Content uploaded by Yvonne Badulescu
Author content
All content in this area was uploaded by Yvonne Badulescu on Jul 06, 2021
Content may be subject to copyright.
LGPP  EPFL
Integrating financial
aspects into, and
performing a risk
analysis on, a supply
chain model
Masters Project 2010
Yvonne Badulescu
Supervisors: Prof. Remy Glardon and Dr MinJung Yoo
ii
SYNOPSIS
The rapid growth of multinational corporations over the past few years has created more
efficient and cost effective supply chains to counteract against global competition. By
increasing responsiveness and leanness of the supply chain the company becomes more
vulnerable and exposed to risks. It is therefore essential to analyse the risks associated with
time and cost and to develop mitigating strategies against exposures.
Through computational modelling it has been possible to simulate the performance of a
supply chain over a period of time. One such model is being developed by LGPP at EPFL in
the context of the project ‘SCOPE’ that extensively simulates the material and information
within a supply chain.
The main objective of this Masters project is to develop a risk analysis method that is
compatible with, but not limited by, the SCOPE simulation platform. In order to be able to
provide a complete and thorough risk analysis methodology, a procedure must be
developed that integrates the most significant financial aspects into the supply chain
simulation.
Therefore the first chapter of this project is dedicated to the identification, development
and integration of the most significant financial aspects into the SCOPE simulation platform.
The procedures outlined in the first chapter allow a comprehensive risk analysis to be
performed within the second chapter of this project.
Chapter 2 develops a methodology to conduct a risk analysis on a supply chain using the
procedure for financial integration from Chapter 1 as a performance indicator. A case study
is analysed using the SCOPE simulation platform.
iii
DECLERATION
The work presented in this thesis has not been submitted, in full or in part, for another
degree at this or any other institution. The contribution of others to the content of this
thesis and all previously published material has been acknowledged.
Yvonne Badulescu
iv
ACKNOWLEDGEMENTS
I would like to thank my supervisors Dr. MinJung Yoo and Prof. Remy Glardon from the
LPGG for their guidance, enthusiasm and support throughout the development of this
project.
I would also like to thank Dr MinJung Yoo as well as Olivier Gobet for their help with the
SCOPE simulation platform.
Finally I would like to thank my family, partner and friends for supporting and encouraging
me throughout my entire university life.
v
CONTENTS
List of Figures and Tables 1
Introduction 4
Chapter 1: Supply Chain Financial Components 5
1.1. Introduction 5
1.2. Financial Statements 8
1.2.1. Balance Sheet 8
1.2.2. Profit and Loss (P&L) Statement 8
1.2.3. Statement of Retained Earnings 8
1.2.4. Cash Flow statement 9
1.3. Introduction to the terminology 10
1.3.1. Gross Profit 10
1.3.2. Net Income 10
1.3.3. Operating Profit (EBIT) 10
1.3.4. EBITDA (nonGAAP) 11
1.3.5. Net Profit 11
1.4. Factors affecting Net Profit 13
1.4.1. Exchange rates 13
1.4.2. Corporate income tax rates 13
1.4.3. Transfer Pricing 13
1.5. Net Profit Calculation Procedure 15
1.5.1. Transfer Price 15
1.5.2. Detailed Procedure of calculating net profit using the simulation output 16
1.6. Validation 18
1.6.1. Scenario #1: 18
1.6.2. Scenario #2: 19
1.6.3. Scenario #3: 22
1.6.4. Results of each scenario 25
Chapter 2: Supply Chain Risk 27
2.1. Introduction 27
2.2. Literature review 29
2.2.1. Definition of risk 29
2.2.2. Supply Chain Risk Management 31
vi
2.2.3. Risk Types 31
2.2.4. Generic SCRM procedure 36
2.2.5. Example of SCRM tool used in industry 36
2.2.6. Summary of literature review 37
2.3. Methodology 39
2.4. Case Study – ABC 45
2.4.1. Step 1: Mapping ABC supply chain 45
2.4.1.1. Production centre locations 46
2.4.1.2. Suppliers 46
2.4.2. Step 2: Identifying the major risks. 47
2.4.3. Step 3: Analysis through sensitivity analysis and finding probabilities. 48
2.4.3.1. Procedure to find probability of exchange rates 51
2.4.3.2. Methodology to determine the probability of tax rates 53
2.4.3.3. Methodology to find probability of supply time and transportation
time 54
2.4.3.4. Procedure chosen to determine the probability of supply and
transportation leadtimes 55
2.4.4. Step 4: Evaluating the risk 55
2.5. Evaluating the Results 59
2.5.1. Conclusion of results 61
Conclusion 63
References 64
Appendix A 65
1
LIST OF FIGURES AND TABLES
FIGURES
Figure 1: SCOPE Simulation Platform
Figure 2: Bill of Material common to all products analysed in this project
Figure 3: Simple way to understand how to set transfer price [1]
Figure 4: Configuration of a supply chain with only one site
Figure 5: BOM of a finished product. The circled items are subassembled in China.
Figure 6: Configuration of a supply chain using two production sites.
Figure 7: Configuration of a supply chain with three production sites
Figure 8: BOM of a finished product. The circles items are subassembled in China and
the squared items are subassembled in Italy.
Figure 9: Key focus areas within logistics and SCM [3]
Figure 10: Risk Map/Matrix [3]
Figure 11: Risk types within a supply chain according to [18]
Figure 12: Risk Sources in Supply Chain  adapted from [18]
Figure 13: Risk in the extended supply chain (adapted from [5])
Figure 14: Risk analysis procedure
Figure 15: Ericsson risk management evaluation tool [4]
Figure 16: Templates for risk assessment and treatment and contingency planning [4]
Figure 17: Methodology to identify most significant risks
Figure 18: Production strategy used in the simulation for the production of all finished
products
Figure 19: The bill of material (BOM) for all products in simulation. Operations
conducted at CHINA are circled.
Figure 20: Sensitivity to net profit
Figure 21: Interactions between factors
Figure 22: Sensitivity to delivery delay
Figure 23: Normal distribution. Coloured parts represent the cumulative distribution
2
Figure 24: Total net profit impact
Figure 25: Total delivery delay impact
Figure 26: Risk matrix based on net profit
Figure 27: Risk matrix based on delivery delay
Figure 28: Risk matrix altered from [15]
Figure 29: Net Profit risk matrix divided into four sections
Figure 30: Delivery delay risk matrix divided into four sections
Figure 31: Prototypes of supply risk utility functions taken from [22]
TABLES
Table 1: Example of a Profit and Loss statement
Table 2: Transfer Pricing Method
Table 3: Supplier locations for scenario 1
Table 4: Supplier locations for scenario 2
Table 5: Supplier locations for scenario 3
Table 6: Net Profit and Transfer Prices of Scenario 1, 2 and 3
Table 7: Types of risk, sources and descriptions [14]
Table 8: Summary of risks [5]
Table 9: Risk effect table to identify risk factors
Table 10: Supplier locations
Table 11: Major risk events and associating risk factors
Table 12: Input factors with minimum and maximum values
Table 13: Design of experiment for sensitivity analysis
Table 14: Exchange rate statistical data from 20032009
Table 15: Exchange rate statistical data from 20072009
Table 16: Cumulative probability of exchange rate minimum and maximum
Table 17: Tax rates from 20032009 for China and Switzerland in percentages
3
Table 18: Probabilities of supply leadtime and transportation time
Table 19: Factors with minimum and maximum probabilities of occurrence
Table 20: Best and worst case scenario of total net profit impact
Table 21: Best and worst case scenarios of total delay impact
Table 22: Configurations of factors for each scenario
EQUATIONS
Equation 1: Harland et al.'s risk calculation
Equation 2: Risk/benefit model [3]
Equation 3: Risk equation used in risk analysis
Equation 4: Factorial design equation
Equation 5: Matrix equation
Equation 6: Matrix manipulation to retrieve vector A
Equation 7: Total impact equations
4
INTRODUCTION
Over the past few decades supply chains have become leaner and faster by reducing costs
and time and increasing quality. This has led to much interest being invested in gaining the
competitive advantage in one’s supply chain with the aim of being more efficient and cost
effective than competitors.
Through computational modelling it has been possible to simulate the performance of a
supply chain over a period of time. One such model is being developed by EPFL in the
context of the project ‘SCOPE’. At this stage the platform has been developed at an
extensive level to include factors such as forecasting, production and procurement planning,
inventory, manufacturing and transportation. Material, information and some financial
flows have been completely integrated into the platform. The platform provides a broad
range of results concerning the material and financial flow in the supply chain however the
financial information is quite elementary.
Leaner supply chain designs have given way to higher uncertainties due to outsourcing,
international production sites and decreasing the amount of suppliers. [4] These
uncertainties are risks that the supply chain must withhold therefore it is integral that these
risks are identified, assessed and managed in order to ensure a robust and resilient supply
chain.
The main objective of this project is to develop a risk analysis method that is compatible
with, but not limited by, the SCOPE simulation platform. One of the major methods in which
a company can gain a competitive advantage is by increasing its profits and its customer
satisfaction. It is hence very important to integrate a comprehensive calculation of net profit
into the simulation in order to financially analyse the supply chain’s comportment to risk.
In order to be able to provide a complete and thorough risk analysis methodology, a
procedure must be developed that integrates the most significant financial aspects into the
supply chain simulation. As the simulation platform is in full flourish, it was not possible to
actually integrate any code to perform the financial calculations automatically therefore all
calculations were executed using the simulation’s outputs and can therefore be used even if
the platform is significantly modified. Hence the first chapter of this project is dedicated to
the identification, development and integration of the most significant financial aspects.
The first chapter allows a comprehensive risk analysis to be performed within the second
chapter of this project. Chapter 2 develops a methodology to conduct a risk analysis on a
supply chain using the financial aspects defined in Chapter 1 as performance indicators.
5
CHAPTER 1: SUPPLY CHAIN FINANCIAL COMPONENTS
1.1 Introduction
The objective of a supply chain is to maximise its profits and ensure the most effective and
efficient transportation of goods and services from suppliers (upstream) to customers
(downstream). [5] Chapter 1 provides a detailed background of a company’s financial
statements and their relationship to key financial performance indicators that measure the
company’s financial standing.
