Demand Forecasting

55,512
-1

Published on

Published in: Economy & Finance, Business
12 Comments
25 Likes
Statistics
Notes
No Downloads
Views
Total Views
55,512
On Slideshare
0
From Embeds
0
Number of Embeds
3
Actions
Shares
0
Downloads
2,594
Comments
12
Likes
25
Embeds 0
No embeds

No notes for slide

Demand Forecasting

  1. 1. Demand Forecasting
  2. 2. SO WHAT “ IS ” DEMAND F ORECASTING?
  3. 3. <ul><li>Forecasting customer demand for products and services is a proactive process of determining what products are needed where , when , and in what quantities . Consequently, demand forecasting is a customer–focused activity. </li></ul><ul><li>Demand forecasting is also the foundation of a company’s entire logistics process. It supports other planning activities such as capacity planning, inventory planning, and even overall business planning. </li></ul>
  4. 4. Characteristics of demand <ul><li>5 main characters of demand are- </li></ul><ul><li>Average </li></ul><ul><li>Demand tends to cluster around a specific level. </li></ul><ul><li>Trend </li></ul><ul><li>Demand consistently increases or decreases over time. </li></ul><ul><li>Seasonality </li></ul><ul><li>Demand shows peaks and valleys at consistent intervals. These </li></ul><ul><li>intervals can be hours, days, weeks, months, years, or seasons. </li></ul><ul><li>Cyclicity </li></ul><ul><li>Demand gradually increases and decreases over an extended period </li></ul><ul><li>of time, such as years. Business cycles (recession/expansion) </li></ul><ul><li>product life cycles influence this component of demand. </li></ul><ul><li>Elasticity </li></ul><ul><li>Degree of responsiveness of demand to a corresponding </li></ul><ul><li>proportionate change in factors effecting it. </li></ul>
  5. 5. TYPES OF FORECASTS <ul><li>PASSIVE FORECASTS </li></ul><ul><li>Where the factors being forecasted </li></ul><ul><li>are assumed to be constant over a </li></ul><ul><li>period of time and changes are </li></ul><ul><li>ignored. </li></ul><ul><li>ACTIVE FORECASTS </li></ul><ul><li>Where factors being forecasted are </li></ul><ul><li>taken as flexible and are subject </li></ul><ul><li>to changes. </li></ul>
  6. 6. Why Study forecasting? <ul><li>Reduces future uncertainties, helps study markets </li></ul><ul><li>that are dynamic, volatile and competitive </li></ul><ul><li>Allows operating levels to be set to respond to </li></ul><ul><li>demand variations </li></ul><ul><li>Allows managers to plan personnel, operations of </li></ul><ul><li>purchasing & finance for better control over wastes </li></ul><ul><li>inefficiency and conflicts. </li></ul><ul><li>Inventory Control-reduces reserves of slack resources </li></ul><ul><li>to meet uncertain demand </li></ul><ul><li>Effective forecasting builds stability in operations. </li></ul><ul><li>Setting Sales Targets, Pricing policies, establishing </li></ul><ul><li>controls and incentives </li></ul>
  7. 7. How? THE FORECAST Step 6 Monitor the forecast Step 5 Prepare the forecast Step 4 Gather and analyze data Step 3 Select a forecasting technique Step 2 Establish a time horizon Step 1 Determine purpose of forecast
  8. 8. LEVELS OF FORECASTING <ul><li>AT FIRMS LEVEL </li></ul><ul><li>AT INDUSTRY LEVEL </li></ul><ul><li>AT TOTAL MARKET LEVEL </li></ul>
  9. 9. KEY FACTORS FOR SELECTING A RIGHT METHOD <ul><li>TIME PERIOD </li></ul><ul><li>SHORT TERM </li></ul><ul><li>3-6 Months, Operating Decisions, </li></ul><ul><li>E.g- Production planning </li></ul><ul><li>MEDIUM TERM </li></ul><ul><li>6 months-2 years, Tactical Decision </li></ul><ul><li>E.g.- Employment changes </li></ul><ul><li>LONG TERM </li></ul><ul><li>Above 2 years, Strategic Decision </li></ul><ul><li>E.g.- Research and Development </li></ul>
  10. 10. DATA REQUIREMENTS <ul><li>Techniques differ by virtue of how </li></ul><ul><li>much data is required to successfully </li></ul><ul><li>employ the technique. </li></ul><ul><li>Judgmental techniques require little or </li></ul><ul><li>no data whereas methods such as </li></ul><ul><li>Time series analysis or Regression </li></ul><ul><li>models require a large amount of past </li></ul><ul><li>or historical data. </li></ul>
  11. 11. <ul><li>PURPOSE OF STUDY </li></ul>It means the extent of explanation required from the study. Some techniques are based purely on the pattern of past data and may do quite well at forecasting, whereas many a times these are not useful by themselves since it is d ifficult to explain the causal factors underlying the forecast.
