Ch. 12 Demand Planning: Forecasting and Demand Management

demand planning
the combined process of forecasting and managing customers demands to create a planned pattern of demand that meets the firms operational and financial goals.
– helps operations managers know what customers they should serve and at what levels of service
demand forecasting
decision process in which managers predict demand patterns
-integrate info from market, internal op, and larger business environment
demand management
proactive approach in which managers attempt to influence patterns of demand
-product pricing
-promotional activities
-incentives(sales comissions)
-coupons
Costs of forecasting should save…
money lost in holding inventory that is never sold
lost capacity making products no one wants to buy
lost wages paying workers who are not needed
forecast demand too low=
lost sales
lowered productivity available for customers
stable pattern
consistent horizontal streams of demands.
seasonality and cycles
regular patterns of repeating highs and lows
trend
identifies general sloping tendency of demand, either upward or downward, in a linear or non linear fashion.
shift of step change
one time change in demand usually due to some external influence on demand such as a major product promotional campaign.
this makes previous data no longer reliable so more judgement based forecasting has to be used.
Autocorrelation
describes the relationship of current demand with past demand
forecast errors
the “un-explained” component of demand that seems to be random in nature
Forecasts have to be :
-usable
-timely
-accurate
judgement based forecasting
built upon estimates and opinions of people (experts).
It incorporates factors of demand difficult to capture in statistical models.
Used when there is a lack of quantitative and historical info (ex new product) or a step change

vs.

Statistical based forecasting
stable product= the more you relay on statistics

Grassroots forecasting
technique that seeks inputs from people who are close in contacts with customers and products.
limitation: “experts” may unconsciously base their forecasts on recent experiences, rather than entire set of experiences
better for developing short term forecasts for individual products
Executive Judgment
Executives better equipped to make judgments regarding long term sales of business patterns.
They have experience and access to sources of information
Historical Analogy
this approach uses data and experience from similar products to forecast demand for a new products
Marketing Research
bases forecasting on the purchasing patterns and attitudes of current potential customers
-consumer surveys
– interviews
– focus groups
Delphi method
develops forecasting by asking a panel of experts to individually respond to a series of questions and then the forecaster shares all answers with the group. After, they are given a chance to revise their responses or ask new questions. THis is repeated until a consensus is achieved .
-prevents any individual to have compete control of process
-reflects input of all
Naive model
assumes that tomorrows demand will be the same as todays
moving average
forecast as the average of demands over a number of immediate past periods.
increasing # of periods reduces impact of atypical demands in isolated periods
but also reduces sensitivity of moving average to actual shifts in demand
weighted moving average
this models assigns a different weight to each periods demand according to its importance. (ex. more importance to recent periods)
Exponential Smoothing
In exponential smoothing an exponentially smaller weight is applied to each demand that occurred further back in time.
-continually changes
-includes all data points
-more weight on recent data
regression analysis
regression analysis
Estimates relationships between leading indicators and demand.
Correlation bet. two data sets
R(squared)=
close to 1=tight correlation
close to 0= no correlation
seasonal index
adjust forecast values for each seasonal time period.
Divides each periods actual demand by an estimate of the average(or base) demand across all periods in a complete seasonal cycle.
causal modeling
concentrate on external factors that are thought to cause demand.
forecast error
=actual demand- forecasted demand (for a given time period)
if positive: overly pessimistic forecast
if negative: overly optimistic forecast
forecast accuracy
measures how closely the forecast aligns with the observation over time.
-every error(too high or too low) reduces accuracy
forecast bias
the average error.
Indicates the tendency of a forecasting technique to continually over-predict or under-predict demand.
postponable product
one that can be configured to its final form quickly and inexpensively once actual customer demand is known.
-largely eliminates need for large and complex forecasting
collaborative planning, forecasting and replenishment
improves collaboration and information sharing
requires buyers and sellers to collaboratively develop their demand plans and collaboratively adjust and execute them with the goal of meeting customer demand with minimal inventories, lead times and transaction costs.
Demand forecasting rules
– short term more accurate
– aggregated demand easy to forecast/less variable
– forecasts developed using multiple info sources are more accurate