CONFIGURING REFERENCE MODELS

An Integrated Approach for Transaction Processing and Decision Support

Ralf Knackstedt, Christian Janiesch, Tobias Rieke

European Research Center for Information Systems (ERCIS), University of Münster

Leonardo-Campus 3, 48149 Münster, Germany

Keywords: Reference models, configuration, transaction processing, decision support, integration, methods.

Abstract: Reference models are of normative, universal nature and provide a solution schema for specific problems by

depicting best or common-practice approaches. The configuration of these reference models has been a field

of research in the past. However, the integrative configuration of different reference models or respectively

reference models serving multiple purposes lacks of applicable methods. In practice this is a common prob-

lem as the simultaneous implementation of an enterprise resource planning system and a management in-

formation system shows. We provide a method that allows the integrative configuration of conceptual mod-

els for transaction processing and decision support by integrating meta models for modeling languages. In

addition, we exemplarily show its application to extend an existing reference model.

1 INTRODUCTION

The conceptualization of business activities into

processes is fundamental to Business Process Man-

agement. The goal is to identify improvement possi-

bilities and weaknesses of processes to generate

propositions for process improvement. Conceptual

models have become accepted as instruments to

attain this goal. The representation of the “as-is” and

“to-be” situations in the form of conceptual models

enhances the structuring of the problem and trans-

forms it into a more solvable form. As a support for

to-be modeling the application of reference models

is discussed alongside other approaches like bench-

marking, process simulation and animation, heuristic

design principles and checklists (Becker et al.,

2006).

Reference models are conceptual models that can

be used for designing specific models (Schütte,

1998, Vom Brocke, 2003). The knowledge which is

provided by reference models can give valuable

contribution to business reengineering and software

engineering purposes. Using reference models as

template for deriving enterprise-specific models

should lead to time and cost reduction (Schütte,

1998). Especially in the field of software engineer-

ing conceptual reference models can be used to

improve the customization process e. g. of ERP

systems (Davenport, 1998) or as a starting and con-

figuration point within the Model Driven Architec-

ture (MDA) procedure (Soley and OMG Staff Strat-

egy Group, 2000). Conceptual reference models

usually represent process and/or data issues of a

specific domain (e. g. industry, retail, insurance)

(Fettke and Loos, 2003).

To attain a broader customer clientele two direc-

tions of reference model evolution can be identified.

The first one is to add additional adaptation support

to the model to achieve an even greater time and

cost reduction of the model application e. g. by de-

fining perspectives on the model or adding variants.

The second direction is to broaden the reference

model to new application areas (area-spanning refer-

ence models) to achieve a more holistic view on the

company (e. g. adding state-of-the-art models for

data management and decision support to the exist-

ing process-oriented reference model). These two

directions are almost completely unrelated.

The research contribution of this paper is to con-

solidate the newer developments in the two direc-

tions to create adaptable and area-spanning (refer-

This work has been funded by the German Research

Foundation (DFG): "Komplexitätsmanagement für die

Gestaltung und Anwendung adaptiver Referenzmodelle",

record no. BE 1422/9-1

135

Knackstedt R., Janiesch C. and Rieke T. (2006).

CONFIGURING REFERENCE MODELS - An Integrated Approach for Transaction Processing and Decision Support.

In Proceedings of the Eighth International Conference on Enterprise Information Systems - ISAS, pages 135-143

DOI: 10.5220/0002492401350143

 SciTePress

ence) models. In this paper we concentrate on adap-

tation support by reference model configuration. The

development of configurable models is reasonable

because configuration rules in reference models

permit the explication of further business knowledge

about the connection of business characteristics and

business processes. Furthermore, configurable mod-

els provide the user with precisely tailored model

variants. Instead of limiting these advantages to one

area of application, future modeling concepts have

to permit an area-spanning consistent definition of

configuration rules. Therefore we provide a model-

ing-language-spanning approach to integrate and

customize different areas of reference models by

meta-model integration and configuration. This

should lead to a more conjoint view on transaction

processing and decision support systems.

