Pronosticos

ININ 6019 ADVANCED PRODUCTION PLANNING AND CONTROL

Viviana I. Cesaní, PhD., PE Associate Professor Industrial Engineering University of Puerto Rico Mayagüez Campus
1

Forecasting Demand
• • •• Predicting characteristics about the future demand for an item Forecasts are always wrong Necessary to predict future demand for an item so as to make replenishment decisions Qualitativeforecasting
– Using human judgment based on past, present, and anticipated future circumstances

•

Quantitative (statistical) forecasting
– Identifying, qualifying, and exploiting patterns in the pastdemand for an item

• • •

Need- Uncertainty and leadtimes Use-production planning and inventory control Impact- inventories, customer service
2

Need for forecasts
• • • • • To setupperformance standards for customer service To plan allocation of total inventory investment To place replenishment orders To identify needs for additional capacity To choose between alternative operatingstrategies

3

Forecasting Steps
• Select an appropriate statistical model
– Chart data and identify patterns

• Estimate model parameters • Use the model
– Generate forecast (s) – Observe actualdemand – Update parameters

• Monitor forecast error
– Alter model as needed – Adjust for unmodeled factors

4

Factors affecting the selection of appropriate forecasting methods
• • • • • •• • Form of forecast required Forecast horizon, period, and interval Data availability Accuracy required Behavior of process being forecast (demand pattern) Cost of development, installation, andoperation Ease of operation Management comprehension and cooperation
5

Quantitative Methods
• Causal models
– Cause-effect relationship (regression) – Dependent variable-sales – Independentvariable- advertising expenditures

• Time series models
– Historical observation of demand
• Sequence of observations of a random process ordered by their time • Assumes they are affected by the...