Forecasting methods like trend extrapolation, exponential smoothing,
and moving averages.
The art and science of anticipating future events is known
as forecasting. Forecasting may involve taking previous Short-range forecasts compared to longer range
information into account and using a mathematical forecasts; this tend to be more accurate.
model in predicting the future data. A subjective or
Types of Forecasts
intuitive prediction might be made (for instance, "This is
not good for a new product and will sell 20% more than Economic forecasts predicting inflation rates, supplies of
the old one"). Some projections are made using money, housing starts, and other planning indicators are
demand driven data, such as future purchase intentions being considered in this type of forecast to address the
of customers. Perhaps the forecast might incorporate a business cycle
combination of these, that is, a mathematical model that
has been modified by an experienced manager. Technological forecasts are concerned about the pace of
Depending on where a product is in the product life technological development, which may lead to the
cycle—whether sales are in the beginning, growing, creation of new products, interesting items that call for
maturing, or declining stages—forecasts may change. The new facilities and machinery.
demand for a connected product can have an impact on Demand forecasts use by most company to check
other products. For instance, navigation systems may projection of goods and services. This drive decisions for
correlate with new automobile sales. We create error operations managers to project the demand of their
measurements or acceptable limits because there are product. Demand-driven projections have the goal of
restrictions on what may be anticipated from projections. quickly detecting and following client preferences.
It can cost money and take time to prepare and monitor
forecasts. Yet, only a small number of organizations can The Strategic Importance of Forecasting
afford to forego the forecasting process in favor of simply Supply-Chain Management Accurate projections are
waiting and taking a chance. A projection of the necessary for good supplier relationships and the
company's product demand is essential for effective benefits that follow in terms of product innovation, cost,
planning over the long and short terms. and time to market
Forecast horizon is the amount of time that forecasts Human Resources Demand projections are used when
must be made for in the future. hiring, training, and terminating employees. The amount
Short-range forecast: of training decreases and the workforce's quality worsens
if the human resources division is forced to hire more
This forecast typically lasts less than three months but employees without warning.
can last up to a year. The workforce, job assignments,
buying, job scheduling, and production levels are all Capacity Lack of capacity can cause shortages that can
planned using it. cost businesses customers and market share.
Medium-range forecast: Seven Steps in the Forecasting System
This prediction typically covers the period from three 1. Analyze what is the use of forecast.
months to three years. It is helpful in the study of various 2. Select which items to be forecasted.
operating plans as well as the planning and budgeting of
sales, production, and cash flow. 3. Identify the time horizon of the forecast: Is it short,
medium, or long term?
Long-range forecast:
4. Choose the forecasting model(s).
Usually used in planning for new goods, capital
expenditures, facility location or expansion, and research 5. Gathering of data in making the forecasts.
and development. They typically have a time horizon of
6. Making the forecast.
three years or more.
7. Validation and implementation of results.
Intermediate and long-range forecasts give management
choices for planning, products, plants, and procedures The forecast is frequently impacted by external
variables that we cannot predict or control.
Short-term forecasting uses methods that are typically
distinct from those used in long-term forecasting. Short The majority of forecasting methods make the
range predictions frequently include mathematical assumption that the system fundamentally stable.
Forecasting Approaches There are two general Quantitative Forecasting Models Time series analysis
approaches to predict future data or forecasting. The first and associative modeling techniques are two recognized
is a quantitative forecasting and the second is a methods or groups in quantitative forecasting.
qualitative forecasting.
Quantitative forecasts make use of a range of
mathematical models that project demand using
historical data and/or associative factors.
