Demand forecast process and inventory management

DEMAND FORECAST PROCESS AND INVENTORY MANAGEMENT

A demand forecast is a central piece of the operations of a modern firm. It is a decision making tool which considers
different factors and justifies decisions. The main idea is to make the prediction and estimations of the future
demand and consequently determinate the potential markets for the product or services for the following period. A
demand forecast states the needed inventory that helps to overcome the fluctuations in demand. According to the
information, a firm can start to plan its upcoming activities in a way that they can most efficiently transform their
inputs into outputs. Additionally, a demand forecast enables a corporation to provide its customer higher value. It
distributes the operations the information including the needed products and stock keeping units (SKU), their
quantities and the facilities required to fulfil the future needs. This way, the firm can gain better profit as forecast
offers them a chance to lower their costs.

-(Keath&Young , 1996, 217-219)

To sum it up, a demand forecast creates a data bank that helps the decision makers settle targets, create plans and
demonstrate the changes in environment. Moreover, it guides a firm to act in the best possible way to increase the
efficiency without decreasing the service value offered to the customers.

Demand Forecast Process

Determination Election of the Selection of
Settling of the
of a forecast products the forecast
time horizon
purpose forecasted model

Revision and
Creation of the Collection of
execution of
forecast the data
the forecast

Demand Forecast Process

Determination of a forecast process
The whole process begins with the determination of the forecast purpose as it is important to be aware of the
purpose of the forecast and the information requirements for it.

Time horizonThe time horizon for the forecast needs to be clarified to the forecasters. The forecasts can be
made for a time period happening within the following three months or then it can be conducted to the one after
ten years; the forecasts made to the near future are more precise while the ones made for further will be less
accurate.

Selection of the forecast modelAfter the baseline has been settled, the appropriate forecasting model is
chosen. Characteristics of the target of the forecast settle limits that should be taken into account when choosing
the method. In other words, a demand might be seasonal or it might be easily affected by a variable like a new
technological invention. Both of these require their own way of forecasting. In some cases, the forecasters solely
investigate the past data. Consequently, it is important to know if it is likely for the same trend to continue in the
future or are there lots of random fluctuations that can not be predicted. Add to this, different states of product life-
cycles might require different forecasting methods. For instance, it is wise to use judgemental forecasting for the
new items as there is no previous data. In contrast, if a rapid growth in the sales is seen or when a product achieves
its mature state, it is more common to investigate the time series data. Yet, it should be remembered that one
forecasting method does not prevent the usage of another.

Forecasting
Methods

Qualitative Quantitative
Methods Methods

Time Series Jury of executive Sales team
Casual Methods
Model opinion opinion

Exponential Regression
Moving Average Opinion Polling Delphi Method
Smoothing Analysis

FORECASTING METHODS

There are two different kinds of methods of investigating the future demand; quantitative and qualitative ones.
Quantitative methods search for mathematical consistencies in the history. They can be divided in to two
subcategories; time series model and correlation model. In contrast, a qualitative forecast is based on humans’
judgements according to their past experiences, premonitions and emotions. Yet, the selection of the most
suitable forecast model does not need to be based solely to either quantitative or qualitative ones; a
combination of models from both approaches is often the most effective one.

Time series

This model is one of the cheapest and fastest ways to conduct a demand forecast. Moreover, it is easy to use. Yet,
the simplicity of the method has its downside: as the model only concentrates on investigating the numbers it
neglects the reasons behind the changes. In other words, it is naïve. Consequently, if a corporation decides to

use this model it should constantly follow the progress of the data in order to notice the deviations and take actions
according to them. The basic idea behind the model is that the past demand patterns will continue similarly and
therefore it can be projected to the future. In this model, time is always an independent variable. Dependent
variable changes according to the issue that is being forecasted. Therefore, in this case it is the demand.

D = F (t)

Where D is the variable to be forecast and f(t) is a function whose exact form can be estimated from the past data
available on the variable. The value of the variable for the future is expressed as a function of its values in the past.

