Marketing Management 16 Global Edition by Philip Kotler test bank.docx
Demand Forecasting
1. University of Moratuwa – MBA in MOT – Session 3
MN 5215 - Supply Chain Management, February 8, 2018
Demand Forecasting
Shanta R Yapa
LinkedIn/Fb: Shanta Rajapaksha Yapa
3. Think of the following..
• Banks forecast demand for cash.
• A restaurant prepares dinner based on
demand forecast.
• ATM is loaded with cash based on a forecast.
• Hotels accept reservations based on forecasts.
• Airlines sell more tickets than available seats.
6. The need to hold stocks
• To keep down production costs
• To accommodate variations in demand
• Variable supply lead times
• Buying costs
• Quantity discounts
• Price fluctuations
• Seasonality
• Smooth operations….
7. Elements of stock holding costs
• Capital cost
• Service cost – insurance, stock management
• Storage – warehousing, handling
• Risk – pilferage, obsolescence,
12. Example
• 3,200 units of double row ball bearings of type
5205 are required for maintenance activities
of the factory. Order placement cost is Rs
10,000-. Each bearing will cost Rs 2,000-.
Annual stock holding cost as fraction of unit
cost is 25%. Compute the EOQ.
13. EOQ constraints
• EOQ assumes a constant demand
• Assume that the ordering cost and holding
cost will remain unchanged.
14. • “All mistakes in forecasting end up as an
inventory problem”. Comment.
17. Criteria for a good forecasting system
• Accuracy
• Plausibility
• Durability
• Flexibility
• Availability
• Economy
• Simplicity
• Consistency
18. Forecasting methods
• Judgmental methods – subjective assessments based
on expert opinion
• Experimental methods – when there is no info. Ex: a
new project
• Causal methods – used when demand is dependent on
many factors (internal and external)
• Projective methods or time series models – use
of historical demand data and trends to project future
20. Common projective methods
• Moving average
– Past observations are weighted equally
• Exponential smoothing
– Used to produce a smoothed time series
– This method assigns exponentially decreasing
weightages as observations get older
21. Exponential smoothing
• New forecast = Last period’s forecast +
Smoothing constant* ( Last period’s actual
demand – last period’s forecast)