This work presents first approaches to a solution

and is structured as follows: In Section 2, we present

an overview of related work and the state-of-the-art

in the field of reference modeling. In Section 3, a

modeling language is constructed from established

modeling approaches. It permits the configuration of

a purpose-spanning model system. Severeal depend-

encies lead to a necessity of a joint and integrated

view on business processes, data objects and man-

agement views, i. e. reports, in reference models. In

Section 4 the application of this modeling technique

is presented. It extends the Retail-H-Model by

means of a small example. So far, the Retail-H-

Model had to be classified as a single-purpose-

model with no configuration support. Finally, future

research needs are discussed in Section 5.

2 RELATED WORK AND

STATE-OF-THE-ART

Existing modeling techniques for reference models

can be classified in two dimensions (cf. Figure 1)

(cf. (Fettke and Loos, 2003)): As a theoretical struc-

turing of business application systems the informa-

tion pyramid has found widespread acceptance

(Mertens, 2000). The systems are divided into trans-

action processing systems and decision support

systems. The formers support the operational execu-

tion of business domains like procurement, produc-

tion, logistics, and distribution. The latter provide

information for management tasks, especially infor-

mation which is decision-relevant concerning the

design of the operational domains. Different model-

ing techniques have been established for both areas.

Process modeling techniques are mainly used in the

field of transaction processing, e. g. Event-Driven

Process Chains (EPCs) (Keller et al., 1992), Petri

Nets (Petri, 1962) or Business Process Modeling

Notation (BPMN) (White, 2004). The operational

databases are described e. g. via Entity Relationship

Models (ERMs) in different variants (Chen, 1976).

Object-oriented approaches like e. g. UML (Booch

et al., 1999) are suited for the description of dynamic

as well as static views on an information system.

The modeling of management views is primarily

addressed by techniques for semantic data-

warehouse-modeling (e. g. (Bulos, 1996, Sapia et al.,

1998, Holten, 2003, Golfarelli et al., 1998)).

The second dimension used to classify reference

modeling techniques is the consideration of ap-

proaches to support adaptation. The aim of adapta-

tion support is to provide reference model content in

a more reusable form. The fundamental mechanisms

of constructing a company-specific or project-

specific model from a reference model are configu-

ration, instantiation, aggregation, specialization and

analogy construction (Vom Brocke, 2003). The main

feature of configuration is the specification of ex-

plicit rules concerning how the reference model will

be adapted to given contexts (e. g. company classes).

Instantiation arranges the use of place holders. Ref-

erence models supporting aggregation are parti-

tioned into modules which can be combined, de-

pending on their gateways, to form new models. In

the case of specialization, the reference model is

kept deliberately on an abstract level. The model

user has to adapt it, whereas in contrast to configura-

tion no explicit rules are provided. Analogy-based

adaptation offers liberty concerning the method of

reusing arbitrary parts of the model which are e. g.

provided in forms of design patterns.

The occurrences of both dimensions, area of ap-

plication and adaptation support concept, are com-

bined to two oppositional values each, in order to

characterize the state-of-the-art of reference model-

ing research regarding a specific development per-

spective (cf. Figure 2): Existing reference models

and reference modeling techniques usually address

only one area of application and do not support con-

figuration. Widespread extensive reference models

like those of Scheer (Scheer, 1997), Becker and

Figure 1: Dimensions of Modeling Techniques.

ICEIS 2006 - INFORMATION SYSTEMS ANALYSIS AND SPECIFICATION

136

Schütte (Becker and Schütte, 2004) and Keller and

Teufel (Keller and Teufel, 1998) address the level of

transaction processing and have to be transferred to

company-specific models via specialization. Hars

(Hars, 1994) focuses on reference modeling with

ERMs, without considering adaptation via configu-

ration. Lang (Lang, 1997) and Remme (Remme,

1997) develop concepts for the aggregation of proc-

ess model parts, Rising (Rising, 2000) and Fernan-

dez and Yuan (Fernandez and Yuan, 2000) provide

design pattern libraries for the transaction processing

level.