Dt+1 = f ( Dt, Dt-1, Dt-2, .....)

Moving averages

It is a simple and widely used method. It is suitable especially for situations where random fluctuation appears. The
model computes the average of the decided number of most recent periods of data to solve the upcoming one. The
correspondence depends on the number of periods chosen; the more periods there are included in the research the
slower is the response. If sudden changes need to be seen quickly, the number of time periods should then be low.

MA:( ∑ demand in previous n period)/n

If a more sophisticated method of forecasting is needed, then a forecaster should consider an exponential
smoothing model. It is often used when there are no bigger changes in the demand. It adds the previous forecast to
the most recent error. Unlike the moving average, exponential smoothing emphasizes the most recent data.
Forecast for the next period (Ft+1) is calculated by counting the difference between the last periods actual demand
(At) and forecast (Ft). This error can be either positive or negative. Then, the error is multiplied by the chosen
smoothing constant (α), which is a fraction between 0 and 1, and in the end added to the previous forecast. The
bigger the fraction is the greater the change between the forecasts will be. A lower fraction offers stability to the
forecast.

Ft+1 = Ft + α (A t − F t)

Where α is the smoothing constant

Casual Models
Causal models investigate the mathematical relationship between demand and another variable, for example a
price. This model takes more time and is more expensive than the previous ones. However, it provides a company
information about the changes that will happen when the other variable changes. Still, finding a suitable variable is a
hard task. One of the most common methods is regression analysis which aims to discover the function that could
state the link between the causal variable and the variable. When doing this the variable, in demand forecast the
demand, needs to be clarified and the causal one or ones. Then, it is necessary to discover lots of data points to
interpret as it is not useful to start with this method if the number of data is limited. Finally, the relationships
between the variables need to be cleared. It can be for example linear or exponential. Hence casual method is used
when :

There is no logical link between the demand in the future and what has happened in the past.
There are other factors which can be logically linked to the demand

Regression analysis
Regression means dependence and involves estimating the value of a dependant variable Y, from an independent
variable X. Regression can be linear or non- linear.

Demand = Trend + Error.

The error part is decomposed to

Seasonal
Cyclical –Similar to seasonal variations except for the fact that the cycle has long period.
Random variations –do not follow any pattern that cannot control or accounted.

Jury of Executive Opinion
Jury of Executive Opinion is a panel consisting of corporate executives mastering in their own field. They gather to
share their thoughts and ideas that result as a forecast. Depending on the situation, the members of the panel might
be from inside a company or consist of people from several companies. Forecasts are made during a few meetings

where the experts meet, determinate the variables and share their expertise with the others. In the end, the
forecast is made according to the knowledge received from these people. However, even though the participants are
experts the outcome might be suffering from bias opinions.

Sales team opinion
The view of individual sales people is another method that is constructed upon the expertise opinion. However,
instead of collecting the members of different business units together, it collects the data from the salespersons that
are assumed to have the most recent and adequate information about the market environment and clientele.
However, quite often the people appear to be biased: they either overestimate or underestimate the situations.
Additionally, they might not be aware of the overall economic conditions which might have an impact on the sales.

Delphi method
Sometimes it might not be reasonable to gather experts as the members of a panel might be affected by each other.
The main idea behind the Delphi method is that all the people involved receive the same questionnaire including
several questions which they should process independently. After the organizer has collected all the answers, the
questions will be modified according to them. People will be asked to make justifications and explanations to some
of their answers. Then, they hold a new questionnaire. In other words, the members never meet to negotiate but
instead the decisions will be made according to the results of several questionnaires. The process will be over when
the participant will not be willing to change their opinion. The greatest advantage that is accomplished by using the
Delphi method is that when there is no hierarchy and group pressure people are more likely to give new and
surprising views and ideas that would not even be discussed in a panel. The down side is that there might be some
difficulties for receiving the answers. Furthermore, the results are based on predictions which might, in the end, be
totally wrong. Additionally, the expenses might increase.

