MULTIDIMENSIONAL REFERENCE MODELS

FOR DATA WAREHOUSE DEVELOPMENT

Matthias Goeken

Frankfurt School of Finance & Management, Sonnemannstraße 9 – 11, 60314 Frankfurt a. M., Germany

Ralf Knackstedt

European Research Center for Information Systems, Westfälische Wilhelms-University Muenster

Leonardo-Campus 3, 48149 Münster, Germany

Keywords: Multidimensional modelling, Data warehousing, Reference modelling, ME/RM.

Abstract: In the area of Data Warehousing the importance of conceptual modelling increases as it gains the status of a

critical success factor. Nevertheless the application of conceptual modelling in practice often remains un-

done, due to time and cost restrictions. Reference models seem to be a suitable solution for this problem as

they provide generic models which can be easily adapted to specific problems and thus decrease the model-

ling outlay. This paper identifies the requirements for multidimensional modelling techniques whose ful-

fillment are a prerequisite for the construction of reference models. Referring to the ME/RM, the concrete

implementation of these requirements will be illustrated.

1 INTRODUCTION

To develop data warehouse systems it is necessary

to identify what kind of data has to be provided to

whom (decision maker) for what kind of manage-

ment decision (Holten 2003). Despite the fact that an

appropriate specification of data warehouse systems

is notably necessary at the beginning of a project

(e. g. for long-term maintenance reasons), the con-

struction of conceptual models, is often neglected

(Vassiliadis, Bouzeghoub, Quix 2000) as data ware-

house engineers often attempt a fast realisation

(Vassiliadis 2000). This seems to be critical because

several studies reveal the importance of determining

information requirements in data warehouse devel-

opment (e. g. Watson et al. 2004; Wixom, Watson

2001).

Reference models can increase the efficiency and

effectiveness of conceptual modelling because they

can be used as a starting point for the construction of

project and enterprise specific models. Thus, refer-

ence models provide best (or common) practice so-

lutions for information modelling projects. They are

blueprints representing a class of domains and can

therefore be seen as reusable requirements. In order

to capture the subjectivity of users’ needs it is neces-

sary to adapt these blueprints according to their re-

quirements. The process of adaptation should pro-

vide mechanisms and modelling constructs that ex-

plicitly represent variability in conceptual models.

In the following article reference models will be

discussed in the context of data warehousing. After

presenting related work the most important require-

ments concerning reference models in data ware-

housing will be developed. A concrete approach for

reference modelling in data warehousing concludes

this paper.

2 RELATED WORK

From a methodical perspective, the debate about

design issues of data warehouse systems is domi-

nated by manifold modelling approaches. For the

multidimensional specification of data warehouse

requirements, a broad variety of modelling tech-

niques exists (Abello et al. 2000; Trujillo et al.

2001). Some of them are closely related to Entity-

Relationship Models (ERM) (Chen 1976) or provide

347

Goeken M. and Knackstedt R. (2007).

MULTIDIMENSIONAL REFERENCE MODELS FOR DATA WAREHOUSE DEVELOPMENT.

In Proceedings of the Ninth International Conference on Enterprise Information Systems - ISAS, pages 347-354

DOI: 10.5220/0002366803470354

 SciTePress

data warehouse specific ERM extensions (Sapia et

al. 1998). Others are derived from modelling ap-

proaches for scientific and statistical data bases

(Chan, Shoshani 1981; Rafanelli, Bezenchek, Tinin-

ini 1996; Rafanelli, Shoshani 1990), are related to

object-oriented modelling approaches (Harren, Her-

den 1999; Trujillo et al. 2001), or present a multidi-

mensional modelling approach which is not based on

an already existing modelling technique (Bulos

1996; Thomsen 1997; Golfarelli, Maio, Rizzi 1998a;

Holten 2003).

The state-of-the-art of reference model application

in the requirements specification phase of data

warehouse projects mostly refers to an ad-hoc modi-

fication of existing information models (Adamson,

Venerable 1998). As the analysis of various multi-

dimensional modelling methods shows, the proposed

modelling methods do not provide constructs for

supporting model adaptation. Libraries comprising

reusable elements of data warehouse reference mod-

els are mostly specialised on particular model ele-

ment types (Spitta 1997).

