FORECASTING

Another important part of operations management decision-making is

forecasting. It is the process of making predictions about the future demand based on
the past events and present data. Prediction is a similar, but more general term. Both
might refer to formal statistical methods employing time series, cross-
sectional or longitudinal data, or alternatively to less formal judgmental methods. Usage
can differ between areas of application: for example, in hydrology, the terms “forecast”
and “forecasting” are sometimes reserved for estimates of values at certain
specific future times, while the term “prediction” is used for more general estimates,
such as the number of times floods will occur over a long period.

Forecasts are known to be significant in two ways. The first is the expected
level of demand, in which it can be a function of some structural variation such as the
trend or seasonal variation. The other one is the degree of accuracy. It is a function of
the ability of the forecasters to correctly model demand, random variation, and
sometimes unforeseen events.

Forecasting is conducted by what are referred to as time horizons. The first one
is the forecast that can be used for a year, however, it is usually used for only three
months or less, known as the short range forecast. This use different methodologies
than the others. Most short term forecasts are quantitative in nature and use existing
data in mathematical formulas to anticipate immediate future needs and impacts. Also,
they are more accurate than medium or long range forecasts because a lot can change
in three months, a year, three years, and longer. Factors that could influence those
forecasts change every day. Short term forecasts need to be updated regularly to
maintain their effectiveness. It is best used for planning purchases, job hiring, job
assignments, production levels and the like. The next one is the medium range
forecast. This is generally three months to three years. Medium range forecasts are
used for sales and production planning, budgeting, and analysis of different operating
plans. Lastly, the long range forecast. Generally used for three years or more in time
span, it is used for new products, capital expenditures, facility expansion, relocation,
and research and development.
 Medium and long range forecasts differ from short range forecasts by other
characteristics as well. Medium and long range forecasts are more comprehensive in
nature. They support and guide management decisions in planning products,
processes, and plants. A new plant can take seven or eight years from the time it is
thought of, until it is ready to move into and become functional.

In operations management, forecasts are

Important for the decision process as they
Provide information on future demand.
 Commonly demand is unknown  a forecast of
Demand is essential for determining how much
Capacity will be needed to meet demand.
In operations management, forecasts are
Important for the decision process as they
Provide information on future demand.
 Commonly demand is unknown  a forecast of
Demand is essential for determining how much
Capacity will be needed to meet demand.
ELEMENTS OF A GOOD FORECAST

1. Timely – Usually, a certain amount of time is needed to respond to the information

contained in a forecast. Forecasting horizons must cover the time necessary to
implement possible changes.
2. Accurate – The degree of accuracy should be stated to provide the basis for
comparing forecast alternatives.
3. Reliable – It should work consistently over different forecasted scenarios. A
technique that sometimes provides a good forecast and sometimes a poor one
will leave users with the uneasy feeling that they may get burned every time a
new forecast is issued.
4. Meaningful Units – The forecast results should be expressed in meaningful units
to provide similar understanding with all the concern.
5. Written – This is to increase the likelihood of using the same technique in
presenting an information. However, it is understood that this will not guarantee
that all of the concerns are using the same information.
6. Simple to understand – Users often lack confidence in forecasts based on
sophisticated techniques; they do not understand either the circumstances in
which the techniques are appropriate or the limitations of the techniques. Misuse
of techniques is an obvious consequence. Fortunately, fairly simple forecasting
techniques enjoy widespread popularity because users are more comfortable
working with them.
7. Cost-effective – The benefits should outweigh the costs.

STEPS IN THE FORECASTING PROCESS

The prediction of future events will surely provide a huge difference to an

organization. However, the process of forecasting is sometimes being overlooked in the
part of the business management. Probably because it is somehow a time consuming
activity that most organizations are just ignoring it. There are six basic steps to follow in
conducting a forecasting process.

1. Determine the purpose and use of the forecast – Identify the problem where the
forecast will it be used and be needed. This step will provide an indication of the
level of detail required in the forecast, the amount of resources (personnel,
computer time, and dollars) that can be justified, and the level of accuracy
necessary.
2. Establish a time horizon – The forecast must indicate a time interval, keeping in
mind that accuracy decreases as the time horizon increases.
3. Obtain, clean, and analyze appropriate data. Obtaining the data can involve
significant effort. Once obtained, the data may need to be “cleaned” to get rid of
outliers and obviously incorrect data before analysis.
4. Select a forecasting technique.
5. Make the forecast.
6. Monitor the forecast errors. The forecast errors should be monitored to determine
if the forecast is performing in a satisfactory manner. If it is not, reexamine the
method, assumptions, validity of data, and so on; modify as needed; and prepare a
revised forecast.
 It's also important to note that additional action may be needed. For example, if
demand was significantly lower than expected, a price drop or a promotion could be
required. Increased output, on the other hand, may be favorable if demand is
significantly higher than expected. Working overtime, outsourcing, or adopting other
measures may be necessary.

