ADEEB EIT

Operations Research - BMIS355

LIU Spring 2016-2017

SIMPLEX METHOD
Simplex Method is an iterative procedure that allows improving the solution at each step. This procedure is
finished when is not possible to improve the solution.

Starting from random vertex value of the objective function, Simplex Method try to find repeatedly another
vertex value that improve the one you have before. The search is done trough the side of the polygon (or the edges
of the polyhedron, if the number of variables is higher). As the number of vertex (and of edges) is finite, it will
always be able to find the result.

Simplex Method is based on the following property: If objective function, f, does not take the max value in the
A vertex, then there is a edge starting at A, along which the value of the function grows.

You should take care about Simplex Method only works with "≤" type inequality and independent coefficients
higher or equal to zero, and you will have to standardize the restrictions for the algorithm. Case after this procedure
"≥" o "=" type restrictions appears (or not modified) you should try other ways, being Two-Phase Simplex Method
the best choice.

PREPARING THE MODEL TO ADAPT IT TO THE SIMPLEX METHOD

This is the standard way of the model:

Objective function: c1·x1 + c2·x2 + ... + cn·xn

Subject to: a11·x1 + a12·x2 + ... + a1n·xn = b1
a21·x1 + a22·x2 + ... + a2n·xn = b2
...
am1·x1 + am2·x2 + ... + amn·xn = bm
x1,..., xn ≥ 0

To do this you must follow these rules:

1. The objective must be maximize or minimize the function.

2. All restrictions must be equal.
3. All variables are not negatives.
4. The independent terms are not negatives.

1
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

Changing the optimization type.

If we want to minimize our model, we can keep it, but we must consider the new criteria for the halt condition
(stop iterations when all coefficients in the value objective function row are less or equal to zero), and the leaving
condition row. In order to not change criteria, we can convert the minimize objective function F to maximize
objective F·(-1).

Advantages: We will not have to worry about halting criteria, or exit condition of rows, since they keep on.

Inconveniences: In the event of the function have his entire basic variables positive, and further the restrictions
are inequality "≤", they become negatives when doing the change and plus signs remain in the row of the value of
the objective function, then Simplex Method obeys the halting condition, and the optimal value that would be
obtained is 0, by default.

Solution: In fact, this kind of problem does not exist, since so that the solution is greater than zero, any
restriction should have the condition "≥", and then we would go into a model for the Two-Phase Simplex Method.

Converting the independent term sign (constants to the right of restrictions)

We will have to arrange our model so that the independent terms of restrictions will be greater or equal to 0, if
not, Simplex Method can not be used. The only thing that would be necessary to do is multiply by "-1" the
restrictions where independent terms be less than 0.

Advantages: With this simple modification of signs in restriction, we can use Simplex Method.

Inconveniences: It can work out in restrictions where we have to modify the signs of constants, the signs of
inequalities be ("=", "≤"), becoming ("=","≥") what in any event we will have to develop the Two-Phase Simplex
Method. This inconvenience is not controllable, although it would be able to benefit us only if terms of inequality
exist ("≤","≥"), and terms "≥" coincide with restrictions where the independent term is negative.

All restrictions become of equality.

If an inequality of the type, "≥” appears in our model, will we have to add a new variable, called surplus
variable si, with restriction si ≥ 0. The new variable appears with coefficient equal to zero in the objective function,
and subtracting in inequalities.

A problem appears to us, let us see how to solve inequalities that contains an inequality type "≥”:

a11·x1 + a12·x2 ≥ b1 a11·x1 + a12·x2 - 1·xs = b1

As all our models based on that all his variables are greater or equal than zero, when we do the first iteration in
the Simplex's model, the basic variables will not be in the base and they will take value zero, and all others will
maintain their values. In this case, our variable xs, after doing zero to x1 and x2, will take the value - b1. The
condition of not negativeness will not come true, so it will be necessary to add a new variable, xr, which will appear
in the objective function with zero coefficients, and adding in the inequality of correspondent restriction. Would be
left of the following way:

2
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

a11·x1 + a12·x2 ≥ b1 a11·x1 + a12·x2 - 1·xs + 1 ·xr = b1

This type of variables are called artificial variables, and they will appear when there are inequalities with
inequality ("=","≥"). This will take us compulsorily to accomplish the Two-Phase Simplex Method that will explain
later on.

