Final Report IME

7/27/2019 Final Report IME

1/88

Integrated ManufacturingExecut ion FunctionalArchi tecture, Costs and Benefits

Alexander Schmidt, Dr. Boris Otto, Dr. Alfrid Kussmaul (EDS)

Report no.: BE HSG/ CC CDQ2 / 17Chair: Prof. Dr. H. sterleVersion: 1.0Date: October 15th, 2009

University of St. Gallen -for Business Administration, Economics,Law and Social Sciences (HSG)

Institute of Information ManagementMller-Friedberg-Strasse 8CH-9000 St. GallenSwitzerlandTel.: ++41 / 71 / 224 2420Fax: ++41 / 71 / 224 2777

Prof. Dr. A. BackProf. Dr. W. Brenner (managing)Prof. Dr. R. Jung

Prof. Dr. H. sterleProf. Dr. R. Winter


2/88


3/88

Table of Contents iv

HSG / IWI / CC CDQ2 / 17

Table of Contents

1 Introduct ion ......................................................................................................1

1.1 Motivation ......................................................................................................11.2 Structure of the Final Report .......................................................................... 3

2 Study Design .................................................................................................... 4

2.1 Research Approach ....................................................................................... 4

2.2 Participating Automobile Manufacturers ........................................................ 7

2.2.1 AUDI AG .................................................................................................. 7

2.2.2 BMW AG ..................................................................................................9

2.2.3

Daimler AG ............................................................................................ 10

2.2.4 Volkswagen AG ..................................................................................... 12

3 Background .................................................................................................... 13

3.1 Manufacturing Execution Systems ............................................................... 13

3.2 Existing MES Standards .............................................................................. 17

3.2.1 Manufacturing Execution System Association (MESA) .......................... 17

3.2.2 Normen-Arbeitsgemeinschaft (NAMUR) ................................................ 18

3.2.3 Verein Deutscher Ingenieure (VDI) ........................................................ 19

3.2.4 National Institute of Standards and Technology (NIST) ......................... 20

3.3 The Automotive Industry .............................................................................. 21

3.3.1 General Characteristics ......................................................................... 22

3.3.2 MES in the Automotive Industry ............................................................. 22

4 Study Findings ............................................................................................... 24

4.1 Current MES Strategy .................................................................................. 24

4.1.1 Strategic Goals ......................................................................................24

4.1.2 Organizational Embedding ..................................................................... 25

4.1.3 Application Landscape and Integration .................................................. 26

4.2 Functional MES Reference Architecture ...................................................... 30

4.2.1 Generic MES Function Map ................................................................... 31

4.2.2 MES Function Map Instantiations .......................................................... 36

4.2.3 Prioritization of MES Functions .............................................................. 39

4.2.4 Parameters Influencing Instantiation of the MES Function Map ............ 44

4.2.5 Non-Functional Requirements ............................................................... 47


4/88

Table of Contents v


4.3 MES Performance Management .................................................................. 50

4.3.1 Initial Business Benefits Framework ...................................................... 51

4.3.2 Business Benefits Framework Comprising Manufacturing Related

KPIs .......................................................................................................52

5 Summary and Outlook ................................................................................... 59

5.1 Study Results ...............................................................................................59

5.2 Critical Acclaim ............................................................................................61

5.3 Outlook on Future Challenges ..................................................................... 61

Literature .................................................................................................................64

Appendix A: Instantiations of the MES Funct ion Map...................................... 68

Appendix B: Detailed MES Functionali ty Defini tion ......................................... 71

B.1. Detailed MES Functionality Definition According to Standards

Investigated ................................................................................................. 71

B.2. Detailed Functionality Definition of the MES Function Map ......................... 76

Appendix C: Contact Persons ............................................................................ 79


5/88

List of Figures vi


List of Figures

Figure11:KeygoalsforintegrationofERPandshopfloor[AberdeenGroup2006,p.3]....3Figure

21:

Research

approach

of

the

IME

Study

.....................................................................

6

Figure31:MESasconnectorbetweenERPandshopfloor(basedon[Albert/Fuchs2007,p.

11,Louis/Alpar2007,p.246])......................................................................................14Figure32:Majorelementsoftheautomotivevaluechain....................................................23Figure41:Simplifiedapplicationlandscapesinassemblyplants.........................................28Figure42:Simplifiedapplicationlandscapesincomponentmanufacturingplants.............29Figure

43:

Generic

MES

Function

Map

...............................................................................

33

Figure44:MESFunctionMapasinstantiatedwithdetailedtasks.....................................37Figure45:FunctionstobecoveredbyMESapplications.....................................................39Figure46:MESFunctionsprioritizeddependingonnumberofassignedlayers................41Figure47:MESFunctionspotentialforimprovementofmanufacturingprocess..............43Figure 48:Differences inMESfunction assignment between componentmanufacturing

plantsand

assembly

plants

.............................................................................................

47

Figure49:EvaluationofcurrentMESsolutionsregardingnonfunctionalrequirements..49Figure410:ImportanceestimationofnonfunctionalrequirementsonMESsoftware.......50Figure411:ManufacturingrelatedKPIsusedbytheautomobilemanufacturers...............54Figure412:CauseeffectnetworkforMESKPIs..................................................................56FigureA1:InstantiationsoftheMESFunctionMap..........................................................70


6/88

List of Tables vii


List of Tables

Table21:MESexpertsinterviewedforquestionnairedesign.................................................6Table

22:

Key

figures

of

AUDI

AG

for

2008

...........................................................................

7

Table23:KeyfiguresofBMWAGfor2008...........................................................................9Table24:WorkshopparticipantsofBMW............................................................................10Table25:KeyfiguresofMBCandDaimlerAG(inbrackets)for2008................................11Table26:WorkshopparticipantsofMBC.............................................................................12Table27:KeyfiguresofVolkswagenAGfor2008................................................................12Table

31:

Organization

profile

of

MESA

..............................................................................

18

Table32:OrganizationprofileofNAMUR..........................................................................19Table33:OrganizationprofileoftheVDI.............................................................................20Table34:OrganizationprofileoftheNIST...........................................................................21Table41:MESfunctionsasspecifiedbydifferentstandards................................................32Table42:NonfunctionalrequirementsonMESsoftware...................................................48Table

43:

Initial

Business

Benefit

Framework

.......................................................................

52

Table44:BusinessBenefitsFrameworkcomprisingmanufacturingrelatedKPIs...............53TableB1:DetailedMESfunctionalitiesasdefinedininvestigatedMESstandards............75TableB2:DetailedMESfunctionality..................................................................................78


7/88

List of Abbreviations viii


List of Abbreviations

AG Aktiengesellschaft(Corporation)

AHM AudiHungariaMotorsKft.

BDN BenefitsDependencyNetwork

BMW BayerischeMotorenWerke

CIP ContinuousImprovementProcess

CNC ComputerizedNumericalControl

CoC CenterofCompetence

DCS DistributedControlSystem

DNC DistributedNumericalControl

EDS ElectronicDataSystemsCorporation

ERP EnterpriseResourcePlanning

HR HumanResources

IME IntegratedManufacturingExecution

IPS

InternationalProduction

System

ISA Industry,Systems,andAutomationSociety

IT InformationTechnology

IWI InstituteofInformationManagement

KPI KeyPerformanceIndicator

MBC MercedesBenzCars

MES

Manufacturing

Execution

Systems

MESA ManufacturingExecutionSolutionsAssociation

NAMUR NormenArbeitsgemeinschaft

NIST NationalInstituteforStandards

OEE OverallEquipmentEffectiveness

OEM OriginalEquipmentManufacturer

PDA ProductionDataAcquisition


8/88

List of Abbreviations ix


PLC ProgrammableLogicController

SCADA SupervisoryControlandDataAcquisition

SPC

Statistical

Process

Control

VDI VereinDeutscherIngenieure

VW Volkswagen

WIP WorkinProgress


9/88

Abstract x


Abstract

One of the challenges in manufacturing companies nowadays is to enable

appropriateITsupportofproductionplanningandexecution. Integratedsolutions

need to cover companywide and even crosscompanybusiness processes (from

order receipt to product distribution) and at the same time live up to the

technologicalcomplexityofmanufacturingprocessesontheshopfloor.Theproblem

is even aggravated in industries, such as the automotive industry, which are

characterizedbynumerous,stronglydivergingmanufacturingprocessesandhighly

versatile products. What such manufacturing companies need is an integrated,

consistent view along their entire value chain, allowing for optimal utilization of

capacitiesandclosingthegapbetweenbusinessandmanufacturingprocesses.

ThisfinalreportoftheIMEstudydescribesessentialcriteriaforfurtheroptimizing

manufacturing execution and control in manufacturing plants of the automotive

industry.BasedoncasestudiescarriedoutatfourOEMs,documentingthecurrent

statusof

MES

related

topics

in

the

respective

companies,

the

study

develops

a

functionalreferencearchitectureforMES tobeused in theautomotive industryas

well as a Business Benefits Framework allowing to measure the impact of an

integratedMESontheperformanceofmanufacturingprocesses.Fromtheseresults,

requirements on and recommendations for future, integrated Manufacturing

ExecutionSystemsarederived.