Collections and definitions of ratios and ratio sys-

tems are widespread in business literature (Cope-

land, Koller, Murrin 1990; Eccles 1991; Lapsley,

Mitchel 1996; Kaplan, Norton 1996). However,

these collections neglect important aspects (mainly

dimensions that have to be analysed for management

tasks) of the data warehouse requirements specifica-

tion (Holten 2003).

In contrast to the area of data warehousing the usage

of reference models for the specification of business

processes is widely accepted. The adaptation of

business models based on configuration patterns is

widely discussed (Nordstrom et al. 1998; Nuseibeh

1994; Nissen et al. 1996; Hofstede, Verhoef 1996;

Kotonya, Sommerville 1995; Becker et al. 2004).

From a practical perspective, corresponding ap-

proaches are particularly established in the context

of customising Enterprise Resource Planning (ERP)-

systems (Rosemann, Shanks 2001; Rosemann 2003).

However, ERP-configuration parameters for report

definitions are mainly restricted to a selection of pre-

defined reports and organisational roles. But the

documentation of underlying configuration rules is

often inadequate since the configuration is con-

ducted on a rather technical level. Thus, end users

are only able to comprehend effects of the configu-

ration in the form of eliminated reports or eliminated

report parts.

The transformation of data warehouse specification

models into design schemes and implementations is

addressed in a broad variety of approaches being

based upon the tool support of data warehousing

(Hahn et al. 2000; Golfarelli et al. 1998b; Blaschka

2000; Burmester, Goeken 2006). These approaches

aim at a (semi-) automatic transformation of data

warehouse requirement specifications into initial

data warehouse implementations. Thus, for further

developments on data warehouse reference model-

ling it seems reasonable to address the requirement

specification layer.

3 REQUIREMENTS FOR REF-

ERENCE MODELS FOR DATA

WAREHOUSE DEVELOPMENT

Data warehouses aim at the satisfaction of users’

information needs. These needs are determined by

several factors and thus imply different requirements

in the design of reference models. To fit these re-

quirements a two-step procedure is proposed (cf.

figure 1).

Step 1 (situational positioning and role orientation):

An important factor impacting the information needs

is the so called “situational positioning”. The situ-

ational positioning of an enterprise is determined by

branch, company type, current life cycle phase and

other enterprise attribute values. (Mertens, Griese

2002). Different types of companies imply company

specific decisions which have to be supported by the

data warehouse system. Only a specific company

type requires specific decisions; the information

need related to this decision is relevant.

Furthermore the information need is affected by the

role of the decision-maker. A role describes the de-

cision rights and accountability of the decision-

maker. To enable a so called “role orientation”, in-

formation needs have to be adapted to the role of the

decision maker (Mertens, Griese 2002).

Figure 1: Process of application of multidimensional

reference models.

Our approach addresses situational positioning and

role orientation by adding a rule basis to multidi-

mensional models. Depending on enterprise attribute

values and roles, the rule basis assigns relevance to

ICEIS 2007 - International Conference on Enterprise Information Systems

348

the elements of information needs. This kind of rule

based adaption of reference models in literature is

discussed explicitly as “configurative reference

modelling” (Becker et al. 2004; Knackstedt, Klose

2005; Delfmann et al. 2006). In the following we

will apply this approach to a specific multidimen-

sional modelling technique.

Step 2 (Personalisation):

Analysis of data warehouse projects identify unsatis-

factory or missing user orientation as a critical suc-

cess factor of data warehouse development (Poon,

Wagner 2001; Mukherjee, D'Souza 2003; Wixom,

Watson 2001). User oriented information delivery

cannot be realised independently from single users,

and the possibilities for standardisation are limited,

since the view of reality or the universe of discourse

is highly subjective. Hence, the users with their in-

dividual preferences should play a significant role in

the adaptation process of the reference models. This

factor is discussed as personalisation (Mertens,

Griese 2002).

In our approach the personalisation is taken into

account by the fact that in specific models, con-

structed on the basis of the reference models, the

individual and subjective needs of the future users

are taken into consideration. This process of adapta-

tion requires a high level of user participation.

