250 • Supply Chain Risk Management: An Emerging Discipline
QUANTIFIED RISK INDEXES
is approach scores risk events or suppliers using algorithms that model
risk. Risk indexes are quantitative models that consider multiple factors
to arrive at a single risk indicator score, similar to the bridge suciency
rating discussed earlier. ese indexes consider more than simply a sup-
plier’s nancial status when arriving at a risk score. To date most risk
indexes or indicators have been internally developed since third parties
have been slow to respond with new tools and approaches, something that
will likely change.
It is no surprise that risk indexes range from basic to complex. On the
basic side, some companies use a variety of simple algorithms to arrive
at a single risk score or index. A company might consider the probability
of a potential risk event occurring using a 1–10 scale. at gure is then
multiplied by the eect a risk event would have on a 1–10 scale. Other fac-
tors can be introduced, such as the ability to detect the risk on a scale of
1–10 (a higher score means less ability to detect the risk). e three scores
would be multiplied together to arrive at an overall risk index, with higher
scores representing higher risk exposure. A later example will further
highlight this approach. A concern about the validity of these simple risk
indexes measures is particularly relevant here.
It is not unusual to translate risk index gures into a red/ yellow/ green
visual system (recall from Chapter6 this method is used with the Z- Score).
is should also happen during new product development as specic risks
are identied, something that Chapter4 addressed. Product development
teams will work to address red and even yellow items prior to product
launch. e goal should be to launch new products with as few red risks as
possible. Why wait until aer the fact?
Interestingly, a body of research is emerging that counters the notion
that complex algorithms and models are automatically more eective than
simple rules of thumb or guidelines when making organizational deci-
sions. is is something to keep in mind when developing risk indexes. A
hypothesis put forward is that complex situations create so many possible
courses of action and become so complex to use that individuals become
confounded, oen to the point where they delay decisions, default to
the safest option, or avoid making choices altogether.
Research suggests that simple rules can equal, and at times exceed, the
eectiveness of more complicated analyses across a range of decision areas.
2
Risk Measurement • 251
While the analysis and data that lead to the rules may be sophisticated, and
at times will even be complex, the rules that result should be elegant in their
simplicity as they provide guidance to users. When the up- front work to
develop the risk index is rigorous, the chances are good the risk index will
have validity. is is something to keep in mind when the temptation exists
to develop overly complex algorithms, models, and risk indexes.
A Risk Index Example
Consider the following example, which is a technique used by a food man-
ufacturer to develop risk indexes to support its risk management eorts.
3
is company examines specic risks from three dimensions: severity of
the risk, the probability of the risk occurring, and the probability of early
risk detection. is approach is consistent with the FMEA (failure modes
and eects analysis) approach, a widely used quality management tech-
nique that considers these three factors. Let’s look at an example:
Risk: Poor Product Freshness
Severity (1 low–10 high) 7
Probability of Occurrence (1 low–10 high) 5
Probability of Early Detection(1 high–10 low) 3
Risk Index (7×5×3) 105
Companies that use this approach identify, evaluate, and then rate all
possible risks, which are then addressed in terms of priority. As men-
tioned, with any risk index we must be concerned about its validity. Here,
what denes the incremental values in each scale (i.e., low to high is a
broad range with a great deal in between)? Is the incremental dierence
between a score of 3 and 4 in a scale the same as between 5 and 6 or 8 and 9?
Should the three categories be equally weighted, which is the case here?
Would three people looking independently at the same risk arrive at the
same or similar score using this tool? (is is called inter-rater reliability).
Too many companies use tools such as the one presented here without
fully performing the up- front work necessary to validate the tool. e vir-
tue of this approach is its simplicity. It is easy to understand and use.
Country Risk Indexes
A variety of risk indexes are available that evaluate country risk. One of
the more comprehensive of these indexes, and one that will be featured
252 • Supply Chain Risk Management: An Emerging Discipline
here, is the International Country Risk Guide (ICRG) composite risk rat-
ing produced by the Political Risk Services (PRS) Group.
4
is composite
risk rating by country, which is updated monthly, includes 22 variables
across three subcategories of country risk—political, nancial, and eco-
nomic. A separate detailed scoring and weighting methodology is created
for each of the 22 subcategories.
e composite risk rating by country is based on 100 total points with
the political risk rating comprising 50% and the nancial and economic
risk ratings contributing 25% each. e composite scores, ranging from
zero to 100, are then broken into categories from very low risk (80 to
100 points) to very high risk (0 to 49.9 points).
At the enterprise risk management (ERM) level, country risk index rat-
ings are valuable when a company is thinking about making foreign direct
investments. At the SCRM level, this information can inuence logistics,
sourcing, and selling activities. Our advice is to not even bother creating
country risk indexes. ird- party indexes are available that are more com-
prehensive than anything a company could construct acting on its own.