10/88

Introduction 1


1 Introduction

1.1 Motivation

One of the challenges in manufacturing companies nowadays is to enable

appropriateITsupportofproductionplanningandexecution. Integratedsolutions

need to cover companywide and even crosscompanybusiness processes (from

order receipt to product distribution) and at the same time live up to the

technologicalcomplexityofmanufacturingprocessesontheshopfloor.Difficulties

result from different levels of detail and accuracy regarding production status

informationneededondifferentcompanylevels.Theproblemisevenaggravatedin

industrieswhicharecharacterizedbynumerous,stronglydivergingmanufacturing

processesandhighlyversatileproducts.Thisisthecaseintheautomobileindustry,

typically involvingbatchproduction inpressplants,highlyautomatedproduction

linesforcarbodyconstruction,andassemblywithitstypicalrequirementsonload

balancing and documentation. What such manufacturing companies need is an

integrated, consistent view along their entire value chain, allowing for optimal

utilizationofcapacitiesbyhavingaccesstorealtimeinformationonmanufacturing

process,quality targetachievement,reworkcostsetc. [Klimm2008,p.4].Classical

ERP systems have proven tobe not capable of meeting this requirement, as they

provide only a coarse granular perspective on companywidebusiness processes.

Therefore,anewcategoryof informationsystems,calledManufacturingExecution

Systems (MES),hasemerged,allowing to consistently collectandprocessdataon

currentmachineandproductionstatuses.WhileparalleloperationofERPsystems

andMESseemsreasonable,anumberofquestionsdoarisethough:

WhatfunctionalscopeshouldMEScover?Whatisactuallymeantbytheterm

ManufacturingExecutionSystems?

Howcandifferentplanningandcontrollingfunctionsaswellasprocessesbe

assignedto

and

covered

by

ERP

systems

and

MES?

What

is

the

scope

each


11/88

Introduction 2


system is supposed to cover? How can functionalitiesbe assigned to the

layers(ERP,MESorShopFloor)inordergenerateminimumoverlap[Lampl

2009,

p.

99]?

How can informationbe smoothly exchangedbetween the various layers?

How can information systems involvedbe integrated to allow for optimal

informationexchange?

How can typical proprietary legacy systems on the shop floor level,which

originally were not designed for this purpose,be connected to the other

layers[Niemietzetal.2009,p.68]?

How can the impact of an additional MES layer on the efficiency and

effectivenessofthemanufacturingprocessbemeasured?

Are there single outofthebox software solutions that are capable of

coveringthediversityofmanufacturingspecificrequirements,particularlyin

anindustrywhichischaracterizedbyhighlydifferentproductionprocesses?

Howcouldoptimalapplicationsupportinsuchanindustrylooklike?

WithregardtointegrationoftheERPlayerandtheMESlayer,theAberdeenGroup

in 2006 conducted abenchmark study across 440 manufacturing companies from

different industries, revealing that the most urgent goals pursuedby integration

effortsareseamlessdataflowfromtheshopfloorlayertotheERPlayer,enterprise

wide access to production data, and improvement of product quality (see Figure

11).


12/88

Introduction 3


Figure11:KeygoalsforintegrationofERPandshopfloor[AberdeenGroup2006,p.3]

1.2 Structure of the Final Report

Afterthisintroductorysectioninwhichthestudymotivationisexpressed,Section2

isdedicatedtothedescriptionofthestudydesignincludingashortpresentationof

thefourparticipatingautomobilemanufacturerswiththecorrespondingworkshop

participants. Thereafter, Section 3 provides background information and the

conceptual foundation on central topics of the study, most notably MES and the

automotive

industry.

Section4presentsthemajorfindingsofthestudyaccordingtothefivesubjectareas

that were investigated. The first part (Section 4.1) is devoted to strategic goals,

organizational embedding, and application landscape. Section 4.2 then aims at

developing a reference architecture for MES functions as one of the pivotal goals

pursuedwith thisstudy.Thegeneric referencearchitecture (namedMESFunction

Map) isderived fromexistingMESstandards (Section4.2.1)and then instantiated

55

51

36

33

28

27

25

0 10 20 30 40 50 60

Pulldatafromtheshop floorprovidingvisibilitytoERP

andenterpriseapplications

Improveproductqualityandreducevariability

Enterprisewideaccesstomanufacturingorders,

inventory,etc.fromERPtoshop floor

Triggers/alertsfromexecutionsystemsignalto

planning/schedulingapplications

ProductcompletionsfromexecutionsystemtoERP

Providevisibilityintoplantfloorfinancials

Providedirectaccesstoproductgenealogyfrom

enterpriseapplicationstoshop floor

PrioritiesinBridgingtheGapBetweenERPandtheShopFloor

%


13/88

Study Design 4


and refined through the four case studies (Section 4.2.2). The instantiations show

functional requirements that the Original Equipment Manufacturers (OEMs) have

on

IT

solutions

for

production

planning

and

control.

Section

4.2.3

then

investigates

thepotentialimprovementstandardizedMESsolutionscouldgeneratewithregard

to cost, time, and quality. In Sections 4.2.4 and 4.2.5, we discuss some factors

influencing the instantiation of the MES reference architecture as well as

requirements which are not coveredby the MES Function Map (nonfunctional

requirements),but which were explicitly expressed asbeing important for MES

solutions.Thefifthsubjectarea,performancemeasuring,isdealtwithinSection4.3,

where we analyze KPIs usedby the OEMs for evaluating their manufacturing

processesinordertoilluminatepossibleeffectsofintegratedMESsolutionsonthe

performanceofthemanufacturingprocesses.

Finally,Section5summarizesthekeyresultsofthestudyandprovidesanoutlook

onfuturechallengesfortheautomotiveindustryaswellasforresearch.

2 Study Design

2.1 Research Approach

Thebasicintentionofthestudyistodescribeessentialcriteriaforfurtheroptimizing

manufacturingexecution inassemblyandcomponentmanufacturingplantsof the

automotive industry. Based on the identification of a functional MES reference

architecture (including adoption criteria for its application), the study aims at

formulating requirements on and developing recommendations for future

integrated Manufacturing Execution Systems. A reference architecture, or more

generallyspeakingareferencemodel,isagenericmodelofaunitofanalysisthatis

applicable as a template for similar cases in the same domain by adapting

predefinedconfigurationparameterstotheparticularsituationinwhichitused.We

willdescribe

reference

models

and

their

adaptation

mechanisms

in

more

detail

in


14/88

Study Design 5


later sections of this report when we present the functional MES reference

architecture and parameters influencing its instantiation (Section4.2). In addition,

the

study

evaluates

potential

benefits

of

an

integrated

Manufacturing

Execution

Systembydevelopingandapplyingaframeworkofbusinessbenefitsconsistingof

MESrelatedKPIs.

Thepivotalelementof thisstudyonIntegratedManufacturingExecution (IME) in

theautomobile industry is the recordingof case studies for the fourparticipating

automobile manufacturers, namely AUDI AG, BMW Group, Daimler AG and

Volkswagen AG1. Basically, case study research can pursue two different goals:

firstly, case studies can examine, describe and explain phenomena in a given

(business)contextinanexplorativemanner,secondly,casestudiesallowtotestand

developnew theories [Eisenhardt1989,p.533,Scholz/Tietje2002,pp.11f.].As the

IMEprojectaimsattheformer,ourcasestudiescanbedefinedasexplorative[Yin

2002, Specht et al. 2004] describing and investigating a complex area of science

[Meyer/KittelWegener 2002, p. 21] and trying to identify and explain

interdependencies or cause effect relations [Yin 2002, p. 15]. The study design is

characterized by an multicase studies approach as a total of four different

companiesareexaminedwithregardto thesametopic(MES)[Yin2002,pp.38ff.].

This leads to increased robustness and generalizability of findings, compared to

individualcasestudies[Benbasat1985,p.58].

Eachof the fourcase studiesdocuments thecurrent status regardingMES for the

respectiveOEM, taking intoaccount fivesubjectareas tobe investigated:strategic

goals,organizationalembedding,applicationlandscape,functionalarchitecture,and

performance measuring. In addition, the individual case studies and the findings

madethereinserveasabasisforthepresentstudyreport.

1

ThecompaniesparticipatingintheIMEstudyarepresentedinthefollowingsection.


15/88

Study Design 6


Figure21:ResearchapproachoftheIMEStudy

Figure21showstheoverallresearchprocesspursuedbytheIMEstudy.Basedon

extensiveliteraturereviewwedesignedaquestionnaireconsistingofbothopenand

closed questions serving as a guideline for the assessment workshops. The

questionnairewas reviewedand refined throughmultiple interviewswithexperts

having substantial experience in MES projects. The interrogated subject matter

expertsarenamedinTable21.

Name Company Department/Role

MichaelSchlecht SAPDeutschlandAG DirectorIndustrySolutions

TonyAschwanden SAPSchweizAG HeadofEMEAPresalesforManufacturingExecution

DieterWormuth EDS,anHPCompany ConsultantforProductionITandMES

Table21:MESexpertsinterviewedforquestionnairedesign

Theworkshopswerecarriedoutassemistructuredonsitefocusgroup interviews

[Cavana

et

al.

