Production & Operations Management IME 316

Created by

@FreshLawrencium6836

Study Notes

Forecasting is the art and science of predicting future events.
It's the foundation for all business decisions related to production, inventory, personnel, and facilities.
Forecasting approaches can be qualitative or quantitative.

Short-range: Less than 3 months, used for purchasing, workforce levels, and production levels.
Medium-range: 3 months to 3 years, used for sales and production planning.
Long-range: 3+ years, used for new product planning and facility location.

Medium/long-range forecasts address broader issues impacting planning, products, plants, and processes.
Short-term forecasts use different methodologies than longer-term forecasts.
Short-term forecasts tend to be more accurate than longer-term ones.

Qualitative: Used when data is limited (new products, new technologies).
- Relies on intuition, experience, and expert opinion.
- Methods include jury of executives' opinion, sales force composite, Delphi method, and consumer market survey.
Quantitative: Used when historical data is available and the situation is stable (existing products).
- Relies on mathematical and statistical methods.
- Methods encompass time-series models (naive approach, moving averages, exponential smoothing, trend projections) and associative models (linear regression).

Time-Series Models: The future is a function of the past.
- Trend: A long-term upward or downward movement.
- Seasonal: Regular fluctuations over a specific period.
- Cyclical: Repeating up and down movements over longer periods.
- Random: Unexpected variations.
Associative Models: Variables or factors influencing the variable being forecast.
- Linear regression.

Disney uses multiple forecasting methods (daily, weekly, monthly, annual, and 5-year forecasts).
Forecasts support labor, maintenance, operations, finance, and park scheduling decisions.
Forecasts help adjust operating hours, rides, shows, staffing, and guest capacity.
Forecast data is gathered from surveys to 1 million guests, employees, and travel professionals each year.
Inputs include airline specials, Federal Reserve policies, market trends, and vacation/holiday schedules.
Average forecast error is 5% for 5-year forecasts, and 0-3% for annual forecasts.

Naive approach: Assumes next period's demand is the same as the most recent period's demand.
Moving averages: A series of arithmetic means, used for smoothing when no trend exists.
Exponential smoothing: A weighted moving average where weights decline exponentially.
- Requires a smoothing constant (α), typically between 0.05 and 0.5.

Trend projections: Fitting a trend line to historical data to project into the medium to long range. Using the least squares technique.
Least squares method: used to get the equation to the regression line.

Associative Forecasting: used when changes in one or more independent variables can be used to predict changes in the dependent variable.
Linear Regression commonly used technique.

Strength of linear relationship between variables.
Does not imply causation, only association.
Ranges from -1 to +1, 1 being a perfect positive correlation, -1 being perfect negative correlation.

Mean Absolute Deviation (MAD): Average absolute difference between forecast and actual values.
Mean Squared Error (MSE): Average of squared differences between forecast and actual values.
Mean Absolute Percent Error (MAPE): Average absolute percentage difference between forecast and actual values.