The financial aspect found most significant in measuring the company’s profitability is net
profit. Net profit can also be easily calculated using the output data from the SCOPE
simulation.
The SCOPE project that has been developed by LGPP at EPFL is a simulation platform that
models a company’s valueadding network (VAN). (Figure 1) The model works by accepting a
number on inputs that determine how many production sites the supply chain has, supplier
locations, bill of material, component, transportation and production costs, as well as the
number and origin of customer orders. This input (in green) is taken into account in the
simulation and an output file is produced containing information concerning transportation,
capacity, inventory, components costs and production.
Figure 1: SCOPE Simulation Platform
The SCOPE simulation will be used to analyse a realistic context of a largescale
manufacturing company. Due to confidentiality issues, the company will be named ABC for
the purpose of this report and the presented output values are slightly scaled. However all
6
other aspects required to model this company are kept the same in order to represent
reality as closely as possible.
ABC is a manufacturing company with its headquarters based in Switzerland. There will be
various configurations tested within this project which determine the number of production
sites in ABC and their international locations. It is assumed that all orders sent to ABC are
partly, if not wholly, produced in Switzerland and shipped to the customers from the Swiss
site. In some cases, certain parts of the finished product may be produced at a site other
than Switzerland. This information is retrieved by looking at the bill of material (BOM).
(Figure 2)
Figure 2: Bill of Material common to all products analysed in this project
7
Five finished products that all have the same bill of material will be analysed in this project.
These are simply named as Product 1, Product 2, Product 3, Product 4 and Product 5. They
differ in number of orders, component costs, production costs, supply time, to name a few.
Several production site configurations are tested in order to ensure that the financial
integration procedures work for elementary as well as complex supply chain configurations.
Chapter 1 identifies the most significant financial aspects to include and provides a basis to
calculate net profit from the simulation output that will be used as a performance indicator
in Chapter 2’s risk analysis.
8
1.2 Financial Statements
Every public company must publish a financial statement in accordance with the GAAP
(Generally Accepted Accounting Principles) which includes four main reports. These are:
1. Balance Sheet
2. Profit and Loss Statement (Income statement)
3. Statement of Retained Earnings
4. Cash Flow statement
1.2.1 Balance Sheet
The balance sheet is a financial statement that shows all of the company’s assets, liabilities
and owner’s equity. Assets are all the tangible and intangible things of positive economic
value that can be owned or controlled by a company. [6] This includes all equipment and
buildings that the company owns but also cash. Liabilities are the “obligation of an entity
arising from past transactions or events, settlement of which may result in the transfer or
use of assets”. [6] This includes the interest paid on funds borrowed from a financial
institution. The owner’s equity is the remaining interest in assets after all liabilities are paid.
So the value of owner’s equity is simply the value of the assets minus the value of liabilities.
1.2.2 Profit and Loss (P&L) Statement
The profit and loss statement (P&L statement) which is also known as the income statement
or the operating statement shows how net revenue, which comes from all sales, is
translated into net income. Net income is the total revenue subtracting the costs and
expenses. The expenses include production costs, depreciation and taxes to name a few.
P&L statements are used by investors or creditors to determine the past financial
performance of a company and then predict its future performance. [7]
1.2.3 Statement of Retained Earnings
‘Retained earnings’ is the amount of net income that is retained by the company rather than
given to its owners as dividends. The statement of retained earnings shows the changes in
the retained earnings of the company over the reporting period. Changes may be due to
dividends being paid or changes in the profit and loss statement. The statement of retained
earning is sometimes combined with the income statement or as an addition to the balance
sheet or it can appear on its own.
9
1.2.4 Cash Flow statement
The cash flow statement basically represents all of the inflow and outflow of cash in the
business from operating, investing and financing activities.
The operating activities are strongly related to the net income from the income statement
and include “the production, sales and delivery of the company’s product as well as
collecting payments from its customers”. [8] Items that are generally added back to the net
income are depreciation, amortisation, and deferred tax. (Deferred tax represents the taxes
owed by the company that is postponed to future periods).
Investing activities include the purchasing of an asset, loans made to suppliers or customers
and payments related to mergers and acquisitions. [8]
Financing activities include inflows from banks and investors used to finance the company’s
activities and outflows in the form of dividends to shareholders. [8]
In the context of this project and in the scope of the simulation platform due to the limited
amount of financial information provided, only operating activities in the cash flow will be
taken into account as a measure of the company’s performance. As stated above, the
operating activities take into account the net income from the income statement and hence
the P&L statement will indirectly also be considered in the scope of this project.
10
1.3 Introduction to the terminology
Companies are mainly interested in their bottom line. They want to ensure that the
company’s profit is as high as possible. This is also in the interest of the shareholder as it
implies larger dividend payments and a continuously profitable company. There are several
definitions of profit which will be described below. Each of these profit measures is used in
different situations to evaluate a specific intention about the company. The measures below
can be evaluated from the company’s cash flow statement and income statement described
in the section above.
Definitions of accounting principles:
1.3.1 Gross Profit
The gross profit is the difference between the net sales revenue and the cost of good sold
(COGS). The COGS are the direct costs of producing the product. The COGS for a car
manufacturer for example would be the sum of the cost of raw materials and the direct
labour costs. Transportation costs are not included in this sum.
1.3.2 Net Income
The net income is similar to the gross profit however it also deducts the indirect operating
expenses alongside the COGS from the sales revenue. This is a more accurate portrayal of
company profit from operations.
1.3.3 Operating Profit (EBIT)
EBIT (Earnings before interest and taxes), used interchangeably with operating profit, is a
measure of a company’s profitability before paying interest and taxes. [9]
EBIT = Operating Revenue – Operating Expenses
Operating Revenue is the sales revenue or the sales cash inflow. The operating expenses
include COGS, depreciation and amortisation, and other general expenses such as
administration and maintenance.
A firm’s fundamental earnings potential is reflected by EBIT and EBITDA (earnings before
interest, taxes, depreciation and amortisation) which is defined in the following paragraph.
11
1.3.4 EBITDA (nonGAAP)
EBITDA (earnings before interest, taxes, depreciation and amortisation) is almost identical
to EBIT however it ignores not only interest and taxes but also depreciation and
amortisation. This measure is not a GAAP metric however it is used by creditors,
shareholders and investors as a measure of company profitability. The advantage of the
EBITDA metric is that it represents the cash earnings without the influence of taxation,
accrual accounting (interest) and effects of different capital structures (depreciation). [10]
EBITDA margin is the EBITDA subtracted from the total revenue and it represents how much
of the operating expenses consume sales revenue.
1.3.5 Net Profit
The net profit is what is left over after all expenses, interest and tax are paid. It is simply
EBIT minus the tax and interest. It is the sum of money the company has made as profit
before paying the dividends. Net profit minus the dividends is the retained earnings which
are described in the previous sections. Net profit includes all of the effects of tax laws in
different countries and the effects of the liabilities that require interest payments.
Net Profit Calculation
This example of the calculation of net profit is in accordance with the International Financial
Reporting Standards (IFRS) and a generic example is provided in Table 1.
Using this same methodology, the output from the SCOPE simulation tool can be used to
calculate the Net Profit of various scenarios or configurations. The financial aspects that can
be determined from the simulation platform are:
Raw materials and services (including transportation costs)
Personnel costs
Sales will be taken as a fixed amount for each machine which will be shown in the example.
The financial aspects that are known and can be applied to manipulate the output from the
simulation in order to calculate net profit are:
Corporate tax rate
Exchange rates
Transfer pricing
12
Table 1: Example of a Profit and Loss statement
in million CHF
Year
2009
2008
Sales
816.6
871.8
Raw Materials and services
485.2
500.15
Personnel costs
266.55
260
Depreciation and amortization
21.25
22.05
Operating profit (EBIT)
43.6
89.6
Interest expenses
7.5
13.25
Profit before income tax
36.1
76.35
Income tax
10.5
23.9
Net Profit
25.6
52.45
These financial aspects are described in further detail in the following paragraphs. The
interest rate will not be included in the calculation of the net profit in the examples as the
amount borrowed for each machine can not be correctly estimated. In any case, it is very
simple to integrate into the calculation.
Using the above financial aspects the Net Profit can be calculated for various configurations
as mentioned earlier. The effects of various changes and of the integration of several
financial aspects are then portrayed and analysed. This will be presented in the results
chapter.
13
1.4 Factors affecting Net Profit
The factors that affect the calculation of net profit are retrieved from Table 1 which
represents a Profit and Loss Statement. These are:
1.4.1 Exchange rates
Exchange rates demonstrate how much one currency is worth in terms of another currency.
They are in continual movement and hence can be a large source of uncertainty when
buying, transferring or selling goods in different currencies. In general foreign exchange
derivatives are used as a form of hedging and in order to reduce the level of uncertainty
however this may not always be the case and therefore they still act as a significant input in
the financial flow.
1.4.2 Corporate income tax rates
The corporate tax is the tax charged by various countries, or in the case of Switzerland,
cantons, on the income or net profit made by organisations. Each country may have a
different corporate tax rate. This fact can be used as an advantage for some organisations to
transfer the majority of their profit in a country that has a low tax bracket in order to
maximise profits. Benefiting from various countries’ tax rate may also be illegal and
therefore there are international regulations put into place to avoid any conflicts.
1.4.3 Transfer Pricing
Multinational companies often disperse their subsidiaries globally in order to reduce costs
and increase profit. Transfer costs or transfer pricing is the price that is set to transfer
something internationally whether it be material, services, or assets, within an organisation.
As the transaction occurs within one organisation, market effects generally do not have an
impact on pricing. Transfer pricing determines how the profit is repartitioned within the
company.
In order to transfer products or components from one country to another (in the same
company) one subsidiary must sell the product to the other therefore this transfer price is
recorded in the annual financial statements as sales or purchases within the subsidiary and
will be taxed accordingly in the case of a sale. Hence there are many restrictions and
regulations on transfer pricing in order to prevent multinationals in profiting from different
tax systems in different countries. There are several pricing methods which will not be
14
discussed here. What is important in this project is to see the effects of external parameters
on the final profit (from production) of the company.