Itself mode, if inequality has "≤" type, we will have to add a new variable, called slack variable s i, with
restriction si "≥" 0. The new variable appears with zero coefficients in the objective function, and adding up in the
inequalities.

To sum up we can let this board, according to the inequality that appears, and with the value that the new
variables must be with.

Type of inequality Type of variable

≥ - surplus + artificial
= + artificial
≤ + slack

3
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

STARTING THE SIMPLEX METHOD

Once we have standardized our model, it can happen to go into the Simplex Method or Two-Phase Simplex
Method. See yourself in the figure, as we must perform on for reach the solution of our problem.

We will explain each method systematically, concretizing the aspects that are necessary to take in account.

4
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

Simplex Method
- First Board Construction: In the board's first column will appear that we will call base. In the second one,
the coefficient that each variable that appears at base has in the objective function (we will call this column C b). In
the third column, the independent term of every restriction (P0), and from this column will appear each variable of
the objective function (Pi). In order to have a more obvious vision of the board, we will include a row that we will
put each one of the names of columns in. On this board we have, we will include two new rows: One that will lead
the board, where the constants of the coefficients of the objective function will appear, and another one that will be
the last row, where the objective function will take value. Our final board will have such rows as restrictions.

Board
C1 C2 ... Cn
Base Cb P0 P1 P2 ... Pn
Pi1 Ci1 bi1 a11 a12 ... a1n
Pi2 Ci2 bi2 a21 a22 ... a2n
... ... ... ... ... ... ...
Pim Cim bim am1 am2 ... amn
Z Z0 Z1-C1 Z2-C2 ... Zn-Cn

Z row's values are obtained this way: Z0 value will be the result of substituting Cim in the objective function
(zero else appears in the base). The left columns are obtained subtracting to this value the one belonging to the
coefficient that appears in the board's front row.

It will be observed when realizing Simplex Method, that slack variables will be in the base, in this first table.

- Halt Condition: Will check if we must do a new iteration or not to do, that it will be know if in Z row
appears any negative value. If this is not the case, it means we have reached the problem's optimal solution.

- Choosing incoming variable: If halt condition has not come true, we must choose one variable to enter the
base in the next board. For it we look for strictly negative values of the Z row, and the minor will be which give us
the incoming variable.

- Choosing the variable that slips out: Once we have obtained the incoming variable, the coming out variable
will be reached, having nothing else to do that select that row whose quotient P0/Pj be the lowest among strictly
positives (considering that only will be done when Pj be greater than 0). The intersection between incoming column
and the coming out row will determine us the pivot element.

- Updating the board: The correspondent rows to the objective function and titles will remain unaltered in the
new board. Left rows will be calculated with two ways:

 If we are trying with pivot row, each element will result from:
New Pivot Row Element = Actually Pivot Row Element / Pivot.

5
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

 The left elements of rows will be reached so:

New Row Element = Actually Pivot Row Element - (Pivot Column Element from actually Row * New
Row Element).

6
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

Two-Phase Simplex Method

This method differs from Simplex Method that first it is necessary to accomplish an auxiliary problem that has
to minimize the sum of artificial variables. Once this first problem is resolved and reorganizing the final board, we
start with the second phase that consists in making a normal Simplex.

1st Phase

At this first phase, all can be done like in Simplex Method, except the first board's construction, halt condition
and preparing the board that will be used in the second phase.