  12. 12. PATTERN OF DATA STUDIED The pattern of historical data is an important factor to consider. Though most of the times, the major pattern is that of a trend, there are also cyclic and seasonal patterns in the data. Certain techniques are best suited for capturing the different patterns in the data. Regression methods incorporates all these variations in data whereas trend analysis methods study these factors individually.
  13. 13. SO ,WE KNOW WHAT IT’S ALL ABOUT!!! NOW LETS ANALYSE THE METHODS OF DEMAND FORECASTING.
  14. 14. 2 MAIN CATEGORIES <ul><li>MICROECONOMIC METHODS </li></ul><ul><li>( QUANTITATIVE) </li></ul><ul><li>- involves the prediction of activity of particular firms, branded products, commodities, markets, and industries. </li></ul><ul><li>- are much more reliable than macroeconomic methods because the dimensionality of factors is lower and often can easily be incorporated into a model. </li></ul><ul><li>MACROECONOMIC METHODS </li></ul><ul><li>(QUALITATIVE) </li></ul><ul><li>- involves the prediction of economic aggregates such as inflation, unemployment, GDP growth, short-term interest rates, and trade flows. </li></ul><ul><li>- is very difficult because of the complex interdependencies in the overall economic factors </li></ul>
  15. 15. <ul><li>QUALITATIVE METHODS </li></ul><ul><li>SURVEY OF BUYERS INTENSIONS </li></ul><ul><li>EXPERTS OPINION METHOD </li></ul><ul><li>DELPHI METHOD </li></ul><ul><li>MARKET EXPERIMENTATION METHOD </li></ul><ul><li>- COLLECTIVE OPINIONS METHOD </li></ul><ul><li>QUANTITATIVE METHODS </li></ul><ul><li>TIME SERIES MODELS </li></ul><ul><li>- TREND ANALYSIS </li></ul><ul><li>- MOVING AVERAGES METHODS </li></ul><ul><li>- EXPONENTIAL SMOOTHING </li></ul><ul><li>CAUSAL MODELS </li></ul><ul><li>- REGRESSION MODELS </li></ul>
  16. 16. BUYERS INTENSION SURVEY <ul><li>FEATURES </li></ul><ul><li>EMPLOYS SAMPLE SURVEY TECHNIQUES </li></ul><ul><li>FOR GATHERING DATA. </li></ul><ul><li>DATA IS COLLECTED FROM END USERS OF </li></ul><ul><li>GOODS - CONSUMER, PRODUCER,MIXED. </li></ul><ul><li>DATA PORTRAYS BIASES AND PREFERENCES OF CUSTOMERS. </li></ul><ul><li>IDEAL FOR SHORT AND MEDIUM TERM DEMAND FORECASTING, IS COST EFFECTIVE AND RELIABLE. </li></ul>