Figure 3 represents a method in setting the transfer price of a product or service between
two sites in country A and B. The first step is to determine whether you are a seller or a
buyer. If you are a seller and the tax rate of country A larger than the tax rate of country B,
you charge a lower transfer price. If the tax rate of A is not larger than B, then a higher
transfer price should be determined. Similarly if you are a buyer and the tax rate of country
A is higher than B then again a lower transfer price should be stated and visa versa.
The important parameters to measure are the effects of external risks on transfer pricing.
The two most important effects on transfer pricing or transfer cost are the exchange rate
and the different tax rates of each country.
Figure 3: Simple way to understand how to set transfer price [1]
15
1.5 Net Profit Calculation Procedure
This section develops a method that determines the net profit including taxes, exchange
rates and transfer pricing. The taxes and exchange rates are easy to determine as they can
be based on historical data however transfer pricing must have a methodology of its own.
As each corporation chooses its own method for transfer pricing whilst complying with the
international regulations, an example was used in this project based on the material,
production and transportation costs as well as a 25% profit margin.
1.5.1 Transfer Price
ABC sets the transfer price for the following year based on forecasted values of material
costs, labour costs, and transportation costs.
The calculation in Table 2 shows how to determine the transfer price to transfer material or
services from a site based in country B to a site based in country A. It is assumed that both
sites are part of ABC and that country A is the home country (where the headquarters are)
and country B is the subsidiary. The transfer price is recorded as an income to site B and an
expense to site A.
Table 2 explains the calculation of the transfer price of a product or part according to ABC.
Table 2: ABC’s Transfer Pricing Method
TRANSFER
PRICE 2010
(Site B)
PRODUCT I
Explanation
Material costs
50,000.00
Obtainable from output data
Labour value
40,000.00
Obtainable from output data
Cost of sales
90,000.00
Subtotal of above values
Markup 25%
22,500.00
25% of cost of sales
TOTAL
112,500.00
Subtotal of marked up price
Packing and
fright
6,000.00
Obtainable from output data
TRANSFER
PRICE
FC
118,500.00
Total of above values in foreign
currency (FC)
CHF
284,400.00
Total in Swiss Francs multiplied by
FC/CHF exchange rate
16
The cost of sales is the sum of the direct material and labour costs for the particular
product. A 25% markup is then added in order to allow a profit margin on which the
packaging and transportation costs are added to determine the final transfer price in the
currency of country B. The transportation costs taken into account are only those of
transporting the goods from site B to site A. It does not take into account the transportation
costs due to the shipment of material from suppliers.
The transfer price is required in order to calculate the net profit of the firm, and all its
subsidiaries, in the home currency of the parent company. Using the exchange rate this
value is converted into the currency of country A, which in this case is Switzerland, therefore
into CHF.
The net profit is calculated following the same reasoning as in the financial statements in
Table 1. The net profit is simply the profit of all the subsidiaries stated in the home currency.
This includes various taxes paid to corresponding governments as well as exchange rates to
transform calculations in the foreign currency into CHF. The following model can be
expanded to more complex configurations which will be presented in the following
chapters.
1.5.2 Detailed Procedure of calculating net profit using the simulation output
In order to construct a net profit calculation the following steps must be performed
alongside the output data given by the simulation.
1. Draw a schema of the configuration by determining all coordinates of the
configurations. This includes:
a. Order destination – where all the orders are sent
b. Origin of shipment of finished products – which in the case of this simulation
is always the same as the order destination
c. Supplier locations
d. Bill of Material – including which operations are conducted in which locations
2. Determine the Transfer Prices
a. Identify what components/subassemblies are being transferred intersite
b. Complete the excel Table that determines transfer price for each product
that is transferred. This includes:
i. ‘Material costs’ directly associated with the product being
transferred.
ii. ‘Labour costs’ directly incurred to manufacture/subassemble the
transferred product. As the simulation does not differentiate between
the direct labour costs and other production costs, ‘Production costs’
is used as the input for labour costs.
17
iii. ‘Transportation costs’ to transport the product from one site to the
other.
The transfer price is stated in the foreign currency which is the currency of the
country in which the site that produced the subassembly is situated.
3. Determine the profit for each finished product at each site.
a. Profit for the main site (parent company), in the currency of the parent
country. This includes:
i. Sales income per finished product
ii. Minus Material costs incurred by this site for this product
iii. Minus Production costs incurred by this site for this product
iv. Minus Transportation costs from external suppliers to this site and
from this site to customers
v. Minus Transfer price which is changed to home currency using current
exchange rate
vi. Once the gross profit is calculated following all the previous steps, it is
multiplied by (1 – corporate tax rate) of the home country to obtain
profit after tax.
b. Profit for the subsidiaries (international sites). This includes:
i. Transfer price in foreign currency
ii. Minus Material costs incurred by subsidiary site for this product
iii. Minus Production costs incurred by subsidiary site for this product
iv. Minus Transportation costs from external suppliers and from
transferring material to other site
v. Once gross profit is calculated following all the previous steps, it is
multiplied by (1 – corporate tax rate) of the subsidiary’s country to
obtain profit after tax in foreign currency
vi. The profit after tax for the subsidiary is then converted into the home
currency of the parent company at the exchange rate.
c. Net Profit is the profit for parent company in home currency plus the profits
from the subsidiaries in the home currency of the parent company.
This method is applied to various supply chain configurations in order to establish that it can
be used for simple as well as complex supply chain systems.
18
1.6 Validation
To illustrate the validity of integrating the calculation of net profit and transfer costs, three
ABC configurations are analysed.
1. Scenario 1: One production site in Switzerland
2. Scenario 2: Two production sites: Switzerland and China
3. Scenario 3: Three production sites: Switzerland, China and Italy.
It will be shown that the methodology can be just as easily applied to a very complex
configuration as it is to a very simplified configuration.
The supply chain of each scenario is simulated using the SCOPE simulation platform
described in section 1.1 and in Figure 1. The BOM is that presented in Figure 2.
1.6.1 Scenario #1:
The first configuration is the simplest. It includes only one production centre in Switzerland
to which all orders arrive and from which all finished products are shipped. (Figure 4)
Figure 4: Configuration of a supply chain with only one site
Transfer pricing is not taken into account in this configuration as there is no material or
service transfer from a foreign subsidiary. The net profit of this configuration is very easily
calculated.
Net Profit = (Sales – Material costs – Transportation costs – Production costs) x (1 – tax)
19
The tax value refers to the countries corporate income tax percentage which for the case of
Switzerland is an average of 21.17% in 2009. [11]
The suppliers in this scenario are located either as close as possible to SWISS which implies
higher material costs or in less expensive economic zones such as Asia and Africa. The
components (BOM: Figure 2) are bought from the suppliers located as in Table 3.
In order to balance the higher transportation costs from ordering further away the cost of
components bought from Asia will be considered as 10% cheaper than Local (Swiss) or
European prices, and that of Africa 15% less expensive than Local or Europe.
Table 3: Supplier locations for scenario 1
Component
Location
Price
C 1
Africa
15%
C 2
Asia
10%
C 3
Asia
10%
C 4
Europe
same
C 5
Local (Swiss)
same
C SP
Local (Swiss)
same
1.6.2 Scenario #2:
Scenario 2 consists of two production centres in Switzerland and China. (Figure 6) The
orders are sent to Switzerland and the finished products are sent from Switzerland to the
customers. Both Switzerland and China have their own suppliers for their own components
that are located based on the Table 4. The prices based on the given original prices are also
shown for the internationally bought components.
This supplier configuration was chosen assuming that there are suppliers all over the world
that can supply any of the components except for the special parts, SP, which are supplied
in Switzerland or ‘local’. All components, subassemblies and special parts are shown in the
bill of material. (Figure 5)
Table 4: Supplier locations for scenario 2
Component
Location
Price
C 1
North America
Same
C 2
Local (CHINA)
10%
C 3
Africa
15%
C 4
Europe
Same
C 5
Local (CHINA)
10%
C SP
Local (SWISS)
Same
20
There are some subassemblies manufactured and assembled at the Chinese site. The BOM
shows which parts are made in Switzerland and which in China.
The net profit can be calculated using the procedure explained in 1.5.2. The Chinese site
must calculate a transfer price including subassembly 3 and 6 (Figure 5). The transfer price
of these two subassemblies that are sent from the Chinese site to the Swiss site must be
determined using the method in 1.5.2.
The profit from the Swiss site for this finished product (FP) can be described in detail as
follows:
Figure 5: BOM of a finished product. The circled items are subassembled in China
21
Where:
Sales: money received for the sale of the finished product FP
Material costs: of C1, C3, C4 and C SP
Transportation costs: from supplier to the Swiss site and to the customer from the
Swiss site
Production costs: to produce SP1, SP2, SP3, SA1, SA2 and FP
Transfer price: which has been converted into CHF from CNY
Tax: Swiss corporate tax rate
Figure 6: Configuration of a supply chain using two production sites
The profit from China is calculated in the same way.
Where:
Transfer price: paid by Swiss site to transfer SA3 and SA6 from China; stated in CNY
Material costs: of C2
Transportation costs: from supplier to Chinese site and from Chinese to Swiss site
Production costs: to produce SA3, SA4, SA5, SA6, SA7
Tax : Chinese corporate tax rate
The transfer price is calculated in the same way as the previous chapter.
22
Where:
Materials: C40
Production: to produce SA3, SA4, SA5, SA6 and SA7
Transportation: to transport goods from the Chinese to Swiss site
Exchange rate (CNY/CHF)
Net profit is the sum of the profits from all subsidiaries and the parent company all
converted into the currency of the parent home country which in this case is CHF.
1.6.3 Scenario #3:
The third scenario is similar to scenario 2 however part of the production is in another
foreign country, Italy shown in Figure 7. Figure 8 is the BOM that shows which
subassemblies are conducted in each of the production sites where the circles elements
represent ‘made in China’ and the elements inside a grey square are ‘made in Italy’.
The suppliers have also been chosen as close as possible to the production sites, as in the
previous scenarios, in order to reduce transportation costs as much as possible, however
exploiting the cost advantage of the purchasing of components from less expensive
economies. The supplier locations are shown in Table 5.
The following locations were chosen in order to reduce transportation costs as much as
possible however also to reduce cost of material in the case of purchasing from Asia.
