Demand Forecasting
JITHIN K THOMAS
Berchmans Institute of
management Studies
Why Demand Forecasting
• Business environment is uncertain, volatile,
dynamic and risky.
• Better business decisions can be taken if
uncertainty can be eliminated or reduced.
• Demand forecasting predicting the future
demand for a firms product, is one way to
reduce uncertainty.
Demand Forecasting
• Forecasting is predication of a future event
• Demand Forecasting is the prediction of a
future situation under given constraints.
Objectives of Demand Forecasting
• Short term objectives
1. Price Policy Formulation
2. Proper control of sales
3. Arrangement of finance
4. Regular supply of raw material
5. Regular availability of labor
6. Formulation of production policy
Objectives of Demand Forecasting
• Long term Objectives
1. Labor requirement
2. Arrangement of Finance
3. Expansion
Types of Demand Forecasting
• Short term demand forecasting
• Long term demand forecasting
• Medium term demand forecasting
Methods of Demand Forecasting
Survey
method
Consu
mer
Survey
Collect
ive
Survey
Statistical method
Trend
Method
Regression
Method
Least
Square
Leading
Indicator
Simultaneous
Equation
Subjective (Qualitative) Method
• Relay on human judgment and opinion.
– Consumers Survey
• Complete Enumeration survey
• Sample survey
• End user method
– Sales Force Composite
– Market Simulation
– Test Marketing
– Expert Opinion
Trend Projection
• Assumption –
– future events are a continuation of the past
– Historical data can be used to predict the future
• Predictions
– Finding a trend for a specific year
– Finding seasonal fluctuations in the variable
– Predicting turning points in the future movement
of the variable
Methods of Finding Trend
1. Fitting Trend Line by Observation
– Plotting of annual sales on a graph and then
estimating by observation where the trend line
lies
2. Time Series Analysis employing Least Squares
Method
– With the help of statistic a trend line is fitted to
the data called ‘the line of best fit’ and then
extrapolated.
Methods of Finding Trend
3. Forecasting through Decomposing a Time
Series
4. Smoothing Methods
– Moving averages
– Exponential Smoothing
5. ARIMA Method
– Auto Regressive Integrated Moving Averages
Barometric or Leading Indicator
Technique
• Relationship can exist among various
economic time series
• Lagging series – Lagging Indicators
– Data moves up and down behind the series being
compared
• Coincident series – Coincident Indicators
– Data moves up and down with some other series
Barometric or Leading Indicator
Technique
• Leading series – Leading Indicators
– Data moves ahead of the series being compared
Application for
housing loan
Demand for
construction
material
Birth rate of
children
Demand of
seats in school
Correlation and Regression Method
• This method recognises the fact a number of
factors like own advertising, competitors
advertising, competitors price, weather
condition etc may affect sales.
• Variables influencing sales is identified
through correlation
• Regression equation specifies relationship
between each variable and sales.
Steps in Demand Forecasting
1. Problem definition
2. Gathering information
3. Preliminary exploratory analysis
4. Choosing the fitting model
5. Using the evaluating model
Econometric Models
• Econometric Models tries to identify all those
economic and demographic variables that
influences the future value of the variable
under forecasting and build up a cause-effect
relationship.
Simulations Equation method
• Simulations Equation method involves
simulations consideration of all the variables,
as it is believed that every variable influences
the other variable in an econometric decision
environment.
Range
Long
Medium
Short
Horizon
5 years
1 – 2 years
Up to 1 year
Application
Facility, capacity,
product planning
Staffing plan,
Aggregate
production Plan
Purchasing ,
Detailed job
scheduling
Methods
Economic,
Demographic,
Market
information,
Technology
Time series,
Regression
Trend exploration ,
Graphical Method,
Exponential
smoothing
Qualities of good Demand Forecasting
1. Simple
2. Accurate
3. Easy Availability
4. Economy
5. Capacity to update forecasts
Limitations
• Change in Fashion
• Consumers Psychology
• Lack of Past Data
• Uneconomical
• Lack of Experienced Experts
Demand forecasting

Demand forecasting

  • 1.
