49

Capital Budgeting Using a Linear

Programming Model*
by N. N. Wijeratne** and F. C. Harris***

Received October 1983

Revised November 1983

A linear programming model designed to determine the most economical arrange­

ment of finance suitable for a contractor is described.
The model requires data in the form ofproject cashflows and information on loan
capital. These are incorporated into a linear program which determines the least costs
of borrowings.
Results from the model can be used by management to decide on the best com­
bination of capital sources, to arrange the best times to start and finish projects and
to select projects to minimise the total net present cost of capital.

Introduction
Generally, contracting firms in the construction industry have capital employed
representing about 25 per cent of annual turnover. For many companies much of this
capital must be obtained from private sources and retained earnings, with working
capital provided through an overdraft facility negotiated with one of the clearing banks.
Thus, unless a firm is attempting to grow rapidly, or has made estimating errors on
one or two major contracts, the cash flow problems are usually manageable by pru-
dent accounting and financial planning.
In recent years, however, particularly with the construction opportunities that have
appeared overseas, contractors have increasingly become involved in design and con-
struct projects. In many cases with such projects much of the capital to finance the
job has had to be provided by the construction company, usually in association with
a merchant bank, either because the client wishes to pay later with the revenues generated
from the operating plant, e.g. a factory, reservoir, toll road, etc, or simply because the
construction work will be paid for in stages of completion. Consequently contractors'
cash flow needs have become significantly more important to plan and arrange.
This paper describes a computer model designed to assist management in making such
decisions. The objectives were to develop a computer model to determine the following:

**Postgraduate student, Department of Civil Engineering, University of Technology,

Loughborough.
***Senior Lecturer, Department of Civil Engineering, University of Technology, Loughborough.
50 IJOPM 4,2

(1) The optimum financing arrangement suitable to meet the cash flow re-
quirements of all the ongoing projects of a contracting organisation
(2) The most feasible incremental financing arrangement to meet the additional
cash flows of a new contract
Initial work on a model of this type was undertaken by Jayeratne[2], whereby cash
flows for a single project were combined in a Linear Program (LP) with data on finan-
cial borrowings, and the least cost of borrowing determined. Subsequently Wijeratne[3]
extended the work to include more than one project, on-going projects and new pro-
jects. A simplified modelling procedure is illustrated in Figure 1 from which it can
be seen that the first step in the model requires a cash flow forecast. This is a finan-
cial document that shows the scheduled flow of funds, both receipts and expenditures
for the company. The most appropriate approach in calculating the cash flows for
the company is first to calculate the cash flows on a project basis and then to ag-
gregate cash flows for all projects and for head office to form the overall company
cash flow[4].
Having made a forecast of the cash flow requirements, the next question is how
to arrange the finance.
When the demand for capital exceeds the available supply the contractor usually
needs to borrow. A bank loan would be a common method, but many other sources
are available, depending upon the need, for long term borrowings — ten years or more,
medium term — three to ten years, or short term — less than three years.
The interest rate, sequence and amounts repayable vary according to the period,
amount needed and type of finance. Thus management has to decide the best ways
of obtaining funds to minimise the costs of borrowing.
Calculating Cash Flow from the Construction Programme (Figure 2)
The construction programme can be in a bar chart form or as a network. Whatever
the format, the values of various activities are firstly reduced into periods. In this
it is assumed that an activity is uniformly distributed over its duration, which means
that the value of work done during each period on any given activity is the same.
Next the total work done on all activities during each period is found by aggregating
the work done on every activity. This is the value of work done on the project during
each period.
The third step is to derive the cost of work done. Here it is assumed that the mark
up, i.e. allowed for profits and overheads, is the same for all activities. When the mark
up is removed from the value of work, the cost of work is obtained. If cash-on-delivery
terms are assumed then this is the actual expenditure of the contractor for each pro-
ject. However, it is usual to obtain materials and services on credit terms. The effect
of these are considered separately as a source of finance in the model.
Generally, there is a time lag between a contractor carrying out the work and receiv-
ing payments for it. For instance, if a one month delay is considered, the payment
for the work done in month 1 will be received only at the end of month 2.
Another complication is the deduction of retention. This means that an agreed
percentage is deducted by the client from each valuation when making payments. Very
often an upper ceiling isfixedfor retention deductions in the contract. The usual clause
Capital Budgeting 51
52 IJOPM 4,2

