Upload
rindangsukmanita
View
232
Download
0
Embed Size (px)
Citation preview
7/27/2019 jurnal GPsepty
1/11
R jeas Research Journal in Engineering and Applied Sciences 2(2) 95-105 R jeas Emerging Academy Resources (2013) (ISSN: 2276-8467)
www.emergingresource.org
95
GOAL PROGRAMMING: - AN APPLICATION TO BUDGETARY ALLOCATION
OF AN INSTITUTION OF HIGHER LEARNING
Ekezie Dan Dan And Onuoha Desmond .O.Department of Statistics,
Imo State University, PMB 2000, Owerri Nigeria.
Corresponding Author:E kezieD an D an__________________________________________________________________________________________
In this paper we discuss the application of Goal Programming in Budgetary allocation of Institutions of higher
learning using Imo State University Owerri as a case study. This paper shall impact positively to the readingcommunity in that, it will enlighten the masses that with the use of Goal programming problems, the aims of an
organization can be achieved financially and otherwise. Data on the budget estimates of Imo State UniversityOwerri were collected from Imo State Ministry of Finance and Imo State University Bursary Department
between 2006 and 2008. Five goals in the budget estimates of the University namely; Personal cost, Overhead
cost, Capital expenditure, Revenue (internally generated) and the Total budget were considered for the study inorder of precedence (priorities). The data collected were used to formulate a goal programming problem and theformulated problem was used solved using Simplex method (using TORA software). Based on the solution
obtained, it was discovered that the optimum value of Z (Z=4.24) satisfied goal 1(the personal cost goal), goal
3(capital expenditure goal) and goal 5(the total budget goal), but failed to satisfy goal 2 and goal 4, which are
Overhead cost and Revenue goals respectively. From the findings, it was concluded that the Imo StateUniversity, Owerri should come within 4.24 billion naira to satisfy goal 2 and goal 4, which are Overhead cost
and Revenue goals respectively. It was also recommended that the University should have a minimum budget of4.24 billion naira in 2010 and should be reviewed upward annually, which should be properly and timely
monitored by active Government budget monitoring team.
Emerging Academy Resources
KEYWORDS: Goal Programming, Budgetary Allocation, Aspiration level and Goal Programming Algorithm.
__________________________________________________________________________________________INTRODUCTION
Budgetary allocation is a complex task that requires
cooperation among multiple functional units in anyinstitution. There is need to have a sound knowledge
of organizational budgeting processes in order todesign an efficient and effective budgetary allocation
model. Despite the fact that such allocation procedure
exists in the University, it is not properly structureddue to the presence of multiple conflicting objectives.
A formal decision analysis that is capable ofhandling multiple conflicting goals through the use
of priorities is the Goal Programming Model.
Goal programming (GP) is an extension of Linear
Programming (LP) which is a mathematical tool tohandle multiple, normally conflicting objectives.
For example,(i) The Vice Chancellor of any NigerianUniversity may decide to increase capital
expenditure and simultaneously reduce
revenue.
(ii) A Governor of any Nigerian State maypromise to reduce the State debt andsimultaneously offer tax relief to workers.
In such situations, it will be difficult to find a single
solution that optimizes the conflicting objectives.
Goal Programming provides a way of strivingtoward such conflicting objectives simultaneously.
According to Ignizio (1978), Goal Programming is a
tool that has been proposed as a model and approachfor analysis of problems involving multiple
conflicting objectives.
He pointed out that actual real world problems
invariably involve non-deterministic system forwhich a variety of conflicting non-commensurable
objectives exist.
The major strength of Goal Programming is its
simplicity and ease of use. This accounts for thelarger number of Goal Programming applications in
many and diverse areas such as in management ofsolid waste, accounting and financial aspect of stock
management marketing, quality control, human
resources, production, transportation and site
selection, space studies, agriculture,
telecommunication, forestry and aviation (Aouni andKettani, 2001).
Goal programming problems can be solved by widelyavailable linear programming computer packages as
either a single linear programming, or in the case of
7/27/2019 jurnal GPsepty
2/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
96
lexicographic variant, a series of connected linearprogramming. Goal programming can therefore
handle relatively large number of variables,constraints and objectives.
TYPES OF GOALS
There are three possible types of goals:
A lower, one-sided goal: - This goal sets a lowerlimit that we do not want to fall under (but
exceeding the limit is acceptable).
An upper, one-sided goal: - This goal sets anupper limit that we do not want to exceed (butfalling under the limit is acceptable).
A two-sided goal: - This goal sets specific targetsthat we do not want to miss on either side.
VARIANTS
Goal programming formulations ordered theunwanted deviations into a number of priority levels,
with the minimization of a deviation in a higher
priority level being of infinitely more importance
than any deviation in lower priority levels. This is
known as lexicographic or preemptive goalprogramming.
Ignizio (1976) gave an algorithm that shows how a
Lexicographic Goal Programming (LGP) can besolved as a series of linear programming model.
Lexicographic Goal Programming should be used
when there is a clear priority ordering amongst thegoals to be achieved.
BUDGETING
Revenue budgeting is an approach to the budget
decision rather than a particular budget system.However, revenue budgeting emphasizes on the
ascendancy of the revenue constraint in budgetingcalculations. Decision makers are constrained by
actual limitation on revenue raising power and/or theperception of impending limitations and fears aboutthe revenue sources.