2001,

pp.

153

159]

with

varying

numbers

of

participants

(between

3

Preliminary Work

Work out general MES Process Map and Business BenefitFramework (BBF)

Align project scope

Design questionnaire

Initial Assessment and Data Collection

Conduct and document Assessment Workshops

Assess MES Process Map and BBF

Discuss improvement and cost saving potentials

Benefit Analysis

Analyze and calculate relevant KPIs

Consolidation of Results

Aggregate benefits; calculate savings and efficiencypotentials (Monetarize KPIs)

Write final report

Presentation and Handover of Results

Final presentation

Best Practice Exchange Workshop (multilateral)

Joint dissemination of project results

1

2

3

4

5


16/88

Study Design 7


and 12) from both IT and manufacturing departments (e.g. plant managers).

Additionally, we analyzed documents provided by the workshop participants,

which

complemented

the

information

gathered

during

the

interviews.

2.2 Participating Automobile Manufacturers

ThefollowingchapterpresentsthefourOEMsthatparticipatedintheIMEstudyas

wellastheobjectsofinvestigation(plants,divisionsetc.).Intherestofthepaperwe

refertotheseOEMsasautomobilemanufacturers.

2.2.1 AUDI AG

2.2.1.1 General Information

AUDI AG is a German automobile manufacturer headquartered in Ingolstadt,

Germany. It has been an almost whollyowned subsidiary (99.7 %) of the

Volkswagen Group since 1964. The company employs about 57,000 employees,

generatingarevenueofmorethan34billionEuros(2008).TheAudiGroupitselfis

subdivided in several national subsidiaries and manufactures cars in seven

internationalmanufacturingsites(IngolstadtandNeckarsulminGermany,Brussels

in Belgium, Gyr in Hungary, Changchun in China, Bratislava in Slovakia, and

AurangabadinIndia).Thecompanyskeyfiguresfortheyear2008aresummarized

inTable22.

AUDIAG

Numberofemployees 57,533

Annualrevenue(inmillionEuro) 34,196

Numberofmanufacturingsites 6

Numberofcountrieswithmanufacturingsites 5

Annualscostsofgoodssold(COGS)(inmillionEuro) 28,848

Manufacturingcosts(inmillionEuro) 28,478

Table22:KeyfiguresofAUDIAGfor2008


17/88

Study Design 8


2.2.1.2 Object of Investigation and Interview Partners

The case studyat Audi did not focuson a singleplant,but instead coveredboth

vehicle and component manufacturing. Consequently, the study covers the

manufacturing plants of AUDI AG in Ingolstadt (with an output of more than

550.000 cars per year and 32,000 employees in 2008), Neckarsulm (approximately

300.000carsperyearand13,000employeesin2008),andAudiHungariaMotorsKft.

(AHM) in Gyr, Hungary (almost 2 million engines and 60,000 cars with

approximately5,900employeesin2008).Ingolstadt,astheheadquartersiteandthe

location forboth technicaldevelopmentanddiversecorporate functions, isAudis

biggestplantwithregardtovehicleoutput.Duetotherelativelysmallfloorspaceof

themanufacturingsite,Ingolstadt typifiesthemodelofpermanentoptimizationof

theplantsmanufacturingandlogisticsprocessesincludingtheunderlyingIT.With

Ingolstadt and Neckarsulm being the two biggest manufacturing plants of the

company (with regard to yearly vehicle production), the important issue of MES

process and system integration is centered on these two sites. AHM is mainly a

component manufacturing plant. however, there is also a small part of vehicle

manufacturing (assembly), producing about 60,000 cars per year. With a total of

almost two million engines produced yearly, the plant is one of the worldwide

biggestenginemanufacturingsites.Duetoitssizeandthefactthatitencompasses

allbusiness functions (e.g. HR, Research & Development etc.), the Gyr location

differsfromothermanufacturingsitesofAudiandisofequalimportancewithinthe

company as the main manufacturing plants in Ingolstadt and Neckarsulm. From

AUDIAG,thefollowingrepresentativesparticipatedintheassessmentworkshops:

JrgGraf,headofITofAHM;

Christoph Lubkoll, responsible for Strategies of Process and System

Integration(CustomerOrderProcess);

EmeseKosar,headoftheProcess,IntegrationandInformationManagement

departmentwithintheITdepartmentofAHM;and


18/88

Study Design 9


Ambrus Zsolt, responsible for IT solutions supporting core manufacturing

processes, most notably the planning production process within the IT

department

of

AHM.

WithintheAudiGroup,AHMtakesauniquerole,asherebothhighvolumeengine

manufacturingandvehicleassemblyrepresentcorebusinessprocesses.Partlythese

divergingprocessesrequirespecificITsolutionsontheERPlayeraswellasonthe

MESlayer.FortheotherAudicomponentmanufacturingplants,thestandardMES

of Volkswagen, as the parent company, areauthoritative, whereas IT solutions at

AHMmaydifferfromthesestandardsbypartlyhavinginstalledlocal,proprietary

systemsinlinewithgoverningITprinciples.

2.2.2 BMW AG


TheBayerischeMotorenWerke(BMW)AGisaGermanautomobileandmotorcycle

manufacturing company, which was founded in 1916. It is headquartered in

Munich, Germany. The company employs approximately 100,000 employees

generatinganannualrevenueofmorethan53billionEuros(2008).TodaytheBMW

GroupistheparentcompanyoftheMINIbrandaswellasRollsRoyceMotorCars.

BMWmanufacturescars in24sitescovering13differentcountries.Thecompanys

keyfiguresfortheyear2008aresummarizedinTable23.

BMWAG






Manufacturingcosts(inmillionEuro) 26,727

Table23:KeyfiguresofBMWAGfor2008


19/88

Study Design 10



In contrast to the original intension of focusing on one single plant only, it was

jointly agreed to do a comprehensive, crossplant analysis of MES related topics.

Thisallowedus toobtainamorecomprehensiveviewon the issueofMESwithin

BMW and to identify and discuss differences between different plants of the

company. Therefore, we had a wide spectrum of workshop participants with

varyingbackgrounds and from different organizational units. The names of the

workshopparticipantsaswellastheirorganizationalassignmentaresummarizedin

Table24.

Name Role Department

RobertPeter CoCLeiterAnlagennaheSysteme EZ24

HaraldScheder ITPLNeue3erBaureihe EZ203

UrsulaRichter Bebauungsplanung,QSysteme EZ201

ThomasPriemuth CoCLeiterLogistischeInformationssysteme EZ241

JakobWersching LeiterSteuerungs Einrichtungs ProzesstechnikMontage TP401

AlbertSextl LeiterInstandhaltungKomponentenfertigung TA334

FranzZurl LeiterSteuerungstechnikKomponentenfertigung TA414

EdmundZuber LeiterAnlagentechnikKarosseriebau TP221

GeraldMeier KompentenzfeldleiterProzessdatenerfassung EZ240LA

RudolfHoefler Programmsteuerung(Mnchen) TM103

HubertPielmaier LeiterInstandhaltungOberflchenbehandlung(Dingolfing) TD311

DieterSchels ProduktionsnaheITMontage(Mnchen) TP401MU

Table24:WorkshopparticipantsofBMW

2.2.3 Daimler AG


Daimler AG is a German automobile manufacturer headquartered in Stuttgart,

Germany.The company employs approximately 273,000employees generating an

annual revenue of more than 95billion Euros (2008). Besidesbeing the worlds

thirteenth largest automobile manufacturer, Daimler is the worlds largest truck


20/88

Study Design 11


manufacturer.ItsdivisionMercedesBenzCars(MBC),onwhichthecasestudyhas

focused, employs approximately 97,000 people generating an annual revenue of

almost

48

billion

Euros

(2008).

Cars

are

manufactured

in

15

sites

covering

six

different countries. The mainbrands of the company are MercedesBenz, Smart,

Maybach,andMcLaren.

MBCskeyfiguresfortheyear2008aresummarizedinTable25.

MBC(DaimlerAG)

Numberofemployees 97,303(273,216)

Annualrevenue(inmillionEuro) 47,772(95,873)

Numberofmanufacturingsites 15(41)

Numberofcountrieswithmanufacturingsites 6(17)


Manufacturingcosts(inmillionEuro)

Table25:KeyfiguresofMBCandDaimlerAG(inbrackets)for2008


In contrast to the original intension of focusing on one single plant only, it was

jointlyagreedtodoacomprehensive,crossplantandcrossdivisionanalysisofMES

relatedtopics.Thecasestudycoversthreedifferentdivisionsofproductionwithin

MBC, namely manufacturing of complex components (engines, gears) for

automobiles (socalledPowertrain),assemblyofautomobiles,andvanproduction.

Therefore,thescopeofthecasestudyisrelativelywide,allowingforcomparisonof

the topics investigated between the different branches and, consequently,

identification and discussion of major discrepancies. Accordingly, findings are

describedforeachofthethreedivisions(Powertrain,Assembly,Van)duringtherest

of thecase study.Theworkshopbrought together representatives fromCoCsand

plantsofallthreebranches.Thenamesoftheworkshopparticipantsaswellastheir

rolesandorganizationalassignmentaresummarizedinTable26.