- First Board Construction: We proceed in same way as Simplex Method, but with some differences.
Objective Function row is different in the first phase, because the objective function changes, due to it will appear
every term with zero value, but them which are artificial variables, which have "-1" value because we are
minimizing the sum of this variables (remember that minimize F is the same that maximize F·(-1)).

The other difference for this first board consists in the way of calculating the row Z. It will have to be
calculated the following way: The Cb·Pj products will be added for all rows and to this sum, we must subtract the
value that appears (according to the column that we are doing) in the objective function row.

Board
C0 C1 C2 ... Cn-k ... Cn
Base Cb P0 P1 P2 ... Pn-k ... Pn
Pi1 Ci1 bi1 a11 a12 ... a1n-k ... a1n
Pi2 Ci2 bi2 a21 a22 ... a2n-k ... a2n
... ... ... ... ... ... ... ... ...
Pim Cim bim am1 am2 ... amn-k ... amn
Z Z0 Z1 Z2 ... Zn-k ... Zn

Being Zj = Σ(Cb·Pj) - Cj and Cj = 0 for all j among 0 and n-k (decision, slacks and surplus variables), and Cj = -1
for all j among n-k and n (artificial variables).

- Halt condition: The halt condition is the same that in Simplex Method. The difference resides in that can
occur two cases when halt condition is reached: the function takes zero value; it means that the original problem has
solution, or function takes a different value, suggesting that our model does not have solution.

- Erasing artificial variables columns: If we have reached the conclusion that the original problem has
solution, we must prepare our board for the second phase. The artificial variables columns will be erased, modify the
objective function row instead original, and calculate Z row at same way that in the 1st phase's first board.

ANOMALOUS CASES AND SOLUTIONS

Obtaining the solution: When halt condition is reached, we can see the basic variables' values that are in the
base, and the optimal value that take the function looking at P 0 column. At case we are minimizing, the optimal
value must be multiplied by "-1".
7
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

Infinite Solutions: Once halt condition is obeyed, if you notice that any variable that does not appear in the
base, has a 0 value at row Z, it means that exists another solution that give us the same optimal value for the
objective function. If we are in face with this case, we have a problem that admits infinite solutions, all them among
the segment (or plane portion, or space region, depending on the number of variables) that defines Ax+By=Z0. If
you wish, you can do another iteration using as incoming variable any of the variables in Z row which have zero
value, and you will have another solution.

Unbounded solution: If we notice that every variable in the incoming variable column has all its elements
negative or void when we are searching the outgoing variable, we are face to face with a problem which have
unbounded solution. There is no optimal concrete value, right now growing the values of variables, the objective
function value also grows, and does not violate any restriction.

Solution does not exist: In case that there seems to be no solution, we will have to solve it using Two-Phase
Simplex Method, so at the end of the 1st phase we will know if we are in such situation.

Tie of incoming variable: You can choose anyone of them, unless it affect the final solution, the
inconvenience that it presents is that according to this choose you will have to do more or less iterations. Is
counseled choosing to favor of the basic variables, since are those that will stay on the base at the end of the method.

Tie of coming out variable: Again, you can choose anyone of them, although it can occur degenerated, case
and entering into loop cycles. In order to avoid them as far as possible, we will have prejudice in favor of basic
variables doing that they remain in the base. At the case to be in the first phase (of the Two-Phase Simplex Method),
we will choose in case of tie to take out the artificial variables.

Curiosity 1st Phase: When first phase finalize, if the original problem has solution, all the artificial variables,
in the Z row, must have the value "1".

Can the pivot be 0? It cannot be zero, because, quotients must be greater than zero.

8
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

Graphical Interpretation of the Simplex Method

The resolution of linear problems which contains two or three decision variables can be illustrated graphically,
pointing like a visual aid to understand a lot of concepts and terms that are employed and formalized with more
sophisticated methods, for example The Simplex Method, needed for resolving problems with several variables.

