Linear Programming in Excel

In Solver software, the objective cell represents the constraints in the model.

When organizing data in a spreadsheet, it is recommended to embed numeric constants in formulas.

When implementing an LP model in Excel, the decision variables should be placed in separate worksheets.

In Solver software, the constraint cells represent the objective function in the model.

When formulating an LP model, the coefficients for the objective function and constraints should be entered in separate spreadsheets.

When designing a spreadsheet, it is recommended to arrange related data and formulas in a scattered manner throughout the spreadsheet.

Relaxation is an approach to finding the optimal integer solution in which you enforce, or strictly follow, the integrality conditions and solve the problem as if it were a standard ILP problem.

For minimization problems, the objective function value at the optimal solution to the LP relaxation represents an upper bound on the optimal objective function value of the original ILP problem.

If we relax all the integrality conditions in an ILP with a minimization objective and solve the resulting LP problem, the objective function value at the optimal solution to the relaxed problem provides a lower bound on the optimal integer solution.

In an ILP problem, the optimal integer solution is always worse than the optimal solution to its LP relaxation.

Many ILP packages allow you to specify a suboptimality tolerance of 10% or more, which tells the B&B algorithm to stop when it finds an integer solution that is exactly 10% worse than the optimal integer solution.

Continuous Variables are variables that are required to assume strictly integer values.

A redundant constraint is a constraint that determines the optimal solution of an LP problem.

Level curves represent the boundary lines of the constraints in an LP problem.

The Solver software is used to solve LP problems in a spreadsheet.

The coefficients of the objective function are organized in separate cells in the spreadsheet when implementing an LP model.

An unbounded solution in an LP problem implies that the problem has no optimal solution.

CPLEX is a type of LP problem, not a software package.