Table 5: Supplier locations for scenario 3
Component
Location
Price
C 1
Asia
10%
C 2
Local (CHINA)
10%
C 3
Local (ITALY)
5%
C 4
Europe
same
C 5
Local (ITALY)
5%
C SP
Local (SWISS)
same
23
Figure 7: Configuration of a supply chain with three production sites
The net profit is calculated using the same methodology as the previous scenario. It can be
seen from the BOM that SA2, SA3, SA4 and SA5 are made in Italy. It is also assumed that the
production cost in Italy is slightly lower (5%) than that in Switzerland otherwise there would
be no advantage in producing in Italy. China produces SA6 and SA7. Therefore there are two
transfer prices set; SA2 being transferred from China and the other for SA6 which is
transferred from Italy.
Where:
Materials: C4, C SP
Transportation: from suppliers to Swiss site and from Swiss site to customers
Production: SA1 and FP
Transfer price CHINA: Paid by Swiss to China for subassemblies transferred
Transfer price ITALY: Paid by Swiss to Italian site for subassemblies transferred
Tax: Swiss corporate tax rate
24
Figure 8: BOM of a finished product. The circles items are subassembled in China and the squared items are sub
assembled in Italy.
The profit of the Chinese site is:
Where:
Materials: C2
Transportation: from suppliers to Chinese site and from Chinese to Swiss site
Production: SA6 and SA7
Tax: Chinese corporate tax rate
Exchange rate CNY/CHF: to convert Chinese profit from CNY into CHF
25
The profit for the Italian site is:
Where:
Materials: C1 and C3
Transportation: from suppliers to Italian site and from Italian to Swiss site
Production: SA2, SA3, SA4 and SA5
Tax: Italian corporate tax rate
Exchange rate EUR/CHF: to convert Italian profit from EUR into CHF
Where:
Where:
Materials: C2
Transportation: from Chinese site to Swiss site
Production: SA6, SA7
Where:
Materials: C1 and C3
Transportation: from suppliers to Italian site and from Italian to Swiss site
Production: SA2, SA3, SA4 and SA5
1.6.4 Results of each scenario
Table 6 presents the net profit and transfer prices results of the previous scenarios.
26
Table 6: Net Profit and Transfer Prices of Scenario 1, 2 and 3
Scenario
Transfer Prices
Net Profit
Product 1
Product 2
Product 3
Product 4
Product 5
All products
1
0
0
0
0
0
110,005,428
2
240,419
17,821,135
6,489,423
6,691,128
21,772,417
85,235,404
3
400,761
26,498,347
8,425,551
9,888,240
36,624,177
87,596,219
As each scenario was very different in terms of suppliers and therefore component costs as
well as production costs due to the site location it is difficult to compare them to each
other. However the results and calculation of net profit for each of the configurations
validates the calculation procedure presented in 1.5.2 for simple and more complex supply
chains.
The same net profit calculation procedure will be used in the following risk analysis chapter
as a performance measure of the supply chain.
27
CHAPTER 2: SUPPLY CHAIN RISK
2.1 Introduction
Over recent years supply chain design has become most cost and time effective. These
leaner designs have given way to higher uncertainties due to outsourcing, international
production sites and decreasing the amount of suppliers. Norrman and Jansson indentify
current business trends that increase the susceptibility of a business to risks [4]:
“Increase use of outsourcing of manufacturing and R&D to suppliers;
Globalization of supply chains;
Reduction of supplier base;
More intertwined and integrated processes between companies;
Reduced buffers, e.g. inventory and leadtime;
Increased demand for ontime deliveries in shorter time windows, and shorter lead
times;
Shorter product life cycles and compressed timetomarket;
Fast and heavy rampup of demand early in product life cycles; and
Capacity limitations of key components” [4]
The goal of supply chain management was to initially reduce costs and time while sustaining
the quality of the product. This has created supply chains that have higher responsiveness,
are leaner and greater agility shown in Figure 9. [4] Norrman and Jansson believe that risk
analysis should be considered as significant as quality as shown in the picture. By increasing
responsiveness, leanness and agility, the company becomes more vulnerable and exposed
to risks. [4] Therefore it is integral to investigate the risks that are associated with time and
cost and eventually quality.
Figure 9: Key focus areas within logistics and SCM [3]
28
Chapter 2 develops a methodology to perform a risk analysis on the supply chain of ABC
using the SCOPE simulation platform. The configuration being tested will be the same as
Scenario #2 in section 1.6.2 and in Figure 6. The BOM is the same as in Figure 5. Once again
there are five finished products that are being considered.
The outputs analysed are the net profit and the delivery delay. Five factors that are changed
either in the input of the SCOPE simulation or in the net profit calculation are considered.
The five factors are varied between their minimum and maximum values which creates a
total of 32 scenarios (25). The effect of these five factors on the output is analysed through a
sensitivity analysis. A complete risk analysis is then done on the output data of 32 scenarios
which are finally ranked on a risk matrix.
This chapter will first introduce the subject of risk and supply chain risk management. A
methodology is then proposed based on the literature review which is then applied to the
ABC case study example.
29
2.2 Literature review
2.2.1 Definition of risk
There is no one definition for the term ‘risk’. Risk concerns the outcome of a future event
which has the possibility of deterring, positively or negatively, from the expected outcome.
In general a positive outcome is considered as an opportunity whilst a negative outcome, a
risk.
Klibi et al. [12] describe a risky situation as being one “where the link between decisions and
outcomes is governed by probability distributions”. [12] That is to say that there is a
possibility that the outcome will deter from the expected outcome. “The risk increases as
the likelihood and the negative impact of possible outcomes increases”. [12]
Risk is measured using three criteria: [9]
1. What is the event that may occur, (rain)
2. The probability that the event will occur (how likely is it that it will rain?)
3. The impact or consequence this event would incur (how will the rain affect my
activities?)
The first and second criteria can be combined as the likelihood or probability that an event
will occur. [9] The likelihood that an event will happen is usually evaluated over three years
in the company, or over the company’s strategic planning period. [13] The probability
distribution of risk is sometimes normal however quite often it is skewed left or right or it is
leptokurtic. [5] In any case the use of historical data can aid in determining the probability
distribution of a risk factor. [5]
The consequence is a measure of the impact that the occurrence of an event may have on
the business. This can be analysed by evaluating a financial value of the company, however
since not all risks can be evaluated financially, it is also possible to approximate risks
qualitatively. [13]
Another definition of risk is given by Harland et al. [14] where risk is roughly defined by the
possibility of danger, damage, or any other type of loss. Mitchell [15] represented risk as a
mathematical formula to measure the probability of a loss, P(lossn), and the significance or
the business impact of the loss, l(lossn) for an event n:.
Equation 1: Harland et al.'s risk calculation
30
Treleven and Schweikhart developed risk/benefit model that quantifies the risk due to a
specific sourcing strategy. A sourcing strategy has been extensively described in their paper
where “Single sourcing is broadly defined as fulfilment of an organisation’s needs for a
particular purchased item from one vendor of choice” [3].
The above model shows the accumulated risks of a sourcing strategy. All the probabilities
and their corresponding impacts were combined to determine the total impact of the
strategy.
They conclude that the risk can be “decreased by decreasing the probability of risk
components or decreasing the impact for each of the risk components” [3].
Both Harland et al.’s and Treleven and Schweikhart’s risk calculations are based on the same
principle: risk equals probability multiplied by the business impact or consequence. Treleven
and Schweikhart use the term ‘total impact’ instead of ‘total risk’ as their model also
considers positive outcomes and the term ‘risk’ implies a negative outcome. However their
equation still has the same principle as Equation 1.
The equation used in the risk analysis of Chapter 2 combines Equation 1 and Equation 2 and
is the combination of the impact and the probability of each risk factor.
Equation 3: Risk equation used in risk analysis
Where ‘Total Impact’ is the total risk of the situation, ‘Business Impact’ is the net profit or
delivery delay output data (‘Business Impact’ is also considered as the ‘consequence’ of the
event) and ‘Probability’ is the probability or the likelihood of the event or scenario.
Equation 2: Risk/benefit model [3]
31
2.2.2 Supply Chain Risk Management
There are various areas that can be analysed using SCRM. The most traditional include
logistics, purchasing and operations management however there is an emerging interest in
analysing the risks in strategy, international business and finance. [16]
As mentioned above it is important for a company to be aware of all the risks it faces in
particular the risks to its supply chain which may have a large impact on its profitmaking
activities. Once the risks are known it is important to develop a strategy to manage these
risks. This practice is called Supply Chain Risk Management.
“Supply Chain Risk Management is to, collaboratively with partners in a supply chain
or on your own, apply risk management process tools to deal with risks and
uncertainties caused by, or impacting on, logistics related activities or resources in
the supply chain.” [17]
The aim of supply chain risk management is to essentially reduce or eliminate the risks or
uncertainties in order to avoid business disruptions or other losses. A byproduct of the risk
analysis is a higher level of information and knowledge of the company.
A common method to represent the level of risk is to present them on a risk matrix where
they are scaled according to the probability of the event or scenario and the business impact
as defined in section 2.2.1. [4] This concept of a risk matrix will be used in the risk analysis of
the case study.
2.2.3 Risk Types
There is a copious amount of information on the types of risks. The following section
provides a literature review on the risk drivers, risk categories and losses due to risk events
from several authors.
Figure 10: Risk Map/Matrix [4]
32
Risk Drivers
According to De Loach [16], risks can either be externally or internally driven or decision
driven. An example of these three risk drivers is provided below:
1. Externally driven – environmental risks (market risks etc)
2. Internally driven – process risks (operations etc)
3. Decision drivers – information risks (erroneous decision support etc)
In the context of this project only external risk drivers that are out of the direct control of
the company are considered.
Risk Categories
Klibi et al. [12] indentifies three categories of risk:
1. Randomness – random variables with known probability distributions related to
business as usual operations. This category includes exchange rate, inflation rates,
demand etc.
2. Hazard – low probability, highimpact events such as natural disasters.
3. Deep uncertainty – where there is a lack of any information to determine the
probability of an extreme event occurring in the future.
The probability distribution of the first type of risk, randomness, is in generally simple to
determine as the design model parameters, such as prices, have historical data. The impact
of business activities of these types of risks are usually the lowest of all three.
Hazard uncertainties are more difficult to evaluate as their probability distributions are less
accurate and their consequences are defined as having a high impact on the business.