In reality, although rarely problems have only two variables of decision, but this solution and interpretation
methodology is very useful, because typical solutions will be showed, like the existence of only one optimal
solution, alternatives optimal solutions, no solution, and unbounded solution. We describe the phases of the Graphic
Method procedure. These are:

1. Draw a coordinate system Cartesian, in which each decision variable be represented by an axis, with
the measure scale, that fits properly to his associated variable.

2. Draw on the coordinate system the restrictions of the problem (including no-negativeness). For it, we
observe that if a restriction is an inequality, circumscribe a region that will be the semi-diagram limited by
the straight line that must be considerate when the restriction regard as equality. If restriction is an equation,
the definitional region sketches like a straight line. The intersection of all these regions determines the
feasible region or space of solutions (that is a convex set). If this region is not empty, go for the following
phase. Alternatively, solution does not exist that can satisfy (simultaneously) all the restrictions and the
problem does not have solution, calling it infeasible.

3. Determine the extreme points (points that are not located in in-line segments that other ones join the
convex set's two points) of the feasible region (that, as we will try at the following section, are candidates to
optimal solution). Evaluate the objective function in these points and that one or those that they maximize
(or minimize) the objective, they correspond to optimal solutions of the problem.

In order to understand easily the application of this method, will show an example of Graphical Method.

9
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

EXAMPLE (Part 1): Simplex method

Resolve using the Simplex Method the following problem:

Maximize Z = 3x + 2y
subject to: 2x + y ≤ 18
2x + 3y ≤ 42
3x + y ≤ 24
x≥0,y≥0

Consider the following phases:

1. Turning the inequalities into equalities

Introduce a slack variable for each restrictions of the type ≤ to turn them into equalities, giving the
following linear equation system:

2x + y + r = 18
2x + 3y + s = 42
3x +y + t = 24

2. Equaling the objective function to zero

- 3x - 2y + Z = 0

3. Writing the initial board simplex

At columns will appear all basic variables of the problem and the slack/surplus variables. At rows, you can
observe, for each restriction the slack variables with its coefficients of obtained equalities, and the last row with
the values resulting of substitute the value of every variable at function objetive, and operate just as was
explained in the theory to get the left values from the row:

Board I . 1st iteration

3 2 0 0 0
Base Cb P0 P1 P2 P3 P4 P5
P3 0 18 2 1 1 0 0
P4 0 42 2 3 0 1 0
P5 0 24 3 1 0 0 1
Z 0 -3 -2 0 0 0

4. Halt condition

10
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

When at the Z row there are not negatives values, the optimal solution of the problem has been reached. In
such a case, the algorithm has finished. If it were not so, the following steps must be executed.

5. Input-output base condition

A. A. First, we must know the variable that enters into the base. For it we choose that value's column that
in the row the Z is the minor of the present’s negatives values. In this case would be the variable x (P1)
of coefficient - 3.

If two or more equal coefficients exist that obey the previous condition (tie case), then we will choose
that variable that be basic.

The column of the variable that goes into the base is named pivot column (In green color).

B. Once the variable that goes into the base was obtained, we are in condition to deduce what will be the
variable that goes out. For it, divide each independent term (P0) among the correspondent element of
the pivot column, taking care that the result must be bigger than zero, and the minimum of this values is
chosen.

In our case: 18/2 [=9] , 42/2 [=21] y 24/3 [=8].

If any less or equal to zero element exist, such divide will not be do, or if every elements that belongs to
pivot column are zero we are in the case of unbounded solution, and the problem just would be finished

The term of the pivot column that gives the lowest positive quotient in previous division, the 3, right
now than 8 is the lower quotient, indicates the row of the slack variable that goes out the base, t (P5).
This row is named the pivot row (In green color).

If two or more quotients are equals when they are being calculated (tie case), make a choice for a non
basic variable (if possible).

