Solver

The Solver is used to find an optimal solution to a model with several input values subject to constraints. This exercise is often called linear programming (LP). Linear programming is most often applied in situations where scarce resources are allocated between competing activities. The scarce resources can be cash, people, raw materials, machine time, etc. The scarce resources are allocated in a way that optimizes some objective function; e.g., maximize profit, minimize costs, minimize development time, etc.

DecisionPro solves linear programming problems using the simplex method. The simplex method was developed in 1947 by George B. Dantzig as part of an Air Force research project. Since then, linear programming has become one of the most widely applied methods in management science.

The simplex method requires that the objective and all its constraints are linear with respect to the input variables. If this condition is not met, DecisionPro warns you and aborts the optimization.

The constraints are equations using the comparison operators <=, >= and ==. For example,

Cash<=456000
Unit Sales+Give Aways<=10000

Constraints can be based on any nodes in a model, not just the input nodes. Note that all input values are implicitly constrained to be greater than or equal to zero.

When the calculation is complete, DecisionPro will modify each input node's definition to reflect the solution found.

To find the solution to an optimization problem:

1) In the Tools menu, click Solve.

2) Enter the name of the output node whose value you want to optimize in the Set node field.

3) Select whether you want to maximize or minimize the value of the output node.

4) Enter the name of the input node(s) whose values can be changed to optimize the output node. Separate node names with commas.

5) Enter all constraints in the subject to the constraints field.

6) Click OK.

See Also

Optimization Example