The current interest in performance measurements has led to a variety of supporting
adages or cliches in the industry, such as:

• “Anything measured improves.”
• “What you measure is what you get.”
• “Anything measured gets done.”
• “You can’t manage what you do not measure.”

These are not new business ideas, but there are a few new twists. Using measurements
to support manufacturing operations has its roots back to the late 19th and
early 20th centuries with ideas espoused by Frederick W. Taylor, the father
of applying scientific methods to running business. His ideas for time and motion
studies of operations were successfully used to scientifically manage production
lines and warehouse operations. These ideas, however, led to exaggerated business
processes that transitioned into “running a business by the stopwatch” with
employers treating human employees as if they were highly reliable, predictable
machines to be monitored and controlled. Over time, the workplace’s view of
performance measurement became more humane and these exaggerated types of monitor
and control methods fell out of favor, replaced by a focus on a measuring a
business’ performance rather than that of the individual.

Throughout the last decade, companies have expended significant amounts of
time and effort to re-engineer their supply chains through business process
change and technology focused on implementing integrated Supply Chain Management
(SCM) principles. While substantial financial and human resources have been
spent on doing this, there has been little sign of realized benefits. While
consultants are recommending supply chain measurement, they generally lack formal
approaches to it. In addition, while SCM software providers are selling solutions
that enable companies to drastically improve their supply chain performance,
these same vendors do not adequately provide tools needed to measure these improvements.
In this report, AMR Research discusses supply chain performance measurement
and the results of research conducted to address the following questions:

• Why is performance measurement important?
• What general approaches are available to measure supply chains?
• What advice can be followed when selecting performance measures?
• What methods are available for setting performance targets?
• What are application vendors doing to support supply chain performance
  measurement?
• How should a company get started?

Why Is Performance Measurement Important?

Measurement is important, as it affects behavior that impacts supply chain performance.
As such, performance measurement provides the means by which a company can assess
whether its supply chain has improved or degraded. The importance of using measures
to help ensure that a supply chain is performing well can be illustrated by
the following anecdotal story:

Tom is driving on a long trip in a car that has a broken speedometer and a
broken gas gauge. He has been traveling for several hours, keeping track of
the time and looking at his odometer to determine how fast he is going. He is
sure that he has been obeying the speed limits – when he is stopped by a patrolman
and given a speeding ticket. Slowing down, he drives for two more hours keeping
track of the time and his odometer, but once again is stopped by a patrolman
and given another speeding ticket. During the remainder of the trip Tom slows
down to a speed that he now believes will avoid getting another speeding ticket.
He drives for one more hour when the car stops all of a sudden. He ran out of
gas!

Not a very good trip for Tom, primarily because he was missing some very important
key measurement devices in his car – the speedometer and the gas gauge. Unlike
Tom, most people would be extremely reluctant to drive this car. In a similar
way, however, there are many companies that run their supply chains without
a good set of measurements in place. Like Tom, the only way they are able to
find out if they are meeting their supply chain goals is after the fact, by
diagnosing poor financial results, or when they lose an important customer –
events similar to Tom’s speeding tickets.

There are several lessons on the importance of measuring supply chain performance
to be learned from this story:

• Measurements are important to directly controlling behavior and indirectly
  to performance – the speedometer reading impacts how hard or soft Tom pushes
  on the gas pedal.
• A few key measurements will go a long way toward keeping a company on track
  towards achieving its supply chain improvement objectives – like those on
  a speedometer and a gas gauge.
• Seemingly relevant, but cumbersome, measurements are of little use, and
  are possibly a hindrance, in helping to improve supply chain performance –
  like the odometer in the car.
• Picking the wrong measures and leaving out important ones could lead to
  supply chain performance degradation – like running out of gas.
• Driving a supply chain based only on after-the-fact measures, like losing
  an important customer or having poor financial performance is not very effective
  – the way getting speeding tickets and running out of gas is an expensive
  way to drive a car.

What General Approaches Are Available To Measure Supply
Chains?

Traditionally, companies have tracked performance based largely on financial
accounting principles, many which date back to the ancient Egyptians and Phoenicians.
Financial accounting measures are certainly important in assessing whether or
not operational changes are improving the financial health of an enterprise,
but insufficient to measure supply chain performance for the following reasons:

• The measures tend to be historically oriented and not focused on providing
  a forward-looking perspective.
• The measures do not relate to important strategic, non-financial performance,
  like customer service/loyalty and product quality.
• The measures do not directly tie to operational effectiveness and efficiency.