C. At intersection of pivot row with pivot column, we have the pivot element, 3.

6. Calculating the coefficients of the new board.

The new coefficients of the pivot row, t (P5), are obtained dividing all of the coefficients from such row
among the pivot element, 3, that is the one necessary to turn into 1.

Following, with Gaussian reduction we do zeros the remainders terms of that column, with it we get the
new coefficients from the other rows including that belong to the objective function row Z.

In addition, it can be done in the following way:

Pivot row:

New pivot row = (Old pivot row) / (Pivot)

11
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

Remainders rows:

New row = (Old row) - (Coefficients from old row placed at incoming variable's colum) x (New pivot
row)

Lets see an example, once the pivot row has been calculated (x's (P1) row at Board II):

Old row P4 42 2 3 0 1 0
- - - - - -
Coefficient 2 2 2 2 2 2
x x x x x x
New pivot row 8 1 1/3 0 0 1/3
= = = = = =
New row P4 26 0 7/3 0 1 -2/3
Board II . 2nd iteration
3 2 0 0 0
Base Cb P0 P1 P2 P3 P4 P5
P3 0 2 0 1/3 1 0 -2/3
P4 0 26 0 7/3 0 1 -2/3
P1 3 8 1 1/3 0 0 1/3
Z 24 0 -1 0 0 1

Can be noticed that we have not attained the halt condition because at Z row, there is one negative, -1. We
must do another iteration:

A. The incoming variable is y (P2), in order to be the variable that corresponds to the column where is the -
1 coefficient.

B. B. To calculate the variable that goes out, we divide the terms from last column among the ones
correspondent to the new pivot column: 2 / 1/3 [=6] , 26 / 7/3 [=78/7] and 8 / 1/3 [=24]
and as the lower positive quotient is 6, we have that the variable that goes out is r (P3).

C. The pivot element, that we must do it 1, is 1/3.

Working of analogous form that before, we obtain the board:

12
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

Board III . 3rd iteration

3 2 0 0 0
Base Cb P0 P1 P2 P3 P4 P5
P2 2 6 0 1 3 0 -2
P4 0 12 0 0 -7 1 4
P1 3 6 1 0 -1 0 1
Z 30 0 0 3 0 -1

As you can see, there is a element with minus sign in Z row, - 1, it means that we have not come still to the
optimal solution. It is necessary to repeat the process:

A. The variable that comes to the base is t (P5) , because is the variable that correspond to the -1
coefficient.

B. To calculate the variable that goes out, we divide the terms from the last colum among the terms
correspondent from new pivot colum: 6/(-2) [=-3] , 12/4 [=3], and 6/1 [=6]
and like the lower positive quotient is 3, we get s (P4) as the variable that goes out from base.

C. The pivot element, that we must do it 1, is 4.

We get the board:

Board IV . 4th iteration

3 2 0 0 0
Base Cb P0 P1 P2 P3 P4 P5
P2 2 12 0 1 -1/2 1/2 0
P5 0 3 0 0 -7/4 1/4 1
P1 3 3 1 0 3/4 -1/4 0
Z 33 0 0 5/4 1/4 0

As in the last row, all the coefficients are positive, then the halt condition is obey, getting the optimal
solution.

The optimal solution is given by the Z value, at the column of the solution values, in this case: 33. In the
same column, you can notice the point where it is reached, watching the rows correspondents to the decision
variables that come at base: (x,y) = (3,12)

13
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

Resolve using the Graphical Method the following problem:

Maximize Z = f(x,y) = 3x + 2y
subject to: 2x + y ≤ 18
2x + 3y ≤ 42
3x + y ≤ 24
x≥0,y≥0

1. Initially we draw the coordinate system correlating to an axis the variable x, and the other axis to
variable y, as can see in the figure.

2. We mark, in these axis, a numerical scale appropriate to the values it can take the variables according to
the constraints of the problem. To do this work, for each constraint we must to void all variables except
the related to a certain axis, so we establishing the right value for such axis. This process must be done
for every axis.