The aims are:
To increase personnel cost (salary andallowance of staff).
To reduce overhead cost; To increase capital expenditure; To increase revenue (internally generated); To reduce the total budget.
STATEMENT OF THE PROBLEM:
The statements of the problem are as follows:
Capital and revenue were allocatedinadequately, and without order ofimportance. This inadequate allocation was
due to not using powerful quantitativeallocation models.
It is observed that allocated funds were notproperly utilized with the result that moneyallocated for the laboratories for example is
diverted into hostel maintenance. In thesame vain, other diversions also take place.
It is not a hidden fact that, the fundsallocated to tertiary institutions are often
mismanaged. Imo State University, Owerri
is not an exception. This retarded the growthof the institution.
There is no active budget monitoring teamwith the result that budgets are allowed to
operate any how. If there were active budget
monitoring teams the problems ofmismanagement and improper utilization
would be reduced.
THE OBJECTIVES OF THE STUDY
The objectives of the study are as follows:
To apply goal programming model to a real-life budgeting situation to find acompromise solution among the different
conflicting goals of the Imo StateUniversity, Owerri.
To minimize the total weights associatedwith meeting the annual budget
requirements of the institution.
SIGNIFICANCE OF THE STUDY
The insight gained from this study will: Guide and assist decision makers of the
institution in achieving the institution goalsof optimum utilization of funds in improving
the institution;
Guide the institution in budgeting; Help the institution to forecast its budget
annually; Assist in Optimization/Operational Research
students for further research.
LIMITATION OF THE STUDY
The study is limited to the budgetary allocation ofImo State University, Owerri. The budget estimates
of the University were used for the study. It is alsolimited to Goal Programming problems.
LITERATURE REVIEW
Multi-Objective and Multi-Criteria Decision
Analysis
Sundaram (1978) and Lee (1972) affirmed that the
distinguishing characteristic of goal programming isthat it allows for ordinal solution.
Ignizio (1978) states that, this approach, known as aGoal Programming, although far from a panacea;often represents a substantial improvement in the
modeling and analysis of multi objective problems.
This presents state of the art in this field now permitsthe systematic analysis of a class of multi-objective
problems that may involve either linear or non-linearfunctions in continuous or discrete variables. Further,
the goal programming model provides a relatively
reasonable structure under which the traditional,
7/27/2019 jurnal GPsepty
3/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
97
single objective tools may be viewed simply asspecial cases.
Sundaram (1978) argued that goal programming
combines the logic of optimization in mathematical
programming with the decision makers desire tosatisfy several goals. According to Winston (1994);
Goal Programming is the technique used to chooseamong alternative optimal solutions.
According to Winston (1994), Suppose a decisionmaker has an additive linear cost function of the
form: C(x1, x2,, xn) = c1x1 + c2x2 + + cnxn.
A decision maker with this type of cost function canoften use goal programming to determine his
decision. He observed that a cost function of theabove form defines the same trade off between each
pair of attributes xi andxj.
Chowdary and Slomp (2002) considered goal
programming as an appropriate powerful and flexible
technique for decision analysis of the troubled
modern decision maker who is burdened with
achieving multiple conflicting objectives undercomplex environmental constraints.
The extensive surveys of goal programming by
Schniederjans (1995), Tamiz, et al (1998), and Aouniand Kettani (2001), have reflected this. Therefore,
goal programming model handles multiple goals in
multiple dimensions.
Taha (2003) also said that goal programmingtechnique is for solving multiple-objective models
and the aim is to convert the original multiple
objective into a single goal. Tipparate (2005) statedthat goal programming extended itself by
reengineering many of the prior single objectivelinear programming with multiple and /or conflicting
(traded-off) objectives.
Wikipedia (2006) described goal programming as a
branch of multiple objective programming (MOP),which in turn is a branch of multi-criteria decision
analysis (MCDA), also known as multiple criteria
decision making (MCDM). It is also thought of as a
generalization of linear programming to handlemultiple conflicting objective measures. Each of
these measures is given a goal or target value to be
achieved.
Managerial and Financial Concepts
The first publication using goal programming was
formulated by Charnes, et al (1955), applied to
constrained regression. That application appliedminimization of the deviational variables as a means
to apply least absolute value regression. The reasonfor that application was that the goal programming
formulation gave the ability to constrain results to
conform to the managerial salary requirement. Thisminimized the difference between market offers and
the competitors. This idea has developed into afruitful research area, least absolute estimation.
The initial development of the goal programming was
due to Charnes and Cooper (1961), in a discussion of
which appeared in their text, although they claim thatthe idea actually originated in 1952.
They proposed a model and technique for dealing
with certain linear programming problems in which
conflicting goals of Management were included asconstraints.
Contini (1968) proposed stochastic goal
programming (SGP) model. He considered the goalsas uncertain variables with normal distribution, that
the model maximizes the probability that theconsequence of the decision will belong to a certainregion encompassing the uncertain goal. Thus, this
model tries to generate a solution that is close to theuncertain goals.
Keown (1978), Keown and Taylor (1980), proposed
an adoption of the chance constrained for stochastic
goal programming model and transforms the modelto a non-linear program, and applied it to budget
allocation production problem.
