location decisions i - institut für...

Lehrstuhl fürIndustrie, Energie und Umwelt

Universität WienFakultät für WirtschaftswissenschaftenLehrstuhl für Industrie, Energie und UmweltBrünner Straße 72, 1210 Wien

IEUInternational Industrial Management I

Location Decisions I| Introduction | Examples | Check List |

| Steiner-Weber Model|

| Florian Pützl © 2009 | International Industrial Management I, WS 2009 Page 2

IEU International Industrial Management ILehrstuhl f. Industrie, Energie und Umwelt | Mag. Florian Pützl |

Overview | Location Decisions

Choice of locations IntroductionA class of locational choice problemsFactors for locationsMethods

Check listsBenefit analysis (simple, additive)Location break-even analysisTransportation methodSteiner-Weber ModelLocation-alloction Model of Cooper and extensionsHotelling Model

Choice of locations with the firmThe basic problemCRAFTLayouts



Location Decisions - Objective

Choose the location that maximizes

the firm’s benefit

There are only three important things concerning locations:1) Location, 2) Location, 3) Location.



Location Decisions

Importance: Location decisions do not only result in real estate and investment costs but influence in particular fixed and variable costs.

Transportation costs

rentals

Wages

Taxes, etc.

Links to the environment (universities, clusters, etc.)



Example: Automobile Industry I

Source: Dyer, J.H.,Dedicated Assets: Japan‘s Manufacturing Edge , Harvard Business Review, Nov.‐Dec., 1994, S.174ff.



Example: Automobile Industry II

Source: Dyer, J.H.,Dedicated Assets: Japan‘s Manufacturing Edge , Harvard Business Review, Nov.‐Dec., 1994, S.174ff.



Example: Automobile Industry IIIComparing locations of production and organizational units of the

automobile industry in Japan and the US (Dyer, 1994) finds and argues the following:Thesis: The huge success of the Japanese car industry is to a large extent due to the close relations with its suppliers and in particular the geographical vicinity.There exist large differences concerning the average distance between locations of Japanese and American car manufacturers. In particular, the entire ring of Toyota and its suppliers fits between two GM locations. Inventory as a share of turnover is much larger at US car producers which implies more bound capital. These distances in turn make it hard to implement Just-in-Time Management (see later), which further increases inventory costs.



Example: Automobile Industry IV

Closeness implies much better contacts between relevant people and organizations (including suppliers). It is a grave misunderstanding that personal contacts become redundant once an industry can actually choose its supply in aglobally integrated world by a mouse click; remember this for the following discussion about clusters. If GM had a similar ratio between inventory and sales it would save US$ 6 billions which amounts to annula cost saving of the order between US$ 400 – 500 Millions depending on costs of capital ranging between 6% to 8%.



Introductory Example

Table: Example of a location decision l

Feasible locations

Distance from procurement center

Transport costs (in 1000 MU)

Distance to center of demand

Transport costs (in 1000 MU)

Sums

Aachen

150

(150*1500*1)=225 650

(650*1000*1,2)= 780

1005= 4th rank

Braunschweig 270 405 625 750 1155= 6th rank

Dortmund 10 15 650 780 795= 1st rank Erfurt 420 630 380 456 1086= 5th rank

Munich 650 975 10 12 987= 3rd rank

Nürnberg 480 720 170 204 904= 2nd rank

Assumptions: Procurement centered around Dortmund. Volume = 1.500 t and transport cost are 1 MU per t and km.

Demand in Munich, Volume = 1000 t, transport cost are 1.2 MU per t and km.



InterdependencesProduktionsprogramm

Produktpreise

Transportwege Transportmengen

TransportmengenInnerbetriebliche Standorte*

Löhne, Steuern usw.

Investitionsprogramm

Investitionsausgaben je Projekt

Innentransportkosten

Kapitaleinsatz

Sonstige aufwandsgleiche Kosten

+ (Aufwand≠Kosten)

Externe Transportkosten

Aufwand

Ertrag

Profitabilityof a firm fora givenconfiguration of locations

*: Optimierungsproblem

* Optimization problem

Source: Lüder, p.31



Assumptions (given):A set of locations:

A set of organisational units:

Location of the organisational units:

Objective:A mapping that assigns to each unit bi a location from the set S:

Mathematical Formulation of Location Decisions I

{ }njsS j ,,1; K==

{ }mibB i ,,1; K==

( ) Ssbs ji ∈=

( ) SbsbSB

ii ⊂→

→∑ :



Mathematical Formulation of Location Decisions II

Further assumptions:O is the set of given locations that are connected to locations bi e.g. customers:

The sets O, S and B are independent (in practice however there are many interdependencies and feedbacks).Further constraints such as those on capacity.

