A New Algorithm for Linear Programming

Dhananjay P. Mehendale
Sir Parashurambhau College, Tilak Road, Pune-411030,
India

Abstract
In this paper we propose a new polynomial-time algorithm for linear programming.
We further extend the ideas used in this new linear programming algorithm for
nonlinear programming problems. The standard linear programming problem consists
of an objective function to be optimized, either maximized or minimized, subject to
certain constraints provided in terms of inequalities. The new algorithm is based on
the idea of treating the objective function as a parameter. We form a matrix of
coefficients, made-up of the coefficients of the variables defined in the problem
itself, and the coefficients of variables defined newly, for converting inequalities into
equations, namely, slack variables if it is the maximization problem, or, surplus
variables if it is the minimization problem. The system of equations we use consist of
the objective equation and equations obtained from inequalities defining constraint
imposed by the problem. We obtain reduced-row-echelon-form, R, for this matrix
containing only one variable, namely, the objective function itself as an unknown
parameter, d, say. This matrix in the reduced-row- echelon-form contains columns
(column vectors) corresponding to basic variables and non-basic variables. If all the
entries in the columns corresponding to non-basic variables in R are already
nonnegative then we will see that we have almost reached to the solution and nothing
much is left to be done. If there are columns corresponding to non-basic variables
which contain some negative entries then we will require to apply suitable row
transformations, at most $m$ in number if there are m rows in R, as we will see
below, to make all the entries in the columns corresponding to non-basic variables
nonnegative. We then proceed to show that the method developed above for linear
programming naturally extends to nonlinear programming problems. For nonlinear
programming problems we use the technique of Grobner bases, since Grobner basis is
an equivalent of reduced row echelon form for a system of nonlinear equations.

1
1. Introduction: There are two types of linear programs (linear programming
problems):
T
1. Maximize: C x
Subject to: Ax ≤ b
x≥0
Or
T
2. Minimize: C x
Subject to: Ax ≥ b
x≥0
where x is a column vector of size n×1 of unknown variables. We
call these variables the problem variables
where C is a column vector of size n×1 of profit (for maximization
T
problem) or cost (for minimization problem) coefficients, and C is a row vector
of size 1×n obtained by matrix transposition of C .
where A is a matrix of constraints coefficients of size m×n.
where b is a column vector of constants of size m×1 representing
the boundaries of constraints.
By introducing the appropriate slack variables (for maximization
problem) and surplus variables (for minimization problem), the above mentioned
linear programs gets converted into standard form as:
T
Maximize: C x
Subject to: Ax + s = b (1.1)
x ≥ 0, s ≥ 0
where s is slack variable vector of size m×1.
This is a maximization problem.
Or
T
Minimize: C x
Subject to: Ax − s = b (1.2)
x ≥ 0, s ≥ 0
where s is surplus variable vector of size m×1.
This is a minimization problem.
In geometrical language, the constraints defined by the inequalities form a region
in the form of a convex polyhedron, a region bounded by the constraint planes,
Axi = bi , and the coordinate planes. This region is called feasible region and it
is straightforward to establish that there exists at least one vertex of this
polyhedron at which the optimal solution for the problem is situated when the
problem at hand is well defined, i.e. neither inconsistent, nor unbounded, nor
infeasible. There may be unique optimal solution and sometimes there may be
infinitely many optimal solutions, e.g. when one of the constraint planes is parallel
to the objective plane we may have a multitude of optimal solutions. The points
on an entire plane or an entire edge can constitute the optimal solution set.
These problems are handled most popularly by using the well
known simplex algorithm or some of its variant. Despite its theoretical
exponential complexity the simplex method works quite efficiently for most of the

2
practical problems. However, there are few computational difficulties associated
with simplex algorithm. In order to view them in nutshell we begin with stating
some common notions and definitions that are prevalent in the literature. A
variable xi is called basic variable in a given equation if it appears with unit
coefficient in that equation and with zero coefficients in all other equations. A
variable which is not basic is called non-basic variable. A sequence of elementary
row operations that changes a given system of linear equations into an equivalent
system (having the same solution set) and in which a given non-basic variable can
be made a basic variable is called a pivot operation. An equivalent system
containing basic and non-basic variables obtained by application of suitable
elementary row operations is called canonical system. At times, the introduction
of slack variables for obtaining standard form automatically produces a canonical
system, containing at least one basic variable in each equation. Sometimes a
sequence of pivot operations is needed to be performed to get a canonical system.
The solution obtained from canonical system by setting the non-basic variables to
zero and solving for the basic variables is called basic solution and in addition
when all the variables have nonnegative values the solution satisfying all the
imposed constraints is called a basic feasible solution. Simplex method cannot
start without an initial basic feasible solution. The process of finding such a
solution, which is a necessity in many of practical problems, is called Phase I of
the simplex algorithm. Simplex method starts its Phase II with an initial basic
feasible solution in canonical form at hand. Then simplex tests whether this
solution is optimal by checking whether all the values of relative profits (profits
that result due to unit change in the values of non-basic variables) of all the non-
basic variables are nonpositive. When not optimal, the simplex method obtains an
adjacent basic feasible solution by selecting a non-basic variable having largest
relative profit to become basic. Simplex then determines and carries out the
exiting of a basic variable, by the so called minimum ratio rule, to change it into a
non-basic variable leading to formation of a new canonical system. On this new
canonical system the whole procedure is repeated till one arrives at an optimal
solution.
The main computational difficulties of the simplex method
which may cause the reduction in its computational efficiency are as follows:
1] There can be more than one non-basic variable with largest value for relative
profit and so a tie can take place while selecting a non-basic variable to become
basic. The choice at this situation is done arbitrarily and so the choice made at this
stage causing largest possible per unit improvement is not necessarily the one that
gives largest total improvement in the value of the objective function and so not
necessarily minimizes the number of simplex iterations.
2] While applying minimum ratio rule it is possible for more than one constraint
to give the same least ratio causing a tie in the selection of a basic variable to
leave for becoming non-basic. This degeneracy can cause a further complication,
namely, the simplex method can go on without any improvement in the objective
function and the method may trap into an infinite loop and fail to produce the
desired optimal solution. This phenomenon is called cycling which enforces
modification in the algorithm by introducing some additional time consuming
rules that reduce the efficiency of the simplex algorithm.
3] Simplex is not efficient on theoretical grounds basically because it searches
adjacent basic feasible solutions only and all other simplex variants which