Opinion polling
Opinion polling helps to recognise and understand the customer’s need by making an interview or making a written
questionnaire. Instead of relying to the experts and consults, the answers are got from the clientele, from a sample
of population carefully chosen. Pitfalls in these cases are the selection of the correct sample population and the
misunderstanding of questions. On the other hand, the recognition of a future trend can be achieved by using
opinion polling.

Execution of Demand Forecast

After the preliminary work has been done, the demand forecast team is then collecting the data required for the
forecast. People who are responsible for doing this are commonly managers in some level or sales people who have
the knowledge related to the markets and demand. The actual creation of the forecast takes place after all the
parties have distributed their information.

EXECUTION
A next step after the demand forecast has been created is to execute it. Furthermore, it needs to be revised. One of
the most important parts of the demand forecast process is to investigate the forecasting error from the historical
data. It is calculated by subtracting the forecasted demand from the actual sales. When this is resolved, it is easier
for a company to compare different forecasting methods and choose the most suitable one as well as to discover the
required stock levels. A rule of thumb is that the greater the forecasting error is, the bigger should the maintained
inventory be to meet the customers’ needs. For a certain period, the error can be calculated by taking the difference
between the actual demand and the forecast demand. Add to this, there are other patterns that are used to
investigate the forecasting error. Yet, the main purpose of the thesis is to see the whole process of forecasting and
thus these will not be introduced here.

Forecasting error = Actual demand − Forecasted demand

Inventory Management
Inventories occupy the most strategic position in the structure of working capital of most business enterprises. It
constitutes the largest component of current asset in most business enterprises. In the sphere of working capital, the
efficient control of inventory has passed the most serious problem to the cement mills because about two-third of
the current assets of mills are blocked in inventories. The turnover of working capital is largely governed by the
turnover of inventory. It is therefore quite natural that inventory which helps in maximize profit occupies the most
significant place among current assets.

The inventory means aggregate of those items of tangible personal property which
(i) are held for sale in ordinary course of business.
(ii) are in process of production for such sales.
(iii) are to be currently consumed in the production of goods or services to be available for sale.

The primary objectives of inventory management are:
(i) To minimize the possibility of disruption in the production schedule of a firm for want of raw material,
stock and spares.
(ii) To keep down capital investment in inventories.

The excessive level of inventories consumes the funds of business, which cannot be used for any other purpose and
thus involves an opportunity cost. The carrying cost, such as the cost of shortage, handling insurance, recording and
inspection are also increased in proportion to the volume of inventories. This cost will impair the concern
profitability further.
On the other hand, a low level of inventories may result in frequent interruptions in the production schedule
resulting in under-utilization of capacity and lower sales. The aim of inventory management thus should be to avoid
excessive inventory and inadequate inventory and to maintain adequate inventory for smooth running of the
business operations. Efforts should be made to place orders at the right time with the right source to purchase the
right quantity at the right price and quality. The effective inventory management should
(i) maintain sufficient stock of raw material in the period of short supply and anticipate price changes.
(ii) ensure a continuous supply of material to production department facilitating uninterrupted production.
(iii) minimize the carrying cost and time.
(iv) maintain sufficient stock of finished goods for smooth sales operations.
(v) ensure that materials are available for use in production and production services as and when required.
(vi) ensure that finished goods are available for delivery to customers to fulfil orders, smooth sales operation
and efficient customer service.
(vii) minimize investment in inventories and minimize the carrying cost and time.
(viii) protect the inventory against deterioration, obsolescence and unauthorized use.
(ix) maintain sufficient stock of raw material in period of short supply and anticipate price changes.
(x) control investment in inventories and keep it at an optimum level.

Inventory Control
Inventory control is concerned with the acquisition, storage, handling and use of inventories so as to ensure the
availability of inventory whenever needed, providing adequate provision for contingencies, deriving maximum
economy and minimizing wastage and losses.

Inventories Control Techniques
ABC Analysis of Inventories
The ABC inventory control technique is based on the principle that a small portion of the items may typically
represent the bulk of money value of the total inventory used in the production process, while a relatively large
number of items may from a small part of the money value of stores. The money value is ascertained by multiplying
the quantity of material of each item by its unit price.