Dade (1981) applied goal programming in revenuebudgeting. He pointed out that the concern with the
process and outcome of the budgeting has regulated
the revenue constraint to a position co-equal withvariables.
Martel and Price (1981) have developed a stochastic
goal programming model for human resource
planning.
De, et al (1982), proposed a chance constrained goalprogramming model for budgeting in the production
area. They considered the following constraint as amathematical programming model:
m
aijxj bi i = 1, 2, .. . , n, j = 1, 2, . . . , m
j=1
xj are the decision variables.
aij are the corresponding constraints and bi
are the right hand side elements.
Helm (1984), Chanchit and Terrell (1993), Nembouand Murtagh (1996), Abdelaziz and Mejri (2001),Easton and Rossin (1996) proposed a stochastic goal
programming model for employee schedulingproblem. Their model integrated the labour
requirements and scheduling decision. They pointedout that this approach allowed reaching satisfactory
result.
Lee and Olson (1985), on the other hand, proposed an
algorithm of deterministic optimization for stochasticcase. This is taken into account in 1992, and the
7/27/2019 jurnal GPsepty
4/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
98
problem of optimal utilization of resources andreservoir operation problem.
Chanchit and Terrell (1993), proposed a model of
stochastic goal programming for the management of
hydraulic resources. This mathematical modelproposed reflects three important characteristics of
the reservoir management problem: Multiple objectives; Stochastic inflow; Large-scale nature of the reservoir systems.
The proposal of Chanchit and Terrell (1993) enabledAbdelaziz and Mejri (2001) to propose a Chance
Constrained Goal Programming (CCGP) model todeal with the operation and the control of Multi-
purpose reservoir system. They suggested a Multi-objective approach to solve the problem consideringthe release for drinking and irrigation purposes, the
pumping cost and the salinity of the water.
Abdulaziz and Majri (2001) considered the control of
drinking water supply goal as a random variable.
Other recent applications of various formulations of
goal programming have been presented by Carrizosaand Romero (2001) and Leung (2001).
Lam and Moy (2002) reviewed the intersection of
goal programming and Discriminant Analysisbeginning with Freed and Glover (1981).
The Concept of Satisfaction Functions
The concept of satisfaction functions in the goal
programming model was introduced by Martel andAouni (1990). Through the satisfactions, the decision
maker may express his/her preferences with regard to
the deviations associated with the goals set for eachobjective.
The concept of satisfaction functions was also used
for modeling the impression related to the value ofthe goals. These are expressed through the intervalsof which the upper and lower limits are set by the
decision maker.
Martel and Aouni (1998) and Aouni (1996)
concluded that the satisfactory functions have a
double role.
Rogers and Bruen (1998) explained the decision
makers sensitivity to the deviations observedbetween the set goal and the level achieved on theother hand. They took into account the imprecision or
uncertainty associated to the value of the goal.
Technology and Planning
The first engineering application of the goal
programming, due to Ignizio in 1962, was the designand placement of the antennas employed on the
second stage of the Saturn V the vehicle used to
lunch the Apollo space capsule which landed the firstmen on the moon.
Stam and Kuula (1991) applied goal programmingtechnique for selecting a flexible manufacturing
system. They, identified production volume, cost andflexibility as multiple objectives.
Shafer and Rogers (1991) used goal programmingmodel for formation of manufacturing cells.
Minimum setup time, minimum intercellularmovements, minimum investment in new equipment
and maintaining acceptable utilization levels were
identified as the multiple objectives.
Premchandra (1993) used goal programming methodfor the activity crashing in project networks. The
multiple objectives identified were as follows: Crashing time of an activity Project cost Project time
Lyu et al (1995) used a goal programming model forwhich the multiple objectives were as follow:
Demand of each boiler; Environmental requirement of the sulphur
oxide emission;
Heating value requirements; Volatile matter requirement; Inventory.
Goal programming approach has been applied for avariety of applications. Problems solved by goal
programming technique are as follows: production
planning in a ship repair company (Tabucanon andMajumder 1989) and production smoothing under
just-in-time manufacturing environment (Kim andSchnierderjans 1993).
Kalpic, et al (1995) successfully applied linear goalprogramming technique for planning objectives
which include calculating the optimum productionMix and achieving the capacity and material balance
while maximizing the contribution and minimizingthe duration of the longest resources management.
Nagarur, et al (1997) illustrated the use of zero-onegoal programming for the development of a
production planning and scheduling model in an
injection molding factory with an objective to
minimize the total cost of production, inventory andstorages.
REVIEW OF RELATED VARIANTS ANDFORMULATIONS/SOLUTIONSDykstra (1984) described that an inferior solution is
not always a problem as long as the model is correct.
It is a problem if the model does not reflect the
decision makers actual preference. In some cases,non-inferior solution may not be intrinsically superior
to an inferior solution especially in the case of naturalresources problem.
Dykstra (1984) argued that the initial run with uniqueweight can be biased, although this technique makes
7/27/2019 jurnal GPsepty
5/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
99
weighting more defensive than a subjectiveweighting. The goal programming solution has a high
tendency to be inferior (dominated) with respect totechnological constraints. To deal with this drawback,
Field et al (1980) presented a technique to ensure
non-inferior goal programming solutions. Thetechnique is to apply goal programming and linear
programming as a complementary package.