  17. 17. ADVANTAGES <ul><li>HELPS IN APPROXIMATING FUTURE </li></ul><ul><li>REQUIREMENTS EVEN WITHOUT PAST DATA. </li></ul><ul><li>ACCURATE METHOD AS BUYERS </li></ul><ul><li>NEEDS AND WANTS ARE CLEARLY </li></ul><ul><li>IDENTIFIED & CATERED TO. </li></ul><ul><li>MOST EFFECTIVE WAY OF </li></ul><ul><li>ASSESSING DEMAND FOR NEW FIRMS </li></ul>
  18. 18. LIMITATIONS <ul><ul><ul><li>People may not know what they are going to purchase </li></ul></ul></ul><ul><ul><ul><li>They may report what they want to buy, but not what they are capable of buying </li></ul></ul></ul><ul><ul><ul><li>Customers may not want to disclose real information </li></ul></ul></ul><ul><ul><ul><li>Effects of derived demand may make forecasting difficult </li></ul></ul></ul>
  19. 19. EXPERTS OPINION METHOD <ul><li>FEATURES </li></ul><ul><li>PANEL OF EXPERTS IN SAME FIELD WITH EXPERIENCE & WORKING KNOWLEDGE. </li></ul><ul><li>COMBINES INPUT FROM KEY INFORMATION SOURCES. </li></ul><ul><li>EXCHANGE OF IDEAS AND CLAIMS. </li></ul><ul><li>FINAL DECISION IS BASED ON MAJORITY OR CONSENSUS, REACHED FROM EXPERT’S FORECASTS </li></ul>
  20. 20. ADVANTAGES <ul><ul><ul><li>CAN BE UNDERTAKEN EASILY </li></ul></ul></ul><ul><ul><ul><li>WITHOUT THE USE OF ELABORATE </li></ul></ul></ul><ul><ul><ul><li>STATISTICAL TOOLS. </li></ul></ul></ul><ul><ul><ul><li>INCORPORATES A VARIETY OF </li></ul></ul></ul><ul><ul><ul><li>EXTENSIVE OPINIONS FROM EXPERT </li></ul></ul></ul><ul><ul><ul><li>IN THE FIELD. </li></ul></ul></ul>
  21. 21. LIMITATIONS <ul><li>JUDGEMENTAL BIASES </li></ul><ul><li>FOR EXAMPLE </li></ul><ul><ul><li>Availability heuristic </li></ul></ul><ul><ul><li>Involves using vivid or accessible events </li></ul></ul><ul><ul><li>as a basis for the judgment. </li></ul></ul><ul><ul><li>Law of small numbers </li></ul></ul><ul><ul><li>People expect information obtained </li></ul></ul><ul><ul><li>from a small sample to be typical of </li></ul></ul><ul><ul><li>the larger population </li></ul></ul>
  22. 22. <ul><li>COMPETATIVE BIASES </li></ul><ul><ul><li>Over reliance on personal opinions. </li></ul></ul><ul><ul><li>Possibility of undue influence in certain cases. </li></ul></ul><ul><li>STATISTICAL INADEQUACY </li></ul><ul><li>Lack of statistical and quantifiable </li></ul><ul><li>data or figures to substantiate the forecasts made. </li></ul>