Deep uncertainties have too little information for their probabilities to be evaluated
quantitatively.
This project will consider ‘randomness’ risks types where the probability distributions are
easily determined and based either on historical data or industry experience.
Juttner et al. [18] also identifies three types of supply chain risks and can be seen in Figure
11 and 12 below:
1. Organisational risks
2. Networkrelated risks
3. Environmental risks
Organisational risk sources are the risks that come from the supply chain parties
themselves. Networkrelated risk sources come from the interactions between SC parties.
33
Environmental risk sources are external uncertainties that arise from the interactions
between the SC and the environment.
The risk types that are mainly dealt with in the case study are networkrelated risks and
environmental risks.
Harland et al. provided a summary of the different types of risks, sources and descriptions
that are present in a supply chain shown in Table 7. The risk types that will be considered in
this project include supply risk, operations risk, fiscal risk and financial risk.
Figure 11: Risk types within a supply chain according to [18]
Figure 12 Risk Sources in Supply Chain  adapted from [18]
34
Table 7 Types of risk, sources and descriptions [14]
Manuj and Mentzer provided a summary table (Table 8) of all categories of risks present in a
supply chain. The first four risks in the table are specifically related to the supply chain and
cause a disruption in operations. [5] Whereas the last four risks: macro, policy, competitive
and resource risks are external to the supply chain and out of the control of the business
however they affect the first four risks. Manuj and Mentzer therefore only take into account
the supply, operational, demand and security risks and identify their position in the supply
chain as seen in Figure 13.
However macro risks such as the exchange rate are considered to be significant in affecting
the financial aspects such as transfer pricing within a supply chain and will therefore be
considered in the risk analysis of ABC’s supply chain.
Figure 13 Risk in the extended supply chain (adapted from[5])
35
Table 8 Summary of risks [5]
Type of risk
Source
Supply Risks
Disruption of supply, inventory schedules, and technology access; quality issues;
technology uncertainty; product complexity; frequency of material design
changes
Operational
Risks
Breakdown of operations; inadequate manufacturing or processing capability;
high levels of process variations; changes in technology; changes in operating
exposure
Demand Risks
New product introductions; variations in demand (fads, seasonality, and new
product introductions by competitors); chaos in the system (the Bullwhip Effect
on demand distortion and amplification)
Security Risks
Information systems security; infrastructure security; freight breaches from
terrorism, vandalism, crime and sabotage
Macro Risks
Economic shifts in wage rates, interest rates, exchange rates and prices
Policy Risks
Actions of national governments like quota restrictions or sanctions
Competitive
Risks
Lack of history about competitor activities and moves
Resource Risks
Unanticipated resource requirements
Losses
Losses due to risk events may be quantitative or qualitative. [14] For example, a quantitative
loss may be lost sales due to supplier unpunctuality and a qualitative loss may be due to
may be negative publicity.
The types of losses are[14]:
Financial loss
Performance loss
Physical loss
Psychological loss
Social loss
Time loss
In general management tends to focus primarily of the gravity of these losses rather than
the probability that they will occur. [4] In other words, a risk with a catastrophic business
impact and low probability is considered more significant than a risk with high probability
and low business impact.
36
2.2.4 Generic SCRM procedure
A very similar SCRM procedure is used in almost all related
literature. It is a multistep technique that efficiently
identifies, evaluates, and manages the risks present within a
specific supply chain shown in the next schema.
1. Mapping: The first step is to map or define the supply
chain to be analysed. This step is not used in all
articles however it is considered to be an essential
step in the risk management process. [14]
2. Identification: The second step is to then identify the
risk types as well as their location in the supply chain.
3. Assessment: The following step is to then assess the
identified risks. This includes determining the
likelihood that the events will occur, the triggers to
the events, the losses if the event occurs and the
gravity of the losses (or consequences).
4. Implementation: Once the risks are known and
measured, a management strategy is then developed
in order to eliminate, reduce or mitigate the risks.
5. Mitigation: Preparing for possible risky events.
Figure 14: Risk analysis procedure
2.2.5 Example of SCRM tool used in industry
Norrman and Jansson studied the supply chain of the telephone company Ericsson. Ericsson
analyses the risks within its supply chain by firstly mapping the supply chain upstream from
the suppliers to the customers. This is done in order to define the critical parts and the risk
sources in these parts. [4] The performance measure used by Ericsson to measure their risk
is the business recovery time (BRT) which calculates how quickly the business activities will
recover from an unforeseen event. The company developed the Ericsson risk management
evaluation tool (ERMET) in Figure 15 which allows them to comprehensively assess various
issues that may be subject to risky events.
Once an analysis of the risks is made they are represented on a risk matrix in Figure 16.
Ericsson regards risks with high consequence and low probability much more significant
than risks with high probability and low consequence. This risk attitude corresponds with
that by Norrman and Jansson [4] implying that although the multiplication sum of a high
probability and low impact risk is the same as a low probability, high impact risk, it is
considered that the latter is much riskier to the company.
37
A high probability, low impact risk event may be a transportation delay due to bad weather
that may result in a delay of 1 day. A low probability, high impact event may be the sinking
of the ship transporting all of the company’s goods. Although this event is unlikely to occur,
the business impact is considered much greater than the first scenario of a 1day delay that
may occur all the time.
In the analysis of ABC’s supply chain risks, a similar reasoning will be considered when
analysing which risks must be reduced in probability or completely transferred to a third
party through insurance for example.
Figure 15 Ericsson risk management evaluation tool [4]
Figure 16 Templates for risk assessment and treatment and contingency planning [4]
2.2.6 Summary of literature review
Section 2.2 provided a literature review on risk determination, identification, categorisation
and application.
38
Firstly a definition was provided with Equation 1 and 2 that determines the degree of risk of
an event depending on the probability of occurrence and negative impact of the event. [3,
14] Equation 3 was deduced from Equations 1 and 2 and will be used in the risk analysis on
ABC’s supply chain.
The risk matrix is then introduced as a manner of representing the risk levels against each
other. This method is used in the results section of this project by graphing the probabilities
of risk events with their business impact or consequence.
A literature review on the different types of risks is provided including the risk drivers, risk
categories and losses due to risk events from several authors. These descriptions are the
basis of identification of the risks that are to be considered for the case study which include
supply risk, operations risk, fiscal risk and macro risk.
A generic supply chain risk management procedure is outlined comprising of five steps:
mapping, identification, analysis, evaluation and the mitigation of risks. This procedure acts
as the basis of the detailed methodology presented in section 2.3. The methodology in this
project only focuses on the identification, analysis and evaluation of risks and does not
include a mitigation plan.
Finally Section 2.2.5 presents an example of a risk analysis tool being successfully
implemented into a multinational company and the importance of such a tool in avoiding a
major negative event in the company’s future.
39
2.3 Methodology
This section provides a methodology to perform a comprehensive risk analysis on a
company’s supply chain. This methodology will then be applied on ABC to demonstrate the
feasibility and utility in a realistic context.
Step 1: Mapping
A map or detailed schema of the company’s supply chain must be drafted with all
production centre, supplier and customer locations clearly defined.
A bill of material that stipulated the location of each operation is required as well as
a table consisting of all supplier locations.
One of the programming conditions in the simulations is that the destination of the
order is also the origin of the shipping of the finished product to the customer.
(Figure 18) For example, if the order is sent to Switzerland, although some
operations may be done at a Chinese site, the finished product will always be sent
back to the customer from the Swiss site.
Step 2: Identification
1. Create a risk effect table like Table 9 including the following factors:
a) Identify the most important key performance indicators. These may include net
profit and customer satisfaction for example.
b) Determine the negative effects of the KPIs which would be a decrease in net
profit or in customer satisfaction.
c) List the events which may cause the negative effects determined in part b).
d) Find the causes of the particular risk event.
e) Determine the factors that are affected in the simulation due to the risk events.
Table 9: Risk effect table to identify risk factors
Effects
Event
Cause
Factors affected in
simulation
Decrease in customer
satisfaction due to late
delivery
Natural disasters
Natural/Unexplained
supply leadtimes,
transportation time
Reduction in Net Profit
Increased income tax
Govt regulations change
tax rates
40
2. From the table deduce the factors that are affected in the simulation. Integrate
these factors as inputs into the simulation to retrieve the output file including the
KPIs previously identified in Table 9. Please refer to Figure 17 for flow chart.
By changing the input factors one by one, determine which make the most
difference in the output KPIs.
3. Identify most significant factors that create the largest deviation in output and
determine the minimum and maximum values between which the factors can
deviate from historical data (when it is available) or expert opinions.
Figure 17: Methodology to identify most significant risks
41
Step 3: Assessment
There are several stages in Step 3. The first is to perform a sensitivity analysis to determine
how sensitive the output KPIs is to the input risk factors identified in Step 2. In this stage it is
also possible to determine the impact or consequence of each scenario which is simply the
output KPI. The second stage is to determine the probability of each factor and in turn each
scenario.
1. Determine to which factors the output KPI is the most sensitive.
a. Take the risk factors that were identified in Step 2 with their minimums and
maximums. Attribute the number 1 to the minimum value and +1 to the
maximum value.
b. Create a table that shows all the possible combinations of the minimum and
maximum values (1, +1). This should always contain 2n scenarios where n is
the number of input risk factors being analysed.
c. Using the factorial design method, the sensitivity of the input risk factors to
the output KPIs can be determined.
FACTORIAL DESIGN METHOD [2]
The factorial design method provides an identification method for additive models with
interactions described by equation 1.
Equation 4: Factorial design equation
Where:
y is the output
x is the input
N is the number of factors
a0, a1... are the coefficients that are found using output of sensitivity analysis
This method standardises the simulation space to that each factor varies between the minimum
and maximum values.
The first step is to organise the inputs into matrix E, which in the case of this project are either 1
or +1 depending on whether a minimum or a maximum is being used.
42
Where N is the number of factors and S is the number of simulations. The matrices Y and A are the
output and coefficient matrices respectively.
Where:
Equation 5: Matrix equation
Y = XA
Where X is the following matrix which is the same as matrix E with an added column of 1 at the
beginning:
Using matrix manipulation, vector A can be found using the following equation:
Equation 6: Matrix manipulation to retrieve vector A
This is the method that was applied to the sensitivity analysis to determine vector A which in the
context of this project is the net profit or delivery delay.