In response to some of these deficiencies in traditional accounting methods
for measuring supply chain performance, a variety of measurement approaches
have been developed, including the following:

• The Balanced Scorecard
• The Supply Chain Council’s SCOR Model
• The Logistics Scoreboard
• Activity-Based Costing (ABC)
• Economic Value Analysis (EVA)
• Balanced Scorecards

The Balanced Scorecard recommends the use of executive information systems
(EIS) that track a limited number of balanced metrics that are closely aligned
to strategic objectives. The approach was initially developed by Robert S. Kaplan
and David P. Norton and was discussed in an article, titled “The Balanced Scorecard
– Measures That Drive Performance,” published in the Harvard Business Review,
January-February 1992. While not specifically developed for supply chain performance
measurement, Balanced Scorecard principles provide excellent guidance to follow
when doing it. The approach would recommend that a small number of balanced
supply chain measures be tracked based on four perspectives:

• Financial perspective (e.g., cost of manufacturing and cost of warehousing
  )
• Customer perspective (e.g., on-time delivery and order fill rate)
• Internal business perspective (e.g., manufacturing adherence-to-plan and
  forecast errors)
• Innovative and learning perspective (e.g., APICS-certified employees and
  new product development cycle time)

An industry has grown around the Balanced Scorecard approach with a variety
of firms that provide consulting and solutions for implementing performance
measurement, such as:

• Renaissance Worldwide, Inc. (Newton, MA) got its start doing this Balanced
  Scorecard consulting and grew to be one of the 30 largest consulting firms.
• Gentia Software Inc. (Boston, MA) markets a software application, Gentia’s
  Renaissance Balanced Scorecard that incorporates Renaissance Worldwide’s performance
  measurement approach.
• Corvu Corp. (Edina, MI) sells a Balanced Scorecard System software application
  that provides interactive scorecard functionality.

Supply Chain Council’s SCOR Model

The Supply Chain Council’s SCOR Model provides guidance on the types of metrics
one might use to get a balanced approach towards measuring the performance of
one’s overall supply chain. The SCOR Model approach advocates a set of supply
chain performance measures comprised of a combination of:

• Cycle time metrics (e.g., production cycle time and cash-to-cash cycle)
• Cost metrics (e.g., cost per shipment and cost per warehouse pick)
• Service/quality metrics (on-time shipments and defective products)
• Asset metrics (e.g., inventories )

In contrast to the Balanced Scorecard, which is focused on executive enterprise-level
measurement, the SCOR Model approach directly addresses the needs of supply
chain management with balanced measurements. Figure 1.0 depicts an illustrative
set of supply chain measures balanced among the SCOR Model’s top-level processes.

Figure 1.0 Illustrative performance measures

The Logistics Scoreboard

Another approach to measuring supply chain performance was developed by Logistics
Resources International Inc. (Atlanta, GA), a consulting firm specializing primarily
in the logistical (i.e., warehousing and transportation) aspects of a supply
chain. The company recommends the use of an integrated set of performance measures
falling into the following general categories:

• Logistics financial performance measures (e.g., expenses and return on
  assets )
• Logistics productivity measures (e.g., orders shipped per hour and transport
  container utilization)
• Logistics quality measures (e.g., inventory accuracy and shipment damage
  )
• Logistics cycle time measures (e.g., in-transit time and order entry time)

Logistics Resources sells a spreadsheet-based, educational tool called The
Logistics Scoreboard that companies can use to pilot their supply chain performance
measurement processes and to customize for ongoing use. The tool and a monograph
(Logistics Performance, Cost, and Value Measures that documents the tool and
its use) are distributed by The Penton Institute (Cleveland, OH). In contrast
to the other approaches discussed, The Logistics Scoreboard is prescriptive
and actually recommends the use of a specific set of supply chain performance
measures. These measures, however, are skewed toward logistics, having limited
focus on measuring the production and procurement activities within a supply
chain.

Activity Based Costing

The Activity-Based Costing (ABC) approach was developed to overcome some of
the shortcomings of traditional accounting methods in tying financial measures
to operational performance. The method involves breaking down activities into
individual tasks or cost drivers, while estimating the resources (i.e., time
and costs) needed for each one. Costs are then allocated based on these cost
drivers rather than on traditional cost-accounting methods, such as allocating
overhead either equally or based on less-relevant cost drivers. This approach
allows one to better assess the true productivity and costs of a supply chain
process. For example, use of the ABC method can allow companies to more accurately
assess the total cost of servicing a specific customer or the cost of marketing
a specific product. ABC analysis does not replace traditional financial accounting,
but provides a better understanding of supply chain performance by looking at
the same numbers in a different way. ABC methods are useful in conjunction with
the measurement approaches already discussed as their use allows one to more
accurately measure supply chain process/task productivity and costs by aligning
the metrics closer to actual labor, material, and equipment usage.

Economic Value-Added

One of the criticisms of traditional accounting is that it focuses on short-term
financial results like profits and revenues, providing little insight into the
success of an enterprise towards generating long-term value to its shareholders
– thus, relatively unrelated to the long-term prosperity of a company. For example,
a company can report many profitable quarters, while simultaneously disenfranchising
its customer base by not applying adequate resources towards product quality
or new product innovation.