According to this approach to inventory control high value items are more closely controlled than low value items.
Each item of inventory is given A, B or C denomination depending upon the amount spent for that particular item.
“A”

or the highest value items should be under the tight control and under responsibility of the most experienced
personnel, while “C” or the lowest value may be under simple physical control.
It may also be clear with the help of the following examples:
“A” Category – 5% to 10% of the items represent 70% to 75%
of the money value.
“B” Category – 15% to 20% of the items represent 15% to 20%
of the money.
“C” Category – The remaining number of the items represent
5% to 10% of the money value.
The relative position of these items show that items of category A should be under the maximum control, items of
category B may not be given that much attention and item C may be under a loose control.

Particulars A item B item C item
Control Tight Moderate Loose
Requirement Exact Exact Estimated
Check Close Some Little
Expenditure Regular Some No
Posting Industrial Individual Group/none
Safety Stock Low Medium Lare

Advantages of ABC Analysis
1. It ensures a closer and a more strict control over such items, which are having a sizable investment in there.
2. It releases working capital, which would otherwise have been locked up for a more profitable investment
3. It reduces inventory-carrying cost.
4. It enables the relaxation of control for the ‘C’ items and thus makes it possible for a sufficient buffer stock.
5. It enables the maintenance of high inventory turn over rate.

Fixation of Norms of Inventory Holdings
Either by the top management or by the materials department could set the norms for inventories. The top
management usually sets monitory limits for investment in inventories. The materials department has to allocate
this investment to the various items and ensure the smooth operation of the concern. It would be worthwhile if
norms of inventories were set by the management by objectives, concept. This concept expects the top
management to set the inventory norms (limit) after consultation with the materials department. A number of
factors enter into consideration in the determination of stock levels for individual items for the purpose of control
and economy. Some of them are:
1. Lead time for deliveries.
2. The rate of consumption.
3. Requirements of funds.
4. Keeping qualities, deterioration, evaporation etc.
5. Storage cost.
6. Availability of space.
7. Price fluctuations.
8. Insurance cost.
9. Obsolescence price.
10. Seasonal consideration of price and availability.
11. EOQ (Economic Order Quantity), and
12. Government and other statuary restriction

Any decision involving procurement storage and uses of item will have to be based on an overall appreciation of the
influence of the critical ones among them. Material control necessitates the maintenance of inventory of every item
of material as low as possible ensuring at the same time, its availability as and when required for production. These
twin objectives are achieved only by a proper planning of inventory levels.

An efficient inventory management, therefore, requires the company to maintain inventories at an optimum level
where inventory costs are minimum and at the same time there is no stock out which may result in loss of sale or
stoppage of production. This necessitates the determination of the minimum and maximum level of inventories.

Minimum Level
The minimum level of inventories of their reorder point may be determined on the following bases:
1 Consumption during lead-time.
2 Consumption during lead-time plus safety stock.
3 Stock out costs.
4 Customers irritation and loss of goodwill and production hold costs.

To continue production during Lead Time it is essential to maintain some inventories. Lead Time has been defined as
the interval between the placing of an order (with a supplier) and the time at which the goods are available to meet
the consumer needs.
There are sometimes fluctuations in the lead-time and/ or in the consumption rate. If no provision is made
for these

variations, stock out may take place-causing disruption in the production schedule of the company. The stock, which
takes care to the fluctuation in demand, varies in lead-time and consumption rate is known as safety stock. Safety
stock may be defined as the minimum additional inventory, which serves as a safety margin or buffer or cushion to
meet an unanticipated increase in usage resulting from an unusually high demand and or an uncontrollable late
receipt of incoming inventory. It can be determined on the basis of the consumption rate, plus other relevant factor
such as transport bottleneck, strikes or shutdowns. In the case of uncertainly, the probabilistic approach may be
applied to determine the safety margin. To avoid stock out arising out of such eventualities, companies always carry
some minimum level of inventories including safety stock.