21/88

Study Design 12


Name Role Department

UweHaag SeniorManager

ITVANManufacturingPlants

HeadofCoCManufacturingControlandLogistics

InformationOfficerforplantsDsseldorfandLudwigsfelde

MarkusHawickhorst Manager ITVANManufacturingPlants

CoCDistributionSystems

GerhardSchiele SeniorManager ITAssemblyPlants

CoCManufacturingControl

ThomasKampfmann Manager ITAssemblyPlants

CoCManufacturingControl(mainlySindelfingen)

StefanRosenwald SeniorManager ITManagementPowertrain(componentplants)

CoCManufacturingControlandLogistics

Table26:WorkshopparticipantsofMBC

2.2.4 Volkswagen AG


Volkswagen (VW) AG is a German automobile manufacturer headquartered in

Wolfsburg,Germany.Withanannualrevenueof113.8billionEurosanda totalof

approximately 370,000 employees in 2008, the Volkswagen AG currently ranks

among the top threeautomobilemakers in theworldand is thebiggestEuropean

automobile manufacturer. It unites numerous automobile brands, among them

Audi, Bentley, Bugatti, Seat, and Skoda. Volkswagen AG currently operates 61

manufacturingsitesin21differentcountries.Thecompanyskeyfiguresfortheyear

2008aresummarizedinTable27.

VolkswagenAG






Manufacturingcosts(inmillionEuro)

Table27:KeyfiguresofVolkswagenAGfor2008


22/88

Background 13



The case study at VW focused on the component manufacturing plants of the

company. Components in this case cover the whole spectrum and include simple

components,suchaspressedorfoundryparts,aswellascomplexcomponents,such

as gears or engines. Within the company, the ITP Components department is

responsibleforITdevelopmentandmaintenanceofallcomponentplants.Fromthis

departmenttworepresentativesparticipatedintheassessmentworkshops:

HansChristianHeidecke,headofITPComponents;

IngoHfer,softwareengineeratITPComponents.

3 Background

3.1 Manufacturing Execution Systems

ManufacturingExecutionSystemsarearelativelynewclassofinformationsystems

designedparticularlytosupportshopfloorprocessesandtheirintegrationintothe

companys

information

system

architecture

[Louis/Alpar

2007,

p.

243].

MES

constitutetheinterfacebetweentheplanning(ERP)layerandtheproductionlayer.

Theyareanessentialcomponentforverticalintegration,asillustratedinFigure31.

The three layerscanalsobe referred toasCompanyManagement (forwhichERP

systemsarethemostcommontools),ProductionManagement(donebyMES),and

Production (supported by systems for machine control and acquisition of

manufacturing

data)

[Gnther

et

al.

2008,

p.

37].

The

latter

usually

contains

hybrid

hardware/software systems, such as Distributed Control Systems (DCS),

Programmable Logic Controllers (PLC), Distributed Numerical Control (DNC),

Supervisory Control and Data Acquisition (SCADA) systems, and other control

systemsdesigned toautomate theway inwhichproductsarebeingmanufactured

[MESA2000,p.1].


23/88

Background 14


Figure31:MESasconnectorbetweenERPandshopfloor(basedon[Albert/Fuchs2007,p.

11,Louis/Alpar2007,p.246])

In contrast to ERP systems, which generally provide a verybroad functionality

coveringallbusinessfunctionsofanenterprisealong itsoperationalsupplychain,

MES aim at enabling companies to quickly respond to events occurring in the

production process (reactive detailed planning). MES take a microscopic, more

granularviewonproductiondata (often restricted toasingleplantorproduction

area),comparedtothemacroscopic,holisticviewofERPsystems,andthereforeare

intended to compensate one of the main shortcomingsof ERP system production

modules: the incapabilityofproviding integrationofrealtimemanufacturingdata

generatedontheshopfloor[Wannenwetsch/Nicolai2004,p.139].Thisincapability

basically results from an inadequacy of ERP production plans to respond to

changing demands or deviations in the manufacturing process. Neither are these

systemscapableofhandling theenormousamountofdatacoming from the shop

floor,nordo theyprovideshort response timesandsufficient levelsofdetail (e.g.

withregardtothemodellingoftheproductionprocess)[Louis/Alpar2007,p.243].

Itis

this

gap

that

MES

are

trying

to

fill.

Production / Automation Systems

Manufacturing

Execution Systems(MES)

ERP

Production (Program)Planning, Master Data

Maintenance

Detailed ResourcePlanning & Allocation,

Production Monitoring,Data Collection, KPIs

Execution, Production

Logistics

Current productiondataPlan variance

Planning data andrestrictions

Production DataAcquisition (PDA)

Reactions onincidents during

production

Business

Partners

Ente

rprisewide

Domain

specific

Levelof

Detail

Planning

Horizon


24/88

Background 15


As MES in the past have notbeen subject of extensive scientific research (some

exceptionsbeing the recentworksofKLETTI [Kletti2006],SAUER [Sauer2004]and

SCHFER

ET

AL.

[Schfer

et

al.

2009]),

a

well

established

definition

of

the

term

has

not

beengivensofar.However, thereare leadingstandardizationorganizationsin the

domain of manufacturing integration, most notably the Industry, Systems, and

Automation Society (ISA) and the Manufacturing Execution Solutions Association

(MESA),thathaveputsomeeffortintofindingacommondefinitionandspecifying

genericMESfunctionality(cf.[ISA2000,ISA2005]and[MESA2000,MESA2004]).2

SoMESaredefinedassystemsthatdeliverinformationenablingtheoptimization

of production activities from order launch to finished goods. Using current and

accuraterealtimedata,MESguide,respondto,andreportonplantactivitiesasthey

occur.Theresultingrapidresponsetochangingconditions,coupledwithafocuson

reducingnonvalueaddedactivities,driveseffectiveplantoperationandprocesses.

[MESA2000,p.1].ThisdefinitionimpliesthefollowingcharacteristicsofMES:

highlevelofdetail(dataacquisitionfrommanufacturingprocesses),

relativelyshortplanninghorizon(reactiveplanning),

bidirectional communication toboth ERP systems and shop floor systems

(interfacing).

TheultimategoalofMEScanthereforebedescribedasincreasingtransparencyon

the manufacturing process and, as a result, establishing horizontal and vertical

(closed) control loops [Kletti 2006, p. 11]. These control loops allow for prompt

reaction to incidents on the shop floor as information is directly fed back to

overlyingplanningsystems(suchasERP)totriggerrespectivemeasuresaswellas

tosubsequentmanufacturingsteps(horizontalintegration).

2

Amoredetailedpresentationofthestandardsisincludedinthefollowingsection.


25/88

Background 16


A major challenge with regard to the model shown in Figure 31 lies in a clear

demarcation of the three layers3. This is a difficult task, as certain enterprise

functions

may

be

supported

by

a

number

of

information

systems

located

in

more

than one layer (e.g. quality management by ERP and by MES applications,

production data acquisitionby control systems on the shop floor andby MES

applications), leading to a high degree of interconnectionbetween the systems.

Nevertheless,acleardistinctionappearsuseful,asthesystemsdifferregardingthe

degreetowhichtheysupportfunctionalityformanufacturingplanningcomparedto

manufacturing execution. We comply with the above given definition by

distinguishing the three layers ERP, MES, and Shop Floor mainlybased on two

parameters(seeFigure31):theplanninghorizon,i.e.theperiodoftimeforwhich

differenttasksarescheduled,andthelevelofdetailoftheinformationmanaged.By

rule of thumb, ERP systems cover the mid and longterm planning for a time

horizonofatleastoneday(uptoseveralweeksormonths),MEShandleproduction

planninginformationrangingfromonehouruptooneday,andontheShopFloor

layertimeintervalsscaledown tothelevelofseveralminutes.Aseveryminuteof

productionstopduetomachineortoolfailuredirectlyleadstolossofmoney,rapid,

adhoc decisions need tobe supported in production management and execution

[Kletti2006,p.11].

We would like to emphasize that our understanding of the term MES is not

limited to thegeneric term fora typeofcommerciallyavailableor selfdeveloped

software,but also includes the functions that are necessary for (more) efficiently

managing the manufacturing process and for establishing the link connecting

commercial order processing at enterprise management level (ERP) with the

3

The problem of demarcating the three layers with regard to their functions is also addressed in this study anddiscussed in more detail with the help of the MES Function Map in chapter 4.2. This part of the study aims at

assigning functionality to the different layers and contributes to a more clear differentiation of ERP, MES and

Shop Floor in the automotive industry.


26/88

Background 17


operationmanagementlevel(ShopFloor)[NAMUR2003,p.6].Fortheremainderof

thereportwethereforeexplicitlydifferentiatebetweenMESfunctionalityandMES

applications,

the

latter

referring

to

software

operated

to

support

MES

related

tasks.