Research dealing with the configuration of refer-

ence models is currently rather area-specific. The

works of Dreiling et al. (Dreiling et al., 2005) and

Becker et al. (Becker et al., 2004b) address the con-

figuration of process models in form of EPCs. Fur-

ther examples of the configuration of ERMs are

provided by Schütte (Schütte, 1998). Schwegmann

(Schwegmann, 1999) deals with configuration of

class diagrams. The configuration of models cover-

ing the decision support-level is discussed in the

works of Knackstedt and Klose (Knackstedt and

Klose, 2005), Goeken (Goeken, 2004) and Becker

and Knackstedt (Becker and Knackstedt, 2004).

Area-spanning modeling is dealt with from dif-

ferent points of view. The derivation of (decision-

relevant) data from operational data bases is e. g.

addressed by Rauh (Rauh, 1992) and Winter

(Winter, 1998). The integration of specifications of

management views into EPCs is e. g. suggested by

and Becker et al. (Becker et al., 2004a). The usage

of process execution data for the process perform-

ance manager is e. g. dealt with by zur Mühlen (zur

Mühlen, 2004). Corresponding software tools e. g.

have been developed by IDS Scheer AG (IDS

Scheer AG, 2002). The works mentioned above

however do not focus on the concept of reference

modeling or in particular on model configuration.

3 META MODEL DEFINITION

AND INTEGRATION

3.1 Introduction to the Modeling

Techniques Utilized and their

Meta Models

It is reasonable that reference models make use of

well-established modeling techniques to allow the

applicability for as many users and as many contexts

as possible. In terms of process modeling and data

modeling current state-of-the-art approaches are

EPC (Keller et al., 1992) and ERM (Chen, 1976) (cf.

section 2). Concerning reports, the MetaMIS ap-

proach to specify management views seems to be

very promising since its integration with the EPC

has already been conducted and evaluated (Becker

and Knackstedt, 2004, Becker et al., 2004a). Thus,

the integration of these three modeling techniques

seems to be the preferable approach. Integral part of

the mentioned modeling techniques is the modeling

language. It is described by a meta model which

defines the model elements and their relationships.

(cf. the corresponding meta models in

Figure 3).

EPCs are a semi-formal modeling technique to

model courses of business. They are an integral part

of the ARIS framework (Scheer, 2000). ER models

were introduced as a semantic data model, as a uni-

fying view, on both, relational and network data-

bases. MetaMIS models (introduced by Holten

(Holten, 2003) center on information objects, i. e.

reports. The first steps to create an information ob-

Figure 2: Portfolio of Related Work.

CONFIGURING REFERENCE MODELS - An Integrated Approach for Transaction Processing and Decision Support

137

ject are the definition of dimension groupings (e. g.

time, product) and its distinct dimensions (e. g. time

by week, weekday; time by month, day of month),

followed by the reduction of dimensions to dimen-

sion scopes (e. g. year 2005; product group one) and

combining them into dimension scope combinations

which define the specific information requirements

of the management. Ratios and ratio systems are

assigned to these dimension scope combinations, in

order to create business facts such as the stock of

products in a certain warehouse at a specific day or

the total revenue achieved by one store. This as-

signment constitutes the navigation space of the

information object. In this way, MetaMIS models

carry the information needed to describe the content

of a report. Exemplary EPC and MetaMIS models of

are depicted in Figure 5,

Figure 3 shows an ER

model.