Therefore, a data warehouse reference model should

enable variants and possibilities for individual adap-

tation.

It has to be noticed, that different variants to analyse

facts and measures must not exclude themselves and

besides can be implemented often in parallel at the

same time. For this the reference models must offer

modelling constructs that mark the relationship be-

tween variants explicitly i.e. represent clearly

whether variants can exclude themselves or can be

implemented in parallel. Therefore, we suggest ex-

tensions to multidimensional modelling languages

for the representation of variants.

Altogether, by serving as a starting point, reference

models support requirements elicitation. Require-

ments Elicitation is the process through which the

users and developers discover, review, articulate,

and finally define the requirements the to-be system

has to fulfil. It is supported, because users and de-

velopers do not start “from scratch”. Instead, the

reference model can be used as a blueprint which is

adapted to subjective and individual information

needs.

4 MULTIDIMENSIONAL

REFERENCE MODELLING

TECHNIQUES

4.1 Basic Modelling Technique

In the following, our extension concept will be ap-

plied to a concrete modelling technique. Therefore a

notation related to multidimensional ERM (ME/RM)

by Sapia et al. is used (Sapia et al. 1998). The

ME/RM extends the traditional ERM by an entity

type, the ‘dimension level’ and two specific relation-

ship types, the ‘fact relationship’ and the ‘rolls-up-

relationship’. The core of ME/RM is represented by

a fact relation, visualised by a three dimensional

square. It represents a set of facts, i. e. an economi-

cally relevant area of interest, substantiated by ratios

as quantitative units of measurement. Like in

ME/RM the ratios are annotated as attributes of fact

relations. The fact relation connects several dimen-

sion levels of different dimensions. Dimensions

characterise the facts and represent qualified aspects,

from which facts and ratios can be analysed. The

dimension levels within a dimension are related in a

hierarchical order and are connected by a directed

acyclic graph, the rolls-up-relationships. It is for the

reason that the ME/RM does not provide an explicit

qualification of dimensions that – following the

DFM by Golfarelli et al. (Golfarelli et al. 1998) – the

dimension is visualised by an oval. Facts and ratios

(as quantitative values) as well as dimensions and

dimension levels (as qualitative values) are the main

components of multidimensional modelling. Further

components of multidimensional modelling– like

dimensional attributes, different types of dimensions

and relationships or heterarchies – are neglected in

following.

4.2 Extensions for Rule Based Configu-

ration

In order to use configurative reference modelling

concepts in practise, the extension of modelling

methods for data warehouse specification is neces-

sary (Knackstedt 2006, Knackstedt, Klose 2006).

Constructs are required to label model components

which are exclusively relevant in a given application

context. Model element types of the modelling

method that are designated for configuration are

connected to configuration parameters.

MULTIDIMENSIONAL REFERENCE MODELS FOR DATA WAREHOUSE DEVELOPMENT

349

Against the background of Category Management,

Promotion

Day

Quarter

Year

Turnover

Product

profitability

Legend

Entity type

here: Dimension level

Relationship type

Fact relationship type

Attribut

Roll-up-relationship type

Month Week

Transaction type (Sales

promotion business)

Report frequency

(Week)

Report frequency

(Month)

Fiscal Year Half year

All Time

(1) Configuration of the configurable reference multidimensional model with actual values of configuration parameters:

report frequency = month; transaction type = not sales promotion business; purchase area = international

Dimension

Promotion typeAll Sales

Product

Product groups

ordered by

countries of

origin

Product groups

ordered by

product lines

Fact:

Product

Analysis

All Products

X V

Time

Fact:

Product

Analysis

Product

Product groups

ordered by

countries of

origin

Product groups

ordered by

product lines

All Products

Purchase area (International)

Day

Quarter

Year

Month

Fiscal Year Half year

All Time

Time

Turnover

Product

profitability

Entity type with build time operator “recursive

relation“ representing an extensible dimension

path

Build time operators indicating

alternative pathes

<term>

Configuration term

Figure 2: Reference model application (part I).

ICEIS 2007 - International Conference on Enterprise Information Systems

350

Concerning our case dimensions, dimension level

attributes, attributes and fact relationship types are

affected. We propose enterprise attribute values and

roles as specialisations of configuration parameters.