is "a retention of ten per cent subject to a maximum limit of five per cent of the con-
tract sum". Because of this, once the cumulative total of retention deductions reaches
the specified ceiling, no more retention will be deducted from future valuations. The
retentions thus deducted will be refunded in one of the following methods.
(1) Release a certain agreed percentage (say 50 per cent) at the time of practical
completion, i.e. at the time of handing over the completed works, and the
balance at the end of maintenance period (normally six months).
(2) Release full amount at the end of the agreed maintenance period.
By considering the above mentioned factors, the payments received by the contractor
are calculated. Finally, cash required to finance the operations of each period is deter-
mined by deducting the payments received from the cost of work to be done in the
period.
Another assumption made is that all payments from the client are received at the
end of a period and hence a payment received in month n will benefit only the opera-
tions of month n + 1.
Cash Flow from the S-Curve (Figure 3)
Hardy[5] observed that the cumulative value versus time curve tended to take a uniform
S-shape for similar contracts handled by a given contractor. Thus using this property,
if the shape of the standard S-curve for the company can be specified then the S-
curves for all the contracts handled by that company can be easily projected knowing
only the end conditions.
The information required for this method is:
(1) Percentage time and cost values to specify the standard curve;
(2) Starting period and ending period;
(3) Contract sum.
 Capital Budgeting 53

Once the value curve for a contract is projected then the cost and cash in curves can
be easily determined by adopting the same method described earlier. The only dif-
ference is that cumulative figures are used now instead of the periodic figures.

With reference to Figure 3 the period cash requirements are given by:
Period 1 — CO1
Period 2 — CO2 - CO1
Period 3 — CO3 - CO2 - CI2
Period 4 — (CO4 - CO3) - (CI3 - CI2) etc.
When calculating cash flow of the company both projects expenditure and the com-
pany overheads have to be considered. Overall projects cash flow is determined by
aggregating the cash flows of individual projects.
In the model two assumptions are made regarding company overheads, viz.
54 IJ0PM 4,2

(1) Company overheads are equally spread over a period of one year.
(2) Rate of annual increase in overheads expenditure is uniform.
In reality a variety of expenses constitute the overheads. Once the projects cash flow
and the overhead expenditure are assessed the cash requirements of the company are
given by the addition of these two.
Development of the Linear Program
Having established the cash flow data and methods and constraints on finance, the
next stage is to optimise the borrowings using a linear programming technique.
In linear programming the same problem can yield different solutions depending
on the definition of the objective. Hence it is important that the objective be properly
defined. Some of the objectives that can be used in a linear programming model,
designed to determine most appropriate financing arrangements, are:
(1) Maximise profits
(2) Maximise rate of return or yield on capital employed
(3) Minimise cost
(4) Maximise turnover
(5) Minimise total interest payments
(6) Minimise the net present value (or more specifically cost) of finance.
Maximising profit would seem to be the most obvious objective; however the mark-
up for each contract is known and therefore minimising the interest on debt which
is paid out of profits would be more meaningful. The same argument would apply
to rate of return on capital. Similarly to minimise costs, requires arranging cash flows
to produce the lowest cost of finance.
Maximising turnover could not be expected to yield a satisfactory solution as the
only bearing it could have on finance would be the absolute limit of borrowing with
total disregard for the cost of finance. Hence, the objective selected was to minimise
the total net present cost of finance. This was considered more appropriate than the
total interest payments since it would correctly assess the time value of money.
It was assumed that this criterion w i l d contribute to the common overall com-
pany objective of maximising real profits by reducing an important item of cost[6].
Selection of Constraints
As highlighted earlier there are many constraints applied to financing arrangements,
such as:
(1) Period cash requirements of working capital
(2) Upper limits of borrowing from individual sources
(3) Lower limits of borrowing from sources at a time
(4) Absolute maximum ceiling on debt finance for the company based on a sound
gearing ratio of loan capital to equity and retained earnings
(5) Limit on the ratio of loan capital interest to profits
 Capital Budgeting 55