3.2 Meta Model-based Integration of

the Modeling Languages

Our approach of integrating MetaMIS and EPC

differs slightly from the one of (Becker and

Knackstedt, 2004, Becker et al., 2004a) to allow a

seamless integration of the ERM as well:

Functions are a core element of business proc-

esses which are performed by persons and/or infor-

mation systems (IS). These persons or IS need in-

formation in forms of reports to make decisions. The

information object of the MetaMIS meta model is

such a report. Thus, the integration of EPC and

MetaMIS models is carried out via the reinterpreted

entity type information object (cf. Figure 3). It can

be regarded as a specialization of the process re-

source of the EPC meta model. In this way it is also

Figure 3: Integration of Modeling Language Definitions.

ICEIS 2006 - INFORMATION SYSTEMS ANALYSIS AND SPECIFICATION

138

linked to the entity-type process chain function by

the relationship type PR-PCF-As (Process Re-

source-Process Chain Function-Association), as

Becker et al. argue in their approach (Becker and

Knackstedt, 2004, Becker et al., 2004a).

The integration of the ERM meta model and EPC

meta model can be performed similarly. The entity

type data specification can also be regarded as a

specialization of the entity type process resource.

Thus, it is also linked to the entity type process

chain function via the relationship type PR-PCF-As.

The core element types of any ERM, entity types and

relationship types, are specializations of the entity

type data specification and therefore are also linked

to process chain function. The data cluster is a

scope of an ER model and is also a specialization of

the entity type data specification. This entity type is

of major importance concerning the following inte-

gration of MetaMIS and ERM.

The integration of MetaMIS models and the

ERM technique cannot follow a similar pattern. The

element types of METAMIS models and ER models

cannot be used interchangeably as with the EPC

technique that has over time developed a universal

connector, an interface called process resource, to

any modeling language, the integration of the data

and report meta model has to be done via sets of

transformation rules (transformation) which specify

the underlying dependencies. This is necessary since

the constructs of both modeling techniques do not

map one to one and rather represent views on the

same data. Whereas the ER models form the analysis

fundament, MetaMIS models try to explicate the

information contained therein. Thus, a transforma-

tion is essential to sophistically configure reference

models for operational and planning purposes (Rauh,

1992). Cf. to Figure 3 for the integration of the meta

models and the description of the transformation

relation between MetaMIS and ER models.

3.3 Consideration of Configuration

Aspects

As stated in Section 1 for the course of this paper we

will focus solely on the configuration of reference

models for adaptation support. Configuration can be

achieved by model projection, the exclusion of

model elements according to configuration rules.

Targets of the model projection can be whole model

types (selection of model type; e. g. to build model

technique specific views), element types of the lan-

guage definition on meta level (selection of element

type; e. g. to support variants of model techniques),

instance model elements (selection of element; e. g.

for the exclusion of specific model elements due to a

specific company class), denotations of the model

elements (variation of denotation; e. g. for exchang-

ing names and labeling of model elements depend-

ing on user groups) and representational aspects

(variation of representation; e. g. for using alterna-

tive symbols or topologies) (for comparable ap-

proaches cf. e. g. (Darke and Shanks, 1996, Engels

et al., 1997, Nissen et al., 1996)).

As a basic mechanism to support detailed con-

figuration, the mechanism of element selection is

depicted in Figure 4 and addressed in this paper.

Targets of the selection of element projection –

which are relevant for this paper – are the following

(cf. to the shaded entity types in Figure 3):

• regarding process aspects: process chain

elements, process chains and the associa-

tion between process resources and process

chain functions (PR-PCF-As) to exclude

whole processes or only parts of them,

• regarding data aspects: data specifica-

tions to exclude whole data clusters or sev-

eral entity and relationship types,

• regarding transformation aspects: trans-

formations and associations between trans-

formations and transformation rules (T-TR-

As) to exclude whole transformations or to

build variants of them,

• regarding report aspects: information ob-

jects and associations between dimension

scopes and dimension scope combinations

(DS-DSC-As), dimension objects and di-

mensions scopes (DO-DS-As), and ratios

and ratio systems (R-RS-As) to exclude

whole information object specifications or

to modify them depending on the configu-

ration context.

