Chapter 7 - Just in Time and Lean Systems

Forecasting

The process of predicting future events that will impact operations.

Qualitative forecasting

Forecasting methods based on subjective opinions and intuition.

Quantitative forecasting

Forecasting methods based on mathematical models and historical data.

Short-range forecast

Forecast covering a period from a day to six months.

Medium-range forecast

Forecast covering 6 months to 2 years.

Long-range forecast

Forecast covering more than 2 years.

Delphi method

A qualitative forecasting method using a panel of experts to reach consensus through rounds of questioning.

Consumer surveys

Forecasting method based on asking customers about their future purchasing plans.

Historical analogy

Forecasting based on the historical pattern of a similar product or situation.

Time series forecasting

A quantitative method that uses historical data to predict future values.

Trend

A long-term upward or downward movement in data.

Seasonality

Regular pattern of up and down fluctuations related to calendar time or holidays.

Cycle

A long-term, wave-like pattern of data that may span years.

Random variation

Unpredictable, short-term fluctuations in data.

Naive forecast

A simple forecasting method where the forecast equals the last actual value.

Moving average

A method that averages a fixed number of past periods to make a forecast.

Weighted moving average

A moving average where more recent data is given more weight.

Exponential smoothing

A forecasting technique where the new forecast is based on the previous forecast plus a fraction of the error.

Smoothing constant (alpha)

A value between 0 and 1 used in exponential smoothing to weight recent observations.

Mean absolute deviation (MAD)

The average of absolute forecast errors.

Mean squared error (MSE)

The average of squared forecast errors.

Mean absolute percent error (MAPE)

The average of absolute percentage errors between forecast and actual values.

Causal forecasting

A method that uses independent variables to predict future demand.

Linear regression

A statistical method that models the relationship between a dependent and one or more independent variables.

Correlation coefficient (r)

A measure of the strength and direction of a linear relationship between two variables.

Coefficient of determination (r²)

A measure of how much variation in the dependent variable is explained by the regression model.

Tracking signal

A measure used to determine if forecast errors are consistently positive or negative over time.