Enterprise attribute values are used as configuration

parameters to cover aspects of situational position-

ing. Roles are used to cover requirements on role

orientation. Figure 2 illustrates the application of the

reference model configuration. Here, the specifica-

tion of a fact for product group analysis is provided.

Within our example, configuration parameters are

enterprise attributes ‘transaction type’, ‘purchase

area’ and ‘report frequency’. An analysis of product

turnovers and product profitability according to sales

promotion types seems to be reasonable only if the

retailer makes use of a ‘sales promotion business’

instead of a permanent ‘low price strategy’. More-

over, a consideration of products according to coun-

tries of origin only makes sense in case of an ‘inter-

national’ purchase area. The report frequency affects

selection possibilities of analysis hierarchies with

respect to the reference object ‘time’.

Figure 3: Decision table.

The underlying rule basis can be presented in alter-

native representation forms. The decision table de-

picted in Figure 3 assigns the stated conditions as

combinations of enterprise attribute values with spe-

cific actions. Actions consist of removing or adding

model elements. The crosses used in Figure 3 illus-

trate which model element is a component of the

derived enterprise-specific model. By means of

analogous extensions we are able to create models

that include perspectives and configurable ratio sys-

tems as well.

An alternative representation form is the use of

parameterisations that can be added to certain model

elements. Depending on configuration parameter

values parameterisations determine which model

elements are parts of the derived project-specific

model. Figure 2 illustrates the application of

parameterisations. Here, the configuration term

‘purchase area (international)’ is annotated to the

entity type ‘product group ordered by countries of

origin’. This rule defines that the entity type ‘prod-

uct group order by countries of origin’ is to be

dropped out in case of an enterprise exclusively pur-

chasing nationally. The syntax of parameterisations

can be defined in the form of a context-free grammar

formulated in the Extended-Backus-Naur-Form

(EBNF) (Hopcroft, Motwani, Ullman 2000) (cf. fig-

ure 4).

Figure 4: Grammar for parameterisations.

4.2 Extensions for Individual Adapta-

tion

For the support of individual adaptation generic ex-

tensions of the conceptual language are used. The

extensions refer to so called build time operators

which represent points where variability takes place.

Using these build time operators one can illustrate

the various variants a reference model contains and

communicate them to users (Goeken 2004; Halmans,

Pohl 2004). By means of these build time operators

the reference dimensions will be adapted to the sub-

jective user needs. The adaptation refers to the num-

ber of dimension levels and paths, their hierarchical

arrangement as well as their naming.

Figure 2 shows a build time operator indicating that

the number of dimension levels has to be adapted

according to the specific conditions of the enterprise.

The “recursive relation” represents the adaptation

point. During the development process this adapta-

tion point has to be solved and bound to a specific

variant. In Figure 5 a concrete model is presented

which was deduced from the blueprint, represented

by Figure 2. It contains two concrete paths within

the product dimension.

In addition, the reference model can give more con-

crete dimension levels, dimension hierarchies and/or

dimension paths. Hence, we suggest  in addition to

MULTIDIMENSIONAL REFERENCE MODELS FOR DATA WAREHOUSE DEVELOPMENT

351

the recursive relations  another extension of the

multidimensional modelling language which helps to

illustrate the relationship between variants.

The possible relationship types can be subdivided

into the inclusive OR (∨) and the exclusive OR (×).

The former means that concerning a business ratio

two dimensions with their dimension hierarchies can

be implemented in parallel whereas the exclusive

OR is mutually exclusive (also according to a se-

lected ratio). In the adaptation process it is required

to accept, rename or drop the levels and the names

which the reference model suggests.

The meaning of the exclusive OR can be illustrated

with the help of a generic time dimension (again fig

2). It shows, that according to the concrete context,

some dimension levels and dimension paths can be

dropped completely, because they contradict each

other or have no relevance. For example, if the fiscal

year starts in January, calendar year and fiscal year

are equivalent. Than there is no need to report the

relevant ratios for the time span October – Septem-

ber. Therefore, we can drop “fiscal year” in this case

(cf. figure 2 and figure 5).