Objective:

Maximize/Minimizee.g. the locations of the organizational units are chosen in order to minimize the transport costs.e.g. the (incremental) benefit from the choice of a particular location or locations should be maximized.

{ }rpoO p ,,1, K==

( ) ( ) ( )[ ]mbsbsbsfZ ,,, 21 K=



Examples of Constraints

Organizational units that must be close to the marketPublic

Police, ambulance, fire brigadepost

Retail and ServicesFast food, gas stations, super marketsPharmacies, shopping malls.

ServicesM.D., lawyers, barbersBanks (?), mechanics, hotels



Evaluation of Locations

Methods

Check lists

(Cost-) Benefit analysis

Location Break-Even Analysis

Transportation Method

Steiner-Weber Model

Location-Alloction Model von Cooper

Hotelling



Factors Crucial for Locations

Market size

Expected profits

Openess of markets

Stability

bureaucracy

Labour (quality and work ethics)

Infrastructure

Technological know how

Labor costs

Resources, primary inputs

Availability of qualified labor

Reserach institutes

Quality of life

Venture-Capital

Costs

Technology

IT-Infrastructure

Bureaucracy

Suppliers & partners

Subsidies

Arthur Anderson Deloitte Touche



Match Product & Country

1. Braun Household Appliances

2. Firestone Tires3. Godiva Chocolate4. Haagen-Dazs Ice Cream5. Jaguar Autos6. MGM Movies7. Lamborghini Autos8. Alpo Petfoods

1. Great Britain2. Germany3. Japan4. United States5. Switzerland6. India7. Italy8. Denmark



Match Product & Country

1. Braun Household Appliances

2. Firestone Tires3. Godiva Chocolate4. Haagen-Dazs Ice Cream5. Jaguar Autos6. MGM Movies7. Lamborghini Autos8. Alpo Petfoods

1. Great Britain2. Germany3. Japan4. United States5. Switzerland6. India



A Few International Comparisons

ProductivityGrowthCountriesCitiesEconomic libertyCorruption



Growth Competitiveness Index: 2006-2007 | Top 20

Source: World Economic Forum



Growth Competitiveness Index – 2006-2007, Rank >100

Source: World Economic Forum

Growth Competitiveness Index: 2006-2007 | Rank > 100



Business Competitiveness Index: 2006-2007



Business Competitiveness



Productivity by Sectors

Martin Neil Baily und Robert M. Solow, International Productivity Comparisons Built from the Firm Level, Journal of Economic Perspectives, 15/3, 151-172, 2001.



Labor Productivity and Wages



International Labor Costs in Automobile Industry (in €/h)

42.2935.52

24.68 28.9436.61 30.71 34.51

21.9912.06

31.46 31.6623.95

€ 0.00€ 10.00€ 20.00€ 30.00€ 40.00€ 50.00

German

y

France Italy

The Netherl

ands

Belgium UK

Sweden

Spain

Portuga

l

Austria

U.S.AJa

pan



Corruption

5 Best:FinlandUSACanadaSingapurAustralia

5 Worst:BangladeshParaguayNicaraguaNigeriaZimbabwe

Corruption index:Finland 9,7Denmark / New Zealand 9,5Canada 9,0UK 8,7USA 7,7Germany / Israel 7,3Japan 7,1Italy 5,2China 3,5Egypt 3,4India / Russia 2,7Nigeria 1,6Bangladesh 1,2

10 = No corruption



Economic Freedom I



Economic Freedom II



Economic Freedom - GraphicalHONG KONG keeps its top position in the Economic Freedom Index compiled by the Fraser Institute, a Canadian think-tank. The index ranks the policies of 141 countries according to how much they encourage free trade, both internally and with other territories. Countries with fewer taxes, strong property rights, low regulation and sound money score best. Britain and America tie with Canada for fifth place in the list. Germany is ranked 18th—on a par with El Salvador but above Japan. Most of the low-ranking countries are African, except Myanmar and Venezuela, which are both in the bottom ten.