3
examine nonadjacent solutions as well have not shown any appreciable change in
the overall efficiency of these modified simplex algorithms over the original
algorithm.
Because of the far great practical importance of the linear
programs and other similar problems in the operations research it is a most desired
thing to have an algorithm which works in a single step, if not, in as few steps as
possible. No method has been found which will yield an optimal solution to a
linear program in a single step ([1], Page 19). We aim to propose a polynomial-
time algorithm for linear programming which aims at fulfilling this requirement in
a best possible and novel way.

2. A New Polynomial-time Algorithm for Linear Programming: In this

section we propose and discuss our new algorithm for solving linear
programming problems. This algorithm is based on treating objective function
as a parameter and aims at finding its optimal value.
We start with the following equation:
CT x = d (2.1)
where d is an unknown parameter, and call it the objective equation. The
(parametric) plane defined by this equation will be called objective plane.
Please note that we are discussing first the maximization
problems. A similar approach for minimization problems will be discussed next.
Given a maximization problem, we first construct the combined
system of equations containing the objective equation and the equations defined
by the constraints imposed by the problem under consideration, combined into a
single matrix equation, viz.,

C(T1×n) 0(1×m)   x d 
  =   (2.2)
A(m×n) I (m×m)   s  b 
 C(T1× n ) 0 (1× m )  d 
Let E =   , and let F = b 
 A( m × n ) I ( m× m )   
Let [E, F] denote the augmented matrix obtained by appending the column vector
F to matrix E as a last column. We then find R, the reduced-row-echelon-form
(rref) ([2], pages 73-75) of the above augmented matrix [E, F]. Thus,

R = rref ([E, F]) (2.3)

Note that the augmented matrix [E, F] as well as its reduced row echelon form R
contains only one parameter, namely, d and all other entries are constants. The
maximization problem of linear programming asks to determine the unique
maximum value for d , for which there exists a feasible solution and this unique
maximal value of the objective function has been used to obtain the feasible
solution and so this solution is in fact the optimal solution. In the case of an
unbounded linear programming problem there is no upper bound (lower bound, in

4
 the case of minimization problem) for the value of d , while in the case of an
infeasible linear program the set of feasible solutions is empty. The steps that will
be executed to determine the optimal solution will also tell by implication when
such optimal solution does not exist in the case of an unbounded or infeasible
problem.
The general form of the matrix R representing the reduced row
echelon form is

 a11 a12 L a1n b11 b12 L b1m c1d + e1 

 a a 22 L a 2n b21 b22 L b2 m c 2 d + e2 
 21 
 a31 a32 L a 3n b31 b32 L b3m c 3 d + e3 
R= 
 M M M M M M M M M 
 a m1 a m2 L a mn bm1 bm 2 L bmm c m d + em 
 
a ( m +1)1 a ( m +1) 2 L a ( m+1) n b( m+1)1 b( m +1) 2 L b( m+1) n c ( m+1) d + e( m +1) 

The first n columns of the above matrix represent the coefficients of the problem
variables (i.e. variables defined in the linear program) x1 , x2 ,L, xn . The next
m columns represent the coefficients of the slack variables s1 , s2 ,L, sm used
to convert inequalities into equalities to obtain the standard form of the linear
programming problem. The last column represents the transformed right hand side
of the equation (2.2) during the process (a suitable sequence of transformations)
that is carried out to obtain the reduced-row-echelon-form. Note that the last
column of R contains the linear form d as a parameter whose optimal value is to
be determined such that the nonnegativity constraints remain valid, i.e.
xi ≥ 0,1 ≤ i ≤ n and s j ≥ 0,1 ≤ j ≤ m . Among first (n + m) columns of
R the columns correspond to basic (pivot) variables are the columns that are unit
vectors and the remaining ones to non-basic variables.
Now, if there are some negative entries in the columns
of R corresponding to non-basic variables then we carry out suitable elementary
row transformations on the obtained R = rref ([E, F]) so that every column among
the columns associated with non-basic variables become nonnegative. We are
doing this because as will be seen below we can then put zero value for these non-
basic variables and can determine the values of all the basic variables and the
linear programming problem will then be solved completely. It is easy to check
that for a linear programming problem if all the coefficients of parameter d in the
last column of R are positive then the linear programming problem at hand is
unbounded since in such case the parameter d can be increased arbitrarily without
violating the nonnegativity constraints on variables xi , s j . Also, for a linear
programming problem if all the coefficients of some non-basic variable
represented by a column of R are nonpositive and are strictly negative in those
rows having a negative coefficient to parameter d that appears in the last column
of these rows then again the problem belongs to the category of unbounded
problems since we can increase the value of d to any high value without violating
the nonnegativity constraints for the variables by assigning sufficiently high value