5 CONCLUSION AND FUTURE

WORK

In this paper we presented extensions for multidi-

mensional modelling techniques to support the usage

of reference models when developing data ware-

house systems. To apply these reference models a

two-step procedure is proposed. The first step com-

prises a rule based configuration of the models

whereas on the basis of the evaluation of this rule

base the individual adaptation of the reference model

takes place in the second step. This procedure con-

sequently fits both on enterprise and role specific

Figure 5: Reference model application (part II)

ICEIS 2007 - International Conference on Enterprise Information Systems

352

impact factors for information needs and further

allows the integration of preferences of data ware-

house users in the specification process.

The presented solution generally can be transferred

to many different types of multidimensional model-

ling techniques. Examinations concerning the trans-

ferability have been successfully performed.

Methodical parts of the approach were tested in sev-

eral projects in practise. It is our aim to develop ex-

tensive reference models for different domains, e. g.

retail information systems, university administration,

and banks which are using our methodical exten-

sions (Knackstedt, Janiesch, Rieke 2006). These

reference models perform a significant contribution

to the explication of knowledge for the construction

of data warehouse systems and can stimulate future

research in this field. In practise these models can

support a faster and more sophisticated development

of data warehouse systems by providing suitable

initial solutions. The analysis of the benefits of these

models in reality will be an important aspect of fur-

ther research.

Another aspect of future research can be seen in the

development of software based modelling tools

which support model-building and their application

(Delfmann et al. 2006). This could lead to a basic

stimulation for the implementation of extensions in

modelling tools because the tools available on the

market lack these functions.

REFERENCES

Abello, A.; Samos, J.; Saltor, F. (2000). A Data Ware-

house Multidimensional Data Models Classification.

Technical Report LSI-2000-6. Dept. Llenguages y Sis-

temas Informáticos (Universidad de Granada), 2000.

Adamson, C.; Venerable, M. (1998). Data Warehouse

Design Solutions. New York 1998.

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

Kuropka, D. (2004). Configurative Process Modeling

– Outlining an Approach to Increased Business Proc-

ess Model Usability. In: Khosrow-Pour, M. (Eds.): In-

formation Through Information Technology. 2004 In-

formation Resources Management Association. Inter-

national Conference. New Orleans, Louisiana, USA.

May 23-36, 2004. Hershey et al. 2004, pp. 615-619.

Blaschka, M. (2000). FIESTA. A Framework for Schema

Evolution in Multidimensional Databases. PhD-

Thesis, München 2000.

Bulos, D. (1996). A New Dimension. OLAP Database

Design. Database Programming & Design, 9 (1996) 6,

pp. 33-37.

Burmester, L.; Goeken, M. (2006). Method for User Ori-

ented Modelling of Data Warehouse Systems. In:

Manolopoulos, Y. et al. (ed.): Proceedings of the 8th

International Conference on Enterprise Information

Systems, ICEIS 2006, May 2006 Paphos/ Zypern.

Chan, P.; Shoshani, A. (1981). SUBJECT – A Directory

Driven System for Organizing and Accessing Large

Statistical Databases. In: IEEE Computer Society:

Proceedings of the 7th International Conference on

Very Large Data Bases (VLDB’81). Cannes, France,

9.-11. September 1981. Cannes 1981, pp. 553-563.

Chen, P. P.-S. (1976). The Entity-Relationship Model –

Toward a Unified View of Data. ACM Transactions

on Database Systems, 1 (1976) 1, pp. 9-36.

Copeland, T. Koller, T.; Murrin, J. (1990). Valuation:

Measuring and Managing the Value of Companies.

New York et al. John Wiley & Sons, 1990.

Darke, P.; Shanks, G. (1996). Stakeholder Viewpoints in

Requirements Definition. Requirements Engineering,

1 (1996) 1, pp. 88-105.

Delfmann, P.; Janiesch, C.; Knackstedt, R.; Rieke, T.;

Seidel, S. (2006). Towards Tool Support for Configu-

rative Reference Modeling – Experiences from a Meta

Modeling Teaching Case. Erscheint in: Proceedings of

the 2nd Workshop on Meta-Modelling and Ontologies

(WoMM 2006). Karlsruhe, Germany. 2006.

