Towards Enterprise Management Systems

A Generic and Flexible Information Representation Approach

Andreas Lux

, Jörg Hess

and Rudi Herterich

Department of Computer Science, Trier University of Applied Sciences, Schneidershof, 54293, Trier, Germany

DHC Business Solutions GmbH & Co. KG, Landwehrplatz 6-7, 66111, Saarbrücken, Germany

Keywords: Enterprise Management System (EMS), Integrated Management System, Information Object Model,

Context Resolution, DHC VISION.

Abstract: This paper describes an enterprise management system which provides an integrated approach to tackle

management relevant topics like business process management, quality management, audit management,

internal risk and control management, enterprise architecture management, compliance management and so

on within a single information system. The new term “enterprise management system (EMS)” is introduced

for such a kind of management system. The core idea is a generic, yet flexible object model for any kind of

relevant information, e.g. audit, risk, measure, process, process step, control. The different kind of

information is related to each other by links with semantic expressiveness. The approach was successfully

implemented in the enterprise management system DHC VISION, which is used today worldwide by

medium- and large-sized enterprises.

1 INTRODUCTION

Nowadays, management systems become a necessity

for multi-site companies operating worldwide. There

is a need to fulfill a vast number of legal

requirements, national and international guidelines

and norms. One the other hand, enterprises always

need to improve their processes and structures to

gain competitive advantages. Business process

management (BPM, vom Brocke and Rosemann,

2010) and governance, risk and compliance (GRC,

Tarantino, 2008) are the key approaches to master

this challenge.

But today, most enterprises only have separate

software solutions for purposes like business process

management, quality management (QM), audit

management (AM), internal risk and control mana-

gement (ICS), enterprise architecture management

(EAM), compliance management and so on. As we

know from our consulting experience, information is

often stored redundantly in several systems with

disadvantages like redundancy, internal and external

inconsistency, reduced changeability and ill-

structuredness. None of the existing systems on an

operational level like ERP software, business

intelligence software, business process management

software or office products (e.g. spreadsheets) fulfill

the functional requirements of a holistic mana-

gement system, especially with respect to integration

aspects. Furthermore, most of them are not usable on

a daily basis within the described management

context.

There is a strong demand for the development of

integrated solutions. In the midst of the 1990s, first

steps have been made towards so-called integrated

management systems (Laudon and Laudon, 2010),

which comprise the three core topics of quality

management, health and safety management and

environmental management. The focus here is on an

integrated management and documentation of

quality handbooks, operating instructions, training

material for norms like e.g. ISO 9001, ISO 14001

and OHSAS 18001 (Pardy and Andrews, 2009,

ISO, 2008).

New approaches try to enhance these integrated

management systems to a so-called enterprise

management system (EMS). The main idea is the

integration of all management-relevant information

in one global database, starting with the description

of strategic goals, business processes and attached

key performance indicators (KPI), and the

corresponding description of the organization with

the roles and responsibilities of its employees (DHC

Business Solutions GmbH, 2013). However, due to

251

Lux A., Hess J. and Herterich R..

Towards Enterprise Management Systems - A Generic and Flexible Information Representation Approach.

DOI: 10.5220/0004419302510257

In Proceedings of the 15th International Conference on Enterprise Information Systems (ICEIS-2013), pages 251-257

ISBN: 978-989-8565-61-7

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

technical and business-oriented barriers, the

development of such a system is far from being

easy. Nevertheless, such a kind of process-driven

enterprise management system provides among

others the following main advantages:

 transparency for the management concerning

the whole enterprise, i.e. structures,

operational and management processes;

 consistent, non-redundant database content;

 avoidance of duplication of work and

reduction of work complexity, especially with

regard to data recording;

 avoidance of media breaks;

 consistent documentation of processes and

support of approval workflows;

 uniform user interface.

By providing the aforementioned functionality,

an EMS bridges the existing information gap

between the management level and the operational

level. This paper describes the approach and the

system architecture of such an EMS. The approach

was carefully conceived within the last five years

and successfully implemented within the software

product DHC VISION (DHC Business Solutions

GmbH, 2013).

2 INFORMATION

REPRESENTATION APPROACH

A database with the description of the enterprise

organization, the business processes and associated

information like forms, handbooks, IT systems, etc.

is the basis of the system. Process orientation is the

backbone of an EMS. Therefore, an EMS must

provide functionalities for the graphical description

and linkage of business processes to documents and

organizational units. Different levels of process

abstraction from a headquarters perspective up to

location-specific variants and different types of

process models (e.g. process maps, process

hierarchies, BPMN diagrams, etc.) must be

representable. Processes can also be seen as a kind

of master data for an EMS. The semantics of

business processes and their implementation can be

described in a hierarchical way in form of a business

process pyramid which provides the procedural basis

for management solutions like e.g. risk management

or audit management (Lux, Hess and Herterich,

2013).