5
 to this non-basic slack variable. Note that the rows of R actually offer expressions
for basic variables in terms of non-basic variables and terms of type
ck d + ek , k = 1,2,L(m + 1) containing the parameter d on the right side.
The rows with a positive coefficient for the parameter d represent those equations
in which the parameter d can be increased arbitrarily without violating the
nonnegativity constraints on variables xi , s j . So, these equations with a positive
coefficient for the parameter d are not implying any upper bound on the
maximum possible value of parameter d . However, these rows are useful in
certain situations as they are useful to find lower bound on the value of parameter
d . The rows with a negative coefficient for the parameter d represent those
equations in which the parameter d cannot be increased arbitrarily without
violating the nonnegativity constraints on variables xi , s j . So, these equations
with a negative coefficient for the parameter d are implying an upper bound on
the maximum possible value of parameter d and so important ones for
maximization problems. Note that actually every row of R is offering us a value
for parameter d which can be obtained by equating to zero each term of the type
ck d + ek , k = 1,2,L(m + 1) . We now define the sub-matrix of R , say R N ,
made up of all columns of R and containing those rows j of R for which the
coefficients c j of the parameter d are negative. Let ci , ci ,L, ci coefficient
1 2 k
of d in the rows of R which are negative. We collect these rows with negative
coefficient for d to form the mentioned submatrix, R N , of R given below. With
this it is clear that coefficients of d in all other rows of R are greater than or equal
to zero.

ai11 ai1 2 L ai1n bi11 bi1 2 L bi1m ci1 d + ei1 

a ai2 2 L ai2n bi21 bi2 2 L bi2m ci2 d + ei2 
 i21
ai31 ai3 2 L ai3n bi31 bi3 2 L bi3m ci3 d + ei3 
RN =  
M M M M M M M M M 
M M L M M M L M M 
 
aik 1 aik 2 L aik n bik 1 bik 2 L bik m cik d + eik 

It should be clear to see that if RN is empty (i.e. not containing a single row) then
the problem at hand is unbounded. Among the first (n + m) columns of RN first n
columns represent the coefficients of problem variables and next m columns
represent the coefficients of slack variables. There are certain columns corresponding
to basic (pivot) variables which are unit vectors. The other columns correspond to

6
non-basic variables. As mentioned, among the columns for non-basic variables those
having all entries nonnegative can only lead to decrement in the value of d when a
positive value is assigned to them. This is undesirable as we aim maximization of the
value of d . So, we can safely set the value of such variables equal to zero. When all
columns corresponding to non-basic variables in R are having all entries nonnegative
then we can set all non-basic variables to zero, set d = min{ d } in every row of R and
find the basic feasible solution which will be optimal, with min{ d } as optimal value
for the objective function at hand.
RN we now proceed to consider those non-basic variables for
In
which the columns of R N contain some (at least one) positive values and some
negative (at least one) values. In such case when we assign some positive value to
such non-basic variable it leads to decrease in the value of d in those rows in which
ck > 0 and increase in the value of d in those rows in which ck < 0. We now need
to consider the ways of dealing with this situation. We deal with this situation as
follows: In this case, we choose and carry out appropriate and legal elementary row
transformations on the matrix R in the reduced row echelon form to achieve
nonnegative value for all the entries in the columns corresponding to non-basic
variables in the matrix R . The elementary row transformations are chosen to
produce new matrix which remains equivalent to original matrix in the sense that the
solution set of the matrix equation with original matrix and matrix equation with
transformed matrix remain same. Due to this equivalence we can now set all the non-
basic variables in this transformed matrix to zero and obtain with justification dmin =
min{ d } as optimal value for the objective function and obtain basic feasible solution
as optimal solution by substitution.

Let us now discuss our new algorithm in steps:

Algorithm 2.1 (Maximization):

1. Express the given problem in standard form:

Maximize: CT x
Subject to: Ax + s = b
x ≥ 0, s ≥ 0
2. Construct the augmented matrix [E F], where
 C(T1× n ) 0 (1× m )  d 
E =   , and F =  
 A( m × n ) I ( m × m )  b 
and obtain the reduced row echelon form:
R = rref ([E, F])
3. If there is a row (or rows) of zeroes at the bottom of R in the first n columns
and containing a nonzero constant in the last column then declare that the
problem is inconsistent and stop.

7
 4. Else if for any value of d one observes that nonnegativity constraint for some
variable gets violated for at least one of the variables then declare that the
problem at hand is infeasible and stop.
5. Else, if the coefficients of d in the last column are all positive or if there
exists a column of R corresponding to some non-basic variable with all entries
negative then declare that the problem at hand is unbounded and stop.
6. Else check whether all the entries in the columns of R corresponding to non-
basic variables are nonnegative.
7. If yes, then find the sub-matrix of R, say RN , made up of those rows of R for
which the coefficient of d in the last column is negative and solve
cir d + eir = 0 for each such a term in the last column of RN and find

the value of d =d ir for r = 1,2,L, k and find d min = min{ d ir }

which represents the maximal value of the objective function.
8. Set the value for all non-basic variables equal to zero.