3. Following, we represent all constraints. We take the first one and we draw the line that is obtained by
considering the constraint as equality. In the figure, this is represented with the A-B edge, and the region
that defines this constraint is shown in YELLOW color. We repeat the process with the other
restrictions, limiting BLUE and RED regions for the second and third constraint respectively.

4. Feasible region is determined for the intersection of every region defined by the constraints and the non-
negativity condition of each variable, that is, both axis. This feasible region is represented by the O-F-
H-G-C polygon, (in VIOLET color).

14
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

5. Since there is a feasible region, we proceed to determine its extreme points, or vertices of the polygon
that represents. These vertices are the candidate points for optimal solutions. In this example are the O-
F-H-G-C points shown in the figure.

6. Finally, we evaluate the objective function (3x+2y) in each point (the results in the table below). As the
G point provides the greatest value of the Z function, and the objective is to maximize, this point is the
optimal solution: Z = 33 with x = 3 and y = 12

Extreme point Coordinates (x,y) Objective value(Z)

O (0,0) 0
C (0,14) 28
G (3,12) 33
H (6,6) 30
F (8,0) 24

15
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

COMPARING GRAPHICAL METHOD VS. SIMPLEX METHOD

Successive tableaus constructed in the Simplex method provide us the value of the objective function at
different vertexes, and at same time, adjusting the coefficients of initial variables.

In the first iteration (Tableau I) have remained all the equal coefficients, has been calculated the objective
function value at the vertex (0,0) that is the value which belong to the basic variables, been 0 the result.

Tableau I . 1st iteration

3 2 0 0 0
Base Cb P0 P1 P2 P3 P4 P5
P3 0 18 2 1 1 0 0
P4 0 42 2 3 0 1 0
P5 0 24 3 1 0 0 1
Z 0 -3 -2 0 0 0

In the Simplex method, the incoming variable establishes which will be the destination vertex. As in this
example P1 (corresponding to 'x') enters to the base, the shifting will be done through of the OF edge until
reach the F vertex, where the Z value is calculated. This step occurs in the second iteration of the Simplex
method, shown in the Tableau II. The value has been calculated for the F vertex, obtaining a function value of Z
= 24.

16
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

Tableau II . 2nd iteration

3 2 0 0 0
Base Cb P0 P1 P2 P3 P4 P5
P3 0 2 0 1/3 1 0 -2/3
P4 0 26 0 7/3 0 1 -2/3
P1 3 8 1 1/3 0 0 1/3
Z 24 0 -1 0 0 1

A new move is done along FH edge, until reach H vertex (you can see these data on Tableau III). In this
third iteration, we calculated the function value at the H vertex, obtaining Z = 30.

Tableau III . 3rd iteration

3 2 0 0 0
Base Cb P0 P1 P2 P3 P4 P5
P2 2 6 0 1 3 0 -2
P4 0 12 0 0 -7 1 4
P1 3 6 1 0 -1 0 1
Z 30 0 0 3 0 -1

17
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

We continue with the process along the HG edge until reach to the G vertex. The obtained data are shown
in the Tableau IV. At this point the process ends, allowing verifying that the solution does not improve by
moving along the GC edge to the G vertex (does not exceed the present value of the function).

Tableau IV . 4th iteration

3 2 0 0 0
Base Cb P0 P1 P2 P3 P4 P5
P2 2 12 0 1 -1/2 0 0
P5 0 3 0 0 -7/4 0 1
P1 3 3 1 0 -3/4 0 0
Z 33 0 0 5/4 0 0

18
 ADEEB EIT
Operations Research - BMIS355
LIU Spring 2016-2017

The maximum value for the objective function is 33, and corresponds to x = 3 and y = 12 (vertex G).

The Graphical method needs to calculate the objective function value at all vertex of the feasible region,
unlike the Simplex, method that finishes when finds the optimal value.

19