To correct this deficiency in traditional methods, some financial analysts
advocate estimating a company’s return on capital or economic value-added. These
are based on the premise that shareholder value is increased when a company
earns more than its cost of capital. One such measure, EVA, developed by Stern,
Stewart & Co., attempts to quantify value created by an enterprise, basing
it on operating profits in excess of capital employed (through debt and equity
financing). Some companies are starting to use measures like EVA within their
executive evaluations. Similarly, these types of metrics can be used to measure
an enterprise’s value-added contributions within a supply chain. However, while
useful for assessing higher-level executive contributions and long-term shareholder
value, economic-value-added metrics are less useful for measuring detailed supply
chain performance. They can be used, however, as the supply chain metrics within
an executive-level performance scorecard, and can be included in the measures
recommended as part of The Logistics Scoreboard approach.

What Advice Can Be Followed When Selecting Measures?

While the approaches described above provide guidance for supply chain measurement,
they provide less help in assessing specific metrics to be used. In this regard,
a key driving principle, as espoused by the Balanced Scorecard, is that measures
should be aligned to strategic objectives. Supply chain strategy, however, differs
for every company and depends upon its current competencies and strategic direction.
Companies, for example, can generally fall into the following developmental
stages that will dictate the types of measures and the degrees to which they
will need to focus:

• Functional Excellence – a stage in which a company needs to develop excellence
  within each of its operating units such as the manufacturing, customer service,
  or logistics departments. Metrics for a company in this stage will need to
  focus on individual functional departments.
• Enterprise-Wide Integration – a stage in which a company needs to develop
  excellence in its cross-functional processes rather than within its individual
  functional departments. Metrics for a company in this stage will need to focus
  on cross-functional processes.
• Extended Enterprise Integration – a stage in which a company needs to develop
  excellence in inter-enterprise processes. Metrics for a company in this stage
  will focus on external and cross-enterprise metrics.

Historically most companies have focused their performance measurement on
achieving functional excellence. With the advent of Supply Chain Management
(SCM) principles aimed at integrating their supply chains, many have objectives
to increase their degree of enterprise-wide integration and extended enterprise
integration. In order to achieve these types of objectives, their performance
measurement systems will need to align to them. Advice for these supply chain
measurement systems falls into five areas that include:

• Function-based measures
• Process-based measures
• Cross-enterprise measures
• Number of measures to be used
• Alignment of executive to management-level measures

A set of measures developed by a leading consumer products manufacturer is
also discussed, providing an illustration of the type that might be selected.

Do Not Focus Only On Function-Based Measures

A major problem encountered with most performance measurement systems is that
they are functionally focused. Within these systems, each functional area measures
its performance in its own terms, with individuals evaluated based on their
ability to meet objectives consistent with their department’s performance measures.
Individuals working under these measurement systems tend to drive operations
toward improving their own area’s performance, frequently at the expense of
the performance of other functional areas. When each functional area sets its
performance measures in isolation from those of others, it often leads to functional
silos and conflicting organizational goals.

Figure 2.0 depicts a typical set of function-based supply chain-related performance
measures used by many manufacturers. These types of measures used in isolation
of each other tend to create conflicting goals among functional areas as follows:

Figure 2.0 Benefits of cross-functional, process-based measures

• Customer Service and Sales – In these functional areas, employees are measured
  by their ability to maintain customer service levels. Measured in this context
  only, these employees tend to drive operations toward satisfying potentially
  smaller-sized customer orders and carrying high levels of finished goods inventories
  by stocking inventories in multiple locations close to customers to shorten
  cycle times
• Logistics – In this functional area, employees are measured by transportation
  and warehousing costs, and inventory levels. Measured in this context only,
  Logistics personnel tend to keep inventories low and batch customer orders
  to ensure that trucks are shipped full and picking operations are minimized.
  On the inbound side, these employees will want to receive full truckloads
  at their warehouse docks to minimize receiving costs, usually at the expense
  of increased inventories.
• Manufacturing – In this functional area, employees are measured in terms
  of manufacturing productivity. Measured in this context only, they want to
  make longer production runs that result in higher levels of finished goods
  inventories. In a make-to-order manufacturing environment there will be a
  tendency to consolidate customer orders into longer production runs, making
  them less responsive to dynamic customer demands.
• Purchasing – In this functional area, employees are typically measured
  by materials costs and supplier delivery performance. Measured in this context
  only, buyers will purchase in large quantities to get volume discounts and
  use more suppliers for each item to ensure a low price. This behavior results
  in purchasing excess, potentially low quality, raw materials.

It is apparent from the behavior described above that use of only function-based
measures could drive employees toward changing functional performance in entirely
different directions. These types of measures alone have reinforced functional
silos, reducing the effectiveness of many supply chains and fostering arms-length
transactions among departments, leading to processes that are slow to respond.
In addition, performance improvement initiatives get focused on a single objective
that frequently runs counter to increasing the efficiency of the total supply
chain. For example, an initiative focused on reducing transportation costs focuses
on filling up outbound trucks. While this seems benign, it may not be best from
a total supply chain perspective when customer orders are held up to fill up
a truck, or if customers are forced to order in greater quantities.