3.2 Existing MES Standards

Asmentioned in theprevious chapter, several standardizationorganizationshave

put some effort into finding a common definition and specifying generic MES

functionality.Thestandardizationorganizationsandtheresultingspecificationsare

brieflysummarizedinthefollowingsubsectionsservingasaconceptualfoundation

forfurther

work

(see

Section

4.2).

Within its S95 standard, ISA specifies four core functionality categories of MES:

Production Management, Inventory Management, Quality Management and

Maintenance Operations Management [ISA 2000]. Each of the four categories is

furthersubdividedintoeightfunctiongroupsandcanserveasabasistodefinethe

functionalscopeoftheMESlayer.However,theISAstandardfocusesmoreonthe

data

models

of

and

interfaces

between

the

three

layers

depicted

in

Figure

3

1,

i.e.

on

verticalintegration.Asitdoesnotprovideconcretefunctionaldefinitions,wedonot

includeitinoursynthesisofMESfunctions.

3.2.1 Manufacturing Execution System Association (MESA)

The Manufacturing Execution System Association is an American industry

association headquartered in Chandler, Arizona. Being established in 1992 by

different

software

companies,

MESA

is

maybe

the

most

experienced

organization

in

thefieldofMESwithitsstandardshavingreachedworldwideacceptance.However,

MESAs guidelines are still strongly influenced by the MES software vendors

perspective.


27/88

Background 18


ManufacturingExecutionSystemAssociation(MESA)

Foundation 1992Headquarters Chandler,Arizona(USA)Relevantindustries AllmanufacturingindustriesScope International; MESonlyMembers Softwarecompanies,usercompanies,andmanufacturingprofessionalsHomepage www.mesa.org

Table31:OrganizationprofileofMESA

The MESA Guidelines, as the most comprehensive MES standard, follow a

pragmaticapproachandidentifieselevenstandardMESfunctionsonafairlycoarse

granularlevel[MESA2000,p.1]:

ResourceAllocationandScheduling,

Operations/DetailScheduling,

DispatchingProductionUnits,

DocumentControl,

DataCollection/Acquisition,

LabourManagement,

QualityManagement,

ProcessManagement,

MaintenanceManagement,

ProductTrackingandGenealogy,

PerformanceAnalysis.

Moreover, with the new concept of a Collaborative MES, MESA is actively

propagating the vision of MES as a companys central data and information hub

[MESA2004].

3.2.2 Normen-Arbeitsgemeinschaft (NAMUR)

NAMUR is a European organization for users of automation technology in the

process industry (Interessengemeinschaft Automatisierungstechnik der

Prozessindustrie).Itsmainfocusisonthechemicalandpharmaceuticalindustry.It


28/88

Background 19


promotes the exchange of knowhow and experiencebetween its members and

designs guidelines in the form of recommendations and worksheets. NAMUR

strongly

relies

on

the

ISA

S95

standard.

NormenArbeitsgemeinschaft(NAMUR)

Foundation 1949Headquarters BadNeuenahr(Germany)Relevantindustries Processindustry(chemical and pharmaceutical industry)Scope European(Germanatitsorigin);notlimitedtoMESMembers UsercompaniesHomepage www.namur.de

Table32:OrganizationprofileofNAMUR

ThemainfunctionalblocksdefinedindetailbyNAMURare:

OperationsneutralProductionPlanning,

DetailedProductionPlanning,

ProductionManagement,

QualityManagement,

StockManagement&MaterialFlowManagement,

ProductionDocumentation.

3.2.3 Verein Deutscher Ingenieure (VDI)

The VDI is Germanys biggest association of engineers and natural scientists,

representing their interests in politics and society. It operates a number of expert

panelswhoworkoutguidelinesfordifferenttopicsofinterest.In2004,VDIstarted

todevelopproperguidelines forMESbasedon the standardspresentedbeforeas

wellasonrecentexperiencesandmarkettrends.


29/88

Background 20


VereinDeutscherIngenieure(VDI)

Foundation 1856Headquarters Dsseldorf(Germany)Relevantindustries AllmanufacturingindustriesScope Germany;notlimitedtoMESMembers EngineersandnaturalscientistsHomepage www.vdi.de

Table33:OrganizationprofileoftheVDI

The5600Guideline focusesonanunambiguousdefinitionof the termMESand

thedistinctionbetweendifferentmanufacturingtypes.Functionsdefinedare:

DetailedPlanning,

EquipmentManagement,

ResourceManagement,

PersonnelManagement,

DataAcquisitionandProcessing,

InterfaceManagement,

PerformanceAnalysis,

QualityManagement,

InformationManagement.

3.2.4 National Inst itute of Standards and Technology (NIST)

TheNISTisameasurementstandardslaboratoryasanonregulatoryagencyofthe

UnitedStatesDepartmentofCommerce. Itsmission is topromoteU.S. innovation

and industrialcompetitivenessbyadvancingmeasurementscience,standards,and

technology.Inthepast,theNISThasactivelybeenworkingonvariousissues,such

asdevelopmentofadistributedshopfloorcontrolarchitecture,developmentofan

activity model of the technical aspects of the product manufacturing process for

discretemetalpartsetc.[Barkmeyeretal.1999]


30/88

Background 21


NationalInstituteofStandardsandTechnology(NIST)

Foundation 1901Headquarters Gaithersburg,Maryland&Boulder,Colorado(USA)Relevantindustries AllmanufacturingindustriesScope USA;notlimitedtoMESMembers GovernmentalrepresentativesHomepage www.nist.gov

Table34:OrganizationprofileoftheNIST

BasedontheMESAstandard,theNISTdefinesthefollowingMESfunctions:

ResourceAllocationandTracking,

Operations/DetailScheduling,

ProductionUnitsDispatching,

SpecificationManagement,

DataCollection/Acquisition,

LabourManagement,

QualityManagement,

ProcessManagement,

MaintenanceManagement,

ProductTrackingandGenealogy,

PerformanceAnalysis,

MaterialManagement.

3.3 The Automot ive Industry

While the definition and illumination of MES in the previous chapter hasbeen

rathergeneralandindependentofanyspecificindustry,itisimportanttonotethat

requirementsonMESdiffersignificantlydependingontheindustryandthetypeof

production given (e.g. process manufacturing in the pharmaceutical industry in

contrasttodiscretemanufacturinginthecomputerindustry).Wewillthereforetake

acloserlookonthecharacteristicsandrequirementsoftheautomotiveindustry.


31/88

Background 22


3.3.1 General Characteristics

The study at hand focuses on the automotive industry. Accordingly, the topic is

investigatedandanalyzedagainst thebackgroundofautomotive industryspecific

characteristics. In general terms, automobile manufacturers currently suffer from

enormouscostpressure,which ispartlydue tostrongpricecompetition,buteven

more to the current economic situation in the automotive market, which is

characterizedbyexcesscapacitiesnecessitatingreductionofproductionoutput.This

costpressurehasadirectimpactonthemanufacturingintermsofdemandingmore

efficient and leaner processes. Furthermore, automobile manufacturing is

characterizedby shortdelivery times,versatileproduction (i.e.numerousproduct

variants), and shortterm change requests. For the manufacturers this means they

needuptodatestatusinformationontheproductionprocessinordertobeableto

reactatshortnotice.

3.3.2 MES in the Automotive Industry

As already suggested in Section 1.1, the automotive industry is characterizedby

numerous, stronglydivergingmanufacturingprocesses:batchproduction inpress

plants,highlyautomatedproductionlinesforcarbodyconstruction,andassembly

with its typical requirements on loadbalancing and documentation. Particularly,

adequate support of different production process types (batch production for

manufacture of parts and simple components, flow or continuous production4 in

assembly,andamixtureofbothinthemanufacturingofcomplexcomponents,such

as engines) constitutes a crucial challenge for OEMs. This heterogeneity of

manufacturing processes is directly reflected on the application level leading to

numerous isolated applications and, thus, to difficulties ensuringboth horizontal

integrationalongtheproductionprocessandverticalintegrationacrossthedifferent

4

Thetermsflowandcontinuousproductionareusedsynonymouslyinthisreport.


32/88

Background 23


layers.Thishas tobeborne inmindwhenevaluatingcurrently implementedMES

andconsideringfutureimprovementpossibilities.

Figure32:Majorelementsoftheautomotivevaluechain

AsdepictedinFigure32MESanditsfunctionalarchitecturedoesnotonlydepend

onthetypeofproductionandthemanufacturedproduct,butisalsoembeddedina

broader architectural framework. It includes most notably the corresponding

processarchitecture,i.e.the(manufacturing)processesinwhichthefunctionalityis

used,aswellastheapplicationarchitectureresultingfromthedistributionofMES

functionstodifferentapplicationsystems.ThedesignofMESwithintheautomobile

manufacturing companies is in strongly influenced by the interdependencies

betweenthesearchitecturallayers.

Functions

Processes

ValueChain

Production

Unit

Systems

Requirements

PlanningGross Planning

Detailed

Planning

Quality

Management

(Production)

Inventory

Management

Material

Requirements

Planning

Production Data

Acquisition

(PDA)

Machine Control

Takeover of

Requirements

Information

Manufacturing

Execution /

Control

Resource/

Equipment

Management

Dynamic RoutingTraceability /

Genealogy

Production

Reporting and

Analysis

MES Master Data Management

Tier2 Tier1 OEM

Customer

Order

Process

Production

Process

Service

Process

Supporting

Processes...