To allow the configuration of these model elements,

the meta model of the modeling language has to be

adapted. The model elements on the meta level have

to be assigned to a configuration term that deter-

mines the elements to be shown in user-specific

views. A configuration term is specified in a gram-

mar which is also depicted in Figure 4. Configura-

tion terms denominate characteristics of company

classes and their instance values. They express that

model elements are part of the configured model if

the company using the reference model is part of the

respective company class.

CONFIGURING REFERENCE MODELS - An Integrated Approach for Transaction Processing and Decision Support

139

4 APPLICATION FOR A RETAIL

REFERENCE MODEL

4.1 Overview of the Reference

Model

In the following, the developed modeling technique

will be applied to expand the reference model for

integrated information systems in retailing compa-

nies of Becker and Schütte (Becker and Schütte,

2004). To provide an overview and a navigational

structure for the numerous models included, the

reference model is structured by the retailing-H-

architecture (cf. the left side of Figure 5; the H

stands for the German term “Handel” which trans-

lates as retail). The architecture identifies the tasks

of retailing companies at an abstract level using a

selected structuring paradigm:

Whereas all tasks concerned with the suppliers

are placed on one leg, all tasks associated with the

customer are placed on the other leg. The logistical

functions covering goods receipt, warehouse and

goods issue are arranged horizontally. The merchan-

dising management-related areas of a retailing com-

pany then form the letter H. The grouping of the

bookings in the general ledger accounting and other

functions of the business-administrative-system

(asset accounting, cost accounting and human re-

sources management) are assigned to the foot of the

H. Long-term planning functions build the roof of

the architecture pictogram.

In accordance with the architecture of integrated

information systems (ARIS) (Scheer, 2000) different

viewpoints on information systems are taken into

account. The original retailing H-model differenti-

ates between functions within the identified tasks of

the architecture (the subtasks to be performed), data

(the structure of the tasks), and processes (the behav-

ior of these areas). Since function and process view

are not completely distinct (the process describes the

temporal and logical sequence of functions) and the

function view is obsolete from an ontological per-

spective (Green and Rosemann, 2002), we substitute

the function view for a reports view. Reports are

needed for basically all tasks of the retailing com-

pany.

Processes

PR-PCF-

Reports

Data

Termtext has to conform to the following grammar for formulation of configuration terms:

<term> ::= <expression> {<operator> <expression>}

<company-attribute value-list> ::= "(" <prefix> <company-attribute value-list> { <operator> <prefix> <company-attribute value-list> } ")"

<company-attribute value-list> ::= <company-attribute value>

<operator> ::= "|" | "+"

Model Element

Compatible for

Term Association

N,P

Process Chain

Element

Data SpecificationDS-DSC-As

DO-DS-As

R-RS-As

Transfor-

mation

T-TR-

ME-CT-

(0,1)

Configuration

Term

(1,1)

Termtext

Process Chain

Information

Object

Figure 4: Extension of the Modeling Language Definition with Configuration Aspects.

ICEIS 2006 - INFORMATION SYSTEMS ANALYSIS AND SPECIFICATION

140

4.2 Example of the Modeling

Language Application

The application of the modeling technique intro-

duced in Section 3 to construct configurable and

area of application-spanning reference models is

shown in the following example where a company

wants to apply the H-Model. The company supports

individual marketing methods to address the cus-

tomers. Thus, the configuration parameter in this

case is the differentiation, whether customers are

known to the company or whether they remain

anonymous.

In this case, account cards are the method con-

sidered to reach customers individually. This deci-

sion influences the company’s processes since full

processes and parts thereof can be omitted (cf. the

right side of Figure 5). For example, if customers are

addressed anonymously only, processes and func-

tions concerning the maintenance of account cards

are omitted. Whether customers are known or un-

known to the company is a distinctive attribute

which not only influences functions which are to be

modeled as process chains but also the resources

which are accessed during the execution of these

processes. For example, the analysis of opening

times cannot be realized based on customer group

differentiation if no customer data is available. In the

process model depicted in Figure 5 this context can

be exemplified by the fact that the specification of

the report used for Weekday Analysis is configur-

able. The configuration term Customer contact

(known) connected to the dimension scope Custom-

ers by “Customer Group” makes sure that the navi-

gation space only contains customer group structures

if the individual customer can be identified via ac-

count cards.

