Using the Information

Decoupling Point to Improve

Supply Chain Performance
Rachel Mason-Jones and Denis R. Towill
Cardiff University

Traditionally, the decoupling point methodology has been associated with the
material flow pipeline. However, to maximize improvement in supply chain
dynamics, information flow is equally important. Many of the problems exhibited in
the material flow pipeline are the result of the distortion of marketplace sales
information as it is transferred upstream through the supply chain. This research
expands on the traditional material decoupling point methodology and establishes
the role of an information decoupling point within the supply chain. The authors
demonstrate the business opportunities generated by first recognizing the
existence of the supply chain information decoupling point and then learning how
to utilize it to gain strategic advantage. In order to make sensible planning and
delivery decisions, a business must be able to separate out contingency from real
orders as they move upstream from the marketplace. It is the basis for supply
chains moving towards continuous flow and away from point-to-point movements
and ultimately, where appropriate, to holistic control.

The difficulties of matching supply and
demand due to the inherent dynamics within
a supply chain structure are a long standing
and widespread problem. Market sales
information notoriously suffers from delay
and distortion as it moves upstream through
the supply chain. This results in production
profiles at the factory, which bear little
resemblance to the end consumer’s buying
behavior. This phenomenon was
demonstrated nearly forty years ago by Jay
Forrester [1]. Unfortunately, real world
supply chains today are still suffering the
often-painful effects of upstream order
Postponing the variant magnification and by much greater ratios than
differentiation of a experienced in the original Forrester
product until the latest simulations [2]. A good illustration of the
possible moment reduces Forrester Effect in action as presented by Fine
the risk and uncertainty [3] is shown in Figure 1. Quite apart from the
imposed by the order magnification as it passes upstream, it
consumer demands. should be noted that other important supply
chain behavioral phenomena are also
evident. These refer to the geographical
(frequently by many miles) and temporal
(usually by many months) separation of cause
and effect within this volatile dynamic
system. Given such masking of what is
happening at the marketplace, the need for
transparent information flow is paramount. It
is therefore not surprising that it should be at
the heart of many successful BPR Programs as
typified by the Carpenter Technology Case
Study [4].
Effective supply chain management
requires companies to develop innovative
strategies that integrate both their logistics
and manufacturing activities [5]. The concept
of postponement has been utilized effectively
by supply chains as a way of limiting the
effect of the debilitating inherent system
dynamics on their ability to meet customer
demands. Postponing the variant
differentiation of a product until the latest
possible moment reduces the risk and
uncertainty imposed by the consumer
demands. A good example is in paint
manufacturing business, where rather than
holding stocks of a whole range of colors,
which then requires a fair amount of
forecasting to ensure the right colors are
available, the paint is produced as a neutral
base and color added only on consumer

Volume 10, Number 2 1999 Page 13

order [5]. Thus, postponement of stages of dynamic magnification is the order
manufacture enables a supply chain to be information distortion as it flows upstream.
more reactive to real changes in consumer The authors have taken the decoupling point
demand. Hence, the product should remain methodology, which is traditionally
generic thereby postponing customization for associated with material flow, and adapted it
as long as possible [6]. to the information flow pipeline. Information
The postponement strategy requires very circulation within the supply chain is not a
careful thought as to the placement of the new idea. What is proposed here is its
material decoupling point. This basically formalization within the decoupling point
establishes a strategic stock that acts as a concept.
planned buffer between each side of the Improved information flow is made
supply chain. Such strategic placement of the possible by EDI. This may be implemented in
material decoupling point enables supply three distinct ways [7]: by automating
chains to cushion the upstream companies traditional, arms-length supply chains; by
from the fluctuating consumer demand thus linking all companies in the chain directly All supply chains have
minimizing some of the self defeating with the marketplace; and by linking all two distinct and basic
dynamics already met in Figure 1. All supply companies in the chain, but implementing flows, known as the
chains have two distinct and basic flows, holistic control over the complete system. material pipeline and the
known as the material pipeline and the Through a generic simulation model information pipeline
information pipeline respectively. However, typical of many real world supply chains the respectively.
the traditional decoupling point approach authors show that in the absence of holistic
only addresses the material flow pipeline. But control to maximize performance
one of the key sources of supply chain improvement the information decoupling

Figure 1
The Supply Chain Bullwhip Effect

Source: Charles Fine, Clockspeed: Winning Industry Control in the Age of Temporary Advantage, Reading, MA: Perseus Books, 1998.

