Hi All,


We are in the process of building a customer risk rating model on existng
data warehouse. Customer's profile data is stored in DimCustomer dimension
table.Customer products (e.g. saving account,credit card,loan..) are stored
in DimCustomerProducts Table.



Now we are to calculate customer Risk (=profile risk+product risk) on the
basis of his demographic profile attributes (e.g. Occupation,Education..) and
products.



Following score matrices are given to us to calculate the customer risk for
each customer in warehouse.



Profile Risk Factors

-------------------------------------------------------------------------------------------------

(1).Occupation Risk Scores

Occupation Risk Score
Self Employed 60
Salaried 10
Unemployed 30




(2).Education Risk Scores

Education Risk Score
Post Graduate 50
Graduate 40
Hsc 10




Product Risk Factors

---------------------------------------------------------------------------------------------------

Products Risk Score
Saving A/c 10
Loan 20
Insurance Policy 30
Credit Card 40






If a customer's occupation status is Salaried,education level is graduate
and he have a saving a/c and credit card then he should be assigned with risk
score (10+40+10+40=100). To keep it simple right now this i have given equal
weightage to each , actually there is 60% weight is given to profile
score,40% to product.



Should we build dimension tables for each type of profile risk to store it's
score like DimOccupation,DimEducation,DimProducts.



Would appreciate if could give us some ideas to achieve it efficiently.



Thanks.

Harry