USING TOTAL COST MEASURES TO MANAGE RISK
Total cost is a topic that companies cannot ignore as they search for new
and better ways to manage supply chain risk. What exactly is total cost?
Total cost includes the expected and unexpected elements that increase
the unit cost of a good, service, or piece of equipment. e logic behind the
development of total cost systems is that unit cost or price never equals
total cost. And, as the gap between unit cost and total cost becomes pro-
gressively larger, so does a company’s risk exposure. Regardless of where
a company applies total cost models, these models all attempt to capture
data beyond unit price.
In uncertain times, the need to understand every element of cost has
never been greater. Total cost systems help management to identify the
impact of dierent cost elements, to track cost improvements in real terms
over time, and to gain management’s attention regarding where cost reduc-
tion eorts will have their greatest payback across the supply chain. While
there is some overlap between sources regarding which cost elements to
include in total cost models, no agreement exists regarding exactly what
these models should contain. is issue becomes more complex once we
Risk Measurement • 253
understand that, like forecasting models, dierent types of cost models
exist. And, like forecasting models, total cost models almost always have
some inherent inaccuracy to them.
Types of Total Cost Models
Every total cost model is part of a family of measurement systems called
cost- based systems. Like any measurement system, cost- based systems oer
advantages and disadvantages. ese systems can be extremely challenging
to develop and use (if they were not challenging then everyone would rou-
tinely use them).
5
Some of the challenges include (1) relying on data that are
derived across global supply chains, (2) maintaining the discipline across
the supply chain to use the models routinely, and (3) using data that are not
known with certainty or are estimates of what might happen. A mean time
between failures (MTBF) estimate, for example, is oen used when evalu-
ating the total expected cost of capital equipment. How close is the estimate
to what actually happens in terms of equipment reliability?
Across a supply chain we generally see three kinds of total cost models
in use, regardless of whether this involves a domestic or international sup-
ply chain. ese include total landed cost models, supplier performance
index models, and life- cycle cost models.
Total Landed Cost Model. A total landed cost model is used when
evaluating suppliers prior to making purchase decisions, although that is
not the only time when a landed cost model should be used. Total landed
cost is the sum of all costs associated with obtaining a product, including
acquisition planning; unit price; inbound cost of freight, duty, and taxes;
inspection; and material handling for storage and retrieval.
6
Each of these
cost categories will also contain numerous subcategories. Best- practice
companies require their commodity teams or buyers to attach spread-
sheets that show the total landed cost of a purchase requirement whenever
they propose a supply strategy or make a supplier selection decision.
Total landed cost models should also be used when doing business with
suppliers on an ongoing basis. e factors that aect the sourcing decision
in the rst place are dynamic and subject to change (think transportation
and exchange rates here). Furthermore, actual costs should replace any
estimated or forecasted costs in the model as they become available. is
helps to validate the assumptions in the model.
When developing total landed cost models it is best to start with the
unit price and then build up the total cost as goods move from origin to
254 • Supply Chain Risk Management: An Emerging Discipline
destination. Ideally every cost element is presented in the same unit of
measure. If a product is priced by the pound, then every corresponding
cost element in the model should appear as a cost per pound. e cost ele-
ments in landed cost models should be divided into categories that reect
a logical progression through the supply chain:
• Unit price—unit price usually appears on the rst line of the cost model
• Within country of manufacture costs—includes materials, storage,
labor, quality, overhead, obsolescence, packaging, risk or disruption,
exchange rates, inventory carrying charges
• In- transit to country of sale costs—includes transportation charges,
fuel surcharges, insurance, port charges, handling, security, bank-
ing fees, broker fees, potential detention charges, duties, handling
agency charges, inventory carrying charges
• Within country of sale costs—includes local transportation and han-
dling, storage fees, taxes, safety stock, inventory carrying charges,
yield, productivity implications, maintenance, quality, overhead
allocation, payment terms
Supplier Performance Index (SPI) Model. Various models attempt to
capture the true cost of doing business with a supplier on a continuous
basis. Perhaps the best known of these models is something called the
Supplier Performance Index (SPI). SPI calculations, which focus largely
on supplier nonconformance costs, are helpful when tracking supplier
improvement over time, quantifying the severity of performance prob-
lems, deciding which suppliers to eliminate from a supply base, and when
establishing minimum acceptable levels of supplier performance.
e SPI is a total cost model that presents its output in the form of an
index or ratio. It assumes that any quality or other infraction committed
by a supplier during the course of business increases the total cost (and
hence the total cost performance ratio) of doing business with that sup-
plier. is approach is more applicable aer supplier selection because it
is populated with cost occurrences that have happened rather than are
expected to happen. e SPI calculation for a specic period is a straight-
forward formula:
SPI = (Cost of material + Nonconformance costs)/(Cost of material)
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.