9. Substitute the value d min = min{ d ir } in place of d =d ir in every

row of R and find the values for all basic (pivot) variables. The set of values
assigned to basic as well as non-basic variables together forms the optimal
solution for the problem.
10. If not, then apply suitable elementary row transformations (which do exist
when the problem at hand is a well-defined legitimate problem) first on R N
and make all the entries in the columns corresponding to non-basic variables
nonnegative. Now, apply suitable elementary row transformations on R if
still there are some negative entries in the column corresponding to non-
basic variables and change R such that now every column corresponding to
non-basic variables will strictly contain nonnegative entries in R . Now
perform steps 7-9 in the new R N contained in the new R , which now
contains nonnegative entries in every column corresponding to non-basic
variable, and find the optimal solution.

The type of the required elementary row transformations: To make an

entry in a column of RN nonnegative we require the following type of elementary

row transformation: Suppose in r-th column the entry bim is positive and suppose the
entry bin in that column is negative then we replace the row containing this negative

b
entry, in , by the row made up of the addition of this same row containing this
negative entry with the row obtained by multiplying the above mentioned row

8
containing positive entry, bim , by the ratio [-( bin / bim )]. This action changes this
negative entry, bin , to zero.

One can now easily see that by proceeding on exactly similar lines one can develop
the formulation as above also for minimization problems and can deal with them
successfully.

The Main Result:

We now state and prove the main result of this paper.

Theorem 2.1: The complexity of the above given algorithm (Algorithm 2.1) for
solving a linear programming problem is of polynomial order and it is same as that of
the complexity required for finding reduced-row-echelon-form.

Proof: The algorithm basically finds the reduced-row-echelon-form (rref) of the

augmented matrix defining the problem, i.e., R = rref ([E, F]). Now, whether the
linear programming problem at hand is either inconsistent, or unbounded, or
infeasible, or a valid one with a solution becomes clear from the nature of R, as
mentioned in the algorithm.
When the linear programming problem at hand is either inconsistent, or
unbounded, or infeasible nothing is to be done and the complexity in this case is
clearly that of the complexity for the algorithm to find rref.
When the linear programming problem at hand is a valid one with a solution
then to find that solution we further require to carry out certain elementary row
transformations to make all the negative entries in the columns corresponding to non-
basic variables nonnegative. Now, to make all the negative entries in a row belonging
to columns for non-basic variables one requires one elementary row transformation.
Since there are in all “m” rows in R, one will require to perform at most “m”
elementary row transformations, so the overall complexity still remains dominated by
the order of the complexity required for rref algorithm.

3. Examples:

We now proceed with some examples for maximization problems:

Example 2.1: Maximize: x + y
Subject to: x + 2 y ≤ 4
− x + y ≤1
4 x + 2 y ≤ 12
x, y ≥ 0
Solution: For this problem we have
R = [ 1, 0, 0, 0, 1/2, -d+6 ]
[ 0, 1, 0, 0, -1/2, -6+2*d ]
[ 0, 0, 1, 0, 1/2, 10-3*d ]
[ 0, 0, 0, 1, 1, 13-3*d ]

9
So, clearly,

RN = [ 1, 0, 0, 0, 1/2, -d+6 ]
[ 0, 0, 1, 0, 1/2, 10-3*d ]
[ 0, 0, 0, 1, 1, 13-3*d ]

For this example the column forming coefficients for non-basic variable s 3 contains
nonegative numbers. So, we set s3 = 0 . Clearly, dmin = 3.3333 = Optimal value for
the objective function. Using this value of optimum we have
x = 2.66, y = 0.66 , s1 = 0, s 2 = 3, s 3 = 0 .

Example 2.2: We first consider the duel of the example suggested by E. M. L. Beale
[3], which brings into existence the problem of cycling for the simplex method, and
provide a solution as per the above new method which offers it directly without any
cycling phenomenon.
Maximize: 0.75 x1 −20 x2 +0.5 x3 −6 x4
Subject to: 0.25 x1 −8 x2 − x3 +9 x4 ≤ 0
0.5 x1 −12 x2 −0.5 x3 +3 x4 ≤ 0
x3 ≤ 0
x1 , x2 , x3 , x4 ≥ 0
Solution: For this problem we have the following

R = [1, 0, 0, 0, −22/3, 38/3, 4/3, (−14/3)d+4/3 ]

[0, 1, 0, 0, −7/24, 11/24, 1/24, (−5/24)d+1/2 ]
[0, 0, 1, 0, 0, 0, 1, 1 ]
[0, 0, 0, 1, 1/18, 1/18, 1/9, 1/9−(1/18)d ]

So, clearly,

RN = [1, 0, 0, 0, −22/3, 38/3, 4/3, (−14/3)d+4/3 ]

[0, 1, 0, 0, −7/24, 11/24, 1/24, (−5/24)d+1/24 ]
[0, 0, 0, 1, 1/18, 1/18, 1/9, 1/9−(1/18)d ]

We perform following elementary row transformations on R: Let us denote the

successive rows of R by R(1), R(2), R(3), R(4). We change

(i) R(1)
R(1) + 132*R(4)

(ii) R(2)
R(2) + (126/24)*R(4)

This leads to new transformed R as follows:

