OM Week 4-1

Looking into the Future:
The Planning Horizon
*A business may have many diff planning horizons depending on the resource in question
-Inventory - Usually very short
-Employee- Generally pretty short
*temps, new hire, etc.
-Equipment- A little longer
*purchasing and installation lead times
-Facilities - Longest
*purchase property, building the building

Qualitative Forecasts
-Do not use past data. Usually used when such data is not available (such as planning for a new product)
*Causal Techniques
-Uses external data to predict
-Looking for the factor that "cause" demand
-Linear regression is often used
*Time Series Techniques
-Use past demand to predict future demand
**There are four potential components of a time series:
-Cycles
--A pattern that repeat over a long period of time (such as 20 yrs)
--Cycles are less important for demand forecasting, since we rarely have 20 yrs worth of data.
-Trend
-Seasonality
-Randomness

Quantitative Forecasts
- Divided into causal forecasting and time series forecasting techniques.

Exponential Smoothing
* A variant of moving average ( weighted average)
-Premise: more recent observations are better indicators of future demand than past observations.
-Use a smoothing costant, 'alpha' (a) to weight the pervious demand and establish the responsiveness of the forecast.
F_t+1 = alphaA_t+(1-alpha)F_t
F_t+1 = the forecast for the next time period
Alpha = A smoothing constant, between 0 and 1
A_t = The actual demand for the most recent period
F_t = The forecast for the most recent period

exam question = Increase alpha = more sensitive, decrease alpha = less sensitive