APPROACHES TO FORECASTING

Forecasting may be categorized into two kinds: qualitative and quantitative. The
q qualitative approaches rely heavily on subjective inputs that are frequently difficult to
quantify. While on the other hand, quantitative approaches entail either historical data
projections or the formulation of associative models that aim to foresee using causal
(explanatory) factors. The human aspects, personal experiences can be included using
qualitative approaches.

In the forecasting process, opinions, hunches are used. Since they are difficult or
impossible to quantify, such aspects are frequently overlooked or minimized when
quantitative approaches are applied. The majority of quantitative procedures involve
evaluating objective, or hard data. Personal biases, which can corrupt qualitative
approaches, are normally avoided. As a matter of fact, a prediction might be developed
using either technique or a mix of the two.

There are a variety of forecasting techniques available. The first technique is the
analysis of subjective inputs received from numerous sources, including as consumer
surveys, sales employees, managers and executives, and expert panels, is used to
make judgmental forecasts. These sources typically give information that would not
otherwise be available. The second forecasting technique is the time-series forecasts.
It essentially try to extrapolate previous data into the future. These methods make use
of historical data, assuming that the future will be similar to the past. Some models only
seek to smooth out random changes in historical data, while others attempt to find
particular patterns in the data and project or extrapolate those patterns into the future
without attempting to understand the patterns' sources. The last one is the associative
models. It make use of equations with one or more explanatory variables that may be
utilized to forecast demand. Paint demand, for instance, might be influenced by factors
such as the price per gallon and the amount spent on promotion, as well as the paint's
special features (e.g., drying time, ease of cleanup).

FORECAST BASED ON JUDGMENTS AND OPINION

Forecasters may make predictions simply on the basis of their own judgment and
opinion. At times, there are only limited time available to collect and analyze quantitative
data if management requires a projection immediately. In most cases, specifically when
political and economic situations are unfavorable, data may become obsolete as time
passes, and more recent data may not yet be accessible. Similarly, the launch of new
items and the redesign of existing products or packaging are being constrained due to
lack of historical data that may be used to foresee. Forecasts are made using CEO
opinions, consumer surveys, salesforce opinions, and expert opinions in certain
situations.

 Executive opinions
The small group of upper-level managers, in the marketing, operations, and
finance department may gather and construct a projection together. This method is
frequently utilized in long-term planning and new product development. It offers the
benefit of bringing together the vast expertise and abilities of many management.
However, there's a chance that one person's viewpoint may win out, and spreading
responsibility for the prediction over the entire group may result in less pressure to
deliver a solid forecast.

 Salesforce opinions
Due to their direct interaction with customers, members of the sales or customer
service personnel are frequently important sources of information. They are frequently
aware of any future plans that clients may be considering. However, there are a few
disadvantages to leveraging salesforce opinions. One possibility is that employees will
be unable to discern between what consumers want to do and what they will really
accomplish. Another factor is that these individuals are sometimes unduly impacted by
recent events. As a result, their forecasts may turn gloomy after multiple periods of poor
sales. They may get overconfident after numerous periods of strong sales. Furthermore,
if projections are utilized to set sales quotas, there will be a discrepancy. Because it is in
the salesperson's best interests to produce low sales projections, there is a conflict of
interest.

 Consumer surveys
Because customers are the ones who ultimately set demand, it seems only
reasonable to seek their feedback. Every client or potential customer may be solicited in
some cases. In most cases, however, there are either too many consumers or no
method to identify all possible clients. As a result, businesses seeking customer
feedback frequently use consumer surveys to sample public perspectives. Consumer
surveys have the apparent advantage of obtaining information that isn't readily available
elsewhere. On the other hand takes a significant amount of expertise and ability to
construct a survey, administer it, and accurately interpret the findings for meaningful
information. In addition, surveys may be costly and time-consuming to conduct.

 Other approaches
A manager may seek input from a variety of different supervisors and employees.
On rare occasions, external expertise are required to assist with a forecast. Advice on
political or economic situations in the United States or a foreign nation, or any other
important topic with which an organization is unfamiliar, may be required. The Delphi
method, an iterative procedure aimed at reaching a consensus prediction. In which
managers and staff complete a series of questionnaires, each developed from the
previous one, to achieve a consensus forecast. This strategy entails sending out a
series of questions to people who have the expertise and skills to make a significant
contribution. The confidentiality of the comments encourages honest responses and
decreases the chance that one person's perspective will win over.
The information gathered is used to create each subsequent questionnaire. From
the preceding one, hence expanding the range of data on which participants may make
choices. The Delphi technique has been used to solve a wide range of problems, not all
of which need forecasting. The focus of this article is on its application as a forecasting
tool. The Delphi method is an effective forecasting technique for technical forecasting,
or examining changes in technology and their influence on a company. The objective is
frequently to anticipate when a specific occurrence will occur.