Winston (1994) stated that; pre-emptive goal
programming problems can be solved by anextension of the simplex known as the goal
programming simplex. To prepare a problem forsolution by the goal programming simplex, we must
compute in rows with the ith
row corresponding togoal i.
Nembou and Murtaph (1996) presented a chanceconstrained goal programming approach to solve a
hydro-thermal electricity generation planningproblem in Port Moresby, Papus New Guinea.
Their model considers explicitly the uncertainty and
seasonal variation inherent in hydro-thermal
electricity demand and supply.
Tamiz et al (1998) suggested straight restoration,
preference based restoration and interactiverestoration which employ cardinal weightings in
lexicographic goal programming to ensure non-inferior solutions.
Tamiz et al (1998) mentioned the case whereby aninefficient objective and an unbounded objective
causes the entire model inefficient and unbounded.
Hillier and Lierberman (2001) said that Goal
programming problems can be categorized accordingto the type of mathematical programming model that
it fits except for having multiple goals instead of asingle objective.
In the case of non pre-emptive goal programming, allthe goals are of roughly comparable importance but
in the case of pre-emptive goal programming, there isa hierarchy of priority levels for the goals.
Hillier and Lierberman (2001) concluded that goal
programming and its solution procedures provide aneffective way of dealing with problems where
management wishes to strive towards several goals
simultaneously. The key is a formulation technique ofintroducing auxiliary variables that enable convertingthe problem into a linear programming format.
Hillier and Lierberman (2001) stated that to complete
the conversion of the goal programming problem to alinear programming model, we must incorporate the
definitions of the deviational variables di+
and di-
directly into the model because the simplex method
considers only the objective function and the
constraints that constitute the model.
According to Romero and Rehman (2003) BothLexicographic Goal Programming (LGP) and
Weighted Goal Programming (WGP) are best knownand widely used as goal programming variants.
However, the variants lie heavily on the great amount
of information that this goal targets, weights as wellas pre-emptive ordering of preferences. These
requirements can cause possible weakness if thedecision makers are not confident of the value of
these parameters.
In many cases, the decision makers could not specify
meaningful information on weights, or the goals areunrelated to each other so that it is not possible to
drive objectively measured preference weights. If it isthe case, the relative value of the shadow prices of
constraints from the unique weight goal programmingrun can be used to drive the real preference weights.
Diaz-Balteiro and Romero (2003) affirmed thatweighted goal programming (WGP) and
lexicographic goal programming (LGP) can be mixed
in a model; for example, weighted lexicographic and
min-max lexicographic goal programming.
Taha (2003) also contributed that the weights and the
pre-emptive methods convert the multiple goals intoa single objective function stating that these methods
do not generally produce the same solution. Neithermethod, however, is superior to the other because
each technique is designed to satisfy certain decision-
making preferences.
RESEARCH METHODOLOGY/DATA
COLLECTION
Goal Formulation
Let fi(x) be the mathematical representation of theobjectives which can be linear or nonlinear (usually
linear). Let gibe the aspiration level, three possiblegoals are
(i) fi(x) gi(ii) fi(x) gi(iii) fi(x) = gi
In regular Linear Programming, these would be hard
constraints but in Goal programming, we measure the
deviation from the goal.
Goal Programming Formulation
The general form of the goal programming model is
given byMinimize Z = (di
++ di
-),
im
Subject to
. . . (1)
n
(aij xj + di-- di
+) = bi
(i = 1,2,,m)j = 1
(j = 1,2,,n)
di+, di
-, xj 0,
7/27/2019 jurnal GPsepty
6/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
100
where di+
is the positive deviation variable fromoverachieving the i
thgoal;
di-
is the negative deviation variable fromunderachieving the i
thgoal and
xj is the decision variables;
aij is the decision variable coefficients;
It is pertinent to note here that; di+
, di-
are calleddeviational variables because they represent the
deviations above and below the right hand side of the
constraint i.
The deviational variable cannot be basic variablessimultaneously, because by definition they are
dependent. This implies that in any simplex iteration,at most, one of them can assume a positive value. If
the original ith
inequality is of the form and its di+
0, then the ith
goal will not be satisfied.
However, di+
and di-allow us to meet or violate the i
th
goal at will. A good compromise solution aims atminimizing the amount by which each goal is
violated.
Goal Programming Algorithms
There are two algorithms for solving goal
programming problems. These algorithms convert the
multiple goals into a single objective function.The methods are:
The weights method; The Preemptive method;
The Weights Method
In the weights method, the single objective function
is the weighted sum of the functions representing thegoals of the problem.
The weights goal programming model is of the form;
n
Minimize Z = {(wi+
+ wi-)di},
i=1
Subject to
. . . (2)m
(aijxj + di+
di-) = gi
j=1
(i = 1, 2, . . ., n)
(j = 1, 2, . . . , m)di
+, di-, xj 0,
where wi+
and wi-
are non-negative constraints and
can be real numbers representing the relative weightsassigned within a priority level to the deviational
variables.
The parameter wi+
represents positive weights thatreflect the decision makers preference regarding the
relative importance of each goal; while wi-
is the
negative weights of the decision makers preference.The determinant of the specific values of these
weights is subjective.
The Pre-Emptive Method
In this method, the decision maker must rank the
goals of the problem in order of importance. Themodel is then optimized using one goal at a time such
that the optimum value of a higher priority goal is
never degraded by a lower priority goal. This variantis known as lexicographic goal.
