International Journal of Computer Applications Technology and Research

Volume 3– Issue 8, 535 - 535, 2014

Role of Bisection Method

Chitra Solanki Pragati Thapliyal Komal Tomar

DIT University DIT University DIT University
Dehradun, India Dehradun, India Dehradun, India

Abstract-: The bisection method is the basic method of finding a root. As iterations are conducted, the interval gets halved. So method
is guaranteed to converge to a root of “f” if “f” is a continuous function at an interval [a,b] and f(a) and f(b) should have opposite sign.
In this paper we have explained the role of bisection method in computer science research. we also introduced a new method which is
a combination of bisection and other methods to prove that with the help of bisection method we can also develop new methods. It is
observed that scientists and engineers are often faced with the task of finding out the roots of equations and the basic method is
bisection method but it is comparatively slow. We can use this new method to solve these problems and to improve the speed.

Key words: continous, absolute error, Iteration, convergence, Newton-Raphson method, Regular- Falsi method

1. Introduction Theorem
Traditional iterative schemes such as Newton’s method and
related classes of algorithms [3,4] often fail to converge to a An equation f ( x)  0 , where f (x) is a real continuous
specific periodic orbit since their convergence is almost
independent of the initial guess. Moreover, these methods
function, has at least one root between x and xu if
are affected by the imprecision the mapping evaluations. It f ( x  ) f ( xu )  0 (See Figure 1).
may also happen that these methods fail due to the
nonexistence of derivatives or poorly behaved partial
derivatives [3,4]. Recently, this method has been applied Note that if f ( x ) f ( xu )  0 , there may or
successfully to various difficult problems; see, for example,
[7–11]. One of the first numerical methods developed to may not be any root between x and xu (Figures 2 and 3).
find the root of a nonlinear equation f ( x)  0 was the If f ( x ) f ( xu )  0 , then there may be more than one
bisection method (also called binary-search method)[1].
Since the method is based on finding the root between two root between x and xu (Figure 4). So the theorem only
points, the method falls under the category of bracketing
methods. Since the root is bracketed between two points, guarantees one root between x and xu .

x and xu , one can find the mid-point, xm between x

and xu . This gives us two new intervals

f (x)
2. THE GRAPHICAL DISCRIPTION-:

What is the bisection method and what is it based on?

One of the first numerical methods developed to find the root
of a nonlinear equation f ( x)  0 was the bisection method
(also called binary-search method). The method is based on
the following theorem. [1]

What is the use of bisection method :

xℓ
 It is used in computer science research to analyze xu
safeguard zero finding methods
 It is simplest of other all methods
 We can safeguard bisection to detect cases where
we don’t have any roots Figure 1 At least one root exists between the two points if
the function is real, continuous, and changes sign.

www.ijcat.com 533
 International Journal of Computer Applications Technology and Research
Volume 3– Issue 8, 535 - 535, 2014

f (x)
f (x)

xℓ xu xu
xℓ

Figure 2 If the function f (x) does not change sign

between the two points, roots of the equation f ( x)  0 may
Figure 4 If the function f (x) changes sign between the
still exist between the two points.
two points, more than one root for the equation f ( x)  0
may exist between the two points.
f (x)

3. PROBLEM DESCRIPTION:- The

bisection method guarantees a root (or singularity) and
is used to limit the changes in position estimated by the
Newton-Raphson method when the linear assumption is
poor. However, Newton-Raphson steps are taken in the
nearly linear regime to speed convergence.
ℓ
x In other words, if we know that we have a root
xℓ xu bracketed between our two bounding points, we first
consider the Newton-Raphson step. If that would
predict a next point that is outside of our bracketed
range, then we do a bisection step instead by
choosing the midpoint of the range to be the next
point. We then evaluate the function at the next
Figure 3 If the function f (x) does not change sign point and, depending on the sign of that evaluation,
between two points, there may not be any roots for the replace one of the bounding points with the new
equation f ( x)  0 between the two points. point. This keeps the root bracketed, while allowing
us to benefit from the speed of Newton-Raphson.
Wrong assumption of Newton-Raphson method can
increase no. of iterations.

An improved root finding scheme is to combine the

BISECTION and REGULAR-FALSI methods.It is
relatively faster then bisection method.