Include Process-Based Measures to Improve Enterprise-Wide Performance

To help integrate their supply chains, companies are starting to break down
the functional silos by organizing around cross-functional processes. This is
done by either creating departments responsible for an overall process or creating
cross-functional teams that drive an overall process, such as:

• Order fulfillment (e.g., order-to-cash)
• New product development/introduction (e.g., concept-to-first sale or production
  batch)
• Total cycle time (e.g., materials purchase to customer payment or cash-to-cash)

To support these organizational changes, companies are supplementing function-based
measures with some process-based performance measures. While this approach does
not advocate the total elimination of function-based measures, it places focus
on the performance of an overall process, using these measures as diagnostic
information to assess what is affecting overall performance.

For example, the perfect order concept measures the percent of customer orders
that are flawlessly fulfilled. This metric is one that measures the effectiveness
of the order fulfillment process, crossing the boundaries of functional departments.
Under this measurement system, a failure during any step in the process or in
any functional department, such as an item shortage on an order line or a wrong
invoice, can result in a failure to meet the overall objective of flawlessly
fulfilling an order. In addition to measuring the overall perfect order process,
diagnostic measures for each task in the fulfillment process would need to be
used.

Figure 3.0 depicts a set of order fulfillment measures based on a perfect
order process concept. It illustrates the hierarchical relationship of process-based
measures with their diagnostic function-based measures. The cross-functional,
process-based measures provide visibility to strategic aspects of supply chain
performance, while the function-based measures are more diagnostic in nature,
useful for pinpointing problem areas.

Figure 3.0 Possible supply chain measures

Include Cross-Enterprise Measures To Improve Extended Enterprise Performance

The cross-functional process approach to measuring supply chains is applicable
for inter- as well as intra-enterprise processes. For example, many would agree
that the two most important bottom-line measures of overall supply chain performance
relate to:

• The availability of the right products at the point of consumption
• The total landed cost to get the products to the point of consumption (including
  all material, manufacturing, transportation, warehousing, and inventorying
  costs along the supply chain)

While these are the penultimate of supply chain measures, it is rare for one
organization to control its whole supply chain’s performance. Supply chains
are typically comprised of many value-adding trading partners that control the
portions in which they transact business. While this might be the case, SCM
principles dictate that significant benefits can accrue when integrated inter-enterprise
processes are in place, to synchronize and optimize the supply chain. These
inter-enterprise processes should also be measured to help ensure that they
are effective.

To ensure the effectiveness of cross-enterprise processes, a company should
measure performance of parts of their supply chain that lie outside their own
enterprise. This leads to the question of “Should you measure what is not within
the domain of your enterprise or what you cannot control?” Some more specific
questions relating this issue are:

• Is a manufacturer responsible for the fact that its products have poor
  availability on the retail shelf?
• Is a shipper responsible for the freight operations of downstream customers
  that pay for their own transportation or pick up products at the shipper’s
  location?
• Is an upstream component parts supplier responsible for the fact that a
  manufacturer’s order could not be produced due to lack of the supplier’s part?
• Is a manufacturer responsible for on-time delivery to the customer after
  it has tendered a shipment to a transportation carrier?

Most people would answer “no” to most of these questions, stating that it
is useless to measure anything on which you have little or no control. In situations,
however, where performance directly or indirectly impacts the availability or
cost of products at the point of consumption, the answer should be “yes” to
all of these questions.

As an illustration, take the case of a leading toy manufacturer’s sales executive
who hired people to visit a sample of some of his customer’s retail stores shortly
after the end of the Christmas holiday season. He had pictures taken of the
shelves to assess the availability of his product following the Christmas rush.
The pictures showed that in many cases the state of the shelves was a mess,
with most items in disarray and most products out of stock – sure to impact
the manufacturer’s post-holiday sales! This executive, who took the position
that his company needed to share some of the responsibility for this, started
initiatives to correct it. He implemented programs that were aimed at working
more closely with customers on joint store-level planning and in-store merchandizing.
The strategy paid off resulting in better product availability on his customer’s
store shelves.