The proposed pre-emptive model is given as
n
Minimize Z = pi (di+
+ di-),
i=1
Subject to
. . . (3)m
(aijxj + di+
di-) = gi
j=1
(i = 1, 2, . . ., n)(j = 1, 2, . . . , m)
di+, di
-, xj 0,
where pi is the preemptive factor/priority level
assigned to each relative goal in rank order (that is p1> p2 > > pn).
The weights goal programming and the preemptive or
lexicographic goal programming can be combined in
a model, for example weighted lexicographic goalprogramming. The weights and rank model according
to Kwak et al (1991) is given byn n
Minimize Z = Pi{wik+di
++ wik
-di
-}
i=1 i=1
Subject to
. . . (4)m
(aijxj + di- di
+) = gi
j=1
(i, k = 1, 2, . . . , n)(j = 1, 2, . . . , m)
di+, di
-, xj 0
where di+ and di
- are deviational variables;
xj = decision variables;
aij = decision variable coefficients;
wi+
and wi-are non-negative constraints representing
relative weights;
pi = preemptive factor/priority level assigned to each
relevant goal in rank order (that is p1 > p2 > > pn).
THE BASIC STEPS IN FORMULATING GOAL
PROGRAMMING MODEL
The basic steps in formulating a goal programmingmodel are as follows:
(i) Determine the decision variables;(ii) Specify goals including goal types (one-
way or two-way goal) and their targets;(iii) Determine the pre-emptive priorities;(iv) Determine the relative weights;(v) State the minimization objective
functions of the deviation;
7/27/2019 jurnal GPsepty
7/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
101
(vi) State other given requirements,example, technological constraints, non-
negativity (linear goal programming);(vii) Finally, make sure that the model can
exactly specify the decision makers
preferences.
SOURCES OF DATA COLLECTION FORANALYSIS
The data used in this research are of the secondary
type as it exists in the published budgets andunpublished budget folder.
The data for this study were collected from the ImoState Ministry of Finance, State Secretariat, Owerri
and Bursary Department of Imo State University,Owerri.
DATA ANALYSIS TECHNIQUEIn this study, the goal formulation and weights goal
programming methods were used.
The simplex method (BigM) by Tora package was
used to analyze the weighted goal programmingformulation.
PRESENTATION AND ANALYSIS OF DATA
This chapter presents and analyzes the data obtained
from the Imo State Ministry of Finance and Imo StateUniversity, Owerri, as indicated above. The five
conflicting goals in the budget estimates of the
institution over the period of three years (2006 2008), were considered.
Summary Of The Budget Estimates Over The
Three Years (2006 2008)
Table 1 gives the budget estimates summary of the
institution over the period from 2006 2008,showing the personnel cost, overhead cost, capitalexpenditure and revenue.
Table 1 Outline of the budget estimates for the three yearsAllocation in Naira/Year
ITEM (GOAL) 2006 2007 2008 TOTAL
Personnel cost 1,189,825,252 1,570,289,718 1,669,380,200 4,429,495,170
Overhead Cost 313,658,446 433,304,000 590,241,950 1,337,204,396
Capital Expenditure 863,000,000 1,019,894,886 803,306,000 2,686,200,886
Revenue (Internally Generated) 316,732,540 529,043,360 728,722,210 1,574,498,110
Total 2,683,216,238 3,552,531,964 3,791,650,360 10,027,398,562
Coded Budget Estimates over the Period of
Three Years (2006 2008)
Table 2 gives the coded budget estimates of the
personnel cost, overhead cost, capital expenditure andrevenue. The reason for coding this budget estimates
is to enable one work with small figures in theanalysis.
Table 2 Coded budget estimate for years 2006 2008
Allocation in Billion Naira/Year
ITEM (GOAL) 2006 2007 2008 TOTAL
Personnel cost 1.19 1.57 1.67 4.43
Overhead Cost 0.31 0.43 0.60 1.34
Capital Expenditure 0.86 1.02 0.8 2.68
Revenue (Internally
Generated)
0.32 0.53 0.73 1.58
Total 2.68 3.55 3.8 10.03
Assignment of Weights to the Goals
The objective function coefficient for the variablesassociated with the goal i is called the weight for the
goal i. The most important goal has the largestweight and so on.
Let wi be the weight for goal i, that could range
from 2, 4, 6, the most important goal has thehighest weight and so on.
Coded Budget Estimates And The AssignedWeights To The Goals
The table below gives the coded budget estimateand the assigned weights to the goals.
Table 3 Coded budget estimates with weights
Allocation in
Billion
Naira/Year
ITEM (GOAL) 2006 2007
2008
TOTAL
Wi
Personnel cost 1.19 1.5
7
1.6
7
4.43 8
Overhead Cost 0.31 0.4
3
0.6
0
1.34 2
Capital Expenditure 0.86 1.0
2
0.8 2.68 6
Revenue (Internally
Generated)
0.32 0.5
3
0.7
3
1.58 4
Total 2.68 3.5
5
3.8 10.03 10
ASPIRATION LEVEL (Target Value) of the
Goals
The goals statements of the budget of the institution
were as follows:
Goal 1: Raise Personnel Cost (Salary andallowances of staff) by at least N3 billion per
annum;Goal 2: Reduce Overhead cost by at most N1.5
billion per annum;Goal 3: Raise capital expenditure by at least N1
billion per annum;
7/27/2019 jurnal GPsepty
8/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
102
Goal 4: Raise revenue (internally generated) by atleast N4.0 billion per annum;
Goal 5: Reduce the total Budget by at least N7billion per annum.