Maximum Level
The upper limit beyond which the quantity of any item is not normally allowed to rise is known as the “Maximum
Level”. It is the sum total of the minimum quantity, and ECQ. The fixation of the maximum level depends upon a
number of factors, such as, the storage space available, the nature of the material i.e. chances of deterioration and
obsolescence, capital outlay, the time necessary to obtain fresh supplies, the ECQ, the cost of storage and
government restriction.

Re-Order Level
Also known as the ‘ordering level’ the reorder level is that level of stock at which a purchase requisition is initiated
by the storekeeper for replenishing the stock. This level is set between the maximum and the minimum level in such
a way that before the material ordered for are received into the stores, there is sufficient quantity on hand to cover
both normal and abnormal circumstances. The fixation of ordering level depends upon two important factors viz, the
maximum delivery period and the maximum rate of consumption.

Re-Order Quantity
The quantity, which is ordered when the stock of an item falls to the reorder level, is know as the reorder quantity or
the EOQ or the economic lot size. Although it is not a stock level as such, the reorder quantity has a direct bearing
upon the stock level in as much as it is necessary to consider the maximum

minimum stock level in determining the quantity to be ordered. The re-order quantity should be such that, when it is
added to the minimum quantity, the maximum level is not exceeded. the re-order quantity depends upon two
important factors viz, order costs and inventory carrying costs. It is, however, necessary to remember that the
ordering cost and inventory carrying cost are opposed to each other. Frequent purchases in small quantities, no
doubt reduce carrying cost, but the ordering costs such as the cost inviting tenders of placing order and of receiving
and inspection, goes up. If on the other hand purchases are made in large quantities, carrying costs, such as, the
interest on capital, rent, insurance, handling charges and losses and wastage, will be more than the ordering costs.
The EOQ is therefore determined by balancing these opposing costs.

Economy Order Quantity
The EOQ refers to the order size that will result in the lowest total of order and carrying costs for an item of

inventory. If a firm place unnecessary orders it will incur unneeded order costs. If a firm places too few order, it
must maintain large stocks of goods and will have excessive carrying cost. By calculating an economic order
quantity, the firm identifies the number of units to order that result in the lowest total of these two costs.

The constraints and assumption followed:
1. Demand is known-- Using past data and future plans a reasonably accurate prediction of demand can often be
made. This is expressed in unit sold in a year.
2. Sales occur at a constant rate-- This model may be used for goods that are sold in relatively constant amount
throughout the year. A more complicated model is needed for firms whose sales fluctuate in response to there
seasonal cyclical factors.
3. Cost of running of goods are ignored-- Cost associated with storage, delays or lost sales are not considered. These
costs are considered in the determination of safety level in the re-order point subsystem.
4. Safety stock level is not considered-- The safety stock level is the minimum level of inventory that the firm wishes
to hold as a protection against running out. Since the firm must always be above this level the EOQ need not be
considered the safety stock level.

Total Ordering Cost (TOC)=(A/Q)*O
Average Inventory=Q/2
Total Carrying Cost (TCC)=(Q/2)*C

Total Inventory Cost=TOC+TCC
Total Cost=(AO/2)+(QC/2)

Where,A=total annual demand
Q=Quantity order in units
O=Order cost per order
C=Carrying cost per unit

The basic formula is EOQ = 2(U)(OC)/CC*PP

Where,
2=mathematical factor that occurs during the deriving of the formula,
U-Units sold per year, a forecast provided by the marketing department.
OC=Cost of placing each order for more inventory provided by cost accounting.
CC% = Inventory carrying cost expressed as a percentage of the average value of the inventory, an estimate usually
provided by cost accounting.
PP = Purchase price per each unit of inventory supplied by the purchasing department.