SAP

APO

SAP

PP

MES1

MES2MES2

PS1

PS2PS3 PS4

ERP

MES

Production/AutomationSystems

Part Component VehicleRaw

Material


33/88

Study Findings 24


4 Study Findings

4.1 Current MES Strategy

Asdescribedin theprevioussection,MESisgenerallyconsideredasanintegrated

process and IT topic within the automobile manufacturing companies, and is

therefore not managed separatelybut within the overall process and application

architecture. Accordingly, explicit MES strategies do not exist. Guidelines for the

longtermaswellasthemid andshorttermdevelopmentofMESrelatedprocesses

andapplicationshavebeenderivedfromtheITstrategy(whichisthencascadedto

thedifferent

core

business

areas,

such

as

manufacturing)

and/or

from

strategic

application maps. The strategic application maps contain the tobe application

architecture,indicatingwhichfunctionalityistobesupportedbywhichapplication,

and theyalso contain the roadmap forallmanufacturing relatedapplications (for

theERPlayerandfortheMESlayer).ITdepartmentsareurgedtoevolvetheirarea

ofresponsibilityaccordingtotheseconstructionplans.

Generally

speaking,

automotive

companies

are

subject

to

high

cost

pressure,

mainly

due to the economic situation in the automotive market, which is currently

experiencing a dramatic downturn. This cost pressure is passed over to

manufacturingaswellastoITdepartments.

4.1.1 Strategic Goals

AsdiscreteMESstrategieshavenotbeendefinedbytheautomobilemanufacturers,

the

top

strategic

goals

for

MES

have

not

been

specified

explicitly.

However,

the

beforementionedITstrategiesandapplicationconstructionplansdofollowgeneral

strategicobjectives,whichalsoapplyfortheissueofMES.Theseobjectivesinclude:

meetingtherequirementsofthevaluechainprocesses(production,customer

order);

achieving operating reliability and ensure robustness to achieve 100%

availability

of

the

manufacturing

processes;


34/88

Study Findings 25


reducingcomplexityandcost(particularlyITcost)by

a) consolidatingtheapplicationlandscapeanddecreasingthenumberof

applications;

b) intelligentstandardizationwherepossibleandprofitable,inorderto

increasereuseofsoftwarefunctionality;

No perse best of breed approach for MES, but rather casespecific

evaluation.

Ingeneral,theassessmentworkshopsshowedthatstandardizationispivotalwithin

allcompanies,particularlywithregardtotheissueofMES.Here,thechallengeisto

reachmaximumstandardizationdespiteaconsiderablespecificityoftasks,whichis

increasing the nearer these tasks are located to the shop floor. This problem

particularly effects component manufacturing, where the production process is

characterizedbyahighdegreeofvarianceandcomplexity,asproductsoftenchange

and the manufacturing process is disrupted quite frequently, making

standardization of manufacturing processes and supporting applications more

difficult.

4.1.2 Organizational Embedding

AsMESisoftenviewedfromarathertechnicalperspective,focusingonapplications

that support (parts of) the manufacturing process, responsibility for the subject is

generally assigned to IT departments. This is also the case in the four companies

examined.However,automobilemanufacturersdoemphasize theprocessviewby

assigningthedesignandimplementationofproductionprocessestothedepartment

responsible for MES, allowing them to develop MES solutions according to the

specificrequirementsoftheproductionprocess.

Moreover, the strategic significance of manufacturing processes entails that the

automobilemanufacturersaimatmanaging the issueofMESonaglobal level in

order to achieve maximum homogeneity among all plants worldwide.


35/88

Study Findings 26


Organizational units appointed to deal with the subject are supposed to be

commonly responsible for aligning the application landscape to process

requirements,

i.e.

they

are

expected

to

develop

guidelines

for

optimal

application

supportofbusinessprocesses.Theseguidelinesare then implementedby the local

ITdepartmentofeachmanufacturingplant.Thegloballyresponsibleorganizational

units do not necessarily focus on MES solely,but are supposed to design the

application landscapeasawhole, including systemson theERP layerandon the

ShopFloorlayer.Duetothedifferencesintheproductionprocess(batchproduction

versusflowproduction),inmostcasestheautomobilemanufacturershaveseparate

unitsforassemblyandcomponentmanufacturingplants.

Some of the automobile manufacturers have implemented socalled Centers of

Competence (CoC)orboards,whichdonotjustmanage the topicgloballybutare

composedofrepresentatives fromvariousplants,divisions(manufacturing,IT),or

even production process types (component manufacturing, assembly). These

organizational units aim at fostering standardization and minimizing the gap

betweencomponentmanufacturingandassembly(asfaraspossible).

4.1.3 Appl ication Landscape and Integration

Concerning automobile manufacturers applications on the MES layer, current

efforts are commonly characterized by the will to consolidate and harmonize

existingapplicationlandscapesinordertoreduceITcosts.Anaveragenumberof70

applicationsbeingoperatedontheMESlayerconstitutesaseriouscomplexitydriver

for integration and, as a consequence, enhances IT costs. Generally, the study

revealed significant discrepancies between assembly and component

manufacturing,whichcanmainlybeattributedtothemanufacturingcharacteristics

given. Being characterized by a high degree of variance and very specific

manufacturing tasks, standardization (of both processes and applications) in

component manufacturing is much harder to achieve than standardization in


36/88

Study Findings 27


assembly. Component manufacturing ranges from relatively simple components

(pressed parts, cast parts etc.) to very complex components (such as engines or

gears),

and

in

many

cases

specialized

applications

are

needed

to

support

complicatedmanufacturingtasks.Accordingly,MESapplicationlandscapestendto

bemoreheterogeneousincomponentmanufacturingplants.

In the manufacturers assembly plants one to maximum three proprietary

applicationscoverallcorefunctionalitiesrequiredontheMESlayer(seeFigure41),

except for very specific tasks, such as screwing technology or testing equipment,

thatareclosetotheshopfloor.Applicationsareeitherselfdevelopedordeveloped

with external partners, and they cover areas such as production logistics and

control, quality management, and machine control. In one case the MES solution

implementedconsistsofseveralfunctionalmodules,whichcanbecombinedflexibly

andallowfortailormadeadaptationtoplantspecificrequirements.Indoingso,this

OEM isable to reuseMES functionalityacrossdifferentplants, thereby increasing

the degree of standardization and at the same time reducing costs for software

developmentandmaintenance.


37/88

Study Findings 28


Figure41:Simplifiedapplicationlandscapesinassemblyplants

Inautomobilemanufacturerscomponentplants,fourtosixselfdevelopedsystems

areoperated,coveringdifferent functionalities (seeFigure42).Forexample, there

areseparateapplications tosupport the (detailed)planningprocess,connection to

thecontrolsystemontheshopfloor,qualitymanagement,orproductionreporting.

BythisthepossibilityofaccomplishingintegratedMESisaggravated,asdataiskept

separately and in different formats. Furthermore, a large number of applications

generallyleadstohigheroperatingITcosts.Onlyinonecaseanewlyimplemented

MES(towhichnewfunctionalityisconstantlyadded)coversallcoreMESfunctions

andcanbedescribedasintegrated.

ERPLayer

MES

Man.

Control

MES

QM

MES

Machine

Control

SAP

ERPLegacy

SF

Control

Systems

ShopFloor

ERPLayer

MESInformationSystem

SAP

ERPLegacy

MESLayer

MESLayer

ShopFloor

ERPLayer

MESManufacturingControl

SAP

ERPLegacy

MESLayer

ShopFloor

MESReporting

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

StandardApplication

NonStandard,Proprietary

Application


38/88

Study Findings 29


Figure42:Simplifiedapplicationlandscapesincomponentmanufacturingplants

In most of the automobile manufacturing companies integration of different MES

applications is accomplished through socalled telegrams (i.e. messagebased),

which are standardized with regard to format and structure. Specification of the

corresponding interfaces is mandatory for integrating newly acquired, additional

applications(e.g.fromcommercialsoftwarevendors).

Ascanbederived from the findingsdescribed,MESapplicationsaremostlynon

standardized, proprietary. They were either selfdeveloped or developed in close

collaborationwithexternal softwarepartners (marked in lightgreen inFigure41

and Figure 42). This canbe explained mainlyby the automobile manufacturers

desire to support the manufacturing process with its specific requirements in the

bestpossiblewayandtointegratewiththeexisting,historicallygrownapplication

landscape. Moreover, in most cases the applications currently operated were

ERPLayer

MES

Man.

Control

MES

Re

porting

MES

QM

MES

Detailed

Planning

ShopFloor

ERPLayer

MESLayer

MESLayer

ShopFloor

ERPLayer

MESInformationSystem

MESLayer

ShopFloor

MES

JIT/JIS

MES

Man.