Ranking of Cities

Source: European Cities Monitor 2006:

0,1671010Zurich

0,181299Munich

0,181588Berlín

0,201777Madrid

0,23566Amsterdam

0,24445Brussels

0,271154Barcelona

0,36333Frankfurt

0,59222París

0,91111London

Index2006199020052006Citiy



Another Ranking of Cities

London and Paris maintain their position as Europe's two top cities to locate a business

Warsaw can expect the biggest influx of international companies over the next fiveyears.

Barcelona is also the city perceived as doingthe most in Europe to improve itself as a business location, followed by Prague and Madrid



Hardship Rating



Austria (WIFO)

Energy costsPolitical stability (?)

Public administration Public safety

Lack of reforms (and willingness to reform)Cultural and leisure activities

Little competition in energy marketsEnvironmental quality

Lack of venture capitalRule of law

Environmental regulationsConsensus seeking society

High costs for unskilled laborAvailability of highly qualified labor

Costs of telecomm (not any more!)Availability of high quality labor

Duration until receiving operating permitsActive in Eastern Europe

BureaucracyPart of the common market

DisadvantagesAdvantages



Austria (US-Investor Confidence Study)

Wages, payroll taxes and other ancillary labor costs

Costs of electricity

Labor lawsSocial peace

Income taxesPolitical stability

Opening hours Culture and leisure activities

Labor permits for Non-EU residentsTransport

Operating permit (length)Qualifications and effort (of labor)

Cost of livingSales in Austria

Real estate pricesAccess to Eastern European markets

Tax deductions of investmentsKnowledge of foreign languages

DisadvatantagesAdvatantages



Topical (Economist, 2008): Eastern Europe vs Asia

• As in the previous five years, economies in eastern Europe and Central Asia have consistently seen the fastest pace of positive reform.• On average, it takes 21 days to register a business in eastern Europe, which is 27 days faster than in East Asia. Setting up a company in Indonesia costs 77.9% of the average annual income per person; in Georgia it costs 4%—though there is the small matter of political risk to factor in. Firing a worker costs an average of 53 weeks’salary in East Asia, compared with 27 in eastern Europe. All this cutting of red tape has brought results: Poland now has as many registered businesses relative to its population as Hong Kong does.• Eastern Europe’s rapid progress due to the accession requirements imposed by the European Union (EU), e.g., the EU requires new members to create a “one-stop shop”. Before Macedonia became a candidate for EU membership in 2005, it took 48 days to start a business there. After three years of reforms, it now takes nine days.• East Asian countries still have the edge in some respects: it is easier to move goods across their borders, for example. Businesses in East Asia also face lower taxes. Taxes on profits in eastern Europe are among the lowest in the world, typically around 10%, but labour taxes and compulsory pension contributions increase the overall tax burden on business.• Of course, a few East Asian economies are still miles ahead of eastern Europe. Singapore 1st, Hong Kong 4th but Georgia, Estonia, Lithuania and Latvia in the top 30, Russia = 120th and Azerbaijan was the top reformer.



(Cost-) Benefit Analysis - Additive

Four steps:

1. Scaling: Evaluation via scale (say 1 – 10) the benefits of location with respect to a particular factor which allows to rank and evaluate also qualitative factors.

2. Weighting: Assign weights {gi} to each individual factor entering the analysis and to differentiate between important and less important ones.

3. Aggregate: Multiply points the scale in (1) with the weights in (2) and add up which gives the measure of benefit associated with a particular location:

N(sj) = n1j*g1 + n2j*g2 + .....

4. Choose the location with the highest benefit = highest value of N.



A Simple Example Labor Sales Subsidies

Locations s1 7 2 1

s2 5 1 9

s3 10 4 6

s4 5 7 8

Weights 0.3 0.5 0.2

N(s1) = 7*0,3 + 2*0,5 + 1*0,2 = 3,3N(s2) = 5*0,3 + 1*0,5 + 9*0,2 = 3,8N(s3) = 10*0,3 + 4*0,5 + 6*0,2 = 6,2N(s4) = 5*0,3 + 7*0,5 + 8*0,2 = 6,6

Therefore, location 4 should be chosen.