Page 14 The International Journal of Logistics Management

 point needs to be positioned as far upstream
as possible, which is by linking all companies
in the chain directly with the marketplace.
This contrasts with the situation where the
material decoupling point is placed as close
to the end consumer as possible dependant
only on the product type, degree of
customization, and supply chain strategy
adopted. The strategic implications of the
information decoupling point are discussed
herein and the dynamic performance
improvements available via repositioning it
upstream are analyzed and demonstrated to
substantially reduce wave form propagation.
Indeed for the industrial example quoted, the
benefit is virtual level scheduling at the
factory, but is additionally coupled with
adequate automatic trend detection.
Information is a
Supply Chain Commodity
Supply chains have at least two distinct
flow pipelines: the order information transfer
pipeline, from point of sale through to raw
material supplier; and, the product transfer
pipeline from raw materials to end customer.
To fully realize Production is activated by demand
performance information therefore speed and fidelity of
improvement potential, order data transference is crucial to enabling
supply chains need to good supply chain dynamics. Figure 2 shows
develop integrated the characteristic “U” shape of the total lead-
strategies for both time pipeline through a simple, four stage,
information and material linear supply chain. It connects end
flow pipelines. consumer demand to goods delivery into the
marketplace with orders following upstream
Figure 2
Traditional Supply Chain with “U” Shaped Total Lead-time
Order information Pipeline
Retailer Distributor
? ?
Material flow pipeline
? Order decision
Volume 10, Number 2 1999
and products flowing downstream. Models of
such apparent simplicity have been shown to
generate improvement strategies, which work
on real world supply chains [8]. Such
validated models also produce outputs,
which match those of the physical system
[9]. Furthermore, the network interference
effects due to value streams competing for the
same resources are readily incorporated in
simulation experiments [10].
Many companies have concentrated
improvement strategies within the material
flow pipeline. This is because it is much
simpler to re-engineer the shop floor under
our control and chase improvement from
immediate suppliers than it is to seriously
practice total supply chain management as
required to redesign the information flow
pipeline. However, Braithwaite [11] warned
of the futility of concentrating on just one
pipeline alone. He quoted the example of a
company striving to reduce manufacturing
cycle times by just one day while not
attempting to tackle two to three week
ordering delays. In such circumstances
customer service levels remain relatively
untouched after what may be frenzied shop
floor reorganization and with no significant
impact on the cash cycle either. To fully
realize performance improvement potential,
supply chains need to develop integrated
strategies for both information and material
flow pipelines.
All supply chains have undistorted order
information available at the marketplace
Warehouse Factory
? ?
Stock
Page 15
interface dealing with the end customer.
However, in many supply chains only the
player closest to the end customer has the
luxury of knowing the true demand because
the information decoupling point is
traditionally at the marketplace – retailer
position. The question is how can this rich
information resource be strategically used to
improve dynamic performance of the supply
chain. If the order decisions within each
player in the supply chain could be enriched
with the undistorted market sales data via
moving the information decoupling point
upstream, the information distortion and
wave form propagation often exhibited in the
real world could be greatly reduced. This
would beneficially impact the behavior of
both ordering and stock level dynamics from
an individual business and from a total supply
chain perspective level [12].
Lipton, discovered the importance of
strategically utilizing information for
successful supply chain management [13].
During a lessons learned analysis conducted
after adopting an Efficient Consumer
Response (ECR) approach at Lipton, three key
components were identified as appropriate to
all levels of the supply chain: competing on
capabilities; changing the way goods are
priced; and changing the way information is
viewed and used.
Unfortunately, sharing in-depth
information sounds easy, but does not come
naturally for most companies. In fact,
company policy has in the past frequently
actively discouraged it [14]. Information has
traditionally been perceived to be power, that
is, those with the knowledge have a strategic
advantage. The traditionally adversial
relationships between supply chain members
where they regard each other as enemies
does not help anybody except competitors
[15]. The information decoupling point
strategy emphasizes the powerful competitive
advantages available only if the data are
shared through the total supply chain. This
paper shows that the dynamic benefits
available from undistorted information on the
total performance of the supply chain is best
achieved when it is treated as a holistic
commodity that can be strategically
utilized to the advantage of all players
therein. An important enabler is to move the
information decoupling point as far upstream
Page 16
as possible.
The Traditional Material The decoupling point is
Decoupling Point a standard term given to
The decoupling point is a standard term the position in the
given to the position in the material pipeline material pipeline where
where the product flow changes from “push” the product flow
changes from “push” to
to “pull”. It should therefore also correspond
“pull”.
to the Demand Penetration Point [16]. The
decoupling point is formally defined [17] as:
The point in the product axis to
which the customer’s order
penetrates. It is where order driven
and the forecast driven activities
meet. As a rule, the Decoupling
Point coincides with an important
stock point - in control terms a main
stock point - from which the
customer has to be supplied.
The material decoupling point thereby
acts as a buffer between upstream and
downstream players in the supply chain. This
enables upstream players to be protected
from fluctuating consumer buying behavior
therefore establishing smoother upstream
dynamics while downstream consumer
demand is still met via a product pull from the
buffer stock. Figure 3 summarizes in flow
diagram from the foregoing material pipeline
decoupling point definition.
The material flow decoupling point can
also be used to establish a postponement
strategy. Postponement basically moves
product differentiation as close to the end
consumer as possible via a strategic stock at
the material decoupling point. The supply
chain can produce a generic product for as
The traditionally
long as possible, thereby further smoothing
adversial relationships
the upstream dynamics via reduced product
between supply chain
variety enabling the chain to both respond
members where they
faster to consumer demands and limit the
regard each other as
effect of obsolete stock. For example,
enemies does not help
companies such as Hewlett-Packard have
anybody except
capitalized on moving the material
competitors.
decoupling point close to the customer by
producing generic printers and only
regionally customizing them at the
distribution centers as opposed to within the
manufacturing plants [18].
The strategic position of the material
decoupling point depends very much on the
product type, consumer demands and supply
chain approach adopted. Figure 4
summarizes the four simplified generic supply
The International Journal of Logistics Management
 Figure 3
Representation of the Material Decoupling Point
Forecast Driven
Distorted order data
Order flow
Forecasting
Feedback
Push
Factory Assembler
chain strategies as defined by Pagh and
Cooper, [19] which are available by altering
the position of the material decoupling point.
These strategies range from providing highly
customized products, which have high
uncertainty (full postponement strategy)
through to providing a standard product with
low demand uncertainty (full speculation
strategy). Despite the fact that the material
decoupling point for each of the previously
mentioned postponement strategies are at
different points of the supply chain, the
governing principle is always to move the
material decoupling point as close to the end
consumer as possible thereby ensuring the
shortest lead-time for the consumer. This
approach also enables full capitalization of
the benefits of divorcing the customer
variability from the demands placed on the
majority of players in the supply chain.
The Information Decoupling Point
As previously stated, there are a
minimum of two pipelines within the supply
chain; the material flow and the information
flow. The authors argue that both flows have
their own significant decoupling point, the
strategic use of which differs due to the
dynamic effect each has on the performance
…each supply chain has of the supply chain, Mason-Jones and Towill
both a material [20]. Hence, each supply chain has both a
decoupling point and an material decoupling point and an information
information decoupling decoupling point each of which is a separate
point each of which is a entity. Only by recognizing this and
separate entity. strategically positioning both can full
performance improvements be realized.
Volume 10, Number 2 1999
Order Based
Rich direct sales data
Sales data
Decoupling
point
Sales
Pull
Market
Sales
Finished goods Retailer
Stock
However, due to the differences in the two
pipelines the information decoupling point
requires a definition of its own and is defined
by the authors as follows:
The point in the information
pipeline to which the marketplace
order data penetrates without
modification. It is here where
market driven and forecast driven
information flows meet.
The order information pipeline
decoupling point is the point at which
information turns from the high value actual
consumer demand data to the typical
upstream distorted, magnified and delayed
order data. Traditionally, in supply chains this
tends to be placed at the same point as the
material decoupling point and is therefore
placed as close to the end consumer as
possible. This positioning is very wasteful
and limits the effectiveness of the high value
resource of undistorted order information
available on the dynamics of the supply
chain. Therefore to maximize the strategic
potential of these data within the supply
chain, in direct contrast to the material
decoupling point, the information decoupling
point should be moved as far upstream as
possible. This enables upstream players to
include within the ordering decisions the
unbiased undistorted, rich information that is
already available downstream. There will be
greater upstream market order penetration
while leaving the point at which the supply
chain directly responds to the customer (the
material decoupling point) intact. Figure 5
Page 17
 Figure 4
Definition of Differing Supply Chain Postponement Strategies