The lesson to be learned from this illustration is that at times it does makes
sense to measure what you cannot control, as you may uncover a deficiency in
your supply chain’s performance. Once found, initiatives can be developed to
address the problem and the performance measures can be used as the “call to
action.” These initiatives usually involve some form of program aimed at taking
some level of control of upstream or downstream supply chain activities – extending
beyond one’s enterprise. Some manufacturers have been implementing SCM programs
to extend their control. These programs and their associated performance measures
include:

• Vendor Managed Inventory (VMI) programs: customer sales, in-stock availability,
  and inventory turns
• Continuous Replenishment Programs (CRP): customer sales, in-stock availability,
  and inventory turns
• Quick Response initiatives: customer sales, in-stock availability, and
  inventory turns
• Forecast-sharing programs: forecast accuracy, order fill rates, and inventory
  turns
• Production scheduling sharing programs: adherence to schedule and order
  cycle time
• Category Management programs: customer category sales and in-stock availability

As more companies implement SCM programs, they will be placing greater emphasis
on cross-enterprise processes, extending beyond their enterprise. This will
lead to the need to implement performance measurement systems that include some
external measures, including some for processes that lie outside of a company’s
domain of control.

Choose A Limited Number of Metrics

A major challenge for many companies when developing a supply chain performance
measurement process is limiting the number of measures. Most companies are involved
in complex business operations that span across multiple business divisions
and geographical boundaries, involving a multitude of sub-processes, tasks,
and organizational departments. Wanting to measure everything, there is a tendency
to measure too much. The number of measures needs to be limited to ensure that
the process is not too cumbersome to administer. One strategy-consulting firm
recommends that their clients limit the number of measures to be tracked in
each area to between three and five, helping to ensure that the measurement
process is not unwieldy.

Align Executive to Management-Level Measures

To ensure that a reasonable number of metrics is defined, an organizing framework
is required to select only those that are most important. A key-enabling concept
taken from the Balanced Scorecard approach is to focus the measurement process
on managing the business, not monitoring and controlling it. Measures should
be aligned to supply chain performance objectives to be achieved, not to whether
employees are adhering to managerial practices and directions. In this way supply
chain performance, not actions, are measured.

To establish a rational set of performance measures, one needs to start with
an understanding of the strategic supply chain objectives of a company’s executive
team. For example, to what degree is the company trying to achieve functional,
enterprise-wide integration and extended enterprise integration excellence?
Once understood, a limited and balanced set of measures that directly aligns
to these strategic objectives needs to be developed. These become the executive
level measures used to provide the executive team with indicators as to whether
or not their supply chain is performing according to their strategic intent.
This set should include a balance of cause and effect type metrics helping executives
determine when a particular process area needs to be improved.

In addition to executives, management personnel also need performance measures
to help ensure that their supply chain activities are performing well. These
measures will be more detailed, tracking both tactical and operational types
of activities. To ensure that the executive and management teams are not driving
the organization in different directions, the management-level measures need
to be aligned with the executive-level measures. Figure 4.0 graphically depicts
the relationship and contrasting nature of executive and managerial measures.
Using the lower level measures, managers can gauge how well they are doing relative
to the overall strategic goals set in place by the executive team. In addition,
the lower-level metrics enable executives to drill-down into the more diagnostic
metrics, detecting where corrective actions are needed.

Figure 4.0 Perfect order process task-related measures

An Illustrative Set of Supply Chain Performance Measures

A number of leading-edge companies are beginning to implement supply chain
performance measurement systems, some calling them scorecards, while others
call them dashboards or cockpits.

While the metrics shown are largely executive-level in nature, this company
has plans to break out each of the metrics into more detailed managerial levels.
This set of measures provides a good illustration of a balanced set of measures
that might be selected to support a manufacturer’s supply chain performance
measurement process.

What Methods Are Available For Setting Performance Targets?

An important issue in performance measurement is how a company can use measures
to gauge its supply chain’s performance. To do this effectively, a target for
each measure needs to be established, providing the framework for determining
the answer to three questions that arise when evaluating a performance metric:

• Has the metric improved from the last time it was reviewed?
• By how much?
• How close is the metric to where it should be?

In order for this evaluation to be meaningful, however, the direction of improvement
needs to be established. Should the metric have gone up or gone down? Frequently,
in looking at productivity-related metrics an increase represents an improvement;
similarly, for cost-related metrics a decrease represents an improvement. This
is not always the case! For example, an increase in manufacturing productivity
and a decrease in cost would normally be considered an improvement. It would
not be an improvement if it caused degradation in customer service performance.

In a way similar to picking a set of balanced metrics, performance targets
need to be jointly, not individually, developed. To achieve objectives some
metrics may need to increase and others may need to decrease. Each metric in
the set has to be viewed in conjunction with the others to determine its proper
target. For example, in a situation where a company is trying to achieve same
day delivery, delivery times should decrease, while warehouse handling and transportation
costs might actually increase.

Thus, while there a variety of ways in which to set performance targets, they
should always be jointly set in the context of strategic objectives. Generally,
there are four methods that can be used to set performance targets, described
in detail below:

• Historically based targets
• External benchmarks
• Internal benchmarks
• Theoretical targets

Historically Based Targets

Historically based target setting is the most frequently used among all the
methods. In using this method, performance targets are based on historical baseline
levels. For example, a company having an historical order fill rate of 90% might
set a performance target at 95%, trying to improve by five percentage points.
This method is the most frequently used because it is the easiest to implement.
Once the baseline metrics are established, the same procedures and systems that
were used to establish the baseline numbers can also be used on an ongoing basis
to measure changes in the metrics.