Eccles, R. G. (1991). The performance measurement

manifesto. In: Harvard Business Review, 69 (1991) 1-

2, pp. 131-137.

Goeken, M. (2006). Entwicklung von Data-Warehouse-

Systemen. Anforderungsmanagement, Modellierung,

Implementierung. Wiesbaden 2006 [in german].

Goeken, M. (2004). Referenzmodellbasierte Einführung

von Führungsinformationssystemen. Grundlagen,

Anforderungen, Methode. In: Wirtschaftsinformatik

46 (2004) 5, S. 353 - 365. [in german]

Golfarelli, M.; Maio, D.; Rizzi, S. (1998a). The Dimen-

sional Fact Model – A Conceptual Model for Data

Warehouse. International Journal of Cooperative In-

formation Systems, 7 (1998) 2-3, pp. 215-246.

Golfarelli, M.; Maio, D.; Rizzi, S. (1998b). Conceptual

Design of Data Warehouses from E/R-Schemes. In: H.

El-Rewini (ed.): Proceedings of the 31st Hawaii Inter-

national Conference on System Sciences). 6.-9. Januar

1998. Los Alamitos et al. 1998. pp. 334-343.

Hahn, K.; Sapia, C.; Blaschka, M. (2000). Automatically

Generating OLAP Schemata from Conceptual Graphi-

cal Models. In: Proceedings of the ACM Third Inter-

national Workshop on Data Warehousing and OLAP

(DOLAP 2000). Washington D. C., USA, 10. Novem-

ber 2000.

Halmans, G., Pohl, K. (2004). Communicating the Vari-

ability of a Software-Product Family to Customers. In-

formatik - Forschung und Entwicklung, 18, 113-131.

Harren, A; Herden, O. (1999). MML and mUML – Lan-

guage and Tool for Supporting Conceptual Data

Warehouse Designs. In: Proceedings 2.GI-Workshop

Data Mining und Data Warehousing (DMDW99), pp.

57-68, Magdeburg (Germany), September 1999.

LWA99. Papers in edited volumes. [in german].

Hofstede, A.H.M., Verhoef, T.F. (1996). Meta-CASE: Is

the game worth the candle? In: Info Systems Journal,

(6), 1996, pp. 41-68.

MULTIDIMENSIONAL REFERENCE MODELS FOR DATA WAREHOUSE DEVELOPMENT

353

Holten, R. (2003). Specification of management views in

information warehouse projects. Information Systems,

28 (2003) 7, pp. 709-751.

Hopcroft, J. E.; Motwani, R.; Ullman, J. D. (2000). Intro-

duction to Automata Theory, Languages, and Compu-

tation. 2 ed. Boston, MA, USA: Addison-Wesley

Longman Publishing.

Kaplan, R. S.; Norton, D. P. (1996). Translating Strategy

into Action. The Balanced Scorecard. Boston, MA,

1996.

Kimball, R. (1996). Data Warehouse Toolkit – Practical

Techniques for Building Dimensional Data Ware-

houses. New York 1996.

Knackstedt, R.(2006). Fachkonzeptionelle

Referenzmodellierung einer

Managementunterstützung mit quantitativen und

qualitativen Daten. Methodische Konzepte zur

Konstruktion und Anwendung. PhD-Thesis, Berlin

2006. [in german]

Knackstedt, R.; Janiesch, C.; Rieke, T. (2006). Configur-

ing Reference Models - An Integrated Approach for

Transaction Processing and Decision Support. In:

Manolopoulos, Y. et al. (ed.): Proceedings of the 8th

International Conference on Enterprise Information

Systems, ICEIS 2006, May 2006 Paphos/ Zypern.

Knackstedt, R.; Klose, K. (2005). Configurative Reference

Model-Based Development of Data Warehouse Sys-

tems. In: Khosrow-Pour, M. (Hrsg.): Managing Mod-

ern Organizations with Information Technology.

IRMA, San Diego, California, USA. May 15-18, 2005.

Hershey et al. 2005, S. 32-39.