Full Speculation Strategy

Production Plant Plant of Optional levels Final Retailer/

central of distribution warehouse customer
warehouse warehouse level

Manufacturing Postponement Strategy

Production Plant Plant of Optional levels Final warehouse Retailer/

central of distribution level customer
warehouse warehouse

Logistics Postponement Strategy

Production Plant Plant of central Retailer/

warehouse customer

Full Postponement Strategy

Production Plant Retailer/customer

Material flows Inventory Manufacturing process Retailer/customer order point

summarizes this perception of the two
decoupling points and their relative positions
within the supply chain structure.
Theoretically, the argument for moving
the information decoupling point upstream in
order to give more players the benefit of
having undistorted data appears quite clear
and straightforward. However, the question
that needs to be answered here is how the
information decoupling point is moved and
what will be the effects on the dynamic
responses of the supply chain. Our view is
that transparency of information is the vital
step forward. How frequently it should be
sampled and used for delivery and capacity …the argument for
planning depends on the individual business. moving the information
For example, a supplier to a supermarket may decoupling point
receive hourly point-of-sale data, but plan upstream in order to
capacity over a weekly horizon. But it can give more players the
make such planning decisions far better on benefit of having
the basis of an undistorted view of what is undistorted data appears
really happening in the marketplace. Indeed, quite clear and
this information would be fundamental in any straightforward.
Continuous Replenishment Program [21].