43
2. Find the probabilities of the minimum and maximum values of the input risk factors.
There are two methods of finding the probability distributions of the input risk
factors. These are basing the decision of historical data or asking expert advice.
a. Historical data: For continuously changing factors such as exchange rates and
interest rates it is possible to fit a probability distribution to the historical
data. The cumulative probability of the minimum and maximum values
occurring can then be determined.
b. If little or unreliable historical data is given then a Delphi analysis can be
conducted where knowledgeable experts in the topic of risks at ABC are
interviewed in the form of surveys. An average is taken of all the answers and
used as the probability distribution of a risk that may not have quantifiable
historical data.
c. If responses that are provided are qualitative then a qualitative scale can be
used on the risk matrix instead of a quantitative scale.
Step 4: Evaluation
The purpose of step 4 is to evaluate the risk based on the data gathered in the previous
steps, namely the probabilities and business impact or consequence of the scenario
(combination of input risk factors) in terms of the output KPI (eg. Net profit).
The total impact or total risk of a scenario is using Equation 3 shown again below.
If there are several input risk factors and hence several probabilities, the probability of the
whole scenario (the combination of minimum and maximum values of the factors) is
determined. This is using the methodology from [3] and Equation 2.
Equation 7: Total impact equations
Where n is the number of factors and y is an output value of a particular scenario S.
The total impact can be graphed for all scenarios to determine quantitatively the best case
and worst case scenario for each output KPI.
44
Another method of measuring this is the risk matrix like in Figure 10 where the cumulative
probability is graphed against the business impact as calculated in the sensitive analysis in
Step 3.
The management and mitigation of these risks are not discussed in this project.
45
2.4 Case Study  ABC
The methodology described in Section 2.3 is applied to the risk analysis of ABC’s supply
chain. The SCOPE simulation platform is used to generate a simulation of the supply chain as
changes are made. ABC and the SCOPE simulation platform are both described extensively
in Section 1.1. The configuration that is to be used in the case study is the same as scenario
#2 in Section 1.6.2 that includes two production sites situated in Switzerland and in China.
The following section is divided into the four steps introduced in the Methodology (section
2.3).
2.4.1 Step 1: Mapping ABC supply chain
ABC consists of two production sites: SWISS and CHINA as in Figure 18. The operations that
occur in each of these sites are described by the BOM (Figure 19). Five finished products are
analysed (called Product 15) and all of them have the same BOM configuration as in Figure
19.
Like any multinational company, ABC strives to increase operating profits and decrease
operating costs such as labour and material costs. Therefore the Chinese site will include
lower production costs and both sites will take advantage of international suppliers that
offer cheaper rates for equivalent components. (Table 10)
Figure 18: Production strategy used in the simulation for the production of all finished products
46
Figure 19: The bill of material (BOM) for all products in simulation. Operations conducted at CHINA are circled.
2.4.1.1 Production centre locations
All orders are sent to the parent company, SWISS, in Switzerland. The operations conducted
in Switzerland include SP1, SP2, SP3, SA2, SA1 and finally FP. The operations conducted in
China include: SA5, SA4, SA3, SA7 and SA6. The operations performed in China are circled in
the BOM schema. The rest is conducted in Switzerland.
2.4.1.2 Suppliers
The origin of the components is also an important factor to be defined. To reduce
transportation costs it is best to locate the component origin as closely to the operation
location as possible. This, of course, may be different in reality as some components may be
specific and may only have one global location from which they can be purchased. It is
47
assumed that this is the case for the component used in the ‘special part’, C SP. The rest of
the components, in green in Figure 19, will be considered as generic components that are
available for purchase at several global locations.
Table 10: Supplier locations
Component
Location
Price
C 1
North America
same
C 2
Local (CHINA)
10%
C 3
Africa
15%
C 4
Europe
same
C 5
Local (CHINA)
10%
C SP
Local (SWISS)
same
2.4.2 Step 2: Identifying the major risks.
Table 11 identifies the major risks that are associated with various input factors in the
simulation. The input factors will be changed from their minimum values to their maximum
values according to a sensitivity analysis.
Besides the financial performance measure of net profit discussed in Chapter 1, another
measure is the delivery delay which is related to the customer satisfaction level.
Table 11: Major risk events and associating risk factors
Effects
Event
Cause
Factors affected in
simulation
Increase in delivery delay
and hence decrease in
customer satisfaction
Natural disasters
Natural/Unexplained
supply leadtimes,
transportation time
Labour strikes
Political stability/
low wages
transportation time
to customers
Supply failure
Due to suppliers
supply leadtimes
Transportation failure
Transportation accident
transportation time
Reduction in Net Profit
Increased income tax
Govt regulations change
tax rates
Increased wages
Govt regulations change
labour costs
Exchange rate losses
Financial market
uncertainty
exchange rates
The input factors to be considered in the risk analysis are:
1. Exchange rate (quoted in CNY/CHF – how many CNY for 1 CHF)
2. Tax rates Switzerland (%)
3. Tax rate China (%)
48
4. Supply leadtime (days)
5. Transportation time (days)
The minimum and maximum values for each of the input risk factors are as follows and have
either been determined using historical data or by consulting experts in the field.
Table 12: Input factors with minimum and maximum values
Factors
Minimum
Maximum
notes
1
Exchange rate
6
7
taken from historical data
2
Tax rates Switzerland
21
25
taken from historical data
3
Tax rates China
25
33
taken from historical data
4
Supply leadtime
original
10 days
consultation
5
Transportation time
original
14 days
consultation
The tax and exchange rates minima and maxima are taken from historical data. [11, 19]
Five input factors were chosen each with a minima and maxima which gives 25=32
simulations shown in Table 13.
2.4.3 Step 3: Assessment: Analysis through sensitivity analysis and finding probabilities.
The aim of step 3 is to perform a sensitivity analysis to determine which factors are the most
sensitive based on Table 13.
Two different outputs are measured: net profit which is calculated as in Part 1 of this project
and delivery delay which is calculated as the difference between the promised delivery date
and the real delivery date.
The sensitivity analysis is conducted using the factorial design method introduced by [2] and
described in detail in Section 2.3.
The output results will confirm which factors are most sensitive and in what order.
The results from the sensitivity analysis can also be used to determine the severity of the
impact of the risk event on the business.
The sensitivity of the risk factors in Table 12 will be tested using the output net profit and
delivery delay. The calculation of net profit is based on the procedure in Chapter 1 of this
report.
It has been decided that it is much easier and faster to use the exchange rate in the
manipulation of the output file rather than changing all the values to the foreign currency in
49
the input files. Once the costs are received in the output file, those that are incurred at a
foreign site (in this case in China) are converted at the average exchange rate, from six years
of exchange data, into CNY and then back into the maximum or minimum exchange rate (1,
1) to convert CNY back into CHF according to the matrix in the previous section.
Table 13: Design of experiment for sensitivity analysis
Scenario
factor 1
factor 2
factor 3
factor 4
factor 5
1
1
1
1
1
1
2
1
1
1
1
1
3
1
1
1
1
1
4
1
1
1
1
1
5
1
1
1
1
1
6
1
1
1
1
1
7
1
1
1
1
1
8
1
1
1
1
1
9
1
1
1
1
1
10
1
1
1
1
1
11
1
1
1
1
1
12
1
1
1
1
1
13
1
1
1
1
1
14
1
1
1
1
1
15
1
1
1
1
1
16
1
1
1
1
1
17
1
1
1
1
1
18
1
1
1
1
1
19
1
1
1
1
1
20
1
1
1
1
1
21
1
1
1
1
1
22
1
1
1
1
1
23
1
1
1
1
1
24
1
1
1
1
1
25
1
1
1
1
1
26
1
1
1
1
1
27
1
1
1
1
1
28
1
1
1
1
1
29
1
1
1
1
1
30
1
1
1
1
1
31
1
1
1
1
1
32
1
1
1
1
1
The sensitivity to the net profit is shown in Figure 20. It can be seen from the results that
the sensitivity of the factors 15 as described in Table 12 is in consecutive order. That is to
50
say that the exchange rate shows to be most sensitive factor in determining the net profit,
followed by the Swiss tax rate, the Chinese tax rate, the supply lead time and finally by the
transportation lead time.
By looking at the results of the interactions between the factors (Figure 21), it can be seen
that the largest interaction is between factors 4 and 5, the supply lead time and the
transportation lead time. This is followed by an interaction between the exchange rate and
the Swiss tax rate, then the exchange rate and the Chinese tax rate, then the interaction
between the Chinese tax rate and the supply lead time which is equal to the interaction
between the Chinese tax rate and the transportation lead time.
Figure 21: Interactions between factors
0
1'000'000
2'000'000
3'000'000
4'000'000
5'000'000
6'000'000
Exchange rate (FX)
Tax rate Swiss
Tax rate China
Supply time
Transport time
Net Profit Factors 15
0
50'000
100'000
150'000
200'000
250'000
300'000
350'000
400'000
Net Profit Factor interactions
Figure 20: Sensitivity to net profit
51
Similarly the results from the delivery delay sensitivity analysis in Figure 22 show that factor
4, supply lead time, followed by factor 5, transportation lead time, and finally the
interaction between these two factors, have the most significant effect on the delivery
delay. This result was expected and is logical
A sensitivity analysis of the calculation of CO2 was also conducted and showed only to be
linked to the transportation lead time. This is expected as this is the only factor influencing
the calculation of CO2 emissions in the simulation.
This however gives light to the fact that all of the output results obtained before the net
profit is calculated depends either on one, both or neither of factors 4 and 5 (changed in the
input files). This is however an exception to the costs as the exchange rate must also be
used in manipulating the costs in the output files.
2.4.3.1 Procedure to find probability of exchange rates
The exchange rate minimum and maximum are determined from historical data from
September 2003 to December 2009. (Table 14) The minimum and maximum values of 6 and
7 respectively were based on the entire data’s mean, minimum and maximum values shown
in the graph below. It was observed that the distribution is quite close to a normal
probability distribution.
However, in order to determine the probability that the random variable (the minimum or
maximum exchange rate) would occur, the data span was shortened to only 3 years from
January 2007 to December 2009. (Table 15) This was done based on the assumption that
the future exchange rate has a stronger correlation with the most recent data.
