Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
FORECASTING IN S&OP
By Larry Lapide
(This is an ongoing column in quantitatively based statistical forecast rapidly changing business environment.
The Journal, which is intended methods are becoming less effective in More market intelligence now needs to be
capturing all that is happening in today’s incorporated during the development of
to give a brief view on a potential forecasts.
topic of interest to practitioners of
business forecasting. Suggestions This means that effective collaborative
on topics that you would like to see environments are necessary to solicit input
covered should be sent via email to from a wider group of people within a
company who have some understanding of
llapide@mit.edu). where a business might be going. This has
T
his May I was invited by my led to greater adoption and interest in the
colleagues at the University of use of S&OP (a process that I have recently
Tennessee to give a presentation on found myself discussing and writing more
the importance of Top-Down & Bottom- about). This is a cross-functional process
Up forecasting in the Sales and Operations that brings together teams of individuals on
Planning (S&OP) process. The last time a routine basis to plan for where businesses
I had covered this type of forecasting in are going on a tactical basis. Each team
one of my columns was eight years ago member brings to the process a specific
in the Summer 1998 issue of The Journal perspective during the development of
of Business Forecasting (JBF). Its title supply and demand plans/forecasts.
was “A Simple View of Top-Down Versus
Bottom-Up Forecasting.” When I wrote In this regard, each S&OP team member
that piece, I felt I had given the topic the may have to generate, review, and revise
LARRY LAPIDE demand forecasts that reflect the aspects of a
coverage it needed for a very long time.
However, having to prepare for this year’s business with which they are most familiar.
Dr. Lapide is a Research Director at This requires leveraging Top-Down &
presentation opened my eyes to the fact that, MIT’s Center for Transportation &
since I wrote that article eight years ago, Bottom-Up forecasting in the process more
Logistics where he manages its Supply than that required eight years ago. The
much has changed. The topic of Top-Down Chain 2020 Project focused on supply
& Bottom-Up forecasting has become even remainder of this article discusses Top-
chain management of the future and Down & Bottom-Up forecasting in terms
more important. its demand management research of what it is, how it supports accountability
initiatives. He has extensive business and commitment in the S&OP process,
The expanding globalization of business, experience in industry, consulting,
the continuing move from push to pull and the use of forecasting hierarchies as
and research, and has a broad range enablers.
manufacturing, and the rise in consumer- of forecasting experiences. He was a
oriented economies, have led to a much more forecaster in industry for many years,
complex forecasting and planning world. has led forecasting-related consulting
Forecasters and planners are being asked projects for clients across a variety of
WHAT IS TOP-DOWN &
to create plans for expanding geographies, industries, and has taught forecasting in BOTTOM-UP FORECASTING?
increased numbers of sales channels, a college setting. In addition, for seven
and broader, more diverse, and shorter years he was a leading market analyst Top-down forecasting is extremely
life cycle product lines. This complexity in the research of forecasting and supply useful for improving the accuracy of
means that markets are more dynamic and chain software. detailed forecasts. As depicted in Figure
Demand
forecasting each component separately, it is
better to first forecast the aggregated group
Time
and then disaggregate the resulting forecast
to derive the forecasts of the individual
components. The good news is that this
principle can be leveraged for any type of
Entity 1 Entity 2 Entity 3
aggregation, such as aggregations across
products, sales channels (e.g., stores),
Demand
Demand
Demand
geographies, and even time itself.