THE GOAL FORMULATION
Let x1 = Amount budgeted in the fiscal year 2006
x2 = Amount budgeted in the fiscal year 2007x3 = Amount budgeted in the fiscal year 2008
x1, x2, and x3 are the decision variables.
The goals can be stated mathematically as follows:
1.91x1 + 1.57x2 + 1.67x3 3 (Personnel costconstraint)
0.31x1 + 0.43x2 + 0.6x3 1.5 (Overhead costconstraint)
0.86x1 + 1.02x2 + 0.8x3 1 (Capital expenditureconstraint)0.32x1 + 0.53x2 +0.72x3 4.0 (Revenue constraint)
2.68x1 + 3.55x2 + 3.8x3 7 (Budget constraint)x1, x2, x3 0.
THE GOAL PROGRAMMING
FORMULATION
Let di+
= amount by which we numerically exceedthe ith goal.
di-
= amount by which we are numerically less thanthe ith goal
di+
and di-are referred to as deviational variables.
Let Z be the weighted sum associated with meeting
the annual budget requirements.Using the weighted goal programming model stated
in (2), the goal programming formulation can be
mathematically stated as follows:Min Z = 8d1
++ 2d2
-+ 6d3
++ 4d4
++ 10d5
-(objective
function) . . . (i)Subject to:
1.19x1 + 1.57x2 +1.67x3 + d1-- d1
+= 3 (ii)
0.31x1 + 0.43x2 + 0.60x3 + d2-- d2
+= 1.5 (iii)
0.86x1 + 1.02x2 + 0.8x3 + d3-- d3
+= 1 (iv)
0.32x1 + 0.53x2 + 0.72x3 + d4-- d4
+= 4 (v)
2.68x1 + 3.55x2 + 3.8x3 + d5-- d5
+= 7 (vi)
x1, x2, x3, d1+, d2
+, d3
+,d4
+, d5
+, d1
-, d2
-, d3
-, d4
-, di
- 0.
The Input Data
Table 4, gives the input data for the analysis of
budgetary allocation in Imo State University,
Owerri from 2006 2008 inclusive.Table 4
Basic variables x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 x13 RHS
Name of variables d1+
d2+
d3+
d4+
d5+
d1-
d2-
d3-
d4-
d5-
Minimize Z 0 0 0 8 0 6 4 0 0 2 0 0 10
Constraint 1 1.19 1.57 1.67 -1 0 0 0 0 1 0 0 0 0 = 3
Constraint 2 0.31 0.43 0.60 0 -1 0 0 0 0 1 0 0 0 = 1.5
Constraint 3 0.86 1.02 0.80 0 0 -1 0 0 0 0 1 0 0 = 1
Constraint 4 0.32 0.53 0.72 0 0 0 -1 0 0 0 0 1 0 = 4
Constraint 5 2.68 3.55 3.80 0 0 0 0 -1 0 0 0 0 1 = 7
Footnote
In the above input data,Let x4 = d1
+, x5 = d2
+, x6 = d3
+, x7 = d4
+and x8 = d5
+.
Let x9 = d1-, x10 = d2
-, x11 = d3
-, x12 = d4
-and x13 = d5
-.
The Tora software was applied on Table 4 to obtain
the table below.
Interpretation Of The Solution
The application of the simplex method (Big M Method) by Tora package, gives the optimum
solution as follows:
Z = 4.24, x1 = 0, x2 = 0, x3 = 1.84d1
+= 0.08, d2
+= 0, d3
+= 0.47, d4
+= 0, d5
+= 0, d1
-= 0,
d2- = 0.39, d3
- = 0, d4- = 2.66 and d5
- = 0.
Since the value of Z is not equal to zero, the solutionsatisfies goal 1, goal 3, and goal 5, but fails to satisfygoal 2 which is the Overhead cost and goal 4 which is
the Revenue goal. Predominantly, for d2-
= 0.39, itmeans that overhead cost level (target) of 1.5 billion
naira has a shortfall of 0.39 billion naira in theOverhead cost; which indicates that the actual
overhead cost should be 1.89 billion naira, and for d4-
= 2.66, it means that the Revenue goal level (target)
of 4 billion naira exceeded the Revenue goal by 2.66
billion naira; which indicates that the actual revenueshould be 1.34 billion naira.
On the other hand, the budget goal of at least 7 billion
naira is not violated as d5- = 0.
DISCUSSION OF FINDINGS
The study examined the problem of conflicting
goals in budgeting situation in Imo State University,
Owerri.
Each of the inequalities in the goal formulationrepresents a goal that the institution aspires to
satisfy. However, compromised solutions among the
conflicting goals were found, and the approach tofinding a compromise solution is to convert each
inequality into flexible goal in which the
corresponding constraints may be violated.
The simplex method (BigM Method) was used tosolve the weighted goal programming model
formulated, and the optimal solution obtained byTORA package is Z = 4.24, x3 = 1.84, d1
+= 0.08, d3
+
= 0.47, d2-
= 0.39, d4-
= 2.66, all the remainingvariables are zero.