(6) Limit on current ratio

(7) Limit on the ratio of turnover to assets
(8) Maximum limit on trade credit for a period
(9) Desired minimum liquid cash balance.
Those constraints which were felt to be essential for a satisfactory definition of the
LP problem were:
(1) Periodic cash requirements of working capital
(2) Upper limits of borrowing from a source
(3) Upper limit on credit per period
(4) Minimum cash balance required per period
(5) Absolute upper limit of total borrowings based on the capital gearing ratio.
1. Periodic Cash Requirements of working capital
This is the main constraint that acts as a monthly "balance sheet" for the model.
All the other constraints are simply conditions and limitations imposed on the balance
sheet.
The factors that are necessary to be considered in building the constraints on periodic
cash requirements are:
(1) Cash flows of all the contracts i.e. cash out minus cash in
(2) Borrowings required for every period
(3) Interest payments on previous borrowings
(4) Repayment of loan sums
(5) Trade credit available for the period
(6) Repayment of previous credit
(7) Excess money generated in the period
(8) Availability of excess cash from previous period to finance the transactions
of current period
(9) Payment of taxes
(10) Payment of dividends.
The full constraint linking all the above factors can be expressed as:
Borrowings - Interest + Credit - Previous Credit — Excess Cash + Previous
Excess Cash = Cash for contracts + Dividends + Tax, or more formally as:
Σ L tk - ΣIk∙L(t - 1)k + Ct - Ct-1 - ECt + ECt-1 = CRQt + D+T
where,
Ltk — Amount borrowed from source k in time t
Ik — Interest rate for source k
L(t - 1)k — Amount borrowed from source k previously
56 IJOPM 4,2

Ct — Trade credit available in this period

Ct-1 — Trade credit obtained in previous period and to be settled in
this period
ECt — Excess cash of this period
ECt-1 — Excess cash from last period
CRQt — Cash required in this period for contracts
D — Dividends to be paid to shareholders
T — Taxes payable.
Some researchers[2, 7] tend to treat dividends as a payment made in each period. In
this model, dividend is accounted for in the more usual way, namely a payment made
once a year.
There are many forms of taxes. Previous researchers have found that it is impossi-
ble to develop an LP model that will accommodate all the types of tax[7]. In this
model tax is computed as a percentage of profits.
On the treatment of tax there seems to be some uncertainty[6] as to whether it should
be considered as a source of credit or not. Here tax is viewed as an outflow of cash,
generally occurring nine months after the end of the company accounting year[8].
Since the dividend and tax payments are considered to occur only once a year, it
is realistic to obtain a solution without considering interest charges and then improv-
ing on this using the interest, by successive iterations. The flow chart for this process
is included in Figure 1.
In practice there are many loan repayment schemes. A realistic model requires pro-
vision for at least the most common variations in repayment schemes.
Features of the most common loan repayment schemes can be summarised as:
(1) Interest paid monthly, quarterly or half yearly
(2) A period of grace of n months before the first interest payment
(3) The loan sum is repaid in one installment or n equal installments at regular
intervals as agreed.
In this model facilities for the above mentioned variations are included.
Overdrafts, which are the main forms of short term borrowing, are technically
repayable on demand. To allow for this uncertainty, in some optimising models over-
drafts are considered as single period loans. This means the amount drawn in period
1 is settled in period 2 and if necessary another amount is drawn in period 2, which
in turn is settled in period 3.
However, an overdraft facility is very seldom revoked in this manner even though
the bank has the right to do so[9]. Hence, the writers feel that the above approach
is unsuitable and could lead to unrealistic results. Instead, to treat an overdraft as
essentially a term loan repayable within the agreed duration is assumed to be closer
to the practical situation.
2. Upper Limits of Borrowing from a Source
This constraint restricts the total borrowings and not the individual borrowings from
a source.
 Capital Budgeting 57

3. Upper Limit on Credit per Period

Normally suppliers of plant and material etc. will not provide unlimited credit to a
contractor. Thus when a contractor has reached the limit he will not be able to make
further purchases without settling part of the previous credit.
However, during the latter part of a contract and during the maintenance period
this credit ceiling could be higher than the total cost of credit items required. Quite
obviously during these periods of low (or zero) activity a contractor will not make
use of all the credit available to him.
Labour wages are not considered as a source of credit as they relate to periods of
less than the basic cash flow period of one month considered in the model. Similarly
other credits of durations less than one month are not considered.