4. RELATED WORK:-
we first analyzed some of the conventional root finding
methods and their limitations. Bisection always converges
but is slow. Newton has quadratic convergence but may fail
in some of the cases. Secant is a good alternative to Newton
but it oscillates in some of the cases and fails to converge.
 It is explained that it is important that we
safeguard bisection to detect cases where

www.ijcat.com 534
 International Journal of Computer Applications Technology and Research
Volume 3– Issue 8, 535 - 535, 2014

we don’t have any roots. The question of Table 1 Comparison

guessing the bound is more intuitive.
 The other method like Newton’s method S.No. Method name No. of iterations
have a disadvantage that higher order roots 1 BISECTION 14
can cause convergence to be slow,and the METHOD
sequence may take undesirable jumps 2 REGULAR-FALSI 5
between roots or take a very large step upon METHOD
encountering an reflection point. One case 3 NEWTON 5
where it fails is when derivative of function RAPHSON
f(x) is either zero or infinite then it fails to METHOD
converge. 4 NEW METHOD 6
 We have proposed a new method by
combining Bisection method with other
methods. So, that we can find roots as well
as the method can be fast in solving.
6. CONCLUSION:
Bisection method is the safest and it always converges. The
 The multidimensional bisection method
bisection method is the simplest of all other methods and is
allows to solve constrained minimization
guaranteed to converge for a continuous function. It is
problem when the feasible region is n-
always possible to find the number of steps required for a
dimensional simplex. This method does not
given accuracy.and the new methods can also be developed
require a differentiability of function and is
from bisection method and bisection method plays a very
guaranteed to converge to the minimize for
crucial role in computer science research.
the class of strictly unimodal function[12]

7. REFERENCES :
5. PROPOSED-METHOD
xi+1=3xi-1f(xi)-xi-1f(xi-1)+xif(xi)-3xif(xi-1) /4[f(xi)-f(xi-1)]
[1] Chapter 03.03 Bisection Method of Solving a Nonlinear
Algorithm for this new method: Equation
The steps to apply the new method to find the root of the
equation Choose xi-1 and xi as two guesses for the root such [2] Book numerical based analysis from DITU library
that f(xi) f(xi-1)<0, or in other words, f (x) changes sign
[3] J.M. Ortega, W.C. Rheinbolt, Iterative solution of
between xi-1 and xi. nonlinear equations in several (1970)

I. Estimate the root lies between xi-1 and xi. [4] J.E. Dennis, R.B. Schnabel, Numerical Methods for
II. xi+1=3xi-1f(xi)-xi-1f(xi-1)+xif(xi)-3xif(xi-1) Unconstrained Optimization and Nonlinear Equations, SIAM,
/4[f(xi)-f(xi-1)] Philadelphia, 1996
III. Now check the following
a) If f(xi+1)<0; then xi-1=xi+1 and the root lies [5] L. Drossos, O. Ragos, M.N. Vrahatis, T.C. Bountis, Phys.
between xi+1 and xi. Rev. E 53 (1996) 1206.
b) If f(xi+1)<0; then xi-1=xi+1 and the root lies
between xi and xi+1. [6] M.N. Vrahatis, T.C. Bountis, M. Kollmann, Inter. J.
c) If, Bifurc. Chaos 6 (1996) 1425.
new xi-1 and xi are same then previous one
then stop and the solution will be xi-1 + xi / 2 [7] M.N. Vrahatis, H. Isliker, T.C. Bountis, Inter. J. Bifurc.
else Chaos 7 (1997) 2707.
goto step I.
[8] H. Waalkens, J. Wiersig, H.R. Dullin, Ann. Phys. 260
(1997) 50.

Comparisons table of new method with existing methods:- [9] N. Buri´c, M. Mudrini´c, J. Phys. A: Math. Gen. 31 (1998)
1875.
For a given problem f(x)=x3 – 7.[2]The comparison is done
between four methods the new method is as faster as Newton- [10] N. Buri´c, M. Mudrini´c, Todorovi´c, J. Phys. A: Math.
Raphson method and Regular -Falsi method and also accurate Gen. 31 (1998) 7847.
as we don’t take any guess
[11] V.S. Kalantonis, E.A. Perdios, A.E. Perdiou, M.N.
Vrahatis, Celest. Mech. Dynam. Astron. (2001), in press.

[12] A multidimensional bisection method for unconstrained

minimization problem

www.ijcat.com 535