34 5

Figure 5: Modified Retailing H-model.

CONFIGURING REFERENCE MODELS - An Integrated Approach for Transaction Processing and Decision Support

141

A model-independent adaptation can cause in-

consistencies in relation to other models. Data ele-

ments that are connected to functions concerning

account card handling will for instance never being

read or written, if these functions were omitted.

Therefore the ER models of the reference model also

have to be adapted. The decision whether customers

are treated anonymously or not does also affect the

entity types customer card and account card and

model elements whose existence is dependent on

them via the configuration term Customer contact

(known).

With the omission of the entity types Customer

and Account Card in the scenario Customer contact

(known), the possibility to derive ER model ele-

ments based on these types is lost. By the omission

of original ERM elements, further transformation

rules are disregarded, too. This is because they are

either directly (e. g. with the transformation of di-

mension scope Customers by “Customer Group”) or

indirectly (e. g. with the transformation of ratios

from derived model elements) dependent on ERM

elements assigned to configuration terms. If the

configuration-based elimination of processes omits

the use of reports in the reference model, complete

report definitions and their assigned transformation

rules are disregarded.

5 FURTHER RESEARCH

The integrative configuration of the retailing H-

Model was described by means of a clear and con-

cise example. Process models, data models and re-

port specifications were configured while preserving

existing dependencies between the views. The ongo-

ing development of the model must consider much

more configuration causes. This can be exemplified

in a compact manner by the company attribute busi-

ness type on the architecture level. The retailing H-

Model shown in Figure 5 describes the classic type

of retailing business which is warehousing business

with procurement, storing and distribution functions.

In addition to the classic warehousing business,

third-party business and the pooled payment busi-

ness have established themselves. A retailing com-

pany can practice one or more types of business,

which have large-scale consequences for the adapta-

tion of the reference model. Greater parts of the

reference model become obsolete in such types of

business.

Beside the enhancements of the retailing refer-

ence model other areas are of interest: The results of

this ongoing research can be employed for other

areas besides reference modeling such as integrated

process and data warehouse documentation (Becker

et al., 2005) or in the context of service companies

(Klose et al., 2005). Furthermore we will focus some

methodical research questions. Our modeling and

configuration approach moves complexity to the

reference model creator. Thus, the creator has to

deal with a highly increased load of complexity. In

particular the management of the dependencies of

several models and their configuration roles seems

to be a matter of special importance. The resulting

effort can be neglected due to economies-of-scale

effects. In this context we will develop methodical

solutions to reduce respectively to handle the com-

plexity of the presented configurable and area-

spanning model systems.

REFERENCES

Becker, J., Brelage, C., Crisandt, J., Dreiling, A., Ribbert,

M. and Seidel, S., 2004a. Methodische und technische

Integration von Daten- und Prozessmodellierungstech-

niken für Zwecke der Informationsbedarfsanalyse. In

Arbeitsberichte des Instituts für Wirtschaftsinformatik,

(Eds, Becker, J., Grob, H. L., Klein, S., Kuchen, H.,

Müller-Funk, U. and Vossen, G.) Münster.

Becker, J., Delfmann, P., Dreiling, A., Knackstedt, R. and

Kuropka, D., 2004b. Configurative Process Modeling

– Outlining an Approach to Increased Business Proc-

ess Model Usability. In Proceedings of the 15

Infor-

mation Resources Management Association Interna-

tional Conference (IRMA 2004). New Orleans, LA,

pp. 615–619.

Becker, J., Janiesch, C., Delfmann, P. and Fuhr, W., 2005.