4. Minimum Cash Balance per Period

No company likes to operate on a hand to mouth cash basis where the finance drawn
from sources exactly equals the cash requirements of the period. Hence this constraint
is necessary to ensure that the solution given by the model provides for the desired
minimum liquid cash balance.
In a net borrowing situation the effect of this constraint will be to increase the bor­
rowing from sources to provide for the additional cash.

5. Absolute Upper Limit on Borrowings

As explained earlier, this will restrict the total amount that can be borrowed by a com­
pany. Ideally, total borrowings in each period should be checked against this limit.
But this will add substantially to the size of the model and hence is not preferred.
Instead, this check will be carried out on an annual basis, which means that the
total borrowings in a year should not exceed this limit. This alternative also makes
more practical sense as normally the financial ratios for a company are revealed only
at the time of publishing the annual balance sheet.

Solving the Model

The objective function for the model, to be minimised, is the net present cost of finance
for the planning period. This can be represented by:
Minimise Σtj Ntj
where Ntj is the net present cost factor for loan Ltj, i.e. loan taken from source j
in time t.
The model was solved by the simplex method using a standard NAG routine (No.
HO1 ADF)[10]. The Numerical Algorithoms Group provide a suite of computer pro­
grams which have proved popular and this particular program was readily available for
use on our computer. However similar standard packages can be obtained as alternatives.
Analysis of Results
A disadvantage with the LP model is the difficulty of testing the results. Under the
circumstances the best compromise was to ensure that accurate data were input to
the LP model and hope that the solution was the true optimum. This was explored
58 IJOPM 4,2

by selecting arbitrary data and making spot checks on the calculations at various stages
in the solution process.
An Example
Company details
Financial planning period = 24 months
Start of planning period = 1
Number of financial sources = 6
Borrowing limit based on debt/equity ratio = £1,500,000
Minimum cash balance per period = £2,000
Annual overhead expenditure = £36,000
Annual increase in overheads = 6.0 per cent
Tax rate on profits = 52.0 per cent
Percentage of profits to be paid as dividends = 30.0 per cent
Total number of contracts = 3
Contract details
Contract 1 Contract 2 Contract 3
Profits and overheads = 9.0 per cent 8.5 per cent 10.0 per cent
Delay in clients' payments = 2 months 1 month 1 month
Retention = 10 per cent 10 per cent 10 per cent
Limit of retention = 5 per cent 5 per cent 5 per cent
Maintenance period = 6 months 6 months 6 months
Release of retention at
practical completion = 50 per cent 50 per cent 50 per cent
Starting time of contract = 0 3 6
Finishing time of contract = 18 months 24 months 15 months
Value of contract = £65,000 £1,000,000 £800,000
S-curve for a typical contract
Time per cent: 0 20 40 60 80 100
Value per cent: 0 9 41 78 92 100
Trade Credit Parameters
Percentage of total Delay for payment
cost of a typical in weeks
contract
Labour wages 14 1
Plant hire 18 6
Material supplies 23 8
Sub-contractors 45 8
 Capital Budgeting 59