0
100
200
300
400
500
600
700
800
900
Sensitivity
Factor
Delivery Delay
Figure 22: Sensitivity to delivery delay
52
Table 14: Exchange rate statistical data from 20032009
from 2003 to 2009
min
5.98
max
7.09
average
6.46
standard deviation
0.28
plus 3sigma
7.29
minus 3sigma
5.62
Table 15: Exchange rate statistical data from 20072009
from 20072009
min
5.98
max
6.87
average
6.39
standard deviation
0.29
plus 3sigma
7.25
minus 3sigma
5.54
Once again in the case of the data from the previous three years, it was observed that the
data best fits a normal probability distribution which is used to identify the probabilities of
the minimum and maximum chosen for the sensitivity analysis.
The excel function ‘NORMDIST’ was used in order to determine the cumulative probability
distribution which is shown in the table below. The cumulative probability distribution
signifies the area under the curve which corresponds to the random variable selected based
on the mean and the standard deviation. Using the mean and the standard deviation, 99.7%
of the area under the graph corresponds to six times the standard deviation.
Based on the results given the probability that the exchange rate will fall to a minimum of 6
CNY/CHF or rise to a maximum of 7 CNY/CHF is approximately 8.42% and 1.71%
respectively. This is also represented as the coloured corners of the normal probability
distribution in Figure 23.
Table 16: Cumulative probability of exchange rate minimum and maximum
random variable
x
cumulative probability
distribution
Probability
min
6
0.084174764
8.42%
max
7
0.982937568
1.71%
53
Figure 23: Normal distribution. Coloured parts represent the cumulative distribution
2.4.3.2 Methodology to determine the probability of tax rates
Determining the probabilities of tax rates however can not be done in the same way as the
exchange rate which is in constant movement. As corporate tax rates are decided by the
government every year it is difficult to apply a model without having inside information
from the governments themselves. It is more likely that the tax rate will stay constant rather
than to change dramatically however an exact probability can not be attributed to this
outcome.
In 2007, China’s government passed a new corporate tax law that unified the tax codes and
set a new corporate tax rate of 25% starting from 2008. [20] Therefore it is highly unlikely
that the tax rate will once again grow to 33% as it was before the law was passed in 2007.
The methodology of finding the probability of the tax rates being at their minimum and
maximum is quite different to that for exchange and interest rates which are in constant
fluctuation. The tax rate is set by the country and although it may change, it has a very high
likelihood that it will remain the same as the previous year therefore some assumptions and
justifications must be made to estimate the probability of the tax rates changing from
minimum to maximum.
The historical corporate tax rates for China and Switzerland are shown in Table 17.
The probabilities that the tax rates will remain at their minimums, which in both cases is the
most recent tax rate of 2009, is almost 100%. This is following the reasoning above. The
reliability of this probability can be improved by conversing with both the Swiss and Chinese
tax administrations to see what the forecast is for the following years. Often laws that will
change the tax rates are in procession and do not come into effect until the following year
as in the case for China described previously. This will provide a more accurate
54
representation of the future. However in the context of this project due to time constraints
it was not possible to communicate with the appropriate tax authorities in each government
therefore a probability of 100% that the tax rate will remain the same as the previous year
was taken for both cases.
Table 17: Tax rates from 20032009 for China and Switzerland in percentages
China
Switzerland
2003
33
25
2004
33
24.1
2005
33
21.3
2006
33
21.3
2007
33
21.32
2008
25
21.17
2009
25
21.17
2.4.3.3 Methodology to find probability of supply time and transportation time
Since there is a lack of historical data in the frequency of a supply or transportation failure it
is difficult to attribute an exact probability to whether the failure event will occur. Therefore
a methodology has been introduced which allows an approximate estimation of the
likelihood that the event will occur to be determined.
For risk factors such as supply time and transportation time where it is very difficult to
determine the probability of a maximum or minimum occurring, one may conduct a Delphi
analysis. This entails developing a questionnaire or survey to be filled out by a number of
experts in the field that have considerable experience in the domain. Once all the answers
are collected, the data may be treated by taking a mean average if the responses are
quantitative, or determine a qualitative scale from unlikely to extremely likely if the answers
are qualitative. [21]
Entchelmeier et al. [22] suggest incorporating a third factor in the calculation of a supply risk
value (SRV):
SRVi = pi x ii x SUVi
Where p is the probability, or likelihood, the event will occur, i is the impact or consequence
that the event will incur on the business and SUV is the supply utility value. The SUV is the
manager’s perception of the risk depending on whether he is risk averse, risk independent
or risk seeking.
55
Appendix A describes the concept of supply utility and risk averse, risk independent and risk
seeking actors.
2.4.3.4 Procedure chosen to determine the probability of supply and transportation lead
times
Experts in the industry were asked for their informed opinions which were provided as:
Supply leadtime: 5%
Transportation leadtime: 10%
It is possible that the supply leadtime is larger than the maximum value of 10 days as well
as the transportation time. It is also possible that the supply leadtime and transportation
times and between the minimum and maximum. That is why the probabilities for the
minimum and maximum values do not add up to 100% as there is a distribution of
probability across time.
Assuming from the simulation output data that 70% of suppliers and transportation are on
time (or at the minimum value) the following probabilities are chosen for the minimum and
maximum values for supply leadtime and transportation time.
Table 18: Probabilities of supply leadtime and transportation time
Factors
Minimum
Min probability
Maximum
Max probability
Supply leadtime
original
70%
10 days
5%
Transportation time
original
70%
14 days
10%
2.4.4 Step 4: Evaluating the risk
Evaluating the risk level of each scenario is quite straightforward considering the business
impact, measured by net profit and delivery delay, have been determined as well as the
probability of each minimum and maximum value occurring for all the risk factors. The
following table shows these minima and maxima as well as their probabilities.
Table 19: Factors with minimum and maximum probabilities of occurrence
Factors
Minimum
Min probability
Maximum
Max probability
1
Exchange rate
6
8.42%
7
1.71%
2
Tax rates Switzerland
21
100%
25
0%
3
Tax rates China
25
100%
33
0%
4
Supply leadtime
original
70%
10 days
5%
5
Transportation time
original
70%
14 days
10%
56
The profit result for each scenario is multiplied with the accumulation of all the probabilities
of that scenario based on the risk/benefit model from Treleven and Schweikhart. [3] The
total probability for each scenario was taken as the multiplication of the output with each of
the factors’ probabilities as shown in the equations below and taken from Equation 7.
Total Impact = (Output x probability1) + (Output x probability2) + (Output x probability3) +
(Output x probability4) + (Output x probability5)
Total Impact = Output x (probability1 + probability2 + probability3 + probability4 +
probability5)
It is a method to measure the degree to which the business impact changes. The total
impact which is the multiplication of the net profit with the probabilities is show in Table 19.
Table 20: Best and worst case scenario of total net profit impact
Scenario
Exchange rate
Swiss tax
rate
Chinese tax rate
Supply lead
time
Transportation
leadtime
1
(worst case)
max
max
max
max
max
16
(best case)
max
min
min
min
min
Figure 24: Total net profit impact

100'000'000
200'000'000
300'000'000
400'000'000
500'000'000
600'000'000
700'000'000
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
total impact = probability x net profit
Simulation number
total impact CHF
57
The same was done for delay time and it was found that the best case scenario or the
shortest delay time is for scenario 4 and the worst case is scenario 29 with the longest
delays.
Table 21: Best and worst case scenarios of total delay impact
Scenario
Exchange rate
Swiss tax
rate
Chinese tax rate
Supply lead
time
Transportation
leadtime
1
(best case)
max
max
max
max
max
32
(worst case)
min
min
min
min
min
Figure 25 shows the total impact on delivery delay of each scenario based on the
corresponding probabilities.
Figure 25: Total delivery delay impact
Another way of representing the total risk is by graphing the net profit of each scenario
against the corresponding probabilities on a risk matrix like in Figure 26 and Figure 27. The
risk matrix for net profit has the xaxis in descending order. This is because in the case of net
profit, a smaller profit has a worse impact of the business and higher profit has a positive
impact on the business. The graph is presented in this way in order to be able to compare
with the delivery delay graph.
0
5'000
10'000
15'000
20'000
25'000
30'000
35'000
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
total impact = probability x delay time
Simulation number
total impact
days
58
Figure 26: Risk matrix based on net profit
Figure 27: Risk matrix based on delivery delay
1
17
2
3
18
19
5
9
21
25
4
20
6
10
7
11
22
26
23
27
13
29
8
12
24
28
14
15
30
31
16
32
0.0000
0.5000
1.0000
1.5000
2.0000
2.5000
3.0000
3.5000
4.0000
1.60E+08 1.65E+08 1.70E+08 1.75E+08 1.80E+08 1.85E+08 1.90E+08
Cumulative Probability of the scenario (%)
Business Impact in terms of Net Profit (CHF)
Risk Matrix Net Profit
1
17
2
3
18
18
5, 9
21, 25
20
4
6, 10
7, 11
22, 26
23, 27
29
13
8, 12
24, 28
14
15
30
31
16
32
0
0.5
1
1.5
2
2.5
3
3.5
4
8500 9000 9500 10000 10500 11000
Cumulative probability of the scenario (%)
Business impact in terms of delivery delay (days)
Risk Matrix Delivery Delay
59
2.5 Evaluating the Results
Depending on whether the risk manager is risk averse, risk independent or risk seeking, he
will divide the risk matrices in different ways. A risk independent manager would view the
risk matrix as in Figure 28 where the matrix is divided into four equal parts. The four parts
recommend to avoid, reduce, transfer or retain the risk depending on their probability and
business impact.
Where:
Avoid: avoiding the risk by using a protective action to eliminate the possibility of the
event occurring. [23]
Retain: decision to accept the risk without any protective actions.
Transfer: “proactive action ... that shifts risks to a third party” [23] like insurance.
Reduce: aim to reduce the probability that the risk will occur.
A risk averse manager who favours less risky events to maximising profit or minimising delay
would move the horizontal divider down and the vertical divider to the left which would
increase the area of the risks to be ‘avoided’.
Conversely, if the manager is a risk taker then the dividers would move in the opposite
directions allowing more risk to be ‘retained’.