  23. 23. DELPHI METHOD <ul><li>PANEL OF EXPERTS IS SELECTED. </li></ul><ul><li>ONE CO-ORDINATOR IS CHOSEN BY </li></ul><ul><li>MEMBERS OF THE JURY </li></ul><ul><li>ANONYMOUS FORECASTS ARE MADE BY EXPERTS BASED ON A COMMON QUESTIONNAIRE. </li></ul><ul><li>CO-ORDINATOR RENDERS AN AVERAGE OF ALL FORECASTS MADE TO EACH OF THE MEMBERS. </li></ul>
  24. 24. <ul><li>3 CONSEQUENCES- DIVERSION, CONSENSUS OR NO AGREEMENT. </li></ul><ul><li>2 TO 3 CYCLES ARE UNDERTAKEN. </li></ul><ul><li>CONVERGENCE AND DIVERSION IS ACCEPTABLE. </li></ul><ul><li>FORECASTS ARE REVISED UNTIL </li></ul><ul><li>A CONSENSUS IS REACHED BY ALL. </li></ul>
  25. 25. ADVANTAGES <ul><li>ELIMINATES NEED FOR GROUP </li></ul><ul><li>MEETINGS. </li></ul><ul><li>ELIMINATES BIASES IN GROUP </li></ul><ul><li>MEETINGS </li></ul><ul><li>PARTICIPANTS CAN CHANGE THEIR OPINIONS ANONYMOUSLY. </li></ul>
  26. 26. LIMITATIONS <ul><li>TIME CONSUMING -REACHING </li></ul><ul><li>A CONSENSUS TAKES A LOT OF TIME. </li></ul><ul><li>PARTICIPANTS MAY DROP OUT. </li></ul>
  27. 27. MARKET EXPERIMENTATION <ul><li>INVOLVES ACTUAL EXPERIMENTS & SIMULATIONS. </li></ul><ul><li>COUPONS ARE ISSUED TO FEW SELECT CUSTOMERS. </li></ul><ul><li>SELECTED CUSTOMERS PURCHASE THE PRODUCTS. </li></ul><ul><li>PROXIMITY WITH CONSUMERS MAKES </li></ul><ul><li>INFORMATION COLLECTED RELIABLE. </li></ul><ul><li>INFORMATION FROM INTERACTIONS BETWEEN SALES PERSONNEL & CUSTOMERS IS USED FOR FORECASTING. </li></ul><ul><li>BEST USED IF SALES PERSONNEL ARE HIGHLY </li></ul><ul><li>SPECIALISED AND WELL TRAINED. </li></ul>
  28. 28. ADVANTAGES <ul><li>USES KNOWLEDGE OF THOSE CLOSEST TO THE MARKET. </li></ul><ul><li>HELPS ESTIMATING ACTUAL POTENTIAL </li></ul><ul><li>FOR FUTURE SALES. </li></ul><ul><li>PROVIDES FEEDBACK FOR IMPROVING CUSTOMIZING & OFFERING MADE TO </li></ul><ul><li>CUSTOMERS. </li></ul>
  29. 29. COLLECTIVE OPINIONS METHOD <ul><li>OPINIONS FROM MARKETING & </li></ul><ul><li>SALES SPECIALISTS ARE COMPILED. </li></ul><ul><li>2 TYPES OF TARGETS ESTIMATED- </li></ul><ul><ul><li>AMBITIOUS TARGETS. </li></ul></ul><ul><ul><li>CONSERVATIVE TARGETS. </li></ul></ul><ul><li>COMBINES EXPERTISE OF HIGHER </li></ul><ul><ul><li>LEVEL MANAGEMENT & SALES </li></ul></ul><ul><ul><li>EXECUTIVES. </li></ul></ul>
  30. 30. LIMITATIONS <ul><li>POWER STRUGGLES MAY OCCUR </li></ul><ul><li>BETWEEN SPECIALISTS. </li></ul><ul><li>CONSENSUS MAY NOT BE REACHED IN GOOD TIME. </li></ul><ul><li>DIFFERENCES AND PREJUDICES IN OPINIONS MAY ALSO EXIST. </li></ul>
  31. 31. “ A RE YOU STILL THERE??”