Control

MES

Re

porting

MES

QM

MES/SF

Connect

SAP

ERPLegacy

SAP

ERPLegacy

SAP

ERPLegacy

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

SF

Control

Systems

StandardApplication

NonStandard,Proprietary

Application


39/88

Study Findings 30


developed severalyearsago leaving thepotentialofnew,uptodate technologies

unexploited.

Commercial

software

vendors,

in

turn,

have

ignored

the

need

for

software

support

betweentheERPlayerand theShopFloorlayerfora longtime,which iswhythe

automobilemanufacturershavebeguntodevelopappropriateapplicationsontheir

own. Although all four OEMs have always recognized the advantages of a

standardizedandintegratedMES(whichismostnotablycharacterizedbyreduced

effort for integration and maintenance as well as increased flexibility), they have

oftenexpressedtheirscepticismregardingfeasibility.

On the ERP layer the degree of standardization is considerably higher. Legacy

systems are being increasingly substituted by standardized, commercial ERP

systems,withSAPbeingthedominantsoftwareprovider(ERP,APO).

4.2 Functional MES Reference Architecture

Reference

models

(and

consequently

architectures

as

specific

models

as

well)

are

characterizedbythreeattributes[Fettke/Loos2004]:Universalapplicabilitydenotes

thepossibility todeployareferencemodel inmore thanonespecificorganization.

This, in turn, fosters reusability meaning that generic conceptual patterns canbe

used againby simply applying predefined adaptation mechanisms reducing the

effortforredevelopment.Finally,referencemodelscontainbestpracticesproviding

recommendations for conducting business. Basically, reference models can be

derivedeitherbygeneralizing findings fromanumberof investigatedcasesorby

adaptinganexistingreferencemodeltoparticularrequirements[Beckeretal.2002,

p.49].Inourstudy,wepursuethefirstapproachbasedonthefourcasestudies.The

functionalMESreferencearchitectureisrepresentedbythesocalledMESFunction

Map.


40/88

Study Findings 31


4.2.1 Generic MES Funct ion Map

TheMESFunctionMapvisualizesdifferentbusinessandmanufacturing functions

andassignsthemtooneofthethreelayers,namelyERP,MES,andShopFloor,or,

morepreciselyspeaking,tothecorrespondingapplicationsassignedtotheselayers.

Itcanservetwoobjectives:firstly,itcanbeusedasameansforcommunication(as

doneinourprojectduringtheassessmentworkshops);secondly,wheninstantiated

theFunctionMapcanbedeployedtodesignorrefinetheapplicationarchitectureby

assigningsoftwarecomponentstothefunctionmapped.

TheFunctionMapwasinitiallydevelopedonthebasisofaliteraturereviewofMES

relatedjournalandconferencearticlesfrombothresearchandbusinesspracticeas

wellasestablishedMESstandards(seeSection3.2),suchasISAS95[ISA2000]and

MESA [MESA 2000, MESA 2004]. Additionally, we analyzed MES related white

papers of more regional standardizationbodies, namely the National Institute of

Standards(NIST,[Barkmeyeretal.1999])intheUnitedStatesofAmericaaswellas

the 5600 Directive of the VDI [VDI 2007] and a guideline publishedby NAMUR

[NAMUR2003]inGermany.

From the specifications of the standardization organizations presented we have

derivedasynthesisofrelevantMESfunctions,whicharedepictedinTable41.The

footnotesrefertotermsusedbythestandardsthatdeviatefromthefunctionnames

usedwithintheIMEproject.

MESA

[MESA2000]

NAMUR

[NAMUR2003]

VDI

[VDI2007]

NIST

[Barkmeyer

etal.1999]

LabourManagement X X

(Material)RequirementsPlanning X1

GrossPlanning X1

DetailedPlanning X X X X

QualityManagement X X X X2

ProductionInventoryManagement X X3 X

ResourceManagement X5 X

4 X

5


41/88

Study Findings 32


MESA

[MESA2000]

NAMUR

[NAMUR2003]

VDI

[VDI2007]

NIST

[Barkmeyer

etal.1999]

EquipmentManagement/

MaintenanceX6 X X

6

ManufacturingExecution/Control X X

Traceability/Genealogy X X

ProductionReportingandAnalysis X7 X X

7

MachineControl X

ProductionDataAcquisition X X X

MasterDataManagement X8 X

8 X

9 X

1OperationneutralProductionPlanning

6MaintenanceManagement

2

QualityAnalysis7

PerformanceAnalysis3StockManagement

8DocumentControl

4MaterialManagement

9InformationManagement

5ResourceAllocationandTracking

Table41:MESfunctionsasspecifiedbydifferentstandards5

BeyondameredefinitionofMEStasks,theVDI5600Guidelinealsoidentifiesand

characterizesobjectstobemanagedontheMESlayer(suchasresourcesordata)and

not

on

the

ERP

layer

or

the

Shop

Floor

layer.

ForthedesignanddevelopmentofanMESfunctionalarchitectureintheautomotive

industrytheinvestigatedstandardswereuniformlyestimatedasinexpedientdueto

theinsufficientdegreeofdetailoftheprovideddefinitionsandthelackingfocuson

specificrequirementsoftheautomotiveindustrybytheinterviewedexperts.

WithregardtoourgoalofderivingafunctionalMESreferencearchitecture,wehave

used the specifications of these standards as a starting point for a more detailed

definitionofMESand related functions,which later resulted in theMESFunction

Map. The generic MES Function Map, as illustrated in Figure 43, comprises the

three layers already specified, namely ERP, MES, and Shop Floor. To each layer,

correspondingprocessesor functionsareassigned.The firstclusterof functions is

5

Amoredetaileddescriptionofthefunctionswithsingletasksforeachofthestandardscanbefound

inAppendixB.1.


42/88

Study Findings 33


thecategoryoftypicalbusinessfunctions,suchassalesanddistributionormaterials

management,whichisassignedtotheERPlayer.TotheShopFloorlayerweassign

functions

that

are

directly

concerned

with

the

control

of

machinery,

such

as

Distributed Control Systems, Remote Terminal Units and Programmable Logic

Controller.

Figure43:GenericMESFunctionMap

The MES layer comprises typical functions for production planning and

manufacturing control, such as Product Traceability and Genealogy, or Dynamic

Routing.

Dynamic

Routing

was

added

although

it

is

not

an

element

of

the

MES

standardsinvestigated,asithasrecentlybeenamuchpropagatedfunctionoffered

by commercial MES software vendors, such as SAP. Moreover, our initial expert

interviewscarriedouttoverifytheMESFunctionMap(seeTable21)confirmedthe

potentialofthisfunctionforcostandtimesavingsinthemanufacturingprocess.

Themainfunctionsareshortlydescribedinthefollowing.Moredetaileddefinitions

foreachfunctionareprovidedinAppendixB.

Shop

FloorLayer

MESLayer

ERPLayer

DNC/CNC

ProgrammeControl

Supervisory,

Control and Data

Acquisition(SCADA)

Remote

Terminal Units(RTU)

Distributed

Control Systems

(DCS)

Programmable

Logic Controller(PLC)

(Internal) Cost

Allocation

ERP Business

Functions

CoreMESFunctionalities

Gross PlanningDetailed

Planning

Quality

Management

(Production)

Inventory

Management

(Material)

Requirements

Planning

Production Data

Acquis ition

(PDA)

Machine Control

Potentially RedundantFunctionalities CompanySpecific Implementation NotwithinMESScope

Resource

Management

Manufacturing

Execution /

Control

Equipment

Management /

Maintenance

Dynamic RoutingTraceability /

Genealogy

Production

Reporting and

Analysis


Labour

Management


43/88

Study Findings 34


LabourManagementprovidesinformationonthestatusofpersonnelinanup

totheminute time frame including time logging, attendance reporting,

qualification

documentation,

and

staff

scheduling.

GrossPlanningderivesprimaryrequirementsfrom the totalcustomerorders

receivedandfromforecastsmadeduringsalesandrequirementsplanning.It

isalsoreferredtoasRoughProductionPlanning.

Detailed Planning takes over the production requirements from Gross

Planning and derives production orders under consideration of production

restrictionswithdetailedtimescheduling(sequencing)andjobsizes(concrete

productionplan/schedule).

Quality Management provides timely analysis of product and process

measuring taken frommanufacturingoperations inorder toensureproduct

quality control, including quality/inspection planning, inspection execution

anddocumentation,andmanagementoftestequipment.

Production Inventory Management, also referred to as Stock Management,

documents all current and available stocks of material (input material,

intermediateandfinalproducts)includingtheirlocation.

ResourceManagement, or Material Management, aims at needbased supply

withanddisposalofmaterialonschedulewithinthemanufacturingprocess,

includingmanagementofworkinprogress(WIP)material,i.e.resourcesthat

arecurrentlyoutsidecentralstockkeeping.The function isoftenreferred to

asMaterialandProductionLogistics.

Equipment Management / Maintenance ensures availability of machinery

equipment on schedule and, therefore, manages all equipment information

includingmaintenanceandrepairrequired.

ManufacturingExecutionandControlguaranteesproductionandmaterialflow

according to thedetailedproduction schedule (productionmonitoring)and


44/88

Study Findings 35


sendsbackinformationontheactualprocess.Itcanalsoaltertheproduction

schedule and provide Electronic Control Station (ECS) and Planning Table

functionality.

