Hierarchical forecasting: issues and use guidelines
| DOI | https://doi.org/10.1108/02635570110365952 |
| Date | 01 February 2001 |
| Published date | 01 February 2001 |
| Pages | 5-12 |
| Author | Gene Fliedner |
| Subject Matter | Economics,Information & knowledge management,Management science & operations |
Hierarchical forecasting: issues and use guidelines
Gene Fliedner
Oakland University, Rochester, Michigan, USA
Introduction
Individuals representing various
management levels and functional
disciplines within an organization have a
variety of forecast information needs. Upper,
middle, and lower level managers have
different forecast information requirements.
Likewise, operations, finance, and marketing
managers each have different forecast
information requirements. For many firms,
their product lines can be used to portray
these different forecast information needs.
For example, L.L. Bean, Inc., a major
cataloger and retailer of outdoor apparel, has
the hierarchical product line structure
illustrated in Figure 1. One can easily
envision how forecast information could be
useful for various users and various
planning activities. For instance, long-term
forecasts of company-wide dollar sales for all
merchandise groups could be useful for
financial budgeting and capacity planning
purposes. Intermediate-term forecasts of
semi-aggregate unit sales for demand centers
could be useful for reserving supplier
capacity and workforce size decisions. Short-
term forecasts of individual item demands
could be useful for inventory planning and
control decisions.
Given the variety of forecast information
needs in large organizations, a centralized
forecast system capable of satisfying all
users' information requirements could be an
important information system. Hierarchical
forecasting (HF), a family-based forecast
methodology, is a centralized forecast
approach capable of satisfying the variety of
forecast information requirements. HF is
able to provide decision support information
to many users, each representing different
management levels and organizational
functions (Fliedner and Mabert, 1992). HF
would be an integral element and easily
integrated within the framework of the
enterprise resource planning (ERP) system.
In addition to the ability of providing
forecast information for numerous users, the
potential of HF is important for two
additional reasons. First, it has the potential
to improve forecast accuracy and support
improved decision making (Fliedner, 1989).
Second, given the proliferation of product
lines in order to enhance customer
satisfaction, HF systems are being used more
extensively in industry as firms attempt to
reduce the magnitude of the forecast
problem. This suggests a need to understand
its capabilities.
As firms pursue cost-cutting measures to
enhance competitiveness, the economic
incentive for research to provide practical
use guidelines regarding HF systems has
become increasingly important. To date,
several studies have offered practical
guidelines for the structural design of HF
systems. The primary purpose of this paper
is to summarize these guidelines. First, an
explanation of the HF process is provided. In
this explanation, important system
parameters and strategic choices, which
allow for the custom configuration of HF
systems, are identified. Second, the relevant
family-based forecast research is reviewed.
The important issues addressed and the
conclusions presented in this research are
identified. Third, practical guidelines
regarding the use of a HF approach that have
been reported in the research literature are
clearly delineated. With much still unknown
regarding the performance impact of various
system parameters and strategic process
choices, the paper concludes with
suggestions for future research.
Hierarchical forecasting process
explained
HF systems are used to provide forecast
information based on a strategy of grouping
The current issue and full text archive of this journal is available
at
http://www.emerald-library.com/ft
[5]
Industrial Management &
Data Systems
101/1 [2001] 5±12
#MCB University Press
[ISSN 0263-5577]
Keywords
Forecasting, Time series,
Hierarchy, Product management
Abstract
In order to provide the appropriate
demand forecast information given
various managerial levels and
functional disciplines within
organizations, reliance on family-
based forecasting is increasing. The
family-based approach, sometimes
referred to as hierarchical
forecasting (HF), is based on a
strategyof aggregating items into
families. HF systems are capable of
providing forecasts for items and
their respective families. The
objectivesof HF systems, include
improved forecast performance and
a reduction in the overall
forecasting burden. To date, several
studies have offered practical
guidelines for the structural design
of HF systems. The primary purpose
of this paper is to summarize these
guidelines. First, an explanation of
the HF process is provided. In this
explanation, important system
parameters and strategic choices,
which allow for the custom
configuration of HF systems are
identified. Second, the relevant
family-based forecast research is
reviewed. The important issues
addressed and the conclusions
presented in this research are
identified. Third, practical
guidelines regarding the use of a HF
approach that have been reported in
the research literature are clearly
delineated. With much still
unknown regarding the
performance impact of various
system parameter and strategic
process choices, the paper
concludes with suggestions for
future research.
To continue reading
Request your trial