What is Linear Programming?

What is Linear Programming?
• It is referred as “programming in a linear structure” (Dantzig, 1998)
• It involves planning of activities to obtain the best or optimal solution to a problem that requires a decision or
set of decisions on how best to use a set of limited resources to achieve a state goal of objectives (Hillier &
Lieberman, 2015)
Linear Programming Problems
Steps in Formulating a Linear Programming Problem
• Determine and define the decision variables and parameters
o What do I need to know in order to solve the problem?
• Formulate an objective that we need to optimize
o What does the owner of the problem want?
o How the objective is related to his decision variables?
o Is it a maximization or minimization problem?
• Formulate each constraint that the solution must satisfy
o What resources are in short supply and/or what requirements must be met?
o What are the connections among variables?
• Use decision variables to write mathematical expressions for the objective function and the constraint
o Should I use inequality or equality type of constraint?

Step 1: Formulate the LP (Linear programming) problem

We have already understood the mathematical formulation of an LP problem in a previous section. Note that this is the most crucial step as all the subsequent steps depend on our analysis here.

Step 2: Construct a graph and plot the constraint lines

The graph must be constructed in ‘n’ dimensions, where ‘n’ is the number of decision variables. This should give you an idea about the complexity of this step if the number of decision variables increases.

One must know that one cannot imagine more than 3-dimensions anyway! The constraint lines can be constructed by joining the horizontal and vertical intercepts found from each constraint equation.

Step 3: Determine the valid side of each constraint line

This is used to determine the domain of the available space, which can result in a feasible solution. How to check? A simple method is to put the coordinates of the origin (0,0) in the problem and determine whether the objective function takes on a physical solution or not. If yes, then the side of the constraint lines on which the origin lies is the valid side. Otherwise the opposite one.

Step 4: Identify the feasible solution region

The feasible solution region on the graph is the one which is satisfied by all the constraints. It could be viewed as the intersection of the valid regions of each constraint line as well. Choosing any point in this area would result in a valid solution for our objective function.

Step 5: Plot the objective function on the graph

It will clearly be a straight line since we are dealing with linear equations here. One must be sure to draw it differently from the constraint lines to avoid confusion. Choose the constant value in the equation of the objective function randomly, just to make it clearly distinguishable.

Step 6: Find the optimum point

Optimum Points

An optimum point always lies on one of the corners of the feasible region. How to find it? Place a ruler on the graph sheet, parallel to the objective function. Be sure to keep the orientation of this ruler fixed in space. We only need the direction of the straight line of the objective function. Now begin from the far corner of the graph and tend to slide it towards the origin.

• If the goal is to minimize the objective function, find the point of contact of the ruler with the feasible region, which is the closest to the origin. This is the optimum point for minimizing the function.
• If the goal is to maximize the objective function, find the point of contact of the ruler with the feasible region, which is the farthest from the origin. This is the optimum point for maximizing the function.

Step 7: Calculate the coordinates of the optimum point.

This is the last step of the process. Once you locate the optimum point, you’ll need to find its coordinates. This can be done by drawing two perpendicular lines from the point onto the coordinate axes and noting down the coordinates.

Otherwise, you may proceed algebraically also if the optimum point is at the intersection of two constraint lines and find it by solving a set of simultaneous linear equations. The Optimum Point gives you the values of the decision variables necessary to optimize the objective function.