Extension: Multi-criteria optimization (or decision making)



Check Lists

Precise evaluation of the benefits associated with each factor.Similarly to the the above benefits analysis – hierarchical



Benefit Analysis

Weights allow to mitigate for outliersRanking without weights:C – B – A - DRanking with weights:A – C – D – B



Location Break-Even Analysis I

Comparison of the costs at different locations depending on the rate of utilization.Example

Production of a new Volkswagen model.3 potential locations with their respective costsAssume a sales priceProfit:

Standort Fixe Kosten Variable Kosten Verkaufspreis E(Menge)Poznan € 3.000.000 € 7.500,00 € 12.000,00 2.000Pamplona € 6.000.000 € 4.500,00Wolfsburg € 11.000.000 € 2.500,00 Standort Gesamtkosten* Gewinn*

Poznan € 18.000.000 € 6.000.000Pamplona € 15.000.000 € 9.000.000Wolfsburg € 16.000.000 € 8.000.000

* bei einer Menge von 2.000 Stk.



Location Break-Even Analysis II

1. Given a fixed quantity of 2000 pieces Pamplona yields the highest profits.

2. Since the assumption of 2000 pieces depends on a marketing studyit is unreliable. Therefore, the management wants to know whether Pamplona remains the most profitable with less or higher production.

Location Break-Even AnalysisThis procedure calculates the intersections of the cost curves associated with the different locations depending on the rate ofproduction.

This yields capacity bounds where which location is efficient as well as the minimal costs over all locations contingent on production.



Location Break-Even Analysis III

PoznanPamplona

Wolfsburg

1.000.000

4.000.000

7.000.000

10.000.000

13.000.000

16.000.000

19.000.000

22.000.000

0 500 1000 1500 2000 2500 3000 3500

Menge (Stk.)

Ges

amtk

oste

n

Poznan Pamplona Wolfsburg

Depending on production, different locations turn out to be the mostprofitable ones:



1) Countries

2) Regions

3) Final location

Sequential (Hierarchical) Location Decisions I



Countries

Regions

Specific location

1. Political risks, regulation, incentives.2. Cultural and economical.3. Location relative to markets (sales and procurement).4. Labor: availability, productivity, costs, behavior.5. Availability of suppliers, communication, energy.6. Exchange rates and their risks.

1. Specifics views and demands of the firm.2. Atractiveness of region (culture, climate, taxes, etc.)3. Labor: availability, productivity, costs, unions.4. Availability and costs of materials.5. Environmental regulation at federal and city level.6. Incentives (by regions, cities).7. Availability and distance to primary inputs.8. Costs for real estates and construction.

1. Location specific costs contingent on the size of the unit.2. Air, rail, road and water transport networks.3. Zoning, in particular restrictions.4. Availability and distance to suppliers and service providers.5. Environment and environmental constraints.

Sequential (Hierarchical) Location Decisions II



Case Study | BMW Spartanburg, USA I

Location decision of BMWProduction of BMW Z3 in the USA, mid 1990ies.Expansion to includeBMW X5 und Z4.



Case Study | BMW Spartanburg, USA II

3) Final decision for Spartanburg

2) Choice of South Carolina

1) Choice of the US



Case Study | BMW Spartanburg, USA III

Factors decisive for choosing USA

Market:Market size: USA is the world‘s largest market for luxury carsGrowing versus declining markets: US growing market due to baby boomers.

Labor:Lower labor and production costs: (US: US$ 17/h, DEU: US$ 27/h)High labor productivity(Holidays: US 11, Germany: 31)

Additionally:Low transport costs (US$ 2.500,- less per vehicle)New facilities improve productivity and thus lower the costs per car (around US$ 2.000-3.000.-)Insurance (‚hedging‘) against high Dollar-Euro exchange rates.



Case Study | BMW Spartanburg, USA IV

Factor in favor of South Carolina

Labor:Low wages in South Carolina

Public subsidies:US$ 135 Mio.in terms of tax deductions.

Free trade zone:Duties neither on imported goods nor on carexports.

Infrastructure:Access to freeway and via the freeway to theairport Charlotte from which Lufthansa operates direct flights to Munich. Port Charleston serves for exports and theimports of engines and gearboxes fromEurope

8585

2626

85



Labor Productivity

Keyword: level of labor costs

Cheaper employees => higher profit?

However, cheap labor is NOT everything – countries with cheap labor are often characterized by low labor productity that can erode partially or totally the advantage in terms of labor cost. Compare: Gregory Clark, Farewell to Alms, textile production in England and colonial India.

‚Skills‘ of workers, good infrastructure, and technology are often more important to many industries than plain labor costs.