Forecasts that are subjective or Qualitative forecasting
take into account the decision maker's intuition, feelings, Time-Series Models. Time-series models predict on the
past experiences, and value system. Some businesses premise that the past influences the future. In other
employ one strategy while others employ the other. In words, they analyze previous events and estimate the
actuality, a mix of the two is frequently most productive. future based on a collection of historical information
Qualitative Forecasting Techniques: Associative Models. Associative models, like linear
regression, include the variables or elements that may
1. Jury of executive opinion: Using this approach, a
have an impact on the quantity being predicted.
group estimate of demand is created by combining the
Decomposition of a Time Series
viewpoints of a number of high-level specialists or
managers, frequently in conjunction with statistical Trend. It is the progressive upward or downward
models. movement of the data over time. Trend movement may
be explained by shifts in income, population, age
Executive Opinion. A technique for forecasting where the
distribution, or cultural perspectives.
insights, knowledge, and technical expertise of one or
more managers are compiled into a single forecast Seasonality. It is a recurring pattern in data that appears
after a few days, weeks, months, or quarters.
2. Delphi method: In this method, there are three
different types of participants: decision makers, staff Cycles. These are data patterns that appear every few
personnel, and respondents. There will be a team of 5 to years. They play a significant role in short-term company
10 experts, who serve as decision makers. A number of analysis and planning and are typically linked to the
questionnaires and survey findings are prepared, business cycle. Forecasting business cycles is challenging
distributed, gathered, and summarized by staff members because they may be altered by political events or by
to help decision makers. The respondents are a collection international turmoil.
of people, frequently dispersed across many locations,
Random variations. These "blips" in the data are the
whose opinions are respected. Before making a forecast,
result of accidental events and peculiar circumstances.
this group offers input to the decision-makers
They exhibit no clear pattern, making prediction
Delphi Method. A method of having a panel of experts impossible
reach a decision while anonymous.
Naive Approach Assuming that demand will be equal to
3. Sales force composite In this strategy, each demand in the most recent period is the easiest method
salesperson predicts the volume of sales for their to forecast. The most economical and effective objective
particular territory. These projections are then examined forecasting model is the naive approach. It offers atleast
to make sure they are reasonable. To create an overall a starting point for comparison with subsequent, more
prediction, they are then aggregated at the district and complex models. A time series method whereby the
national levels. forecast for the next period equals the demand for the
current period.
4. Market survey This approach seeks feedback on future
purchase plans from current or potential clients. It can Moving Averages A moving-average forecast creates a
assist with forecasting as well as product design and the forecast by averaging many historical real data values. If
planning of new products. Overly optimistic projections we can expect that market demands will remain largely
resulting from customer feedback, however, can hurt the constant over time, moving averages can be effective. In
consumer market survey and sales force composite a 4- month moving average, we can do it by just adding
approaches. up demand over the previous four months and dividing
by four yields. The simple moving average (which acts as
Market Survey. remaining Use data collection surveys in
a forecast of the demand for the following period) can be
order to ascertain external client interest in a service or
represented mathematically as follows
product.
Weighted Moving Averages
When a detectable trend or pattern is present, weights
can be used to place more emphasis on recent values.
This practice makes forecasting techniques more
responsive to changes because more recent periods may
be more heavily weighted A weighted moving average
may be expressed mathematically as:
Exponential Smoothing
Another weighted-moving-average forecasting technique
is exponential smoothing. It is comparatively simple to
operate and requires very little historical data archiving.
As may be seen, the fundamental exponential smoothing
formula is as follows:
Trend Projections For medium- to long-range projections,
this method fits a trend line to a set of historical data
points before projecting the slope of the line into the
future. While exponential and quadratic trend equations
can be generated mathematically, we will only consider
linear (straight-line) trends in this method The height at
which a least-squares line intercepts the y-axis and the
expected change are used to define the line (slope). The
line can be represented by the following equation if we
can determine the slope and y-intercept: ŷ = a + bx
Linear Regression Analysis
Associative forecasting models usually consider several
variables that are related to the quantity being predicted.
Once these related variables have been found, a
statistical model is built and used to forecast the item of
interest. This approach is more powerful than the time-
series methods that use only the historical values for the
forecast variable. The most common quantitative
associative forecasting model is linear-regression
analysis. ŷ = a + bx