  32. 32. <ul><li>THAT FINISHES THE QUALITATIVE METHODS. </li></ul><ul><li>NOW LETS LOOK AT THE </li></ul><ul><li>“ QUANTITATIVE </li></ul><ul><li>METHODS” </li></ul>
  33. 33. TIME SERIES MODELS <ul><li>Past data is used to make future </li></ul><ul><li>predictions . </li></ul><ul><li>Known or Independent variables are </li></ul><ul><li>used for predicting Unknown or </li></ul><ul><li>dependent variables, using the trend </li></ul><ul><li>equation- “ Predictive analysis” </li></ul><ul><li>Based on trend equation, we find </li></ul><ul><li>‘ Line of Best Fit’ and then it is </li></ul><ul><li>projected in a scatter diagram, </li></ul><ul><li>dividing points equally on both sides </li></ul>TREND ANALYSIS
  34. 34. <ul><li>TREND EQUATION </li></ul><ul><li>Y^ = a + bX + E </li></ul><ul><li>Y^ = Estimated value of Y </li></ul><ul><li>a = Constant or Intercept </li></ul><ul><li>b = slope of trend line </li></ul><ul><li>X = independent variable </li></ul><ul><li>E = Error term </li></ul>
  35. 35. = EXPLAINED VARIATION 1 - = UNEXPLAINED VARIATION Explained variation - means the extent to which the independent variable explains the relative change in the dependent variable. Higher the explained variation, lower the error value leading to accurate forecast
  36. 37. MOVING AVERAGE METHOD <ul><li>Data from a number of consecutive </li></ul><ul><li>past periods is combined to provide </li></ul><ul><li>forecast for coming periods.Higher </li></ul><ul><li>the amount of previous data, better </li></ul><ul><li>is the forecast. </li></ul><ul><li>Since the averages are calculated </li></ul><ul><li>on a moving basis, the seasonal and </li></ul><ul><li>cyclical variations are smoothened </li></ul><ul><li>out. </li></ul>
  37. 38. EXPONENTIAL SMOOTHING <ul><li>Used in cases where the variable </li></ul><ul><li>under forecast doesn’t follow a </li></ul><ul><li>trend. </li></ul><ul><li>2 Types- Simple and Weighted </li></ul><ul><ul><li>Simple smoothing- simple average of specific observation called order. </li></ul></ul><ul><ul><li>Weighted smoothing- weights assigned in decreasing order as one moves from current period observations to previous observations. </li></ul></ul>
  38. 39. <ul><li>The equation for exponential </li></ul><ul><li>smoothing follows a Geometric </li></ul><ul><li>Progression.Values may be written as- </li></ul><ul><li>a, a (1-a), a(1-a)^2….. a(1-n) where, </li></ul><ul><li>a = value of weight assigned </li></ul><ul><li>to the observation </li></ul><ul><li>a(1-a) = weight assigned to 1 period </li></ul><ul><li>previous observation </li></ul><ul><li>a(1-a)^2 = weight assigned to 2 </li></ul><ul><li>periods previous observation </li></ul><ul><li>Sum of all weights always equals Unity. </li></ul>
  39. 40. CASUAL MODELS <ul><li>It is a statistical technique for </li></ul><ul><li>quantifying the relationship between </li></ul><ul><li>variables. In simple regression analysis, </li></ul><ul><li>there is one dependent variable (e.g. </li></ul><ul><li>sales) to be forecast and one independent </li></ul><ul><li>variable. The values of the independent </li></ul><ul><li>variable are typically those assumed to </li></ul><ul><li>&quot;cause&quot; or determine the values of the </li></ul><ul><li>dependent variable. </li></ul>REGRESSION MODEL
  40. 41. <ul><li>For example </li></ul><ul><li>Assuming that the amount of advertising dollars spent on a product determines the amount of its sales, we could use regression analysis to quantify the precise nature of the relationship between advertising and sales. For forecasting purposes, knowing the quantified relationship between the variables allows us to provide forecasting estimates </li></ul>
  41. 42. STEPS IN REGRESSION ANALYSIS <ul><li>1.Identification of variables </li></ul><ul><li>influencing demand for product </li></ul><ul><li>under estimation. </li></ul><ul><li>2.Collection of historical data on </li></ul><ul><li>variables. </li></ul><ul><li>3.Choosing an appropriate form of </li></ul><ul><li>function </li></ul><ul><li>4.Estimation of the function. </li></ul>
  42. 43. REGRESSION EQUATION <ul><li>Y = </li></ul>Where Y= value being forecasted = constant value = coefficients of regression = independent variable
  43. 44. <ul><li>HIGHER REVENUES </li></ul><ul><li>SALES MAXIMIZATION </li></ul><ul><li>REDUCED INVESTMENTS FOR SAFETY </li></ul><ul><li>STOCKS </li></ul><ul><li>IMPROVED PRODUCTION PLANNING </li></ul><ul><li>EARLY RECOGNITION OF MARKET TRENDS </li></ul><ul><li>BETTER MARKET POSITIONING </li></ul><ul><li>IMPROVED CUSTOMER SERVICE LEVELS </li></ul>BENEFITS OF EFFECTIVE DEMAND FORECASTING

×