10
R = [1, 0, 0, 132, 0, 20, 1 6, 16-12d]
[0, 1, 0, 21/4, 0, 3/4, 5/8, 5/8-(1/2)d]
[0, 0, 1, 0, 0, 0, 1, 1 ]
[0, 0, 0, 1, 1/18, 1/18, 1/9, 1/9−(1/18)d ]

In the transformed R we have nonnegative columns for all non-basic variables, which
are now those corresponding to x4 , s2 , s3 . So, by setting x4 = s2 = s3 = 0 and
setting all expressions of type ci d + ei = 0 in the last column we find
r r
d min = min{ d i r } = 1.25. Using this value in the last column of the newly obtained
transformed R we have: x1 = 1.0000, x2 = 0, x3 = 1, x4 = 0, s1 = 0.7500, s2 = 0,
s3 = 0, and the maximum value of d = 1.2500.
Example 2.3: We now consider an unbounded problem. The new method directly
implies the unbounded nature of the problem through the positivity of the coefficients
of d in matrix R for the problem.
Maximize: − x + 3 y
Subject to: − x − y ≤ −2
x − 2y ≤ 0
− 2x + y ≤ 1
x, y ≥ 0

Solution: The following is the matrix R :

R=[ 1, 0, 0, 0, -3/5, (1/5)d-3/5 ]
[ 0, 1, 0, 0, -1/5, (2/5)d-1/5 ]
[ 0, 0, 1, 0, -4/5, (3/5)d-14/5 ]
[ 0, 0, 0, 1, 1/5, 1/5+(3/5)d ]

Here, all the coefficients ofd are positive. So, by setting variable s3 = 0 we can see
that we can assign any arbitrarily large value to variable d without violation of
nonnegativity constraints for variables. Thus, the problem has an unbounded solution.

Example 2.4: We now consider a problem having an infeasible starting basis. We see
that new algorithm has no difficulty to deal with it.
Maximize: 3 x + 2 y
Subject to: x+ y≤4
2x + y ≤ 5
x − 4 y ≤ −2
x, y ≥ 0

11
Solution: The following is the matrix R:
R =[ 1, 0, 0, 0, 1/7, (2/7)d-2/7 ]
[ 0, 1, 0, 0, -3/14, 3/7+(1/14)d]
[ 0, 0, 1, 0, 1/14, 27/7-(5/14)d]
[ 0, 0, 0, 1, -1/14, 36/7-(9/14)d]

We perform

(i) R(4)
R(4) + R(3)

(ii) R(2)
R(2) + 3*R(3)

This leads to getting new transformed R as follows:

R =[ 1, 0, 0, 0, 1/7, (2/7)d-2/7 ]
[ 0, 1, 3, 0, 0, 12- d ]
[ 0, 0, 1, 0, 1/14, 27/7-(5/14)d]
[ 0, 0, 1, 1, 0, (63/7) – d ]

The first two columns of R correspond to basic variables x, y . Since the columns
corresponding to non-basic variables s1,s3 contain nonnegative entries in R N , so
we set these variables to zero. From last row we have s2 = 0 and
dmin = 9. Also from first and second rows, x = 1, y = 3

Example 2.5: We now consider an infeasible problem.

Maximize: 3 x + 2 y
Subject to: 2 x − y ≤ −1
− x + 2y ≤ 0
x, y ≥ 0
Solution: The following is the matrix R :
R= [ 1, 0, 0, -1/4, (1/4)d ]
[ 0, 1, 0, 3/8, (1/8)d ]
[ 0, 0, 1, 7/8, (-3/8)d-1 ]

Here, the coefficient of d is negative only in the last row and so

RN = [ 0, 0, 1, 7/8, (-3/8)d-1].

We perform following elementary row transformations on R: Let us denote the

successive rows of R by R(1), R(2), R(3), R(4). We change

(i) R(1)
(2/7)*R(4) + R(1)

This leads to
R= [ 1, 0, 2/7, 0, 1/7*d-2/7]
[ 0, 1, 0, 3/8, (1/8)d ]
[ 0, 0, 1, 7/8, (-3/8)d-1]

12
Setting s1,s2 equal to zero, we have for consistency of last row d = ─(8/3) and using
this value for d we have y = ─(1/3). Thus, this problem is infeasible.

Remark 2.1: Klee and Minty [4], have constructed an example of a set of linear
programs with n variables for which simplex method requires 2 − 1 iterations to
n
reach an optimal solution. Theoretic work of Borgwardt [5] and Smale [6] indicates
that fortunately the occurrence of problems belonging to the class of Klee and Minty,
which don’t share the average behavior, is so rare as to be negligible. We now
proceed to show that there is no problem of efficiency for new algorithm in dealing
with the problems belonging to this class.