Details of Financial Sources

RE OD S1 S2 S3 S4
Repayment period in
months = 36 24 12 18 9 18
Annual percentage interest
on borrowings = 14 9 9.5 10 9 10.5
Interval of interest
payments in months = – 1 1 3 1 3
Grace period in months = – – – – –
at the start for
interest payments 3
Number of instalments =
for repayment of
loans 1 1 1 1 1 3
Upper limit on loan
in £ = 750,000 300,000 50,000 150,000 25,000 700,000
Note: RE — retained earnings
OD — overdrafts
S1 — loan source 1
S2 — loan source 2 etc.
TC — Trade credit per period = £7,500
It can be seen that three contracts are to be started within six months and the finan-
cial demands are to be analysed over a twenty-four month period.
The computer program uses the S-curve profile and the trade credit parameters to
produce the cash flow pattern for the combined contracts.
Comments on Output
It can be seen from Table I, which is extracted from the output of the computer package
that during the twenty-four months considered, cash flow requirements are financed
by drawing from four loan sources and using trade credit and a bank overdraft. Each
column shows the month that the finance was needed, and the Net Present Cost of
the total financing is also stated at the bottom of the table. Table II shows the schedule
of repayments of the interest and capital. For example, loan sourceS1was drawn upon
three times (months 1, 2 and 4), interest was paid monthly and the capital repaid in
months 13, 14 and 16. A similar procedure is evident from loan source S3, while the
interest on sources S2 and S4 are repaid quarterly, with a grace period of three months
for source S4. Only the first repayment instalment of the initial loan of £4393.70 for
S4 is indicated at month 24.
The table also illustrates the heavy use of the overdraft and trade credit facilities.
In the example shown above, a company is provided with a solution for present
contracts only, and based on the solution negotiates loans. However, when the package
is run again with additional contracts, the solution given by the model will probably
be entirely different from the previous solution. The model must, however, be forced
60 IJ0PM 4,2

Table I. Borrowing Requirements

Month Retained Overdraft Trade Loan 1 Loan 2 Loan 3

Earnings Credit

1 0 0 7,500 12,408 0 0
2 0 0 7,500 18,003 0 0
3 0 0 7,500 0 0 18,140
4 0 12,189 7,500 19,586 0 6,859
5 0 61,779 7,500 0 0 0
6 0 0 7,500 0 0 0
7 0 64,353 7,500 0 0 0
8 0 0 7,500 0 107,993 0
9 0 0 7,500 0 0 0
10 0 108,138 7,500 0 0 0
11 0 55,277 7,500 0 42,007 0
12 0 0 7,500 0 0 0
13 0 0 7,500 0 0 0
14 0 0 7,500 0 0 0
15 0 0 7,500 0 0 0
16 0 0 7,500 0 0 0
17 0 0 7,500 0 0 0
18 0 0 7,500 0 0 0
19 0 0 7,500 0 0 0
20 0 0 7,500 0 0 0
21 0 0 7,500 0 0 0
22 0 0 7,500 0 0 0
23 0 0 7,500 0 0 0
24 0 0 7,500 0 0 0

Month Loan 4 Loan 5 Loan 6 Loan 7 Loan 8 Loan 9

1 0 0 0 0 0 0
2 0 0 0 0 0 0
3 0 0 0 0 0 0
4 0 0 0 0 0 0
5 0 0 0 0 0 0
6 42,939 0 0 0 0 0
7 0 0 0 0 0 0
8 0 0 0 0 0 0
9 197,760 0 0 0 0 0
10 18,592 0 0 0 0 0
11 0 0 0 0 0 0
12 0 0 0 0 0 0
13 0 0 0 0 0 0
14 0 0 0 0 0 0
15 0 0 0 0 0 0
16 0 0 0 0 0 0
17 0 0 0 0 0 0
18 0 0 0 0 0 0
19 0 0 0 0 0 0
20 0 0 0 0 0 0
21 0 0 0 0 0 0
22 0 0 0 0 0 0
23 0 0 0 0 0 0
24 0 0 0 0 0 0
Net present cost of finance for the period = 67,210
 Capital Budgeting 61

Table II. Schedule of Repayments

Month Retained Overdraft Trade Loan 1 Loan 2 Loan 3

Earnings Credit

1 0 0 0 0 0 0
2 0 0 7,500 94 0 0
3 0 0 7,500 231 0 0
4 0 0 7,500 231 0 131
5 0 88 7,500 380 0 180
6 0 533 7,500 380 0 180
7 0 533 7,500 380 0 180
8 0 997 7,500 380 0 180
9 0 997 7,500 380 2,604 180
10 0 997 7,500 380 0 180
11 0 1,776 7,500 380 0 180
12 0 2,175 7,500 380 3,617 18,320
13 0 2,175 7,500 12,788 0 6,909
14 0 2,175 7,500 18,288 0 0
15 0 2,175 7,500 149 3,617 0
16 0 2,175 7,500 19,735 0 0
17 0 2,175 7,500 0 0 0
18 0 2,175 7,500 0 3,617 0
19 0 2,175 7,500 0 0 0
20 0 2,175 7,500 0 0 0
21 0 2,175 7,500 0 3,617 0
22 0 2,175 7,500 0 0 0
23 0 2,175 7,500 0 0 0
24 0 2,175 7,500 0 1,013 0