Page 18 The International Journal of Logistics Management

 Figure 5
Comparison of Material and Information Decoupling Point Positions Within a Supply Chain

Forecast Driven Order Based

Less uncertainty due to forecast
enrichment of undistorted data

Material
Push Decoupling Pull
point

Market
Sales
Factory Assembler Finished goods Retailer
Stock
Information Direct Market Sales Information
Decoupling
point

Simulating the Information Automatic Pipeline Inventory Order Based

Decoupling Point
The use of simulation as a systems
engineering analysis tool to research
and understand the impact of supply
chain dynamics on business performance
was established 40 ago by Jay Forrester
[22]. Simulation models are widely
regarded as good demonstration vehicles for
design and development of system design
strategies by management consultants [23].
They also support general redesign principles,
which may be applied to a wide range of real
world supply chains [24]. The information
decoupling point simulation model
used here is based on the well documented
Production Control System (APIOBPCS) [25].
This in turn embraces a number of
control structures proposed elsewhere
in the literature [26]. The corresponding
simulation model order decision rule for each
echelon of the supply chain is represented in
causal loop format in Figure 6 [26].
Many supply chain dynamics
scenarios can be adequately mimicked
by using the APIOBPCS format to represent the
activities at each echelon within the chain.
By careful selection of parameters, each
echelon can be simulated with decision rules
ranging from level scheduling; make-to-stock;
passing on orders; through to fully integrated

Figure 6
Causal Loop of the Decision Support System (DSS) Utilizing Market Sales Information
Enrichment Mechanism
Our customers
order
Actual market
Smoothing time sales
constant Smoothing time
Dispatch to
our customers constant

Smoothed Smoothed market

Actual inventory dispatch sales
level Utilized sales
information

Inventory Order Order quantity to

correction rate Decision our supplier

Inventory Orders in pipeline

shortfall correction
Inventory recovery
time Pipeline recovery
Our desired Orders in pipeline time
inventory level shortfall

