A method of identifying a best quote from a pool of RFP responses for implementing a business scenario, the method to be performed by a computer system including a relational database and user interfaces, the method comprising the steps of:
- gathering data from the RFP responses and existing circuitry, including a plurality of supplier identities s, identities of access circuits n, technologies t and discount levels d associated with each said supplier identity s;
- specifying a total cost function C of the business scenario as
- C
- =
- ∑
- n
- ,
- s
- ,
- i
-
-
- x
-
-
- (
- n
- ,
- s
- ,
- t
- )
- ·
- (
- ∑
- d
-
-
- (
- mrc
-
-
- (
- n
- ,
- s
- ,
- t
- ,
- d
- )
- ·
- m
- +
- otc
-
-
- (
- n
- ,
- s
- ,
- t
- ,
- d
- )
- )
- ·
- y
-
-
- (
- s
- ,
- d
- )
- +
- (
- 1
- -
- cursp
-
-
- (
- n
- ,
- s
- )
- )
- ·
- (
- mig
-
-
- (
- n
- ,
- s
- ,
- t
- )
- +
- incent
-
-
- (
- n
- )
- )
- )
- +
- ∑
- n
-
-
- (
- 1
- -
- ∑
- n
- ,
- t
-
-
- opt
-
-
- (
- n
- ,
- t
- )
- ·
- x
-
-
- (
- n
- ,
- s
- ,
- t
- )
- )
- ·
- c
-
-
- (
- n
- )
- ·
- m
- wherein x(n,s,t) is a binary variable indicating whether a bid by supplier s with technology t is the optimal solution for circuit n;
- mrc(n,s,t,d) is a bid monthly price for circuit n by supplier s using technology t at discount level d over m months;
- otc(n,s,t,d) is a one time charge for bid pricing for circuit n by supplier s using technology t at discount level d;
- y(s,d) is a binary variable indicating whether discount level d is offered by supplier s;
- cursp(n,s) is a binary variable indicating whether a current supplier for circuit n is supplier s;
- mig(n,s,t) is a one time migration cost incurred for moving a circuit n to supplier s with technology t;
- incent(n) is a one time customer incentive for circuit n;
- opt(n,t) is a technology feasibility indicator for technology t on circuit n and c(n) is a current monthly cost of access circuit n;
- specifying bid constraints as mathematical constraint rules that are functions of the gathered data;
- linearizing by the computer system the total cost function C by assigning a possible value to the binary variable y(s,d) indicating discount level y for supplier s;
- minimizing by the computer system the linearized cost function to find a lowest total cost supplier s at the possible value of the binary variable y(s,d), using a linear programming solver subject to the mathematical constraint rules;
- repeating by the computer system the linearizing and minimizing steps using additional values of the binary variable y(s,d), until the total cost C is minimized;
- and
- selecting the best quote based on the minimized total cost C.