Forecasting

Forecasting

1. What is?

   • Process of predicting the future
     • Long term
       • + 3 years
       • Planning new product
       • I+D
     • Mid- term
       • 3 month - 3 years
       • Budget
       • Sales & production planning
     • Short term
       • Less than 1 year, usually 3 months
       • Jobs scheduling
       • Workforce level
       • Production level
   • And is the base of all business decisions
     • Influence of prodcut life cycle
       • Intro
         • I+D critical
       • Growth
         • Strengthen niche
       • Maturity
         • Defend market position
       • Decline
         • Cost control critical

2. Types of forecasts

   • Economic forecast
     • Inflation rates, money supply & so on
   • Tech forecast
     • Impacts development of new product
   • Demand forecast
     • Predict sales

3. Strategic importance of forecasting

   • H.R.
     • Hiring
     • Training
     • Firing
   • Capacity
     • Shortages
       • Loss clients
       • Loss market share
   • Supply chain management
     • Good supplier relationships

4. 7 steps in the forecasting system

   • 1. Determinate the use
   • 2. Select the intems
   • 3. Determine the time horizon
   • 4. Select the forecasting model
   • 5. Gather the data
   • 6. Make the forecast
   • 7. Validate the results

5. Forecasting approaches

   • Qualitative methods
     • Little data, inaccurate
     • Intuition & EXP
     • Executive opinions - Delphi mothod - Sale force - Survey
   • Quantitative methods
     • Historical data exist & stable situation
       • Math techniques
     • 1. Naive approach - 2. Moving averages - 3. Expo smoothing - 4. Trend projection
     • 5. Linear regression

6. Time-series forecasting

   • Trend
     • Upward or Downward
     • Change in population, tech, culture & so on
   • Cyclical
     • Changes in political, economic factors, business cycle
     • Associative w/ relationships
   • Seasonal
     • Occurs once a year
     • Due weather, customs
   • Random
     • Residual fluctuations
     • Due for unforeseen events

7. Formulas

   • Moving average method
   • Weighted moving average
   • Expo. smoothing
   • Forecast Error = Act. D -- Forecast value
   • MAD = Σ(Act - Forecast) / n
   • MSE = [Σ(Forecast Errors)^2] / n
   • MAPE = [Σ(Act - Forecast) / Act] / n
   • Tracking signal = (RSFE / MAD)
     • How well the forecast predict real values
   • RSFE + MAD = Low values = good tracking signal
   • RSFE =Σ Error

8. Controlling forecasts

9. In the service sector

   • Unusual challenges
     • Needs changes around industry and product
     • Holidays & calendar events affect

Imagens e arquivos do mapa