Volume 10, Number 2 1999 Page 19

 Figure 7
Information Decoupling Point Position for Each of the Simulated Designs
Information flow pipeline
1
Retailer
2 Distributor
Design
3
Material flow pipeline
4
Market sales
information
Design 1
Design 2
Design 3
Design 4
order/stock control policies. Such a simulation
model has been many times verified on real
world supply chains including the electronics
products industry and the automotive industry
[28]. The APIOBPCS based simulation model
may be used confidently as a benchmark to
demonstrate performance enhancement for a
wide range of practical supply chains. Note
from the Causal Loop Diagram that we have the
freedom to mix the marketplace information
with our direct customer orders in any
combination we wish. The important point is
that by minimizing downstream order
uncertainty, we have the data needed to make
an informed judgement [29].
The demonstrator model utilized is a
four level supply chain constituting a retailer,
distributor, warehouse and factory. For
convenience, the material flow lead times are
set the same at each stage. If anything, due to
the presence of resonance effects, such
equality tends to highlight potential areas for
performance improvements [30] Figure 7
shows a representation of the optimized order
information pipeline. This adaptation we call
the information enrichment model, and it is
achieved via an electronic point of sales link
(EPOS) to all players down stream of the
information decoupling point. This enables
Page 20
Warehouse Brewery
Distorted order
information
=Information
decoupling point
all such players to have an undistorted view
of the consumer buying behavior thereby
enabling the supply chain to operate
effectively without the added burden of the
distortion of end customer demand
characterized by the Forrester Effect.
In order to analyze the strategic benefits
of the information decoupling point and In order to analyze the
determine where in the supply chain its strategic benefits of the
optimal position might be, four competing information decoupling
designs were simulated. Figure 7 illustrates point and determine
the selected relative information decoupling where in the supply
point positions in the supply chain. Design 1 chain its optimal
simulates the traditional supply chain where position might be, four
only the retailer accesses the actual buying competing designs were
pattern. This design provided data against simulated.
which the other designs could be
benchmarked. Design 2 and 3 test the
dynamics of the supply chain as the
information decoupling point is moved
further upstream enabling an increasing
number of players to utilize the consumer
sales information. Finally, Design 4 analyzes
the potential dynamic improvements
available if the information decoupling point
is moved to the furthest point upstream
thereby enabling all four members of the
supply chain to have a direct line of sight to
the consumer.
The International Journal of Logistics Management
 We find that for synthesis purposes it is
easiest to use a step demand to simulate a
shock event at the marketplace. Analyzing
the response to a step increase in demand in
the marketplace is an essential starting point
when simulating a supply chain because of
the relative ease of analysis and the extensive
background of previous research. The reason
for this is that the step response provides an
extremely rich picture of the supply chain
dynamics and which can be used as a
predictor of system behavior under a wide
range of circumstances. Therefore, the
simulation model was injected with a 20%
step increase in consumer demand for each of
the designs to be tested. Using the step
response as a performance benchmark it is
possible to thereby predict system responses
for a wide range of operating scenarios. [31].
Moving the Information
Decoupling Point to Maximize
Competitive Opportunity
The dynamic behavior of the supply
chain for each of the four designs was
assessed and analyzed via the order rate and
stock level behavior patterns at each echelon.
Since it is the factory which experiences the
worst buffeting in a traditional supply chain,
this would appear to be an important statistic
for measuring the improvement in dynamic
behavior of the whole chain. The peak
factory order rate is a simple but realistic
performance metric since improvement here
will benefit the whole chain.
The percentage of improvement of
the factory order rate for each of the four
designs is summarizd in Figure 8. For
convenience, the improvement scale selected
is linear and in the range 0% for the
datum design to 100% for the idealized
best imagined dynamic response. Using such
a linear scale gives a very conservative
estimate of likely performance improvement.
What is clear from the simulation results is
that the further the information decoupling
point is moved upstream the better the
…the further the improvement in the dynamic behavior of the
information decoupling supply chain, as observed via factory order
point is moved upstream rate. This is due to improved knowledge of
the better the the actual end consumers buying behavior
improvement in via a direct link to the marketplace for all
the dynamic behavior of players downstream of the information
the supply chain… decoupling point.
Volume 10, Number 2 1999
The delay of initial response to a change
in the marketplace experienced by the
baseline traditional supply chain (Design 1) is
reduced through strategic use of the
information decoupling point. The further
upstream it is placed, the more stages of
the supply chain have unbiased data of
the consumers’ buying pattern, and can
further help to improve the dynamic response
of the whole chain. When the
information decoupling point is placed at the
factory, the potential performance
improvement due to this particular action is
maximized; Figure 8 shows a gain of 80%
of the theoretical best attainable
dynamic performance. This is because the
whole chain begins to respond to a consumer
change as soon as it happens since the factory
can see changing consumer habits as they
occur. This is in contrast to the much
smaller benefit gained when the information
decoupling point is placed at the
distributor, as might be the case for
Vendor Managed Inventory (VMI). But
with full disclosure upstream, players can
now endeavour to plan ahead and pre-empt
future dispatch fluctuations, thereby
reducing the distortion effects. Hence, via the
strategic placement of the information
decoupling point at the furthest point
upstream, the factory is able to make an
informed decision at a much earlier point in
time than the factory at the end of the
traditional supply chain.
Using Both Decoupling Points:
An Industry Example
More companies are becoming aware of
the dynamic benefits available via better use
of the information within the supply chain.
Some organizations make the mistake of
thinking that adopting Information
Technology (IT) is in itself sufficient to realize
the available benefits. But IT is an enabler
and must be utilized properly, and in itself is
not the total answer [32]. While managers in
many companies see the need for information
transparency perhaps less have realized that
what they need is a policy on the strategic
placement of the information decoupling
point. The optimal place for the information
decoupling point is at the furthest point
upstream possible, thereby, enabling the
maximum number of members in the chain to
Page 21
 Figure 8
Dynamic Performance for Different Information Decoupling Point Designs
0 20 40
% of dynamic improvement
have available the actual marketplace
information for better forecasting and
decision making.
This section summarizes a substantial
reengineering program aimed at streamlining
an electronics products supply chain, which
included a strategic placement of both the
material and information decoupling points
[33]. The process of supply chain redesign
consisted of a series of overlapping business
process reengineering (BPR) programs that
began in the early 1980’s and which continue
to be enabled through to the present day. As
part of its BPR initiative, the electronics
products company concentrated the later
stages on establishing and strategically
placing both a material and an information
decoupling point. The optimized strategy is
illustrated in Figure 9 where the positioning of
both decoupling points can be clearly
identified in what has since become
recognized as the legal supply chain [34]. As
can be seen, the material decoupling was
placed as close to the consumer as possible
(at the finished goods assembler), while
available postponement opportunities were
maximized. The information decoupling
point was placed as far upstream as possible,
thereby communicating the actual consumer
demand direct to the components assembler.
The component assembler is forewarned of
changes in sales profile and is able to meet
the necessary demands of the supply chain.
The company’s BPR program was
carried out in four key stages as identified in
Figure 10. Each stage of the BPR program
actively improved the dynamic performance
Page 22
UPSTREAM MOVEMENT OF
DECOUPLING POINT
60 80 100
of the supply chain. The dynamic
improvements estimated via the verified
aggregate product six-echelon simulation
model realized for each BPR stage is
summarized in Figure 10. Although in
practice some BPR stages overlapped, for the
purpose of illustration the simulation is based
on sequential implementation. For
consistency, judgements were made on the
basis of dynamic response at the most
upstream echelon to a step change in the
marketplace. The linear scale was selected so
that 0% is the base point and 100% is the
idealized best possible response. Once all
four stages of improvement were completed,
the supply chain model predicted an
impressive resulting estimated dynamic
improvement of 92% in comparison to its
initial baseline performance. The final system
design eventually generated the desirable
upstream pattern of virtual level scheduling.
At the same time the system retains the highly
desirable automatic trend following
capability throughout the chain.
What is apparent from Figures 9 and 10
is that the latter two stages of the BPR
program involve establishing and strategically
placing both material and information
decoupling points. The utilization of both the
EDI enabled information decoupling point
(which occurs in stage D) and product design
material decoupling point (stage E)
approaches enabled the supply chain to
further improve its dynamic performance due
to these sources alone by approximately 42%
as shown in Figure 10. In this supply chain
information transparency was in place before
The International Journal of Logistics Management
 Figure 9
The Reengineering Electronic Products Supply Chain