Case Study | Labor productivity

Quality Coils Inc., CT, USA Coil production for electrical equipments 2 potential locations:

Expanding a the present location in Connecticut New factory in Juarez (one of the fastest growing cities in Mexiko along the US border due to the Maquiladoras)

Comparison based on labor costs:

Historical Example: Alfa Romeo - Alfasud

US$ 1.25/unit20 unitsUS$ 25.-Juarez Plant

US$ 1.17/unit60 unitsUS$ 70.-Connecticut Plant

Labor cost/piece

Output/worker

Wage/d

unit per costday) per (units tyProductivi

dayperCostLabortyProductivi Labor ==



Four International Operations StrategiesC

ost R

educ

tion

Con

side

ratio

nsC

ost R

educ

tion

Con

side

ratio

ns

HighHigh

LowLowHighHighLowLow

Local Responsiveness ConsiderationsLocal Responsiveness Considerations(Quick Response and/or Differentiation)(Quick Response and/or Differentiation)

Cos

t Red

uctio

n C

onsi

dera

tions

Cos

t Red

uctio

n C

onsi

dera

tions

HighHigh

LowLowHighHighLowLow

Local Responsiveness ConsiderationsLocal Responsiveness Considerations(Quick Response and/or Differentiation)(Quick Response and/or Differentiation)

Standardized productEconomies of scaleCross-cultural learning

ExamplesTexas InstrumentsCaterpillarOtis Elevator

Global Strategy Transnational StrategyMove material, people, ideas across national boundariesEconomies of scaleCross-cultural learning

ExamplesCoca-ColaNestlé

International Strategy

Import/export or license existing product

ExamplesU.S. SteelHarley Davidson

Multidomestic StrategyUse existing domestic model globallyFranchise, joint ventures, subsidiaries

ExamplesHeinz The Body ShopMcDonald’s Hard Rock Cafe

Standardized productEconomies of scaleCross-cultural learning


Global StrategyStandardized productEconomies of scaleCross-cultural learning


Global Strategy Transnational StrategyMove material, people, ideas across national boundariesEconomies of scaleCross-cultural learning


Transnational StrategyMove material, people, ideas across national boundariesEconomies of scaleCross-cultural learning
















Steiner-Weber Model I

Geometric, continuous Model based on the works of the German geometricians and regional scientists:

Launhardt (1882), Steiner-Weber (1909) & Lösch, Christaller (1940er)

Model assumptions:The set location where to delivery to, is given.The set of potential locations (S) is the entire plane (infinitely many, actually uncountable many).The decision involves only a single location s.Transport costs are proportional to the Euclidean distance.

Problem – minimize transport costs:Find the optimal location determined by the coordinates x*, y*that minimizes from the chosen location the transport costs to the sinks Op with the coordinates xp, yp (p=1,…r).

{ }rpoO p K,1, ==

Ss ∈*



Steiner-Weber Modell II

The objective function depends on:Volume (delivered or for delivery) from the given location p, ap,Normalized transport cost (e.g. EUR/ton/km) kt,Distance measures in terms of direct connection (Euclidean), dp.

Objective

∑∑==

==r

pppt

r

ppTT dakKK

11**

( ) ( )22ppp yyxxd −+−=

dp

xp

x

ypy

(xp, yp)

(x, y)



Steiner-Weber Modell III

Objective min:

Differentiation with respect to x and y and equating to zero yields the first order optimality condition.

Result: two nonlinear equations that require numerical means (e.g. Newton‘s iteration scheme)Fortunately, Excel does it! (via the Solver).

( ) ( )∑=

−+−=r

pppptTyx

yyxxakK1

22

,*min

( ) ( )

( ) ( )∑

∑

=

=

=−+−

−=

∂∂

=−+−

−=

∂∂

r

p pp

ppt

T

r

p pp

ppt

T

yyxx

yyak

yK

yyxx

xxak

xK

1

!

22

1

!

22

0)(

0)(



Steiner-Weber Modell IV | Increments

Nonlinear transport costs but distance depending, e.g., quadratic, or including set up costs for each connection: It can make sense to link the transport costs to particular sinks p to account for differences in deliveries (e.g., scrap versus sensitive final prducts):A further extension is to allow for the deliveries of multiple goods from the source (i.e., the sought location) s to sinks O.

Accounting for the last two recommendation leads to the following cost minimization:

pqkqq

qpT sink toproduct ofcost transport products,,1

:where

K=

( ) ( ) ∑∑ ∑∑== ==

−+−==q

qqpqpT

q

q

r

ppppqpqpT

r

pTyx

akyyxxdakK11 1

22

1,***min

( )ppTpT dkk =

( )pkk tt =

location decisions i - institut für...

Documents