Kotonya, G.; Sommerville, I. (1995). Requirements Engi-

neering With Viewpoints. Technical Report: CS

EG/10/1995. Cooperative Systems Engineering

Group. Computing Department. Lancaster 1995.

Lapsley, I.; Mitchel, F. (1996). The Accounting Chal-

lenge: Accounting and Performance Measurement.

London, 1996.

Mertens, P.; Griese, J. (2002). Integrierte

Informationsverarbeitung. (Integrated Information

Processing) 9. eds, Wiesbaden 2002. [in german].

Mukherjee, D.; D'Souza. D. (2003). Think phased Imple-

mentation for successful Data Warehousing. In: In-

formation Systems Management, 20, 82 -90.

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

Huber, H. (1996). Managing Multiple Requirements

Perspectives with Metamodels, IEEE Software, 13

(1996) 3, pp. 37-48.

Nordstrom, G.; Sztipanovits, J; Karsai, G.; Ledeczi, A.

(1998). Metamodeling – Rapid Design and Evolution

of Domain-Specific Modeling Environments. In: Pro-

ceedings of the IEEE ECBS’98 Conference. Nashville,

Tennessee, April 1998. Nashville 1998, pp. 68-74.

Nuseibeh, B. A. (1994). A Multi-Perspective Framework

for Method Integration. Dissertation, University of

London 1994.

Poon, P.; Wagner, C. (2001). Critical success factors re-

visited: success and failure cases of information sys-

tems for senior executives. In: Decision Support Sys-

tems, 30, 393-418.

Rafanelli, M.; Bezenchek, A.; Tininini, L. (1996): The

Aggregate Data Problem – A System for their Defini-

tion and Management. Sigmod Record, 25 (1996) 1,

pp. 8-13.

Rafanelli, M.; Shoshani, A. (1990). STORM – A Statisti-

cal Object Representation Model. In: Z. Michaelewicz

(Eds.): Proceedings of the 5th International Confer-

ence on Statistical and Scientific Database Manage-

ment (SSDBM’90). Charlotte, USA, 3.-5. April 1990,

pp. 14-29.

Rosemann, M. (2003). Application Reference Models and

Building Blocks for Management and Control. In:

Handbook of Enterprise Architecture. Eds.: P. Bernus,

L. Nemes, G. Schmidt. Berlin et al. 2003, pp. 595-615.

Rosemann, M.; Shanks, G. (2001). Extension and Con-

figuration of Reference Models for Enterprise Re-

source Planning Systems. Proceedings of the 12th

Australasian Conference on Information Systems

(ACIS 2001). G. Finnie, D. et al. (Eds.). Harbour, 4-7

December, pp. 537-546.

Sapia, C.; Blaschka, M.; Höfling, G.; 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’98). Lecture Notes in Computer Science.

Singapur, 19.-20. November 1998, pp. 105-116.

Spitta, T. (1997). Wiederverwendbare Attribute als

Ausweg aus dem Datenchaos. In: HMD, 34 (1997)

195, pp. 38-55. [in german].

Thomsen, E. (1997). OLAP Solutions – Building Multi-

dimensional Information Systems. New York 1997.

Trujillo, J., Palomar, M., Gómez, J., Song, I.-Y. (2001).

Designing Data Warehouses with OO Conceptual

Models. IEEE Computer 34, 66-75

Vassiliadis, P. (2000). Gulliver in the land of data ware-

housing: practical experiences and observations of a

researcher, Proceedings of the International Workshop

on Design and Management of Data Warehouses

(DMDW’2000), June 5-6, Stockholm, Sweden.

Vassiliadis, P.; Bouzeghoub, M.; Quix, C. (2000). To-

wards Quality-Oriented Data Warehouse Usage and

Evolution, Information Systems 25 (2000) 2, pp. 89-

115.

Watson H., Fuller C., Ariyachandra T. (2004). Data ware-

house governance: best practices at Blue Cross and

Blue Shield of North Carolina. Decision Support Sys-

tems, 38, 435-450.

Wixom, B. H.; Watson, H. J. (2001). An Empirical Inves-

tigation of the Factors Affecting Data Warehousing

Success. In: MIS Quarterly, 25, 17 - 41.

ICEIS 2007 - International Conference on Enterprise Information Systems

354