External Benchmarks

Using external benchmarks to help set performance targets is currently popular,
but difficult to use in practice. In general, benchmarking has been in the business
limelight for almost ten years, with companies looking outside their operations
for best practices and performance comparisons. This method relies on collecting
information on performance metrics of companies internal and external to one’s
industry. A few organizations have collected some benchmarking data in a formal
way, including:

• Herbert W. Davis and Company (Fort Lee, NJ) is a small consulting firm
  that has been conducting logistics cost and service surveys for over 20 years,
  including its most recent survey that is comprised of information on around
  300 North American manufacturer, distributor, and retailer companies. (Davis
  uses the results to support its consulting business.)
• Pittiglio, Rabin, Todd & McGrath (PTRM) (Weston, MA) conducts a supply
  chain a benchmarking study, its Integrated Supply-Chain Benchmarking Study.
  The last published study, fourth in the series, included information on about
  225 worldwide participants. (The company uses the results to support its consulting
  business.)
• The Demand Activated Manufacturing Architecture (DAMA) project, a part
  of the American Textile Partnership, has collected benchmark data on the performance
  companies, mostly in the ‘soft goods’ industries. This data was obtained from
  Kurt Salmon Associates, Inc. (Atlanta, GA) and The Garr Consulting Group,
  a division of Deloitte & Touche, Inc. (New York, NY), and The Logistics
  Institute of the Georgia Institute of Technology.

Once external benchmarking metrics are collected, a company’s internal metrics
are generated and a gap analysis is done – typically looking at the best-in-class
within their own industry as well as external to it. This is followed by more
analysis to assess the degree to which the company can achieve these performance
levels, including what business practice changes are necessary to close the
gaps.

While appealing, the external benchmarking method has a major shortcoming
to it as to which set of companies are comparable. A substantial amount of analysis
is required to ensure that external benchmarks are meaningful, especially when
using data from companies that operate within different business environments
(e.g., differing products or sales channels). This makes the use of external
benchmarks difficult, since comparable external benchmarks may not be available
or too controversial. On the other hand, external benchmarks, especially from
one’s competitors, may be extremely important towards keeping an organization’s
supply chain competitive.

Internal Benchmarks

Performance target setting using internal benchmarks is a common approach,
since it requires only internal measures. Within this method, comparable functional
departments, processes, and facilities within a company are measured in the
same way. For example, there may be a set of metrics in use for all warehousing
facilities, another set for all manufacturing plants, and another set for all
customer service departments. Similar to the external benchmarking approach,
“best-in-class” functional organizations are identified and their benchmark
metrics are used as the basis for establishing performance targets for other
functional organizations.

In contrast to external benchmarking, internal benchmarking data is easier
to collect. The method is less controversial when comparing business operations
since internal organizations usually operate in similar business environments.
While this internal benchmarking method is easier to implement, it too has some
serious drawbacks to it. The major one involves stretching the organization
to achieve better performance. That is, using a “best-in-class” internal organization
to set targets may limit the company’s performance relative to its competitors.

Theoretical Targets

The use of theoretical target setting is a relatively new method advocated
by some consultants. Under this method a company conducts an analysis to theoretically
determine how its supply chain performance could be improved. It would then
implement the business changes necessary to achieve these improvements and put
a set of performance targets in place based on estimates made during the analysis.

In particular, one consultant, Paul Bender of Bender Consulting/SynQuest,
Inc. (Atlanta, GA), advocates the use of supply chain optimization to help set
theoretical performance targets. Using his approach, a company would first undergo
an analysis to determine how it should optimize supply chain performance. It
would then use the estimates made during the analysis to set its performance
targets. For example, a company might determine that in order for it to maximize
its long-term profits, it should increase on-time order due-date performance,
while increasing its manufacturing costs and decreasing its air freight charges.
The company would then use the results of the analysis to increase its performance
targets for manufacturing costs and on-time order due-date performance, while
appropriately decreasing the target on its airfreight charges.

While conducting an optimization analysis is an intuitively appealing method
for determining performance targets, it is not always the easiest to do. Another
alternate approach involves the use of supply chain simulation analysis that
includes conducting what-if studies on initiatives to improve performance. The
results of these studies could then be used to set theoretical targets. For
example, a “what-if” study might be conducted to assess inventory reductions
that might accrue from statistical safety stock setting. The study’s estimated
reductions would be used to reset performance targets for inventory turns.

Setting performance targets on a theoretical basis is most useful for insuring
that a balanced set of metrics is developed. Often, only by doing a thorough
analysis can one assess how an initiative would impact various aspects within
a supply chain.

