Term
|
Definition
| Process of predicting future events |
|
|
Term
| Types of Forecasts by Time Horizon |
|
Definition
Short Range- Usually less than one year. used to schedule jobs and worker assignments
Medium Range_ between 1-3 years. involves the planning or sales and productions as well as budgeting
Long Term forecast- Over 3 years. usually used to plan new products or expansion of large assets such as facility location or opening a new branch |
|
|
Term
| Principles of Forecasting |
|
Definition
Forecasts are rarely perfect
The shorter the period the more accurate the forecast will be
forecasts for individual items are less accurate than forecasts for grouped data |
|
|
Term
| Importance of Forecasting |
|
Definition
It is the only estimate of demand before the actual demand is derived.
It impacts human resources, capacity and supply chain management |
|
|
Term
|
Definition
Qualitative Methods: Educated guesses or generated subjectively by the forecaster
Quantitative methods: Generated through mathematical means |
|
|
Term
|
Definition
- Executive Opinion: Managers meet and create a forescast. Usually good for strategic or forecasting a new product. one persons opinion can dominate it, causing a weakness in the forecast
- Market Research: utilises surverys, interviews and questionnaires to gather cprefrences of consumers. but developing these questionnaires can be difficult.
- Delphi Method: Develops concensus among experts. great for forecasting long term product demand but it can be time consuming
|
|
|
Term
|
Definition
- Time Series Model: Assumes the future will follow the same pattern as the past. Information needed is contained in a time series of data
- Casual/ Associative Model: Explores cause and effect relationships. Also uses leading indicators to predict
|
|
|
Term
|
Definition
- Level or Horizontal Pattern: data follows a horizontal pattern around the mean
- Trend Pattern: Data is progressively increasing or decreasing
- Seasonal Pattern: Data displays a regular repeating pattern
- cycle- data increases then decreases over time
|
|
|
Term
| Selecting the right forecast model |
|
Definition
1.The amount & type of available data: Some methods require more data than others
2.Degree of accuracy required:Increasing accuracy means more data
3.Length of forecast horizon: Different models for 3 month vs. 10 years
4.Presence of data patterns: Match the forecasting model to the time series pattern. (trend, seasonal, cyclical, random) |
|
|
Term
| Forecasting Across the Organization |
|
Definition
Forecasting is critical to all functions of a business :
- Marketing: forecasting allows them to predict deman as well as future sales
- Finance: forecasts stock prices, potential financial performance and capital investment needs
- Human Resources: forecasts future hiring requirments
|
|
|
Term
| Forecasting Calculations using Exponential Smoothing |
|
Definition
Next Period Forecast= (Smoothing Constant* Actual)+( Remaining Smoothing Constant* Previous Forecasted amount)
Smoothing Constant+ Remaining Smoothing Constant= 1
Absolute Error= difference between the forecasted amount and the actual amount
|
Time
|
Forecast
|
Actual
|
Absolute
Error
|
Calculations for the next periods Forecast
|
|
Jan
|
250
|
261
|
|
Smoothing Constant has to equal one ie .7 remaining .3
|
|
Feb
|
255
|
253
|
2
|
|
|
Mar
|
248
|
257
|
9
|
Next Fore = (SC*Actual) + (RSC * Previous Forecast)
|
|
Apr
|
255.2
|
|
|
255.2 = .8 * 257 + .2 * 248
|
|
May
|
|
|
|
|
|
June
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Term
| Forecasting Calculations using Moving Average |
|
Definition
(For example a 2 month moving average was used, you can use any amount of months for moving average)
Calculations start in march because you have 2 months worth of information.
To calculate, find the average or the actual of the two months previous of the month that is being forecasted.
April Forecast= March Actual + February Actual / 2
|
Time
|
Forecast
|
Actual
|
Absolute Error
|
Calculations for the next periods Forecast
|
|
Jan
|
250
|
261
|
11
|
|
|
Feb
|
255
|
253
|
2
|
|
|
Mar
|
248
|
257
|
9
|
Next Forecast March Actual + February Actual / 2
|
|
Apr
|
255
|
|
|
257 + 253 / 2 = 255
|
|
May
|
|
|
|
|
|
June
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Term
| Forecasting Calculations using Weighted Moving Average |
|
Definition
Method is similar to moving average but a weight is put on each time period (month)
Period 1 = . 5
Period 2 = . 3
Period 3= . 2
(NB: Period 1 would be the month closest to the one being forecasted and period 3 being the furthest away: uning a 3 month weighted average)
|
Time
|
Forecast
|
Actual
|
Absolute Error
|
Calculations for the next periods Forecast
|
|
Jan
|
250
|
261
|
11
|
|
|
Feb
|
255
|
253
|
2
|
|
|
Mar
|
248
|
257
|
9
|
Next Fore (March * .5) + (February *.3) + ( January *.2)
|
|
Apr
|
256.6
|
|
|
257 *. 5 + 253 * .3 + 261 * .2 = 256.6
|
|
May
|
|
|
|
|
|
June
|
|
|
|
|
|
|
|
|
|
|
|
|
|