Month Loan 4 Loan 5 Loan 6 Loan 7 Loan 8 Loan 9

1 0 0 0 0 0 0
2 0 0 0 0 0 0
3 0 0 0 0 0 0
4 0 0 0 0 0 0
5 0 0 0 0 0 0
6 0 0 0 0 0 0
7 0 0 0 0 0 0
8 0 0 0 0 0 0
9 1,085 0 0 0 0 0
10 0 0 0 0 0 0
11 0 0 0 0 0 0
12 6,084 0 0 0 0 0
13 479 0 0 0 0 0
14 0 0 0 0 0 0
15 6,084 0 0 0 0 0
16 470 0 0 0 0 0
17 0 0 0 0 0 0
18 6,084 0 0 0 0 0
19 470 0 0 0 0 0
20 0 0 0 0 0 0
21 6,084 0 0 0 0 0
22 470 0 0 0 0 0
23 0 0 0 0 0 0
24 20,397 0 0 0 0 0
62 IJOPM 4,2

to accept the existing loans and thus a facility was provided to store a previous solu-
tion in a file. From this the loans already committed were fed into a new solution
at the discretion of the user.
An Appraisal of the LP Package
The following rather unsatisfactory features emerged in the results obtained during
the many trials carried out with the package:
(1) Heavy reliance on short term loans
(2) Maximum use of credits
(3) Arbitrary nature of loan sums
(4) Repeated withdrawals from the same source
(1) Heavy Reliance on Short Term Loans
This is to be expected as generally:
— Short term loans carry lower interest rates
— Use of net present cost for the objective function tends to favour the short loans.
However, unrestrained reliance on short term loans is not advisable when overdrafts
are available, as this could reflect a lesser commitment by the company towards bet-
ter financial management which might be detrimental to attracting new investors.
This may be rectified by the following measures:
— Specify a period of more than one year for the settlement of overdrafts
— Impose an additional constraint based on the balance sheet current ratio.
(2) Maximum Use of Credits
The model tends to use credits to the maximum possible and if there is no constraint
restricting the maximum amount of credit available at any period then the solution
given by the model will be wholly based on credits. This is not surprising as trade
credit is treated as an interest free short term loan.
But to depend too much on trade credits can create supply problems and can be
expensive if the loss of cash discounts is taken into account[9]. An obvious solution
to this was to make a provision for the costs associated with the loss of cash discounts.
But the nature and variability of these discounts are too complicated to be provided
in the model. Hence the best practical alternative was to impose an upper limit to
the use of trade credit. This limit should be kept at a level considered reasonable by
the contractor with a bias towards the minimum.
(3) Arbitrary Nature of the Loan Sum
Another drawback in the solution given by the model was the arbitrary nature of loan
sums. In particular the following unsuitable features were noted:
— Withdrawal of very small sums
— Loan sums not rounded off
— Repeated withdrawals from the same source.
 Capital Budgeting 63

Most of the loan schemes available are subject to a minimum limit and are offered
in steps of £100, 500, etc. Such a constraint is not available with the NAG routine
and the writer believes that this feature should be incorporated in a redesigned NAG
routine together with a rounding-off procedure.