Perspectives on Process Documentation - A Case

Study. In Proceedings of the 7th International Confer-

ence on Enterprise Information Systems (ICEIS 2005).

Miami, FL, pp. 46-56.

Becker, J. and Knackstedt, R., 2004. Referenzmodel-

lierung im Data-Warehousing – State-of-the-Art und

konfigurative Ansätze für die Fachkonzeption.

Wirtschaftsinformatik, 46, pp. 39-49.

Becker, J., Kugeler, M. and Rosemann, M. (Eds.) (2006)

Process Management: A Guide for the Design of Busi-

ness Processes, Springer, Berlin, Heidelberg, New

York. To appear.

Becker, J. and Schütte, R., 2004. Handelsinformationssys-

teme, Redline Wirtschaft, Frankfurt am Main, 2

Edi-

tion.

Booch, G., Rumbaugh, J. and Jacobsen, I., 1999. Unified

Modeling Language User Guide, Addison-Wesley,

Boston, MA.

Bulos, D., 1996. A New Dimension. Olap Database De-

sign. Database Programming & Design, 9, pp. 33-37.

Chen, P. P.-S., 1976. The Entity-Relationship Model.

Toward a Unified View of Data. ACM Transactions

on Database-Systems, 1, pp. 9-36.

ICEIS 2006 - INFORMATION SYSTEMS ANALYSIS AND SPECIFICATION

142

Darke, P. and Shanks, G., 1996. Stakeholder Viewpoints

in Requirements Definition: A Framework for Under-

standing Viewpoint Development Approaches. Re-

quirements Engineering, 1, pp. 88-205.

Davenport, T. H., 1998. Putting the Enterprise into the

Enterprise System. Harvard Business Review, 76, pp.

121-131.

Dreiling, A., Rosemann, M., van der Aalst, W. M. P.,

Sadiq, W. and Khan, S., 2005. Model-Driven Process

Configuration of Enterprise Systems. In Wirtschaftsin-

formatik 2005, (Eds, Ferstl, O. K., Sinz, E. J., Eckert,

S. and Isselhorst, T.) Physica-Verlag, Heidelberg, pp.

687-706.

Engels, G., Heckel, R., Taentzer, G. and Ehrig, H., 1997.

A View-Oriented Approach to System Modelling

Based On Graph Transformation. ACM SIGSOFT

Software Engineering Notes, 22, pp. 327-343.

Fernandez, E. B. and Yuan, X., 2000. Semantic Analysis

Patterns. In Proceedings of the 19

International Con-

ference on Conceptual Modeling (ER 2000). Salt Lake

City, UT, pp. 183-195.

Fettke, P. and Loos, P., 2003. Classification of reference

models: a methodology and its application. Informa-

tion Systems and e-Business Management, 1, pp. 35-

53.

Goeken, M., 2004. Referenzmodellbasierte Einführung

von Führungsinformationssystemen. Grundlagen, An-

forderungen, Methode. Wirtschaftsinformatik, 46, pp.

353-365.

Golfarelli, M., Maio, D. and Rizzi, S., 1998. The Dimen-

sional Fact Model - A Conceptual Model for Data

Warehouses. International Journal of Cooperative In-

formation Systems, 7, pp. 215-247.

Green, P. and Rosemann, M., 2002. Perceived Ontological

Weaknesses of Process Modeling Techniques: Further

Evidence. In Proceedings of the 10

European Con-

ference on Information Systems (ECIS 2002). Gdańsk,

pp. 312-321.

Hars, A., 1994. Referenzdatenmodelle. Grundlagen ef-

fizienter Datenmodellierung, Gabler, Wiesbaden.

Holten, R., 2003. Specification of Management Views in

Information Warehouse Projects. Information Systems,

28, pp. 709-751.

IDS Scheer AG, 2002. ARIS Process Performance Man-

ager. White paper.

