Вы находитесь на странице: 1из 5

Forecasting

1) First step is to prepare AOP (annual operating plan), which is the


annual commitment from market to the organization about sales
no.s. This will be done in the month of Oct-Nov of current year for
the AOP of next year. Inputs are mostly from the regional sales
heads, and senior management who has revenue targets and
clarity about growth strategy.
2) Next critical step is to have Rolling Forecast (RoFo), which is done
quarterly. RoFo in the month of March will give clarity of Q2, June
will give clarity of Q3. The RoFo is primarily used for material
planning resulting in optimum inventory. Probability of meeting
RoFo no.s are much higher than AOP. As in AOP there are some
assumptions which gets validated during the course of year as
time progresses
3) After RoFo, to gain month on month clarity, monthly S&OP (Sales
and operation planning) meeting happens. This gives clarity of
immediate next month. So every third week of the month we will
discuss about next month no.s and revenue. S&OP no.s are almost
confirmed from sales team where the confidence is highest.

Forecasting Techniques
There are a wide variety of forecasting techniques that can broadly be
classified in three main approach
a) Judgmental Forecasts: Useful when forecasts must by done in a
short period of time, when data is out dated, unavailable, or
there's limited time to collect it.
b) Time Series Forecasts: Most Common, are used to identify specific
patterns in data and use them to project future forecasts
c) Associative models: identify related variables in order to predict
necessary forecasts.
Expert Selection- Expert Selection takes the guesswork out of
forecasting. The built-in expert system analyzes your data, selects the
appropriate forecasting technique, builds the model and calculates
the forecastsit even explains its reasoning in ordinary English!
Exponential Smoothing- Twelve different Holt-Winters exponential
smoothing models are provided to accommodate a wide range of data
characteristics. The robustness of exponential smoothing makes it
ideal when there are no leading indicators, and when the data are too
short or volatile for BoxJenkins. You can select the model and set the
parameters yourself or let Forecast Pro do it automatically.

Forecasting Techniques
Box-Jenkins- For stable data sets, Forecast Pro supports a
multiplicative seasonal Box-Jenkins model. The model can be built
completely automatically or interactively using a full range of screenoriented diagnostics.
Dynamic regression- If there are important leading indicators, use
Forecast Pro's dynamic regression. You can include independent
variables, lagged or transformed variables and build generalized
CochraneOrcutt models. Using Forecast Pro's selfinterpreting
diagnostics, you can build and compare alternative models with a few
clicks of the mouse.
Event models- Event models extend exponential smoothing by
providing adjustments for special events such as promotions, strikes
or other irregular occurrences. You can adjust for events of several
different types such as promotions of varying sizes or types, or
movable holidays like Easter and Rosh Hashanah. Creating, adjusting
and maintaining events is easy using Forecast Pro's intuitive Event
Manager.
Multiple-level models- Multiple-level models allow you to
aggregate data into groups that can be reconciled using a topdown
or bottom-up approach to produce consistent forecasts at all levels of
aggregation. Seasonal and event indexes can be extracted from the
higherlevel aggregates and applied to lower-level data.

Forecasting Techniques
New Product Forecasting- Forecast Pro includes several methods
for forecasting new products, including forecasting by analogy, item
supersession and the Bass diffusion model.
Seasonal Simplification- This is a useful technique if you are
forecasting data with more than 12 observations per year. Seasonal
Simplification reduces the number of seasonal indexes used to model
the data and often substantially improves forecast accuracy.
Low Volume Models- Croston's intermittent demand model and
discrete data models are provided to accommodate low volume and
"sparse" data (i.e., data where the demand is often zero).
Curve Fitting- Curve fitting provides a quick and easy way to
identify the general form of the curve which your data are following.
Forecast Pro supports four types of curvesstraight line, quadratic,
exponential and growth (Scurve).
Simple Methods- This set of "simple" models can be extremely
useful. Moving average, "same as last year," percentage growth and
fixed forecast value models are included.

Application:
1. It is important to know how to calculate a forecast error:Error =
Actual - Forecast. There are three ways of measuring the
accuracy of forecasts: MAD, MSE, and MAPE. MAD weighs all errors
evenly. MSE weighs errors according to their squared values.
Lastly, MAPE weighs according to relative error.
2. Qualitative forecasting is subjective, while quantitative forecasting
involves projecting historical data, or developing associative
models. Forecasts are never 100% accurate; hence, there is always
room for improvement.
Forecasting Tools:
Forecast pro: http://www.forecastpro.com
Designed specifically for business forecasters, Forecast Pro has an
intuitive interface that displays hierarchical data in a tree-like
structure, allowing you to work at any level of detail with a simple
click of the mouse. As you navigate from item to item, synchronized
windows displaying graphical and tabular reports are instantly
updated. Sharing your work is easy with Forecast Pro's flexible output
formats, including oneclick, directtoExcel reporting.