Since the optimum is not equal to zero, this
indicates that at least one of the goals is not
satisfied. Z is the weighted sum associated withmeeting the annual budget requirements.
7/27/2019 jurnal GPsepty
9/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
103
All but goal 2 and 4 were fully satisfied, which is
the overhead cost and the revenue goalsrespectively. Specifically, d2
-= 0.39; shows that
Overhead cost level (target) of 1.5 billion naira has a
shortfall of 0.39 billion naira and d4- = 2.66, indicates
that the revenue goal level (target) of 4 billion
exceeded the revenue goal by 2.66 billion. On thecontrary, the budget goal of at least 7 billion target is
not violated as d5-= 0.
The Z value of 4.24 billion indicates that if goal 1, 3
and 5 were satisfied, the institution should comewithin 4.24 billion naira budget in order to meet goal
2 and 4, which are Overhead cost and Revenue goalsrespectively.
Hence, the minimum budget of the University shouldbe 4.24 billion naira in the next fiscal year 2010 and
should be reviewed upward annually.
CONCLUSION
This study, examined the budgeting system of Imo
State University using goal programming model.
The results demonstrated that all the goalsformulated were met, except the overhead cost and
revenue target.
The Universitys minimum budget should be 4.24billion naira to meet goal 2 and 4 which are the
overhead cost and the revenue goals. Optimistically,
it can be said that the University has not performedbelow expectation, the institution should continue
with their budget allocation formula with increasedadaptation to new scientific techniques.
RECOMMENDATIONBased on the findings of this study, the following
recommendations are made:(1)The personnel cost should be raised by at least
3 billion. The overhead cost should beincreased to at most 0.39 billion.
(2)The capital expenditure should be increased byat least 1 billion.
(3)The minimum budget of the University shouldbe 4.24 billion per annum and should be
reviewed upward annually.
(4)Budgeting (capital and resources allocation)should be based on new technology (preferable
goal programming model).
(5)The budget should be properly managed andutilized.
(6)Timely and active budget monitoring teamshould monitor the budget of the institution.
(7)In all, the minimum budget of 4.24 billionshould cover all the University needs.
REFERENCES
Abdelaziz, B.F and Mejiri, S. (2001). Application of
goal programming in multi-objective reservoir
operation model in Tunisia. European Journal OfOperational Research 133(2), 352 361.
Aouni, B (1996). Lin estimate des expressionquandratiques en programmation math ematique:
Des bornes plus efficacies, Administration SciencesAssociation of Canada, Management Science 17(2),
38 46.
Aouni, B. and Kettani, O (2001). Goal Programming
Model: A glorious history and promising future.European Journal Of Operations Research 133(2), 1
7.
Carrizosa, E. and Romero, M.O. (2001). Combining
Maxmin and Minmax: A goal programmingapproach; Operations Research 49(1), 169 174.
Cavalier, T. M., and Ignizio, J.P., (1994). Linear
Programming, Pretence Hall.
Chanchit, C., and Terrell, M. P. (1993). A multi-
objective reservoir operation model with stochasticinflows. Computers and Industrial Engineering 24
(2), 303 313.
Charnes, A., Cooper, W.W. and Ferguson (1955).
Optimal estimation of executive compensation bylinear programming, Management Science 1, 138
151.
Charnes, A. and Cooper, W.W. (1961).Management Models and Industrial application of
Linear Programming, Wiley, New York.
Charnes, A., and Cooper, W.W. (1952). Chance
constraints and normal deviates. Journal ofAmerican Statistics Association 57, 132 148.
Chowdary, B.V., and Slomp, J. (2002). ProductionPlanning under Dynamic Product Environment: A
Multi-objective Goal Programming Approach.Management and Organization. Pp 112.
Contini, B. (1968). A stochastic approach to goalprogramming. Operations Research 16(3), 576
586.Dade, C. (1981). Revenue Budgeting: A
Decremental Application, P.I Jerry Mccaffery,
Indiana University, Bloomington.
De, P. K, Acharya, D. and Sahu, K.C (1982). A
chance-constrained goal programming model for
capital budgeting. Journal of the OperationalResearch Society. 33(7), PP. 635.
Diaz-Balteiro, L and Romero, C (2003). Forest
Management optimization Models when carbon
captured is considered: A Goal ProgrammingApproach. Forest Ecology Management. 174:447
457.
Dykstra, D.P (1984). Mathematical Programming
for National Resource Management. MCGraw-Hill,New York.
7/27/2019 jurnal GPsepty
10/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
104
Easton, F.F and Rossin, D. F, (1996). A stochasticgoal programming for employee scheduling.
Decision Science. 27(3), 54 568.
Field, R.C, Dress, P.E and Fortson, J.C (1980).
Complementary Linear and Goal Programmingprocedures for Timber Harvest Scheduling, Forest
Science. 26: 121 133.
Freed, N. and Glover, F. (1981). Simple but
powerful goal programming models for discriminateproblems, European Journal of Operations
Research, 7, 44 66.
Helm, J. C (1984). A stochastic goal programmingfor employee scheduling. Decision Science 7(5), 13
18.
