SSCM CP 5

• Demand planning


  The practice of creating a forecast that predicts the future need for a particular good or service
• If demand planning is done correctly organisations can deliver excellent consumer services while meeting financial goals
• Demand planning is considered to be a part of the supply chain management process which contributes towards the delivery of reliable goods and services that satisfy the needs of consumers
• Forecasting


  A prediction of demand that is based on statistical and mathematical analysis of past demand
• Forecasting starts with predetermined assumptions based on experience, knowledge and judgement
• Organisations require


  • SHORT TERM FORECAST: Are required for the scheduling of personnel, production and transportation
  • MEDIUM TERM FORECAST: Are required for the determination & scheduling of the future resource requirements in order to acquire raw materials, employ personnel or purchase equipment and machinery
  • LONG TERM FORECASTS: Are required for strategic planning. Decisions must take into account market opportunities, environment factors as well as internal resources
• What to forecast


  • Is it for every product line, or for groups of products
  • Is it for every store or for stores grouped by province or only for total sales
  • Is it weekly data, monthly data or annual data
• Organisations should consider


  • Time horizon
  • Frequency forecast
  • Efficiency and productivity of personnel
• Forecasting techniques


  • Qualitative
  • Quantitative
• Quantitative forecasting


  Uses mathematical and statistical methods during forecasting process. The input for the forecast is historical data which can be utilised to predict for the future
• Time series


  Quantitative forecasting techniques that use historical data to predict future values
• Moving average


  A forecasting method that assumes the demand for the good or service will remain fairly constant over a period of time. It uses the most recent set of data and measures the average information within a specific time frame
• Weighted moving average


  A forecasting method that gives more weight to the most recent data, reducing the weight of data that is less important
• Exponential smoothing


  A forecasting method that requires the most recent forecasts, actual demand and the smoothing constant alpha. The new forecast will be based on the previously utilised forecast and actual demand
• Exponential smoothing is a broadly accurate forecasting method for short-term forecasts but may produce slightly unreliable long-term forecasts
• It is important that there is no trend in the data before the exponential smoothing forecasts, as the forecasts may not be accurate if there is a trend
• The most recent forecast has the greatest weight in exponential smoothing and therefore should be the most accurate in predicting demand, as opposed to the moving averages method where the weight for each period is fixed
• Causal forecasting methods


  Quantitative forecasting techniques that use independent variables to predict the dependent variable
• Linear regression


  A causal forecasting method that expresses the forecasted variable as a linear function of an independent variable
• Multiple regression


  A causal forecasting method that allows for more than one independent variable to predict the dependent variable
• The development of software packages assists organisations when computing regression and multiple regression
• Qualitative forecasting


  Forecasting based on the judgement of those involved, using soft data and human factors rather than mathematical techniques
• Qualitative forecasting methods


  • Customer surveys
  • Jury of executive opinion
  • Sales force opinion
  • Delphi method
• Forecasting process


  1. Determine the purpose of the forecast

  2. Identify the correct time horizon

  3. Choose correct technique

  4. Collect and analyse data

  5. Make the forecast

  6. Continuously monitor the forecast
• Forecasting accuracy


  Determined by evaluating the forecast against actual data, with a smaller variance indicating higher accuracy
• Forecasting errors


  • Forecasting error
  • Mean absolute deviation
  • Mean square error
  • Mean absolute percentage error
• Cloud-based forecasting


  The use of cloud-based systems for forecasting
• Forecasting accuracy


  The accuracy of a forecast is determined when it is evaluated against the actual data, which are only derived after the forecast has been established
• An organisation makes plans based on its forecast, and therefore accuracy can be only be determined after a period of time
• If the actual data are close to the forecasted data

  Accuracy is high
• If there is a significant variance between the actual data and the forecasted data


  There is a low level of accuracy
• Variance between the actual data and the forecasted data


  Reveals the relevance of the forecast
• When there is a smaller variance


  The forecast is deemed to be relevant
• When there is significant variance


  The forecast is considered not to be as relevant and is revised accordingly
• Forecasting error


  The difference between the actual quantity and the forecasts
• Mean absolute deviation (MAD)


  The sum of the absolute values of the individual forecasting errors divided by the number of periods
• Mean squared error (MSE)


  The average of the squared differences between the forecasted and observed values
• Mean absolute percentage error (MAPE)


  The absolute difference between the forecast and actual values, expressed as a percentage of actual value
• Forecasting errors include mean absolute deviation, mean square error, and mean absolute percentage error
• Cloud-based forecasting


  The use of cloud computing technology to predict future events or trends over the internet