Chapter 3 - Forecasting Flashcards
**
Occurs when inventory is positioned in the supply chain processeses or entities to operate independently.
Decouping points
The demand for a product or serviced caused by the demand for other products or services
Raw materials, component parts, sub-assemblies
Dependent Demand
Demand that **cannot be delivered directly from that of other products **
Independent demand
What are the 2 types of forecasting?
Qualitative and quantitative
It is based on the idea that data relating to past demand can be used to predict the future
Time series analysis
What are the 5 components of demand?
Trend, Seasonality, Cyclical variations, autocorrelation, random variation (TSCAR)
Long term movement in data (growth in a business)
Trend
Short-term regular and repetitive variations in data
Seasonality
Long term, occasionally caused by unusual circumstances (war, economic downturn)
Cyclical valriation
Denotes persistence of occurence (momentum driven)
Autocorrelation
caused by a chance
Random variation
Used mainly for tactical decision such as replenishing inventory - usually less than 3 months
Short-term forecasting
Used to develop a strategy which will be implemented over the next 6 to 8 months such as meeting demand.
Medium term forecasting
Useful for detecting general trandes and identifying major turning points - greater than 2 years
Long-term forecasting
It is a forecasting method based on average demand over the most recent periods - useful when demand is not growing or declining rapidly, and no seasonality is present.
Simple moving average
It is a a method where it allows unequal weighting of prior time periods and the sum of weights must be equal to 1
Weighted Moving Average
This method is based on the importance of data that diminishes as the past becomes more distant and is the most logical and easiest method to use
Exponential Smoothing
Functional relationship between 2 or more correlated values, usually from observed data.
Regression
Predicted for a given values of the other variable (the independent variable||)
One variable
Special case which assumes the relationship between the variables can be explained with a straight line
Linear Regression
Past data and future projectsions are assumed to fall around straight line.
Linear Regression Forecasting
Determines the parameters a and b such that errors is minimized the least squares
Least Squares Method
Process of identifying and separating time series data into fundamental components: trend and seasonality.
Decomposition
Difference between the forecast value and what actually occured
Forecast error
When consistent mistake is made
Bias
Errors that are not explained by the model being used
Random
Scales the forecast error to the magnitude of demand
Mean Absolute Percentage Error (MAPE)
Indicates whether forecast errors are accumulating over time (either psotive or negative errors)
Tracking Signal (TS)
A forecasting uses indepedent variables other than the time to predict the demand.
Causal relationship Forecasting
Generally used to take advantage of expert knowledge. For example, market research, panel consensus, dephi method and historical analogy.
Qualitative Forecasting