In practice, a combination of the four performance target-setting methods
described above should be used. No one method is practical for determining targets
since one cannot always get a full set of comparable benchmarking information
or conduct the extensive analyses needed to develop a full set of theoretical
performance targets.

What Are Application Vendors Doing To Support Supply
Chain Performance Measurement?

Application vendors are faced with a challenge when trying to provide supply
chain performance measurement functionality within their software products.
Users often wish to include metrics relating to information not residing within
the vendor’s application database. (Figure 5.0 shows the potential sources from
which a supply chain performance measurement system may have to draw data).
This is especially the case when measuring the performance of cross-functional
and inter-enterprise processes, which involve drawing information about any
functional department within a company, or about customer/supplier activities.
Also, most advanced planning and scheduling (APS) applications focus on the
future, rarely concerned with what went on in the past (except relative to measuring
forecast errors). While Enterprise Resource Planning (ERP) vendors that offer
SCM functionality have more of the necessary data within their product suites,
they have not focused on supply chain-related performance measurement. Until
recently they have focused most of their historical reporting functionality
on providing transactional auditing, tracing, and tracking.

Figure 5.0 Balanced metrics developed by a CPG manufacturer

Given the interest shown in supply chain performance measurement, substantially
driven by business consultants, vendors have started to consider supplementing
their product suites by offering enhanced supply chain performance reporting
capabilities.

Traditional SCM Vendors Provide Reporting On A Limited Number of Metrics

Traditionally SCM application vendors have focused their development efforts
on enabling planning, scheduling, and execution, targeted toward supporting
decision-making, not tracking historical performance. Some SCM vendors have
functionality to report historical supply chain performance focused around either
functional or planning-related metrics. For example, SynQuest provides standard
reporting on a variety of metrics such as forecast error, inventory turns, and
order completeness, while also allowing users to define related metrics within
each of its function-based modules. Similarly, Manugistics (Rockville, MD) provides
standard reporting functionality on forecasting performance and on-time delivery,
as well as some general purpose capabilities to report on user-defined metrics
such as order fulfillment, factory floor conformance, warehouse space utilization,
and promotional effectiveness.

In contrast to these two vendors, i2 Technologies (Irving, TX) provides a
solution, RHYTHM Reporter, which is enabled by incorporating OLAP technology
from Arbor Software’s (Ann Arbor, MI) Essbase product. The solution allows users
to create reports on information contained within the Supply Chain Planner application
and includes a variety of standard reports focused on master planning, profit
optimization, and demand fulfillment. Similar to i2, Logility (Atlanta, GA)
provides some reporting capability by incorporating OLAP technology from Cognos
(Ottawa, Canada). Logility provides standard interfaces into Cognos’ solution
to support users doing customized analysis. Logility also provides standard
reporting capabilities within its functional modules that can report on a variety
of measures such as forecast error, assembly line utilization, late orders,
and warehouse receipts.

Similar to i2 and Logility, AMR Research expects that other SCM vendors will
start to provide OLAP-based solutions to allow their customers to track additional
supply chain performance measures. For example, webPLAN (Ottawa, Canada), formerly
called Enterprise Planning Solutions, is developing a product, onPLAN, which
will provide users with a KPI business report card. The product, which was released
in 1998, imbeds OLAP technology from InterNetivity, Inc. (Ottawa, Canada) that
enables numeric and graphical reporting, and the drilling down and slicing and
dicing of data.

There are two new noteworthy SCM vendors that focus exclusively on performance
measurement. One is VIT (Palo Alto, CA), which offers the SeeChain software
application suite that consists of five modules: Demand Accuracy, Raw Materials,
Production Performance, Finished Goods, and Fulfillment. The vendor has a unique
graphical user interface that allows users to drill down from higher, aggregated
measures to lower detailed measures, enabling them to easily diagnosis non-performing
supply chain elements. This vendor comes the closest to offering an application
that enables users to view their overall supply chain, limited only by data
that can be drawn by a company into a single database structure the company
provides. Another SCM vendor that provides unique performance measurement functionality
is Maxager Technology (San Rafael, CA). The company specializes in an application
that reports on historical product-level profitability and production performance.
The vendor provides users with shop floor data collection applications and uses
constraint-based costing methods to assess product level production performance.
Similar to other SCM solutions, the application provides refined estimates of
manufacturing performance representing only a portion of the overall supply
chain performance measures needed.

Figure 6.0 Function-based measurements and related goals

ERP Vendors Offering SCM Functionality Are Starting to Address Supply Chain
Performance Measurement

Some ERP vendors also offering SCM functionality have put more focus on historical
performance measurement than the SCM vendors; having specialized in providing
solutions focused more on transactional history. While these vendors have built
some OLAP-based performance reporting functionality into their product suites,
they are still a long way from fully meeting the needs of supply chain performance
measurement.