Global
DEMAND INFO
Material
Planning STOCK INFO STOCK
System INFO
PUSH STOCK PULL
PLAN
INFO
STOCK
INFO PULL
PULL PULL PULL PULL PULL

PULL
D.P.
PULL

Components Sub-Assembly Finished Goods Distribution Retailers

Assembly lead-time Assembly centers lead-time
lead-time (weeks) (weeks) lead-time (days) lead-time (days) (days)
Flow of information

finalization of the material flow decoupling
point in its present form. This is because in
the electronic products supply chain, the
material flow decoupling point location is
heavily dependent on product design [35],
which generally takes longer to achieve,
compared to the installation of EDI. It is the
combination of information transparency and
product modularization that enables mass
customization [36].
Conclusion
Supply chains are constructed around
two principal pipelines: the information and
material flows. These require operational
strategies, which are sympathetic and which
integrate and compliment the individual
characteristics of both pipelines. The
decoupling approach has been successfully
utilized within the material pipeline by many
companies. However, utilizing a decoupling
point can create benefits in the information
flow pipeline. The information decoupling
point is the point in the supply chain at which
the order information changes from
marketplace sales data to forecast driven data.
It is the basis for supply chains moving
towards continuous flow and away from
point-to-point movements and ultimately to

Figure 10
Estimated Improvements in Dynamic Performance of a Electronics Products Supply Chain
at Various Stages of BPR Program

(A) Baseline operation

1981
(B) Plus continuous flow manufacturing
1984
(C) Plus global material planning systems (MPS)
1987
(D) Plus vendor integration into global (MPS)
1990
(E) Plus strategic use of decoupling point & total pipeline control
1993