Trial and error approach
Select a number of possible lot (Order) sizes to purchase, then determine the total cost for each lot size chosen, now
select the ordering quantity that minimizes the total cost.
Quantity Discount and Order Quantity
The standard EOQ analysis is based on the assumption that the price per unit remains constant irrespective of the
order size. When quantity discount are available which is often the case, price per unit is influenced by the ordered
quantity. This violates the applicability of the EOQ formulas. However
205

the EOQ framework can still be used as a starting point for analyzing the problem.

To determine the optimal order size when quantity discount is available, the following procedures may be followed:

1. Determine the order quantity using the standard EOQ formula assuming no quantity discount.
2. If Q enable the firm to get quantity discount then it represents the optimal order size.
2
3. If Q is less then the minimum order size required for quantity discount (call it-G ) compute to change in
1 2
profit as a result of increasing the order quantity from O to O as follows.
1 2 2 1
=AD+[A/Q -A/Q ] O-[Q ((P-D)/2-(Q PC/2)
= Change in profit,
A = total demand,
D = discount per unit when quantity discount in available,
1
Q = EOR assuming no discount,
2
Q = minimum order size required for quantity discount,
O = order cost, P = Purchase price without discount,
C = carrying cost
2
4. If change in profit is positive = Q
1
If change in profit is positive = Q

Reorder Point
The reorder point is the level of inventory at which the firm places an order in the amount of EOQ. If the firm places
the order when the inventory reaches the reorder point, the new goods will arrive before the firm runs out of goods
to sell.
In designing reorder point subsystem, three items of information are needed as inputs to the subsystem.
1. Usage rate-- This is the rate per day at which the item is consumed in production or sold to customers. It
is expressed in units. It may be calculated by dividing annual sales by 365 days. If the sales are 50,000 units
the usage rate is 50,000/365 = 137 Units per day.
2. Lead time-- This is the amount of time between placing an order and receiving goods. This information is
usually provided by the purchasing department. The time to allow for an order to arrive may be estimated
from a check of the company’s record and the time taken in the past for different suppliers to fill orders.
3. Safety stock-- The minimum level of inventory may be expressed in terms of several days’ sales. The level
can be calculated by multiplying the usage rate and time in the number of days that the firm wants to hold
as a protection against shortages.
Re-order point = (Usage rate)(Lead time + Days of safety) = (Lead Time x Consumption rate) + Safety stock.

Factors Influences the Level of each Component of Inventory
Raw Material Inventory:
1. The volume of safety stock against material shortages that interrupt production.
2. Considerations of economy in purchase.
3. The outlook for future movements in the price of materials.
4. Anticipated volume of usage and consumption.
5. The efficiency of procurement and inventory control function.
6. The operating costs of carrying the stocks. .
Work-in-process Inventory:
1. The length of the complete production process.
2. Management policies affecting length of process time.
3. Length of process in runs.
4. Action that speed up the production process, e.g. adding second or third production shifts.
5. Management’s skills in production scheduling and control.
6. Volume of production.
7. Sales expectations.
8. Level of sales and new orders.
9. Price level of raw materials used, wages and other items that enter production cost.
10. Customer requirements.

11. Usual period of aging.

Finished Goods Inventory:
1. The policy of the management to gear the production to meet the firm order in hand.

2. The policy to produce for anticipated orders and stock keeping.
3. Goods required or the purpose of minimum and safety stocks.
4. Sales policies of the firm.
5. Need for maintaining stability in production.
6. Price fluctuations for the product.
7. Durability, spoilage and obsolescence.
8. Distribution system.
9. Ability to fill orders immediately.
10. Availability of raw material on seasonal basis while customer’s demand spread throughout the year.
11. Storage capacity.
Stores and Spares Inventory:
1. Nature of the product to be manufactured and its lead-time of manufacture.
2. State of technology involved.
3. Consumption’s patterns.
4. Lead time of supply.
5. Indigenous or foreign.
6. Minimum and safety stock and ordering quantities.
7. Capacity utilization.
8. Importing formalities.

Demand forecast process and inventory management

More Related Content

What's hot (20)

Similar to Demand forecast process and inventory management (20)

Recently uploaded (20)

Demand forecast process and inventory management