Keller, G., Nüttgens, M. and Scheer, A.-W., 1992. Seman-

tische Prozeßmodellierung auf der Grundlage "Er-

eignisgesteuerter Prozessketten (EPK)". In Veröf-

fentlichungen des Instituts für Wirtschaftsinformatik,

(Ed, Scheer, A.-W.) Saarbrücken.

Keller, G. and Teufel, T., 1998. SAP R/3 Process Oriented

Implementation – Iterative Process Prototyping, Har-

low et al.

Klose, K., Knackstedt, R. and Becker, J., 2005. Process

Modelling for Service Processes - Modelling methods

extensions for specifying and analysing customer inte-

gration. In Proceedings of the 7th International Con-

ference on Enterprise Information Systems (ICEIS

2005). Miami, FL, pp. 260-265.

Knackstedt, R. and Klose, K., 2005. Configurative Refer-

ence Model-based Development of Data Warehouse

Systems. In Proceedings of the 16

Information Re-

sources Management Association International Con-

ference (IRMA 2005). San Diego, CA, pp. 32-39.

Lang, K., 1997. Gestaltung von Geschäftsprozessen mit

Referenzprozessbausteinen, DUV, Wiesbaden.

Mertens, P., 2000. Integrierte Informationsverarbeitung 1

- Administrations- und Dispositionssysteme in der In-

dustrie, Gabler, Wiesbaden, 12

Edition.

Nissen, H. W., Jeusfeld, M. A., Jarke, M., Zemanek, G. V.

and Huber, H., 1996. Managing Multiple Require-

ments Perspectives with Metamodels. IEEE Software,

13, pp. 37-48.

Petri, C. A., 1962. Kommunikation mit Automaten, Uni-

versität Bonn, Bonn.

Rauh, O., 1992. Überlegungen zur Behandlung ableitbarer

Daten im Entity-Relationship-Modell. Wirtschaftsin-

formatik, 34, pp. 294-306.

Remme, M., 1997. Konstruktion von Geschäftsprozessen.

Ein modellgestützter Ansatz durch Montage

generischer Prozesspartikel., Gabler, Wiesbaden.

Rising, L., 2000. The Pattern Almanac 2000, Addison-

Wesley, Boston et al.

Sapia, C., Blaschka, M., Höfling, G. and Dinter, B., 1998.

Extending the E/R Model for the Multidimensional

Paradigm. In Proceedings of the International Work-

shop on Data Warehouse and Data Mining (DWDM

1998). Singapore, pp. 105-116.

Scheer, A.-W., 1997. Wirtschaftsinformatik – Refer-

enzmodelle für industrielle Geschäftsprozesse, Berlin

et al., 7

Edition.

Scheer, A.-W., 2000. ARIS – Business Process Modelling,

Berlin et al., 3

Edition.

Schütte, R., 1998. Grundsätze ordnungsmäßiger Refer-

enzmodellierung. Konstruktion konfigurations- und

anpassungsorientierter Modelle, Gabler, Wiesbaden.

Schwegmann, A., 1999. Objektorientierte Referenzmodel-

lierung. Theoretische Grundlagen und praktische An-

wendung, DUV, Wiesbaden.

Soley, R. and OMG Staff Strategy Group, 2000. Model

Driven Architecture, Draft 3.0, Object Management

Group, Needham

Vom Brocke, J., 2003. Referenzmodellierung. Gestaltung

und Verteilung von Konstruktionsprozessen, Logos,

Berlin.

White, S. A., 2004. Business Process Modeling Notation.

Version 1.0, Business Process Management Initiative.

Winter, R., 1998. Informationsableitung in betrieblichen

Anwendungssystemen, Vieweg, Braunschweig, Wies-

baden.

zur Mühlen, M., 2004. Workflow-based Process Control-

ling - Foundation, Design and Application of Work-

flow-driven Process Information Systems, Logos, Ber-

lin.

CONFIGURING REFERENCE MODELS - An Integrated Approach for Transaction Processing and Decision Support

143