0 25 50 75 100
% of dynamic improvement

Volume 10, Number 2 1999 Page 23

holistic control [37].
Management has attempted to address
information flow difficulties by implementing
Information Technology (IT) based solutions
such as EDI. But IT driven solutions
frequently fail to delivery the full potential
benefits. So the question many managers are
asking is why despite implementing IT have
the traditional problems associated with
information flow not been eradicated? The
answer is that IT is not the unique solution
because it is just one of many tools required.
What is of crucial importance is the
information usage methodology adopted
when using IT. If the actual information
transference strategy remains the same then
the bolted-on IT will only succeed in moving
distorted data faster! This is because the
newly adopted IT is utilizing the existing
information strategy and hence still contains
the same decision points that led to the
original upstream information distortion.
Significant dynamic performance
benefits will accrue if the value of the
information transferred with the supply chain
is improved thereby reducing bias, noise, and
staleness. There is evidence that this is now
happening on a major scale [38]. The
approach is to establish an information
decoupling point and place it as far upstream
as possible enabling as many members of the
supply chain as possible to access the actual
marketplace data. Businesses at each
echelon can decide how best to exploit
undistorted information in their delivery and
planning processes. They usually have to
overcome the functional silo mentality where
information is regarded as power to the
individual, and replace it with a philosophy
that transparent information flow means more
power and more profitability for the entire
chain.
References
[1] Forrester, Jay W., Industrial
Dynamics, Cambridge, MA: MIT Press, 1960.
[2] Holmstrom, J., A. P. Hameri, N. H.
Nielson, J. Pankakoshi and J. Slotte,
“Examples of Production Dynamics Control
for Cost Efficiency,” International Journal of
Production Economics, Vol. 48, (1997), pp.
109–119.
[3] Fine, Charles, Clockspeed: Winning
Industry Control in the Age of Temporary
Page 24
Advantage, Reading, MA: Perseus Books,
1998.
[4] Goldman, Steven, Roger Nagel and
Kenneth Preiss, Agile Competitors and Virtual
Organizations, Reinhold, NY: Van Nostrand,
(1995).
[5] Pagh, Janus D. and Martha C.
Cooper, “Supply Chain Postponement and
Speculation Strategy: How to Choose the
Right Strategy,” Journal of Business Logistics,
Vol. 19, No. 2 (1998).
[6] Bowersox, Donald, Philip Carter and
Robert Monczka, “Material Logistics
Management,” International Journal of
Physical Distribution & Logistics
Management, Vol. 23, No. 5 (1993), pp.
46–51. What is of crucial
[7] Hewitt, Fred, “Supply or Demand? importance is the
Chains or Pipelines? Coordination or information usage
Control?” Proceedings 4th International methodology adopted
Symposium on Logistics in the Information when using IT.
Age, 1999, pp. 785–790.
[8] Towill, Denis R. and Peter McCullen,
“The Impact of an Agile Manufacturing
Program on Supply Chain Dynamics,” The
International Journal of Logistics
Management, Vol. 10, No. 1 (1999), pp. 83-
96; Joakim Wikner, Denis R. Towill and
Mohammed Naim, “Smoothing Supply Chain
Dynamics,” International Journal Production
Economics, Vol. 22, (1991), pp. 231-248.
[9] Vecchio, Anthonio del and Denis R.
Towill, “Conceptual Modeling of Industrial
Production Distribution Systems via Standard
Forms,” Math. Comp. Modelling, Vol. 14,
(1990), pp. 1029-1034.
[10] Towill, Denis R., Gary N. Evans and
Preetinder Cheema, “Analysis and Design of a
Minimum Reasonable Inventory Control
System,” International Journal Production
Planning Cont, Vol. 8, No. 6 (1997), pp. 545-
547.
[11] Braithwaite, Alan, “Logistic Systems
or Customer Focused Organization: Which
Comes First,” Logistics Information
Management, Vol. 6, No. 4 (1993), pp. 26-37.
[12] Mason-Jones, Rachel and Denis R.
Towill, “Information Enrichment: Designing
the Supply Chain for Competitive
Advantage,” Supply Chain Management: An
International Journal, Vol. 2, No. 4 (1997), pp
137-149.
[13] Collins, Richard, “ECR – Breaking
the China in the US Supermarket Industry,”
The International Journal of Logistics Management
 Supply Chain Management, Vol. 2, No. 3
(1997), pp. 92-98.
[14] Towill, Denis R., “The Seamless
Supply Chain - The Predators Advantage,”
International Journal of the Technology
of Management, Vol. 13, No. 1 (1997),
pp. 37-56.
[15] Macbeth, Douglas K. and N.
Ferguson, Partnership Sourcing: An Integrated
Supply Chain Approach, Financial Times,
London: Pitman Publishing, (1994).
[16] Christopher, Martin, “Logistics and
Supply Chain Management,” 2nd Edition,
Financial Times, London: Pitman Publishing,
1998.
[17] Hoekstra, Sjoerd and Jac Romme,
Integrated Logistics Structures: Developing
Customer Oriented Goods Flow, London,
McGraw-Hill, (1992).
[18] Davies, Tom, “Effective Supply
Chain Management,” Sloan Management
Review, Summer, (1993), pp 35-45.
[19] Pagh, Janus D. and Martha C.
Cooper, “Supply Chain Postponement and
Speculation Strategy: How to Choose the
Right Strategy,” Journal of Business Logistics,
Vol. 19, No. 2 (1998).
[20] Mason-Jones, Rachel and Denis R.
Towill, “Time Compression in the Supply
Chain: Information Management is the Vital
Ingredient,” Logistics Information
Management, Vol. 11, No. 2/3 (1998), pp.
93–105.
[21] Andraski, Joseph “Foundations for
Successful Continuous Replenishment
Programs,” The International Journal of
Logistics Management, Vol. 5, No. 1 (1994),
pp. 1-8.
[22] Forrester, Jay W., Industrial
Dynamics, Cambridge, MA: MIT Press, 1960.
[23] Stalk, George and T. Hout,
Competing Against Time, Free Press: Michael
Balle, The Business Process ReEngineering
Action Kit, London: Kogan Page, 1995.
[24] Towill, Denis R. and Peter
McCullen, “The Impact of an Agile
Manufacturing Program on Supply Chain
Dynamics,” The International Journal of
Logistics Management, Vol. 10, No. 1 (1999),
pp. 83-96; Joakim Wikner, Denis R. Towill
and Mohammed Naim, “Smoothing Supply
Chain Dynamics,” International Journal
Production Economics, Vol. 22, (1991), pp.
231-248.
Volume 10, Number 2 1999
[25] John, Simon, Mohamed Naim and
Denis R. Towill, “Dynamic Analysis of a WIP
Compensated Decision Support System,”
International Journal of Manufacturing Systems
Design, Vol. 1, No 4 (1995), pp. 283-297.
[26] Aken, Van, J. E., On the Control of
Complex Industrial Organizations, London,
Martinus Nijhoff, (1978); John Sterman,
“Modeling Managerial Behavior:
Misperceptions of Feedback in a Dynamics
Decision-Making Experiment,” Management
Science, Vol. 35, No. 3 (1989), pp. 321-339;
Ann Ackere, Erik R. Larsen and John W.D.
Morecroft, “Systems Thinking and Business
Process Re-design: An Application to the Beer
Game,” European Management Journal, Vol.
11, No. 4 (1993), pp. 412-423; Marc
Lambrecht and Jeroin Dejonckeere,
“Exending the Beer Game to Include Real-Life
Supply Chain Characteristics,” Proceedings
EUROMA Conference on Managing
Operations Networks, (1999), pp. 237-243;
Keith Popplewell and Maurice C. Bonney,
“The Application of Discrete Linear Control
Theory to the Analysis and Simulation of
Multi-Product, Multi-Level Production
Control Systems,” International Journal of
Production Research, Vol. 25, No. 1 (1987),
pp. 45-56.
[27] Mason-Jones, Rachel, Mohamed
Naim and Denis R. Towill, “The Impact of
Pipeline Control on Supply Chain Dynamics,”
The International Journal of Logistics
Management, Vol. 8, No 2 (1997), pp. 47-62.
[28] Berry, Danny, Mohamed Naim and
Denis R. Towill, “Business Process Re-
Engineering an Electronic Products Supply
Chain,” IEE Proc-Sci Meas. Technology, Vol.
142, No. 5 (September 1995); and Vecchio,
Anthonio del and Denis R. Towill,
“Conceptual Modeling of Industrial
Production Distribution Systems via Standard
Forms,” Math. Comp. Modelling, Vol. 14,
(1990), pp. 1029-1034.
[29] Mason-Jones, Rachel, Ph.D. Thesis,
Logistics Systems Dynamics Group, Cardiff
University, 1998; and Ann Van Ackere, Erik
Larsen and John Morecroft, “Systems
Thinking and Business Process Re-design,”
European Management Journal, Vol. 11, No.
4 (1993), pp. 412-423.
[30] Towill, Denis R. and Anthonio del
Vecchio, “The Application of Filter Theory to
the Study of Supply Chain Dynamics,”
Page 25
International Journal Production Planning and
Control, Vol. 15, No. 1 (1994), pp. 82-96.
[31] Mason-Jones, Rachel, Mohamed
Naim and Denis R. Towill, “The Impact of
Pipeline Control on Supply Chain Dynamics,”
The International Journal of Logistics
Management, Vol. 8, No 2 (1997), pp. 47-62.
[32] Davenport, Thomas, “Saving IT’s
Soul: Human Centred Information
Management,” Harvard Business Review,
(March-April 1994), pp. 119-131; Marshall
Fisher, “What is the Right Supply Chain for
Your Product?,” Harvard Business Review,
(March-April 1997), pp. 105-116; and,
Fred Hewitt, “Supply or Demand? Chains or
Pipelines? Coordination or Control?”
Proceedigns 4th International Symposium on
Logistics in the Information Age, 1999, pp.
785–790.
[33] Berry, Danny, Ph.D. Thesis,
Logistics Systems Dynamics Group, Cardiff
University, 1994.
[34] Naylor, J. Ben, Mohammed Naim
and Danny Berry, “Leagility: Integrating the
Lean and Agile Manufacturing Paradigms in
the Supply Chain,” International Journal
Production Economics, Vol. 62, (1999), pp.
107-118.
[35] Lee, Hau and Marguerita Sasser,
“Product Universatility and Design for Supply
Chain Management,” International Journal of
Production Planning and Control, Vol. 6, No.
3 (1995), pp. 270-277.
[36] Childerhouse, Paul and Denis R.
Towill, “Engineering Supply Chains to
Match Customer Requirements,” Logistics
Information Management, 1995; Michael
Dell with Catherine Fredman, Direct From
Dell: Harper Collins Business, 1999.
[37] Hewitt, Fred, “Supply or Demand?
Chains or Pipelines? Coordination or
Control?” Proceedigns 4th International
Symposium on Logistics in the Information
Age, 1999, pp. 785–790.
[38] Hewitt, Fred, “Supply or Demand?
Chains or Pipelines? Coordination or
Control?” Proceedings 4th International
Symposium on Logistics in the Information
Age, 1999, pp. 785–790.

