Forecasting involves using past data to estimate future events. It is a vital function that impacts management decisions. There are qualitative and quantitative forecasting methods. Qualitative methods include expert opinions, surveys, and the Delphi technique. Quantitative methods include time series models, decomposition, and regression. Time series models assume the future is related to the past and use trends in historical data. Decomposition breaks down time series into trend, seasonal, cyclical, and random components. Regression establishes relationships between variables to forecast a dependent variable.
2. • Forecasting is the process of estimating a future
event by casting forward past data
• Forecasting is the scientific function to the
operations management, to understand the
customer demand
• Forecasting is a vital function and affects every
significant management decision
4. Qualitative Methods
• Jury of Executive opinion-
Opinions of Higher levels of managers in combination with statistical models are
utilized to group the estimate of demand.
• Sales Force Composite-
Each sales person estimates what sales will be in his/her region. This forecast is
then reviewed to ensure they are realistic and then combined at the district and
national levels to reach an overall forecast.
• Consumer Market Survey-
Utilizing inputs from potential customers and consumers regarding future
purchase plans. This method is also handy in improving product design, and
planning for new products.
5. Qualitative Methods
• Delphi Method-
process intended to achieve consensus forecasts, specifically avoiding direct
inter-personal relations.
Primarily, three different participants in Delphi methods are- i) Decision
makers/ expert ii) Staff personal/ coordinator, iii) Respondents.
Procedure followed stepwise is as follows:
• Posing questions to participants
• Writing brief prediction
• Co-coordinator collating, and editing the prediction inputs together
• Requisitioning on the basis of input responses received.
• Feedbacks in writing
• Re-updating of the feedback and synthesizing the consensus
• Nominal Group Discussions:
Process is similar to Delphi technique, only difference is – experts are
allowed to sit in a group, discuss, debate and synthesize the consensus.
6. Quantitative Methods
time series model
smoothing
• Assumptions : future is the function of past.
• The past trends and historical data is utilized to for the forecasting.
• Some of the basic tools of time series smoothing models are
i) Simple Average
ii) Moving Average
iii) Exponential Smoothing.
7. 1. Simple Average
• Depends on the detecting the central tendency of demand
If, Di – demand of the ith period
n- no. of periods
X= sum of demands for all the periods number of periods/
number of periods
•One of the disadvantage of this method, extreme outliers and values affect
the central tendency, and affect the results.
8. January February March April May June
200 250 260 280 270 290
Example,
Consider the Tata Sky monthly new connections in a city are as follows:
Forecast, the demand for July month
9. 2. Moving Average
A moving average forecast uses a number of recent actual data
values from several of the most recent periods to generate a
forecast.
Month Actual Sales Forecast
January 200
February 268
March 285
April 280 (200+268+285)/3= 251
May - (268+285+280)/3= 277.66 ≈ 278
10. 3. Exponential Smoothing
• The pattern of weights of demand follows the exponential
demand.
• Demand for the most recent period is weighted most heavily
• The basic exponential smoothing formula for creating a new or
updated forecast (Ft) uses two pieces of information:
i) actual demand for the most recent period (Dt-1)
ii) Most recent demand forecast (Ft-1).
11. 3. Exponential Smoothing
Exponential smoothing based demand is forecasted using:
Ft=αD(t−1)+(1−α)F(t−1) ------------------------- (1)
Where,
F= Forecast
𝛼 = Smoothing coefficient (0≤ 𝛼 ≤ 1)
D = Demand
t = is the period
t-1 = immediate previous period.
Expanding the exponential form, the equivalent form of equation (1)
becomes,
Ft=α(1−α)0 D (t−1) + α(1−α)1 D(t−2) + α(1−α)2 D(t−3)
12. 3. Exponential Smoothing
Example,
Company develops and launches new product in August 2021. The actual
sales of product in September and October 2021 were 200 and 350 units
respectively. Forecast for month September was 200 units. Considering the
given forecasts and sales values, predict the demand for November
month. (take α =0.7)
Step 1: Calculation of Forecast for immediate previous month
F October = αD September + (1- α) F September
F October = (0.7*300) + [(1-0.7) * 200]
F October = 270 units
Step 2: Calculation of Forecast for the required period
F November = αD October + (1- α) F October
F November = (0.7 * 350) + [(1-0.7) *270]
F November = 326
13. Quantitative Methods
time series model
Decomposition
• directs the separation of series into the basic components that are likely to
have predictable or more recognizable pattern.
• Four basic types of time series components are-
a) Trend, b) Cyclical, c) Seasonal, and d) Random.
• General forms of the time series decomposition model are:
i) Multiplicative Model (Forecast is done by multiplying time series components)
ii) Additive Model (Forecast is done by adding the time series components).
14. Quantitative Methods
time series model
Decomposition
If, TF = time series forecast
T= trend component
S= measure of seasonality
C= measure of cyclical adjustment
R= random component
Then,
• Multiplicative model,
TF= T x S x C x R
• Additive Model,
TF = TF= T + S + C + R
15. Quantitative Methods
Causal / Regression Model
• Regression model is a causal forecasting technique output that
establishes a relationship between variables. There is one dependent
variable and one or more explanatory variables.
• Historical data establishes a functional relationship between the two
variables.
• If there is one explanatory variable it is called simple regression,
otherwise, it becomes multiple regression.
• If the model takes the shape of a linear equation, we call it simple linear
regression or multiple linear regression models.
• Normally least square curve fitting technique is used to develop the
models for fitting a line to a set of points.
16. Quantitative Methods
Causal / Regression Model
• The objective in linear regression is to obtain an equation of a straight
line that minimizes the sum of squared vertical deviations of data points
from the line.
• This least square curve has the equation:
Y = a + bX
Y- Dependent variable (Predicted)- Normally drawn on Y axis
X- Explanatory variable - Normally drawn on X axis
a- value of predicted variable (Y) when x = 0 (Y-intercept value of the line)
b- slope of line (Change in Y corresponding to a unit change in X).
17. Quantitative Methods
Causal / Regression Model
•The coefficients a and b of the regression line are computed using the following two
equations:
•
18. Quantitative Methods
Causal / Regression Model
Example:
A firms sales for a product line during the 12 quarters of the past three years were as
follows.
Quarter Sales Quarter Sales
1 600 7 2600
2 1550 8 2900
3 1500 9 3800
4 1500 10 4500
5 2400 11 4000
6 3100 12 4900
Forecast the sales for the 13, 14, 15 and 16th quarters using a regression equation.
19.
20. Quantitative Methods
Causal / Regression Model
Y = 400 + 382X
Quarter Forecast
13 400 + 382(13) = 5366
14 400 + 382(14) = 5748
15 400 + 382(15) = 6130
16 400 + 382(16) = 6512
The forecasts for quarters 13 to 16 are