Example 2.6: We now consider a problem for which the simplex iterations are
exponential function of the size of the problem. A problem belonging to the class
described by Klee and Minty [4], containing n variables requires 2
n
− 1simplex
steps. We see that the new method doesn’t require any special effort
Maximize: 100 x1 +10 x2 + x3
Subject to: x1 ≤ 1
20 x1 + x2 ≤ 100
200 x1 +20 x2 + x3 ≤ 10000
x1 , x2 , x3 ≥ 0
Solution: The following as the matrix R :

R = [1, 0, 0, 0, 1/10, -1/100, -90+(1/100)d]

[0, 1, 0, 0, -1, 1/5, 1900-(1/5)d]
[0, 0, 1, 0, 0, -1, 2d-10000 ]
[0, 0, 0, 1, -1/10, 1/100, 91-(1/100)d ]

We perform following elementary row transformations on R: Let us denote the

successive rows of R by R(1), R(2), R(3), R(4). We change

(i) R(2)
10R(1) +R(2), and

(ii) R(4)
R(1) +R(4)
(iii) R(1)
10*R(1)
(iv) R(1)
R(1) + R(2)
(v) R(3)
R(3) +10R(2)

This changes R to

R = [10, 1, 0, 0, 1, 0, 100 ]
[0, 1, 0, 0, 0, 1/10, 1000-(1/10)*d ]
[0, 10, 1, 0, 0, 0, d ]
[1, 0, 0, 1, 0, 0, 1 ]

which produces (after setting non-basic variables to zero as usual) the solution:

13
x1 = 0, x2 = 0, x3 = 10000, s1 = 1, s2 = 100, s3 = 0, and the maximum value of the
objective function, d = 10000.

Example 2.7: Maximize: 4x + 3 y

Subject to: x + 3 . 5 y ≤ 9
2x + y ≤ 8
x+ y ≤ 6
x, y ≥ 0
For this problem we get following R
R=[ 1, 0, 0, 0, -3, d-18 ]
[ 0, 1, 0, 0, 4, 24-d ]
[ 0, 0, 1, 0, -11, -57+5/2*d]
[ 0, 0, 0, 1, 2, 20-d ]

We perform following elementary row transformations:

R(1)
R(1) + (3/2)R(4)

R(3)
R(3) + (11/2)R(4)

this leads to:

R=[ 1, 0, 0, 0, 0, 12 – 1/2d]
[ 0, 1, 0, 0, 4, 24-d ]
[ 0, 0, 1, 11/2, 0, 53-3d ]
[ 0, 0, 0, 1, 2, 20-d ]

which leads to maximal basic feasible solution: d = 17.667 ,

x = 3.1665, y = 1.667, s1 = 0, s 2 = 1.1665, s3 = 0
Example 2.7: Maximize: 3x1+2x2+3x3+4x4+x5

Subject to:

4x1+3x2-2x3+2x4-x5 ≤ 12
2x1+3x2+x3+3x4+x5 ≤ 15
3x1+2x2+x3+2x4+5x5 ≤ 20
2x1+4x2+x3+6x4+x5 ≤ 25
x3 ≤ 3

A=
[ 3, 2, 3, 4, 1, 0, 0, 0, 0, 0, d ]
[ 4, 3, -2, 2, -1, 1, 0, 0, 0, 0, 12]
[ 2, 3, 1, 3, 1, 0, 1, 0, 0, 0, 15]
[ 3, 2, 1, 2, 5, 0, 0, 1, 0, 0, 20]
[ 2, 4, 1, 6, 1, 0, 0, 0, 1, 0, 25]
[ 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 3 ]

14
R = rref(A)

R=
[1, 0, 0, 0, 0, 0, 10/27, -2/27, -13/27, -34/27, 13/27*d-317/27]
[0, 1, 0, 0, 0, 0, 7/9, -1/18, -1/9, 5/9, 85/9-7/18*d ]
[0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 3 ]
[0, 0, 0, 1, 0, 0, -16/27, 1/54, 10/27, -5/27, 5/27+7/54*d ]
[0, 0, 0, 0, 1, 0, -8/27, 7/27, 5/27, 11/27, 178/27-5/27*d ]
[ 0, 0, 0, 0, 0, 1, -79/27, 37/54, 46/27, 166/27, 1157/27-65/54*d]

and its sub-matrix RN is

RN =
[0, 1, 0, 0, 0, 0, 7/9, -1/18, -1/9, 5/9, 85/9-7/18*d ]
[0, 0, 0, 0, 1, 0, -8/27, 7/27, 5/27, 11/27, 178/27-5/27*d ]
[0, 0, 0, 0, 0, 1, -79/27, 37/54, 46/27, 166/27, 1157/27-65/54*d]

With the appropriately chosen three elementary row transformations this RN

becomes new

RN =
[0, 35/27, 0, 0, 7/9, 0, 7/9, 7/54, 0, 28/27, 469/27-35/54*d ]
[0, 8/21, 0, 0, 1, 0, 0, 5/21, 1/7, 13/21, 214/21-1/3*d ]
[0, 79/21, 0, 0, 0, 1, 0, 10/21, 9/7, 173/21, 1646/21-8/3*d ]

After appropriately re-embedding the rows of this new RN in original place in R we

get

R=
[1, 0, 0, 0, 0, 0, 10/27, -2/27, -13/27, -34/27, 13/27*d-317/27]
[0, 35/27, 0, 0, 7/9, 0, 7/9, 7/54, 0, 28/27, 469/27-35/54*d ]
[0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 3 ]
[0, 0, 0, 1, 0, 0, -16/27, 1/54, 10/27, -5/27, 5/27+7/54*d ]
[0, 8/21, 0, 0, 1, 0, 0, 5/21, 1/7, 13/21, 214/21-1/3*d ]
[0, 79/21, 0, 0, 0, 1, 0, 10/21, 9/7, 173/21, 1646/21-8/3*d ]

Further two more elementary row transformations leads us to

R=
[1, 79/21, 0, 0, 0, 1, 10/27, 76/189, 152/189, 1319/189, -59/27*d+12595/189]
[0, 35/27, 0, 0, 7/9, 0, 7/9, 7/54, 0, 28/27, 469/27-35/54*d ]
[0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 3 ]
[0, 35/27, 0, 1, 7/9, 0, 5/27, 4/27, 10/27, 23/27, 158/9-14/27*d ]
[0, 8/21, 0, 0, 1, 0, 0, 5/21, 1/7, 13/21, 214/21-1/3*d ]
[0, 79/21, 0, 0, 0, 1, 0, 10/21, 9/7, 173/21, 1646/21-8/3*d ]

15
Using this R and setting all non-basic variables to zero we can easily find the maximal
value of the objective function, namely, d = 26.8 and by further substitution of this
value of objective function in we can easily find the optimal solution.