Hillier, F.S and Lieberman, G.J (2001). Introduction
to Operations Research. Seventh Edition; McGraw-Hill, New York, 7, 331 340.
Ignizio, J. P (1976). Goal Programming and
Extensions, Lexington books, Lexington, MA.
Ignizio, J.P (1978). A review of Goal Programming:
A tool for Multi-Objective Analysis. Pennav LvaniaState University, University Park, pp 1109.
Kalpkic, D., Mornar, V., and Baranovic, M, (1995).
Case study based on multi period. Multi-Criteria
Production planning Model. International Journal OfProduction Economics, 87, 658 669.
Keown, A. J., (1978). A chance-constrained goal
programming model for bank liquidity management.
Decision Sciences, 9(7), 93 106.
Keown, A. J., and Martin J. D. (1974). A chanceconstrained goal programming model for working
capital management.
The Engineering Economist 22(3), 153 174.
Keown, A. J., and Taylor III, B. W., (1980). A
chance-constrained integer goal programming
model for capital budgeting in the production area.
Journal of the Operational research Society, 31(7),579589.
Kim, G. C. and Schniederjams, M. J., (1993). Amultiple objective model for a just-in timemanufacturing system environment. International
Journal of Operations and Production Management
13(12), 49-61.
Kwak, N. K., Schnierderjams, M. J. and
Warkenstin, K. S. (1991), An application of lineargoal programming to the Marketing distribution.
European Journal of Operations Research, 52, 334
344.
Lam, K. P., and Moy, J. W. (2002). Combinationdiscriminant methods in solving classification
problems in two-group discriminant analysis.European Journal of Operational Research, 138, 294
301.
Lee, S. M. and Clayton, E. R., (1978). A goal
programming model for academic resourcesallocation. Management Science 18(8), 395 408.
Lee, S. M. and Olson, D. L. (1985). Gradientalgorithm for chance-constrained nonlinear goal
programming. European Journal of OperationalResearch Society, 22(3), 359 369.
Leung M. T. (2001). Using investment portfolio
return to combine forecasts; A multi objectiveapproach. European Journal of OperationalResearch, 134, 84 102.
Lyu, J., Gunasekaram, A., Chen, C. Y., and Kao, C.
(1995). A goal programming model for the goal
blending problem; Computer and Industrial
Engineering, Vol. 28, No. 4, 861 868.
Martel, J. M., and Aouni, B., (1990). Incorporating
the decision makers preferences in the goalprogramming model. Journal of Operational
Research Society, 4(12), 1121 1132.
Martel, J. M., and Aouni, B., (1998). Diverse
imprecise goal programming formulation. Journal ofGlobal Optimization, (12), 24 34.
Martel, A., and Price, W., (1981). Stochastic
programming applied to Human resource planning.
Journal of the Operational Research Society, 32(3),187 196.
Nagarur, N., Vrat, P., and Duongsuwan, W. (1997).
Production Planning and Scheduling for injectionmolding of pipe fittings; a case study. InternationalJournals of Production Economics, 53, 157 170.
Nembuo, C. S., and Murtagh, B. A., (1996). A
chance-constrained programming approach in
modeling hydrothermal electricity generation in
Papua New Guinea, Asia Pacific. Journal ofOperations Research, 13,105 144.
Premchandra, I. M., (1993). A goal programmingmodel for the activity crashing in project networks.International Journal of Operations and Production
Management, 13(6) 79 85.
Rogers, M. and Bruen, M. (1998). Choosingrealistic values of indifference, preference and veto
threshold for use with environmental criteria withinElectra. European Journal of Operational Research,
107, 542 551.
7/27/2019 jurnal GPsepty
11/11
Research Journal in Engineering and Applied Sciences (ISSN: 2276-8467) 2(2):95-105Goal Programming: - An Application To Budgetary Allocation Of An Institution Of Higher Learning
105
Romero, C and Rehman, T. (2003) (Second edition):Multiple Criteria Analysis for Agricultural
Decisions. Development in Agricultural Economics,5. Elsevier Science Publishers B.V.
Shafer, S. M., and Rogers, D. F. (1991). A goalprogramming approach to the cell function problem.
Journal of Operations Management. 10(1), 28 43.
Stancu Minansian, I. M., and Giurgifiu, V. (1985).
Stochastic Programming with multi objectivefunctions. D. Reidal publishing Company,
Dordrecht.
Stam, A. and Kuula, M. (1991). Selecting a flexibleManufacturing system using Multiple criteria
analysis. International Journal of productionResearch, 29(4), 803 820
Steuer, R. E., (1986). Multiple criteria optimization:Theory, Computation, and Application. John Wiley
and Sons, New York.
Taha, H. A (2003). Operation Research: An
Introduction, seventh edition, prentice-Hill delhi,India. 8, 347 360.
Tabucanon, M. T and Majumda, S. (1989).
Production planning in a ship repair company inBangladesh, proceedings of international conference
on MCDM: Application in industry and service;
Asia Institute of Technology, Bangkok, 47 61.
Tamiz, M., Jones, D. and Romero, C. (1998). Goalprogramming for Decision making; An overview of
the current state of the art. European Journal of
Operational Research, 111, 569 581.
Winston, W. L. (1994). Operations Research,Applications and Algorithms. Third edition,
Duxbury Press, Belmont, California. 14, 772 782.