    Demand Forecasting JITHIN KTHOMAS Berchmans Institute of management Studies
  • 2.
    Why Demand Forecasting •Business environment is uncertain, volatile, dynamic and risky. • Better business decisions can be taken if uncertainty can be eliminated or reduced. • Demand forecasting predicting the future demand for a firms product, is one way to reduce uncertainty.
  • 3.
    Demand Forecasting • Forecastingis predication of a future event • Demand Forecasting is the prediction of a future situation under given constraints.
  • 4.
    Objectives of DemandForecasting • Short term objectives 1. Price Policy Formulation 2. Proper control of sales 3. Arrangement of finance 4. Regular supply of raw material 5. Regular availability of labor 6. Formulation of production policy
  • 5.
    Objectives of DemandForecasting • Long term Objectives 1. Labor requirement 2. Arrangement of Finance 3. Expansion
  • 6.
    Types of DemandForecasting • Short term demand forecasting • Long term demand forecasting • Medium term demand forecasting
  • 7.
    Methods of DemandForecasting Survey method Consu mer Survey Collect ive Survey Statistical method Trend Method Regression Method Least Square Leading Indicator Simultaneous Equation
  • 8.
    Subjective (Qualitative) Method •Relay on human judgment and opinion. – Consumers Survey • Complete Enumeration survey • Sample survey • End user method – Sales Force Composite – Market Simulation – Test Marketing – Expert Opinion
  • 9.
    Trend Projection • Assumption– – future events are a continuation of the past – Historical data can be used to predict the future • Predictions – Finding a trend for a specific year – Finding seasonal fluctuations in the variable – Predicting turning points in the future movement of the variable
  • 10.
    Methods of FindingTrend 1. Fitting Trend Line by Observation – Plotting of annual sales on a graph and then estimating by observation where the trend line lies 2. Time Series Analysis employing Least Squares Method – With the help of statistic a trend line is fitted to the data called ‘the line of best fit’ and then extrapolated.
  • 11.
    Methods of FindingTrend 3. Forecasting through Decomposing a Time Series 4. Smoothing Methods – Moving averages – Exponential Smoothing 5. ARIMA Method – Auto Regressive Integrated Moving Averages
  • 12.
    Barometric or LeadingIndicator Technique • Relationship can exist among various economic time series • Lagging series – Lagging Indicators – Data moves up and down behind the series being compared • Coincident series – Coincident Indicators – Data moves up and down with some other series
  • 13.
    Barometric or LeadingIndicator Technique • Leading series – Leading Indicators – Data moves ahead of the series being compared Application for housing loan Demand for construction material Birth rate of children Demand of seats in school
  • 14.
    Correlation and RegressionMethod • This method recognises the fact a number of factors like own advertising, competitors advertising, competitors price, weather condition etc may affect sales. • Variables influencing sales is identified through correlation • Regression equation specifies relationship between each variable and sales.
  • 15.
    Steps in DemandForecasting 1. Problem definition 2. Gathering information 3. Preliminary exploratory analysis 4. Choosing the fitting model 5. Using the evaluating model
  • 16.
    Econometric Models • EconometricModels tries to identify all those economic and demographic variables that influences the future value of the variable under forecasting and build up a cause-effect relationship.
  • 17.
    Simulations Equation method •Simulations Equation method involves simulations consideration of all the variables, as it is believed that every variable influences the other variable in an econometric decision environment.
  • 18.
    Range Long Medium Short Horizon 5 years 1 –2 years Up to 1 year Application Facility, capacity, product planning Staffing plan, Aggregate production Plan Purchasing , Detailed job scheduling Methods Economic, Demographic, Market information, Technology Time series, Regression Trend exploration , Graphical Method, Exponential smoothing
  • 19.
    Qualities of goodDemand Forecasting 1. Simple 2. Accurate 3. Easy Availability 4. Economy 5. Capacity to update forecasts
  • 20.
    Limitations • Change inFashion • Consumers Psychology • Lack of Past Data • Uneconomical • Lack of Experienced Experts