4. A New Algorithm for Nonlinear Programming: We now proceed show that we

can deal with nonlinear constrained optimization problems using the same above
given technique used to deal with linear programming problems. The algorithms
developed by Bruno Buchberger which transformed the abstract notion of Grobner
basis into a fundamental tool in computational algebra will be utilized. The
technique of Grobner bases is essentially a version of reduced row echelon form
(used above to handle the linear programs made up of linear polynomials) for
higher degree polynomials [7]. A typical nonlinear program can be stated as
follows:
Maximize/Minimize: f (x )
Subject to: h j ( x) = 0, j = 1,2,L, m
g j ( x) ≥ 0, j = m + 1, m + 2,L, p
xk ≥ 0, k = 1,2,L, n
Given a nonlinear optimization problem we first construct the following nonlinear
system of equations:
f ( x) − d = 0 (3.1)
h j ( x) = 0, j = 1,2,L, m (3.2)

g j ( x) + s j = 0, j = m + 1, m + 2,L, p (3.3)
where d is the unknown parameter whose optimal value is to be determined subject
to nonnegativity conditions on problem variables and slack variables. For this to
achieve we first transform the system of equations into an equivalent system of
equations bearing the same solution set such that the system is easier to solve. We
have seen so far that the effective way to deal with linear programs is to obtain the
reduced row echelon form for the combined system of equations incorporating
objective equation and constraint equations. We will see that for the nonlinear case
the effective way to deal with is to obtain the equivalent of reduced row echelon form,
namely, the Grobner basis representation for this system of equations (3.1)-(3.3). We
then set up the equations obtained by equating the partial derivatives of d with
respect to problem variables xi and slack variables si to zero and utilize the standard
theory and methods used in calculus. We demonstrate the essence of this method by
solving certain examples. where d is the unknown parameter whose optimal value is
to be determined subject to nonnegativity conditions on problem variables and slack
variables. For this to achieve we first transform the system of equations into an
equivalent system of equations bearing the same solution set such that the system is
easier to solve. We have seen so far that the effective way to deal with linear
programs is to obtain the reduced row echelon form for the combined system of
equations incorporating objective equation and constraint equations. We will see that
for the nonlinear case the effective way to deal with is to obtain the equivalent of
reduced row echelon form for the set of polynomials, namely, the Grobner basis
representation for this system of equations (3.1)-(3.3). We then set up the equations

16
obtained by equating the partial derivatives of d with respect to problem variables xi
and slack variables si to zero and utilize the standard theory and methods used in
calculus. We demonstrate the essence of this method by solving an example:
These examples are taken from [8], [9]. These examples sufficiently illustrate the
power of this new method of using powerful technique of Grobner basis to
successfully and efficiently deal with nonlinear programming problems.
Example 3.1: Maximize: − x12 + 4 x1 + 2 x2
Subject to: x1 + x2 ≤ 4
2 x1 + x2 ≤ 5
− x1 + 4 x2 ≥ 2
Solution: We build the following system of equations:
− x12 + 4 x1 + 2 x2 − d = 0
x1 + x2 + s1 − 4 = 0
2 x1 + x2 + s2 − 5 = 0
− x1 + 4 x2 − s3 − 2 = 0
such that: x1 , x2 , s1 , s2 , s3 ≥ 0
We now transform the nonlinear/linear polynomials on the left hand side of the above
equations by obtaining Grobner basis for them as follows:
486 − 81d − 18s2 − 16 s22 + 36s3 − 8s2 s3 − s32 = 0 (3.1.1)
9 − 9 s1 + 5s2 − s3 = 0 (3.1.2)
− 9 + s2 − 2 s3 + 9 x2 = 0 (3.1.3)
− 18 + 4 s2 + s3 + 9 x1 = 0 (3.1.4)
In order to maximize the objective function represented in terms of parameter d the
partial derivatives of d with respect to problem variables must vanish, i.e. they must
be equal to zero. therefore,

∂d ∂d
By setting = 0 and = 0 we get equations:
∂s2 ∂s3
32s2 + 8s3 = −18
8s2 + 2 s3 = 36
∂d
a rank deficient system. Note that for maximization of d if we set = 0 we get
∂s2
the value of s2 that maximizes d , namely, s 2 = −(18 / 32) − (8s3 / 32) , a
negative value for any nonnegative value of s3 . So, we set s2 = 0 . Similarly, for

17
 ∂d
maximization of d if we set = 0 we get the value of s3 that maximizes d ,
∂s3
namely, s3 = 18 − 4 s 2 ( = 18) , setting s 2 = 0 . But, by setting s2 = 0 in the
second equation above the largest possible value for s3 that one can have (is obtained
by setting s1 = 0 and it) is 9, when s2 = 0 . Thus, setting s2 = 0, s3 = 9 in the
first equation we get d = 9. From third and fourth equation we get x2 = 3, x1 = 1 .