(4) Repeated Withdrawals from the Same Source

Repeated withdrawals are suitable for bank overdrafts but not for other loan sources.
Lenders are unlikely to prefer to have a large number of loans with the same customer.
Even from the contractor's considerations it could be rather embarrassing to have a
loan negotiation with the same lender every month.
Again this would best be provided for as a facility in a revised NAG routine.
In certain test runs the optimal solution given by the model included borrowings
during the maintenance period of a contract. At first glance this may appear to be
erroneous. But what was actually happening was that fresh loans were taken to ser-
vice the previous loans. Clearly this indicated the usefulness of the package as a con-
tractor would not have anticipated the need for finance after completing a contract.
Normally the cost of equity would be higher than the cost of debt capital. Under
such circumstances the model would tend to use debt capital to the maximum possi-
ble limit governed by the financial ratios. Retained earnings would be used only when
this limit was reached.
The suggestion of borrowing when a contractor has sufficient retained earnings
may sound absurd. But the interpretation is that it is cheaper to raise debt capital
rather than use available retained earnings. Then the natural question is, "What is
he to do with the retained money?" An answer to this is that the contractor should
try to get more contracts and create an additional need for working capital.
On the other hand the situation is reversed when the cost of equity is less than the
cost of debt which may happen during a period of national economic stagnation.
When this happens the model will first make use of all the retained earnings before
resorting to any borrowing.

Conclusions
Building and civil engineering contractors are highly vulnerable to liquidation and
hence financial planning is indispensable to them. Thus financial planning packages
could have an important role to play in the construction industry.
The computer package described in this paper is a useful tool for organising the
finances of a contractor in the least costly manner.
In particular the package can be effectively used in the following situations:
(1) To find the best arrangement of borrowings required to finance the contracts
in hand
(2) To find the additional borrowings required for new contracts in the presence
of existing borrowings
(3) To prepare a financial plan for the contractor if prospective new contracts could
be predicted
(4) At the tender stage to estimate the cost of capital for the contract
64 IJOPM 4,2

(5) To determine the best time to start a new contract

The power of the package lies in the fact that it facilitates a contractor arranging bor-
rowings sufficiently in advance so that the possibility of liquidity crisis is reduced.
Also the financial report provided by the package could provide substantiating
documents during loan negotiations.
References
1. Churchman, C. W., Ackoff, P. L. and Arnoff, E. L., Introduction to Operational Research, Wiley
Inc., USA, 1961.
2. Jayaratne, W. H. D. W., "Capital Budgeting and Linear Programming", MSc Project Report, Depart-
ment of Civil Engineering, Loughborough University of Technology, UK, 1979.
3. Wijeratne, N. N., "Financial Optimisation for Construction Contractors", MSc Project Report,
Department of Civil Engineering, Loughborough University of Technology, UK, 1981.
4. Harris, F. C. and McCaffer, R., Modern Construction Management, Granada Ltd., St. Albans, UK,
1977.
5. Hardy, J. V., "Cash Flow Forecasting in the Construction Industry", MSc Project Report, Depart-
ment of Civil Engineering, Loughborough University of Technology, UK, 1970.
6. Hancher, D. E. and Gebert, D. K., "Optimising Capital Investments in Construction", Journal of
the Construction Division, ASCE, December 1976.
7. Ashton, D. J., "The Construction and Use of Mathematical Programming Models for Corporate
and Financial Planning", PhD Thesis, University of Warwick, UK, 1978.
8. Leaflets of Lloyds Bank, Williams & Glynns Bank, Midland Bank and ICFS London, UK, 1980.
9. Money for Business, Bank of England, 1980.
10. NAG, (HOI ADF) NAG (USA)Inc.1250 Grace Court, Downers Grove, Illinois, USA or NAG Cen-
tral Office, 7 Banbury Road, Oxford, UK.

Additional Reading
Merret, A. J. and Sykes, A., The Finance and Analysis of Capital Projects, Longman, UK, 1966.
Mackay, I. B., "Feasibility of Cash Flow Forecasting by Computers", MSc Project Report, Department
of Civil Engineering, Loughborough University of Technology, UK, 1971.
Sizer, J., An Insight into Management Accounting, Penguin Books Ltd, London, UK, 1969.
Bhaskar, K., "Linear Programming and Capital Budgeting: The Financing Problem", Journal of Business
Finance and Accounting, Summer 1978.
Sealey, C. W., "Financial Planning with Multiple Objectives", Financial Management, Winter 1978.
Jamieson, R. G., "Project Planning using Capital Investment Optimisation", MSc Project Report, Depart-
ment of Civil Engineering, Loughborough University of Technology, UK, 1980.
Makower, M. S. and Williamson, G., Operational Research, Teach Yourself Books Ltd., London, UK, 1972.