African Journal of Mathematics and Computer Science Research Vol. 4 (7), pp. 266-269 July 2011
Available online at http://www.academicjournals.org/AJMCSR
ISSN 2006-9731 ©2011 Academic Journals
Full Length Research Paper
A Branch and Bound approach to a state government
capital budgeting
N. P. Akpan
1
, E. H. Etuk
2
and I. D. Essi
2
*
1
Department of Mathematics and Statistics, University of Port Harcourt, Port Harcourt, Rivers State, Nigeria.
2
Department of Mathematics and Computer Science Rivers State University of Science and Technology, Port Harcourt,
Rivers State, Nigeria.
Accepted 18 April, 2011
The capital budgeting problem is a multi-constraint Knapsack problem. Hence this paper provides a
solution to a capital budgeting problem of Akwa Ibom State, Nigeria using a Branch and Bound
approach. The capital projects are summarized in four broad groups: The Economic sector ( )
1
x , The
Social Service sector ( )
2
x , The Environmental/Regional Development sector ( )
3
x , and The
Administration sector ( )
4
x . Using Tora software (version 11) for the analysis, it was observed that the
optimal solution was found in node 60 while the upper bound and the lower bound were 154.46 and
129.76 respectively. The result shows that the second and the third sector will be selected
( 1 , ; 0 ,
3 2 4 1
= = x x x x ).
Key words: Capital budgeting, branch and bound, Knapsack problem, capital projects.
INTRODUCTION
Branch and Bound method is one of the deterministic
techniques of solving a Knapsack problem. It is seen as
the most efficient tool among other deterministic methods
as it is based on a restriction of the search tree growth;
(Sedova and Seda, 2008). They also observed that
avoiding much enumeration depends on the precise
bounds (the lower and the upper bounds, the faster the
finding of the solution is). Lawler and Wood (1966)
observed Branch and Bound as algorithm for finding
optimal solutions to combinatorial problems. The method
produces convergent lower and upper bound for the
optimal solution using an implicit enumeration scheme.
Branch and bound method does not go through iterative
partial solution until an optimal solution is found like the
case of dynamic programming neither does it enumerate
all feasible solutions which may lead to 2
n
different
solutions and select the one with the highest objective
function value; brute force approach. The concept of
Branch and Bound is based on an intelligent complete
*Corresponding author. E-mail: ddmetra_utibe@yahoo.com.
enumeration of the solution space since in many cases,
only small subsets of the feasible solutions are
enumerated explicitly. It is however guaranteed that the
parts of the solution space which were not considered
explicitly cannot contain the optimal solution (Kellerer et
al., 2004). In this work, we will apply this method in a
capital budgeting problem in order to maximize the value
of some selected sectors among the other sectors. We
are using a state budget; Akwa Ibom State, Nigeria, as
our case study for the period of ten years. We are
considering the entire Capital budget in terms of their
sectors grouping. Each sector has various areas in it. The
sectors are, the Economic Sector, the Social Services
Sector, the Environment/Regional Development sector
and the General Administrative sector. The problem is a
multidimensional Knapsack Problem and the model will
be stated as we progress.
Assumptions
(I)All the projects are independent of each other.
(ii) All the sectors have equal importance to the