Example 3.2: Maximize: − 8 x1 − 16 x2 + 24 x1 + 56 x2

2 2

Subject to: x1 + x2 ≤ 4
2 x1 + x2 ≤ 5
− x1 + 4 x2 ≥ 2
x1 , x2 ≥ 0
Solution: We build the following system of equations:
− 8 x12 − 16 x22 + 24 x1 + 56 x2 − d = 0
x1 + x2 + s1 − 4 = 0
2 x1 + x2 + s2 − 5 = 0
− x1 + 4 x2 − s3 − 2 = 0
We now transform the nonlinear/linear polynomials on the left hand side of the above
equations by obtaining Grobner basis for them as follows:
504 − 9d + 8s2 − 16 s22 + 56s3 − 8s32 = 0 (3.2.1)
9 − 9 s1 + 5s2 − s3 = 0 (3.2.2)
− 9 + s 2 − 2 s3 + 9 x 2 = 0 (3.2.3)
− 18 + 4 s2 + s3 + 9 x1 = 0 (3.2.4)
from first equation (3.2.1), in order to maximize d , we determine the values of
s2 , s3 as follows:
∂d 1
If we set = 0 we get the value of s2 that maximizes d , namely, s2 = .
∂s2 4
∂d
Similarly, if we set = 0 we get the value of s3 that maximizes d , namely,
∂s3
7
s3 = . Putting these values of s2 , s3 in the first and second equation we get
2
3
respectively the maximum value of d = 67 and the value of s1 = . Using
4
further these values in the third and fourth equation we get x1 = 1.5, x2 = 1.75 .

18
Example 3.3: Minimize: ( x1 − 3) 2 + ( x2 − 4) 2
Subject to: 2 x1 + x2 = 3
Solution: We form the objective equation and constraint equations as is done in the
above examples and then find the Grobner basis which yields:
5 x12 − d − 2 x1 + 10 = 0
2 x1 + x2 − 3 = 0
∂d
Setting = 0 we get the value of x1 that minimizes d , namely, x1 = 0.2 .
∂x1
This yields d = 9.8 and x2 = 2.6
Example 3.4: Minimize: x1
2
− x2
Subject to: x1 + x2 = 6
x1 ≥ 1
x12 + x22 ≤ 26
Solution: We form the objective equation and constraint equations as is done in the
above examples and then find the Grobner basis which yields:
2d − 14s1 − s2 + 8 = 0
x2 + s1 − 5 = 0
x1 − s1 − 1 = 0
− s2 − 8s1 + 2 s12 = 0
For minimizing d we should set the values of s1 , s 2 equal to zero (as they have
signs opposite to d ) which yields d = −4 . From other equations we get
x1 = 1, x2 = 5 .
Example 3.5: Minimize: − 6 x1 + 2 x1 − 2 x1 x2 + 2 x2
2 2

Subject to: x1 + x2 ≤ 2
Solution: We form the objective equation and constraint equations as is done in the
above examples and then find the Grobner basis which yields:
6 x22 − d + 6 x2 s1 − 6 x2 + 2s12 − 2 s1 − 4 = 0 (3.5.1)
x1 + x2 + s1 − 2 = 0 (3.5.2)
∂d ∂d
Setting = 0 and = 0 we get equations:
∂x2 ∂s1
12 x2 + 6 s1 =6
6 x2 + 4 s1 =2

19
There solution gives s1 = −1, which is forbidden so we first set s1 = 0 in the
∂d
initial equations (3.5.1) and (3.5.2) and again set = 0 which yields the value
∂x2
1 3
of x2 that minimizes d , namely, x2 = . This in turn produce x1 = and
2 2
11
d =− .
2
5. Conclusion: Condensing of the linear form (to be optimized) into a new parameter
and developing the appropriate equations containing it is a useful idea. This idea is
useful not only for linear programs but also for nonlinear programming problems and
provides new effective ways to deal with these problems.

Acknowledgements

The author is thankful to Dr. M. R. Modak for useful discussion.

References

1. Hadley G., Linear Programming, Narosa Publishing House,

New Delhi 110 017, Eighth Reprint, 1997.
2. Goldberg Jack L., Matrix Theory with Applications, McGraw-Hill,
International Editions, 1992.
3. E. M. L. Beale, Cycling in the Dual Simplex Algorithm, Naval Research
Logistics Quarterly, 2, pages 269-275, 1955.
4. Klee V., Minty G., How Good is the Simplex Algorithm?, Inequalities III, O.
Shisha (ed.), Academic Press, pages 159-175, New York, 1972.
5. Borgwardt K. H., The Average Number of Pivot Steps required by the
Simplex-Method is Polynomial, Zeitschrift fur Operations Research, 26, pages
157-177, 1982.
6. Smale S., On the Average Speed of the Simplex Method, Mathematical
Programming, 27, pages 241-262, 1983.
7. William W. Adams, Philippe Loustaunau, An Introduction to Grobner Bases,
Graduate Studies in Mathematics, Volume 3, American mathematical Society.
8. Ravindran A., Phillips D. T., Solberg J. J., Operations Research Principles and
Practice, John Wiley & Sons, Pvt. Ltd., 2000.
9. Williams H. P., Model Solving in Mathematical Programming, John Wiley &
Sons, Pvt. Ltd., 1993.

20