Quantitative methods in marketing 237
Heuristic programming
Heuristics is commonly defined as the use of
rule of thumb for solving problems. Therefore,
heuristic programming techniques are based on
an orderly search procedure guided by these
rules of thumb and are mainly applied to
problems when mathematical programming
techniques are either too expensive or compli-
cated. However, they do not guarantee optimal
solutions. In the past, heuristic programming
has been applied fairly extensively in certain
areas of marketing.
Taylor (1963) devised a graphical heuristic
procedure to derive solutions to the problems
of media scheduling that suggested the optimal
number of advertisements to be placed in each
medium and the size of each insertion. The
number of insertions was determined by
graphical methods, which attempted to find the
point where the marginal returns to the last
insertion equalled the marginal cost of the
advertisements for each medium.
A summary of the main characteristics,
advantages, limitations and applications of
dynamic and heuristic models in marketing is
presented in Table 9.14.
Stock control models
The distribution side of marketing has been
successfully modelled using quantitative meth-
ods for a number of years. The objective of the
distribution system is to get the right goods to
the right places at the right time for the least
cost, and involves decisions on such problems
as the number, location and size of warehouses,
transportation policies and inventories. In this
section the inventory decision will be dis-
cussed, but it must be realized that inventory
represents only one part of the local distribu-
tion network, a complete analysis of which is
outside the scope of this chapter. The inventory
decision has two parts to it: when to order
(order point) and how much to order (order
quantity). These are not independent and can
be deduced from a stock control model. The
ordering of goods involves costs, such as
transportation and handling, which increase
the number of orders placed.
On the other hand, the storage of goods
also involves costs such as storage space char-
ges, insurance costs, capital costs and deprecia-
tion costs. The first two decrease and the last
two types of costs increase with the order
quantity.
The simplest model assumes that demand
is constant, shortages are not allowed (no
stockouts), immediate replacements of stocks
and a regular order cycle. If C= cost of holding
of one unit of stock/unit time, Co= cost of
placing an order, d= demand rate (units/unit
time),q= order quantity in units and t= order
cycle time (order point), mathematical analysis
shows that the total costs (ordering plus stock-
holdings) are minimized when the following
order quantity is used:q=dt=
Cod
Cq* = is often referred to as the economic order
quantity (EOQ). The simplest model can be
improved by relaxing some of the
assumptions.Network programming models
Network programming models are the meth-
ods usually used for planning and controlling
complex marketing management projects.
There are two basic methods: the critical path
method (CPM) and the performance evaluation
and review technique (PERT). The two meth-
ods together are also called critical path analy-
sis (CPA) techniques.Critical path analysis
Critical path analysis, in its various forms, is
one of the techniques developed in recent years
to cope with the increased need for planning,