Traceability andGenealogyprovides transparencyas to theproduction status

including localizing of manufactured goods at any time during the

manufacturingprocess.Moreover,ittracks,collectsandverifiesinformation

about subcomponentsbeing assembled to parent components. This record

allowsfortraceabilityofcomponentsandusageofeachendproduct.

DynamicRoutingprovidesalgorithmstoroute inrealtime intermediaryor

workinprocessmaterialtoappropriatestationsand,hence,achieverealtime

loadbalancing inorder to increasemanufacturingperformancewithregard

tothroughput,workloadbalanceandworkinprocessqueues.Thefunctionis

oftenincludedinManufacturingExecutionandControl.

ProductionReporting andAnalysisallows foruptotheminute reportingand

analysisofactualmanufacturing.The functionallows forStatisticalProcess

Control (SPC), i.e. to compare planned manufacturing and actual

manufacturing using performance indicators which can be graphically

visualized.

Machine Control constantly monitors the status of all machinery collecting

machinedata(MachineDataAcquisition).

Production Data Acquisition ensures eventdriven acquisition, storage, and

update of data from the production process, including manual acquisition,

data preprocessing, and automated data transfer. Usually, this function

provides initialplausibilitychecksareprovided inorder tominimizeentry

errors.

Master DataManagement is a synthesis of the two functions Information

Management (defined in the VDI standard) and Production


45/88

Study Findings 36


Documentation (definedbothby MESA and NAMUR), as these tasks are

fairlysimilar,allowing tomanage records (ofproductorequipmentmaster

data)

that

preferably

should

be

kept

within

the

production

unit.

Some functionshavebeendeliberatelypositioned inbetween the layers.Although

someMESstandardsdoassignthesefunctionstotheMESlayer,inpracticetheyare

notalwaysimplementedononelayerortheotherunambiguously.Implementation

of these functions often is determined by company specific or production site

specific factors. Quality Management, for example, canbe assigned to either ERP

systemsortoMES;insomecasesevenoverlappingimplementationsonbothlayers

exist.IfthisisthecasetherehastobeagoodmatchingbetweentheERPsystemand

the MES, as the entire process is running across systems [Lampl 2009, p. 99].

Therefore, we left this question tobe answeredby our project partners and the

FunctionMaptobeinstantiatedflexibly.

However,wehaveprovidedabasicdefinitionforeachfunctioninordertoallowfor

a common understanding, so that assignment to one or more of the layers is

facilitated.It is important toemphasize that thefunctionsdescribedarenot totally

disjunctivebutmayoverlaptosomeextent.

4.2.2 MES Function Map Instantiations

MESFunctionMapsastheywereinstantiatedbytheautomobilemanufacturersare

enclosedinAppendixA.Duetothefactthat inonecasestudy threeplantsofone

OEMwithdifferentproductionprocess typeswereexamined,FigureA1contains

sixdifferentMESFunctionMaprepresentations.

Duringtheassessmentworkshopsitturnedoutthatsomeofthefunctionshadnot

been sufficiently detailed, preventing unambiguous assignment to the layers. We

thereforedecidedtofurtherspecifyeachMESfunctionintermsofdeterminingsub

functions and corresponding tasks. Again, the abovementioned standards (see

Section 4.2.1) served as a starting point for definition. Appendix B.1 gives a


46/88

Study Findings 37


definition for each MES functionby listing all tasks definedby the respective

standard. Based on this analysis we developed a consolidated list that includes

essential

tasks

of

each

MES

function

(see

Appendix

B.2).

The

list

was

validated

and

slightlycomplementedwithfindingsfromexpertinterviewsandthenusedtoverify

theMESFunctionMapsinstantiatedduringtheassessmentworkshops,enablingus

tomorepreciselyassignMESfunctionsandrelatedtaskstothethreelayers.Figure

44showsanexemplaryMESFunctionMapwithassignmentofthetasksdefinedin

Table B2. Each of the letters denotes a single task of a MES function with tasks

belongingtothesameMESfunctionhavingthesameletter.

Figure44:MESFunctionMapasinstantiatedwithdetailedtasks

The automobile manufacturers functional MES architectures illustrated by the

respective MES Function Maps have revealed some functional requirements that

needtobemetbyMESsoftwareproducts.Onemajorfindingfromthecomparison

of the different instantiations is that a generalized statement with regard to

functional requirements across all automobile manufacturers cannot be made.

Requirements are rather company specific or production specific, with the MES

ShopFloorLayer

ME

SLayer

ERPLayer

ShopFloorLayer

MESLayer

ERPLayer

Equipment

Management

B

A

A

A

A

A

LabourManagement

B

B

BB

BB

B

Detailed

Planning

C

C

C

C C

Gross

Planning

D

D

D

D

D

D

D

D

D

QualityManagement

E

E

E

E EProduction

Inventory

Management

F

F

FF

F

F

Resource

Management

G

G

G

G

G

G

H

H H

H

H

H

H

Manufacturing

Execution /Control

H

H

H

H

I

II

I

Traceability /

Genealogy

K K

KK

Dynamic

Routing

L

L

L

L

Production

Reporting &

An alysi s

M

M

M

Machine

ControlN

NPDA

N

N

O

O

O

Master Data

Management

L


47/88

Study Findings 38


layer (and corresponding applications) covering different functions, raising the

questionwhetherthenotionofstandardizedMESsolutionsupportisrealisticatall.

Moreover,

the

heterogeneity

identified

leads

to

another

question

about

the

factors

influencing the assignment of functions to the different layers. This topic is

addressedinmoredetailinSection4.2.4.

Nevertheless,basedonthetotalnumberofMESFunctionMapinstantiationssome

generaltrendsonfunctionalMESrequirementscouldbeidentified(seeFigure45).

For instance, Detailed Planning, Traceability and Genealogy, Dynamic Routing,

ProductionReportingandAnalysisaswellasManufacturingExecutionandControl

are mostly seen as core functions tobe addressed on the MES layer6. For other

functions,suchasResourceandEquipmentManagement,andQualityManagement,

MES applications need to provide support. Here, the topic of integration with

applications from the ERP layer and the Shop Floor Layer, covering some of the

tasks of these functions, is predominant. Evaluation of MES functions from an

integrationperspectiveisdiscussedinmoredetailinthefollowingsection.

Figure45showsthefrequencydistributionofallfunctionsofthegenericfunctional

referencearchitecturewithregardtobeingassignedtotheMESlayer.

6

These conclusions are made independently from the importance of each function for each OEM investigated.

The topic of relevance is discussed separately in chapter 4.2.3.


48/88

Study Findings 39


Figure45:FunctionstobecoveredbyMESapplications

4.2.3 Priori tization of MES Functions

Beyond thegoalofachievingterminologicalclarity,theMESFunctionMapserved

for prioritizing MES functions, depending on their relevance for the respective

Company as well as on their estimated contribution for improving the

manufacturing process.Although theautomobile manufacturers revealed a rather

heterogeneouspicturehere, theassessmentworkshopsyielded some findings that

canbegeneralized.

Firstly,ageneraltendencycouldbeidentifiedthatMESapplicationsshouldfirstand

foremost supportbasic MES functions, such as Detailed Planning, Manufacturing

Execution and Control, Production Data Acquisition, or Production Reporting.

Although the potential for efficiency gains of functions such as Traceability and

Genealogy or Dynamic Routing has been recognized and even first positive

experiences with corresponding implementations havebeen made in some of the

plants (most notably the component manufacturing plants), these functions are

assessedrather

as

supplementary

nice

to

have

functions

by

the

majority

of

the

Function

Number of assignments to the MES layer0 1 2 3 4 5 6 7

(Material) Requirements Planning

Product ion Inventory Management

Gross Planning

Production Worker Guidance

Labour Management

Equipment Management


Resource Management

Dynamic Routing

Traceability / Genealogy

Machine Control

Manufacturing Execution/Control

Production Data Acquisition (PDA)

Production Reporting and Analysis

Quality Management

Detailed Planning


49/88

Study Findings 40


workshopparticipants.Moreimportanceisattachedtononfunctionalrequirements

suchasguaranteeing sufficient integrationwithexistingapplications (be iton the

MES

layer

or

on

the

ERP

and

the

Shop

Floor

layer).

This

point

was

a

typical

example

for the importance of nonfunctional requirements, which in many cases were

valued more highly than purely functional requirements. They are discussed in

moredetailinsection4.2.5.

Secondly, thereseems tobeasignificantpotentialforimprovementwithregardto

functionsprovidedby severalapplicationsondifferent layers. A largenumber of

functionswasassignedtomorethanonelayer,implyingthatapplicationsystemsof

different layers and with different planning horizons support execution of these

functions (seealso instantiatedMESFunctionMaps inAppendixA).This leads to

the need for effective integration of these systems in order to enable continuous

planning and control. Consequently, the MES functions canbe prioritized with

regardtotheneedforbeingintegrateddependingontheirassignmenttoone,two

or,allthreelayers(seeFigure46).

7/27/2019 Fina

Documents

Final Report IME