In the context of this case study it is assumed that the manager is risk independent and will
hence divide the risk matrices into four equal parts.
Figure 29 shows the net profit risk matrix divided into four sections. Each section however
does not have the same meaning as in Figure 28. This is due to the xaxis, business impact in
terms of net profit. Comparing the xaxis of the net profit risk matrix with the matrix on its
left, it can be seen that a low profit is catastrophic and a higher profit does not bear any
Figure 28: Risk matrix altered from [15]
60
negative impact on the business. Therefore each of the sections obtains a different
management strategy than in Figure 28.
Figure 29: Net Profit risk matrix divided into four sections
The strategy to retain, transfer or avoid the risk obtains the same meaning as in Figure 28.
“Increase” implies increasing the probability of the scenarios situated in this section as they
already yield a high profit but are not likely to happen.
According to Figure 29, the best case scenarios are those in the top right hand corner which
include scenario 8, 12, 13, 14, 15 and 16. These all occur when the exchange rate is at its
maximum value.
Those in the bottom left hand corner should be completely avoided or a management
strategy needs to be developed to reduce the risk of these scenarios. It shows that scenarios
17, 18 and 19 than scenario 1 which was stated the worst case scenario are much more risky
as they have a higher probability of occurring and a more serious business impact (smaller
profit) than scenario 1.
This shows that it is not sufficient to simply multiply the business impact with the probability
as sometimes it may not be show the riskiest situations as perceived by the manager. This
agrees with the previous literature that states that low probability, high impact events are
considered riskier or less desirable than high probability, low impact events even if they
have the same quantitative risk rating. [4]
Figure 30 presents the delivery delay risk matrix divided into four management strategies.
These strategies are the same as the original presented in Figure 28. According to Figure 30,
the best case scenario is to be found in the bottom left hand square which consists of
61
scenario 3, 4, 7, 11, 18 and 20. As most of these scenarios yield approximately the same
business impact, the best case would be that which also obtains the lowest probability of
occurring which is scenario 3.
This result is contrary to the quantitative results provided in Figure 25 which stipulates that
the best case scenario is in fact scenario 1 which proves again that it is important to
represent the scenarios on a risk matrix.
2.5.1.1 Conclusion of results
The results show that the net profit is most sensitive to changes in the exchange rate and
delivery delay is most sensitive to changes in the supply time. A combination of the five risk
factors to determine the total impact as per Equation 7 shows that the best case scenarios
for net profit and delivery delay are 16 and 1 respectively and that the worst case scenarios
are 1 and 32 respectively. All scenarios are shown in Table 22.
This information does not however correspond to the risk matrices of net profit and delivery
delay. The worst case scenario for net profit is considered to be 17 and that for delivery
delay is considered to be 14 and 30.
Therefore it is important to graph the business impact and probabilities on a risk matrix in
order to make an informed decision on which scenarios to avoid and which to favour.
Figure 30: Delivery delay risk matrix divided into four sections
62
Table 22: Configurations of factors for each scenario
Scenario
Exchange
rate
Swiss Tax
rate
China Tax
rate
Supply
leadtime
Transportation
time
1
max
max
max
max
max
2
max
max
max
max
min
3
max
max
max
min
max
4
max
max
max
min
min
5
max
max
min
max
max
6
max
max
min
max
min
7
max
max
min
min
max
8
max
max
min
min
min
9
max
min
max
max
max
10
max
min
max
max
min
11
max
min
max
min
max
12
max
min
max
min
min
13
max
min
min
max
max
14
max
min
min
max
min
15
max
min
min
min
max
16
max
min
min
min
min
17
min
max
max
max
max
18
min
max
max
max
min
19
min
max
max
min
max
20
min
max
max
min
min
21
min
max
min
max
max
22
min
max
min
max
min
23
min
max
min
min
max
24
min
max
min
min
min
25
min
min
max
max
max
26
min
min
max
max
min
27
min
min
max
min
max
28
min
min
max
min
min
29
min
min
min
max
max
30
min
min
min
max
min
31
min
min
min
min
max
32
min
min
min
min
min
These scenarios are deemed to obtain the most favourable scenario for delivery delay and
net profit independently of each other. However if the manager is interested in obtaining
the best scenario of both, using Figures 29 and 30, they can decide upon a compromise such
as scenario 11 or 12 that ensure reasonably high profits as well as reasonably low delivery
delay.
63
CONCLUSION
This project presents a methodology to conduct a risk analysis on a supply chain by
identifying the main factors affecting the net profit and delivery delay.
Data from the SCOPE simulation platform developed by the LGPP at EPFL was employed to
perform a risk analysis on company ABC using the methodology presented in Chapter 2. Five
major risk factors were identified and then analysed using a sensitive analysis to determine
which factors have the most significant effect on two key performance indicators of a supply
chain. These are net profit and delivery delay.
The calculation of net profit has not yet been integrated into the SCOPE simulation platform
therefore a procedure to determine the net profit is presented in Chapter 1 including
external risk sources like exchange rates and corporate tax rates as well as internal risk
sources such as transfer pricing. A calculation method for transfer pricing was also
introduced. Both the net profit and transfer pricing calculations were used as output in the
risk analysis of ABC.
The results subsequently provide quantitative risk matrices that are used in the risk
management decisionmaking process. The worst case scenarios can be quantitatively
identified and hence hedged against or managed.
The main objective of developing a risk analysis method is to identify risk factors are the
effect they have on the company’s supply chain performance. This project presents one
such method which proves to be successful in determining the level of danger of a
combination of risk factors.
64
REFERENCES
1. Abdallah, W.M., Critical concerns in transfer pricing and practice 2004: Greenwood
Publishing Group.
2. Wang, X., Etude de variantes de configuration stratégique de la Supply Chain de BOBST par
simulation mulltiagent, in Institute de Génie Méchanique. 2009, EPFL: Lausanne.
3. Treleven, M. and S.B. Schweikhart, A risk/benefit analysis of sourcing strategies: Single vs.
multiple sourcing. Journal of Operations Management, 1988. 7(4): p. 93114.
4. Norrman, A. and U. Jansson, Ericsson’s proactive supply chain risk management approach
after a serious subsupplier accident. Journal of Physical Distribution & Logistics
Management 2004. 34(5): p. 434456.
5. Manuj, I. and J. Mentzer, Global Supply Chain Risk Management. Journal of Business
Logistics, 2008. 29(1): p. 133154.
6. Wikipedia. Assets. 2010; Available from: http://en.wikipedia.org/wiki/Assets.
7. Wikipedia. Income Statement. 2010; Available from:
http://en.wikipedia.org/wiki/Income_statement.
8. Wikipedia. Cash Flow Statement. 2010; Available from:
http://en.wikipedia.org/wiki/cash_flow_statement.
9. Wikipedia. Earnings before interest and taxes. 2010; Available from:
http://en.wikipedia.org/wiki/Earnings_before_interest_and_taxes.
10. Wikipedia. Earnings before interest, taxes, depreciation and amortization. 2010; Available
from:
http://en.wikipedia.org/wiki/Earnings_before_interest,_taxes,_depreciation_and_amortizati
on.
11. KPMG, KPMG's Corporate and Indirect Tax Rate Survey 2009. 2009.
12. Klibi, W., A. Martel, and A. Guitouni, The design of robust valuecreating supply chain
networks: A critical review European Journal of Operational Research, 2010. 203: p. 283293.
13. Outil 6: Exemple de liste de questions types liées à l'identification et l'evaluation des risques.
2006, KPMG.
14. Harland, C., R. Brenchley, and H. Walker, Risk in supply networks. Journal of Purchasing &
Supply Management, 2003. 9: p. 5162.
15. Mitchell, V.W., Organisational risk perception and reduction: a literature review. British
Journal of Managment, 2005. 6: p. 115133.
16. De Loach, J.W., Enterprisewide Risk Management: Strategies for linking risk and opportunity
in Financial Times. 2000, Prentice Hall.
17. Supply Chain Risk, ed. C. Brindley. 2004, Hampshire: Ashgate Publishing Limited
18. Juttner, U., H. Peck, and M. Christopher, Supply chain risk management: outlining an agenda
for future research. International Journal of Logistics Research and Applications, 2003. 6(4):
p. 197210.
19. ORANDA. Historical Exchange Rates. 2010 [cited 2010; Available from:
http://www.oanda.com/currency/historicalrates.
20. Council, T.U.C.B. Foreign Investment in China. 2007; Available from:
http://www.uschina.org/info/forecast/2007/foreigninvestment.html.
21. Wikipedia. Delphi Method. 2010; Available from:
http://en.wikipedia.org/wiki/Delphi_method.
22. Entchelmeier, A., E. Hartmann, and M. Henke, Supply Risk Assessment: A Utility Value Based
Concept, in 22nd Industrial Marketing and Purchasing (IMP) Conference. 2006: Milan, Italy.
23. Khemani, K., Bringing Rigor to Risk Management, in Supply Chain Management Review.
2007. p. 6768.
65
APPENDIX A: DEFINITION OF UTILITY THEORY
The supply chain utility concept is based on the cardinal utility theory proposed by
Neumann and Morgenstern in 1947. The utility theory suggests that “each individual
attempts to optimise the expected value of something which is defined as utility, and that
for each individual a relationship between utility and payout can be found”[22].
This utility concept has been extended to a manager or partner in a supply chain by [22]. It
assumes that the manager can allocate their risk preferences “in order to evaluate the trade
off between risk and expected return” [22] or net profit in the case of this project.
Figure 31 shows the utility function of a risk averse, risk indifferent and risk seeking actor. A
risk averse actor favours exchanging a lower payout against lower risk. The graph implies
that for a risk averse actor, “a gain of the amount of payout increases the utility value less
than a loss of the same amount decreases the utility value” [22]. This means that as the
payout increases so does the risk which implies that the risk averse actor has a decreasing
desire, or utility, to bet on this payout while the payout, as well as the risk, is increasing.
The risk indifferent actor will not change their decision based on the amount of payout or
risk.
The graph implied that for a risk seeking actor, “a gain of the amount of possible payout
would increase the utility value more than a loss of the same amount decreases the utility
value” [22].
Figure 31: Prototypes of supply risk utility functions taken from [22]