Rachel Mason-Jones is a Ph.D. researcher in the Logistics Systems

Dynamics Group at the University of Wales. Her Ph.D. deals with the analysis,
simulation and optimization of real world supply chains via the Industrial Dynamics
approach. Her experience includes design and development of mechanical
hardware and project management in the gas exploration/platform and electronic
products industries and first hand experience of Japanese manufacturing
techniques. She can be reached at Logistics Systems Dynamics Group,
Department of Maritime Studies and International Transport, University of Wales,
P. O. Box 924, Cardiff, CFI 3, UK. Phone: 44 1222 874271 ext. 6312.
Fax: 44 1222 874301.

Denis R. Towill is the Lucas Research Professor and Director of the Logistics
Systems Dynamics Group at Cardiff University. He holds the Higher Doctorate of
DSc from the University of Birmingham. His experience inlcudes the design of a
wide range of hard and soft systems, including supply chains, in aerospace,
process, automotive, electronic products and maritime industries. He has published
over 350 papers, 12 of which have been awarded institution prizes. He is past
Chairman of the Management and Design Board of the Institution of Electrical
Engineers. In 1988 he became only the second manufacturing systems engineer to
be elected to the Royal Academy of Engineering. He can be reached at Cardiff
University, 65-68 Park Place, Cardiff CFI 3YP, Wales, U.K. Phone: 41 1222 874271.
Fax: 44 1222 874301. E-mail: http://www.cf.ac.uk/uwcc/masts/

Page 26 The International Journal of Logistics Management