FORECASTS BASED ON TIME-SERIES DATA

A time series is a time-ordered collection of observations recorded at regular

intervals (hourly, daily, weekly, monthly, quarterly, and yearly, for example). Demand,
earnings, profits, shipping, accidents, output, precipitation, productivity, and the
consumer price index are examples of data that can be measured. On the premise that
future values of the series may be inferred from previous values, forecasting techniques
based on time-series data are used. Despite the fact that an attempt is made to
discover factors that impact the series; these approaches are commonly utilized, and
the results are frequently satisfactory.

When analyzing time-series data, the analyst must first determine the series'
underlying behavior. Often, simply graphing the data and visually inspecting the figure
will suffice. Trends, seasonal fluctuations, cycles, or changes around an average are
some of the patterns that may arise. There will also be random and maybe irregular
variances. These behaviors can be described as follows:

1. Trend refers to a long-term upward or downward movement in the data. Population

shifts, changing incomes, and cultural changes often account for such movements.
2. Seasonality refers to short-term, fairly regular variations generally related to factors
such as the calendar or time of day. Restaurants, supermarkets, and theaters
experience weekly and even daily “seasonal” variations.
3. Cycles are wavelike variations of more than one year’s duration. These are often
related to a variety of economic, political, and even agricultural conditions.
4. Irregular variations are due to unusual circumstances such as severe weather
conditions, strikes, or a major change in a product or service. They do not reflect
typical behavior, and their inclusion in the series can distort the overall picture.
Whenever possible, these should be identified and removed from the data.
5. Random variations are residual variations that remain after all other behaviors have
been accounted for.

ASSOCIATIVE FORECASTING TECHNIQUES

Associative techniques rely on the discovery of related variables that may be

used to predict the value of the target variable. For example, beef sales may be
influenced by the price per pound charged for beef as well as the prices of substitutes
such as chicken, pork, and lamb; real estate prices are usually influenced by location
and square footage; and crop yields are influenced by soil conditions as well as the
amounts and timing of water and fertilizer applications. The construction of an equation
that describes the impacts of predictor variables is at the heart of associative
approaches. Regression is the most used type of analysis. A basic explanation of
regression should be sufficient to put this method in context. The various forecasting
methods outlined in this chapter.

 Simple Linear Regression

To model the relationship between two continuous variables, simple linear

regression is utilized. The goal is frequently to anticipate the value of an output variable
(or responder) based on the value of an input variable (or predictor).

We are usually interested in determining the link between a numbers of factors.

To investigate potential links between pairs of data, scatterplots and scatterplot matrices
can be utilized. Correlation measures the linear link between two variables, but it
doesn't reveal more complicated correlations. If the link is curvilinear, for example, the
correlation may be approaching zero. Regression can help you gain a more formal
grasp of the connections between variables. We seek to describe the link between an
output variable, or a response, and one or more input variables, or factors, in regression
and statistical modeling in general. In addition, regression may also be used to forecast
the values of a response variable based on the values of the key predictors. Predictive
modeling is the term used to describe this process. Alternatively, we may utilize
regression models to improve a response by determining the settings of components.
The purpose of optimization can be to discover parameters that result in a maximum or
minimal response. Alternatively, the objective may be to hit a target within a reasonable
time frame.

 Nonlinear and Multiple Regression Analysis

Because a linear model is incorrect or because more than one predictor variable
is involved, simple linear regression may be insufficient to address certain issues. When
there are nonlinear relationships, you should utilize nonlinear regression; models with
more than one predictor should use multiple regression analysis. While these analyses
are outside the scope of this article, they are often employed. Multiple regression
forecasting considerably increases data needs.

References:

Simple Linear Regression. (n.d.). Introduction to Statistics | JMP.

https://www.jmp.com/en_ph/statistics-knowledge-portal/what-is-regression.html#:

%7E:text=What%20is%20simple%20linear%20regression,input%20(or

%20predictor)%20variable.

Stevenson, W. J. (n.d.). Operations Management (Twelfth ed.). McGraw-Hill

Companies, Inc.,. https://drive.google.com/file/d/1J1XXhfJUtqwIQ7-

95GAGhlqRwfQXh0Li/view?
fbclid=IwAR13dPO_b_ENNYLP8QUo0CrLeHRauRP68MLoEIPQXGsZXCVKaXh

0vSR-PAo