SAP (Walldorf, Germany) has been offering a logistics-related performance-reporting
product, its Logistics Information System (LIS). This R/3-based application
allows a user to assemble information from a range of R/3 modules and provides
functionality to analyze the data in a number of ways, including in tables and
graphs. As part of its new Advanced Planner and Optimizer (APO) product initiative,
the company is planning to provide users with functionality that measures a
set of pre-defined supply chain KPIs within its Supply Chain Cockpit product.
The company has also recently laid out ambitious plans to develop more general-purpose
functionality under its Business Intelligence initiative, announced at SAPPHIRE
’98. In addition, SAP recently purchased a minority share in ABC Technologies,
Inc. (Beaverton, OR), a leading player in activity-based costing and management
solutions. ABC Technologies will integrate its functionality into R/3 to enable
general-purpose performance measurement based on the ABC approach. This functionality
will also be applicable to measuring supply chain performance.

Oracle (Redwood Shores, CA), a database vendor that has always provided some
performance measurement functionality, has plans to expand it. Similar to SAP,
it will offer ABC costing functionality, having recently purchased PriceWaterhouse-Coopers
LLP’s (New York, NY) activity-based costing, budgeting, and management software,
ACTIVA. While more general purpose in nature, some of the ABC functionality
should be applicable to measuring supply chain performance.

PeopleSoft (Pleasanton, CA) is also adding general-purpose performance measurement
that will be applicable to supply chain measurement. The vendor is currently
working with on an initiative to provide a solution based on the Balanced Scorecard
and ABC approaches.

How Should A Company Get Started?

Based on our research, we have concluded that there is no one recommended
approach or set of measures to be used to measure one’s supply chain performance.
While espousing the importance of measuring supply chain performance, leading
consultants have no definitive set of metrics to recommend. All agree, however,
that approaches such as the Balanced Scorecard, the SCOR Model, The Logistics
Scoreboard, and others discussed herein, provide excellent guidance when developing
a supply chain performance measurement system. In addition, although the software
vendors we polled enable a limited range of supply chain performance measures,
they are improving and planning to add more functionality to their product sets.
Over time, we expect vendors to offer more complete packaged applications for
supply chain performance measurement.

All this should not dissuade users, however, from starting to measure their
supply chain’s performance in the context of assessing the success of initiatives
aimed at achieving strategic objectives. It is too important! With substantial
resources – dollars and people – being applied towards implementing various
supply chain programs, users should measure performance to insure desired change
happens.

For those users just getting started, implementing supply chain performance
measurement should not be done all at once. For example, one could start by
first implementing executive-level scorecard measures in a manual fashion. This
could then be followed up with more automation, through the use of database
tools and the addition of managerial-level metrics. As application vendors develop
more capabilities in the area of performance measurement further automation
of the process can be implemented over time. Based on the research done for
this report, we would recommend the following steps be taken when implementing
supply chain performance improvement and measurement:

• Have executives articulate the strategic supply chain vision and company
  objectives, including the degree of focus to be placed on achieving functional,
  enterprise-wide integration and extended enterprise integration excellence.
  For example, the functional excellence portion of the vision might be that
  “we will reduce our manufacturing costs over the next two years” and related
  objectives would be to reduce manufacturing operating and material costs.
• Define executive level measures for each objective for their scorecard.
  The total number of measures used should be limited to up to 20 or so. For
  example, these might be metrics such as material cost per pound purchased
  and operating manufacturing cost per unit produced.
• Establish managerial level objectives and measures that align to the executive
  level ones. These should be more tactical and operational, providing diagnostic
  information on whether executive objectives are being met. Breaking down the
  higher-level measures typically does this. For example, these might be measures
  for a particular plant’s cost per ton purchased for a specific class of material.
• Identify supply chain initiatives that specifically address the executive
  and managerial performance improvement objectives. For example, this might
  include a core supplier program reducing the number of material suppliers
  to ones with the lowest cost, meeting quality standards.
• Establish targets for all metrics defined, using a combination of historical
  performance, external/internal benchmarks, and theoretical estimates (often
  obtained from operational quantitative analysis of the supply chain initiatives).
  A timeline for achieving the targets needs to be established for each metric,
  consistent with the schedules developed for the supply chain initiatives.
  For example, while a program might be expected to ultimately reduce material
  costs by 3%, targets for its first year of implementation might be only 1%,
  with an additional 2% improvement the next year.
• Implement new initiatives in concert with a formal measurement system to
  keep track of performance improvement over time, using a combination of whatever
  technology makes sense; be it based on spreadsheets, database products or
  a vendor’s suite of packaged applications.

While these steps are useful for getting started, ongoing supply chain performance
measurement requires that the steps be revisited on a routine basis, as objectives
change and new programs and initiatives are undertaken. Keeping the measurement
process aligned to supply chain objectives and activities will provide the information
needed to drive your supply chain’s performance, helping to ensure that resources
are appropriately applied and desired strategic change is happening.