the following conditions are followed
    • The users loyalty factor is greater than 3.
    • All the orders purchased by the user so far should amount to more than 100,000.
    • User should have at least made more than fifteen successful orders.
    • The user should have visited the site more than 25 times every month.
  6. Give 0% discount in any other case.

Inputs

Here the inputs will be of three different types as they are coming from three different data sources/APIs.

User Basic Info

This input has information like

  • Name
  • Country
  • Email
  • Credit history
  • Loyalty factor
  • Sum of the purchases made by the user till date.

Users Order Information

This input is a summarization of the orders made by the user so far. This input has information like

  • Total number of orders
  • Recurring items in those orders if any

Users Telemetry Information

This input is a summarization of the telemetry information collected based on the users visit to the site. This input has information like

  • Number of visits to the site per month
  • Percentage of the number of times the user purchased something to the number of times the user visited