In this paper, we concentrate on the approach to

represent master data of an EMS in a generic and

very flexible manner. In the following, we denote

any kind of master data as information resp.

information object.

Because of the heterogeneity of information and

the goal of integrating several stand-alone

management solutions in one EMS, the following

functional requirements must be addressed:

 A generic representation concept for any kind

of information must be provided - is it a

process, an organizational item or any other

kind of item (document, risk, measure, etc.).

 There must be a versioning mechanism to log

information changes.

 Information must be available in different

languages.

 Information must be modifiable according to

different locations/sites of an enterprise.

 Information must be related to each other, e.g.

it must be possible to link a document to a

process step.

 A sophisticated and flexible search routine

must exist to handle the different versions,

language- and location-specific occurrences of

certain information.

 Traceability of any kind of information object

must be guaranteed.

Besides these, the following non-functional

requirements should be addressed:

 Security: Only users with corresponding rights

are allowed to read, create, modify or delete

information objects.

 Scalability / performance: The 4-tier client/

server architecture (see section 4 in more

detail) of the EMS is scalable and is able to

handle hundreds of users with acceptable

response time behavior.

 Maintainability and administration: The EMS

should be easy to maintain. This is realized by

mechanisms like Active Directory linkage and

Single-Sign On (SSO).

Within a world-wide operating enterprise,

information of an EMS can be created, modified,

read or deleted depending on the location, language

and role of a user. Location, language and role of an

EMS user should be encapsulated in one or more

To efficiently manage any arbitrary information,

it has to be associated with meaningful object types

like e.g. audit, risk, measure, process, process step,

control, operating procedure etc. The main idea of

our approach is that each kind of information is

managed in a so-called information object, which

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems

252

has a certain object type. The object type classifies

the information. Each object type has a freely

configurable set of attributes, which describes the

type in more detail. Free configuration both means

selection of pre-defined attributes as well as

definition of new, enterprise-specific attributes

(customizing).

Information objects can be related to each other

by typed links. Typical examples are “is process-

oriented superior”, “is reference from”, “is informed

by”, “has risk”, etc. As object types, link types are

also freely configurable and can be described by

attributes. Link types are used to represent process-

oriented relationships (e.g. is executed by),

hierarchies (e.g. is superior), topic-oriented

relationships (e.g. has risk) or general connections

(e.g. is reference from).

Furthermore, information can be structured by

so-called structural units, which are comparable to a

folder structure in a file system. Structural units have

an object type, too. All system functions regarding

data management are realized on the basis of

information objects and structural units.

An EMS object – either an information object or

a structural unit – now consists of exactly one object

wrapper (logical view) and one or more object

contents (physical view) with respective attributes,

see figure 1.

Figure 1: EMS Information Object Model (simple).

The object content contains the proper information

and is identified by version, status (e.g. published, in

progress/draft), language and location.

It has the following attributes:

 Version, status, language, and location are the

identifiers of the object content within the

object wrapper and are required for context

resolution (details see below).

 Links: Typed links have their anchor on the

object content itself, but refer to all derived

information collected within an object

wrapper (details see below).

 Change history: All changes to object content

can be logged as well.

 Properties: Via the object type, attributes can

be freely defined. These attributes are then

manageable on the object content.

 KPIs: Information objects can manage,

compute and aggregate KPIs.

 Application/Document: The file or application

associated to the information object is

managed here.

 Display format: If the attribute

“application/document” contains a document,

the document is automatically converted into a

display format (HTML or PDF) for display.

 Print format: If the application/document

attribute contains a document, the document is

automatically converted into a print format

(PDF or source application format) for

printing.

The object wrapper contains the different object

contents which can be seen as “variants” of the same

information. The object wrapper of an EMS object

has the following attributes:

 Object type: This attribute is used for typing

objects. Most of the configuration settings are

executed on the object type.

 Access Rights: Read permission is set for all

object contents. Write, delete and translate

rights can be defined according to users of

different locations/enterprise sites.

 Versioning mode: The versioning mode

defines the procedure how new versions of an

information object are build.

Information objects are connected with other

information objects. The relationships are also

represented via typed links (see figure 2). Such a

link always leads from the object content to the

object wrapper of the target object. The most

appropriate object content within the target object is

determined via a so-called context resolution

algorithm (see below) and is displayed to the user.

Figure 2: Links between Information Objects.

Typed Links

Information object

Context Resolution

TowardsEnterpriseManagementSystems-AGenericandFlexibleInformationRepresentationApproach

253

The object wrapper and its attached information

contents allow for a decoupling of links between

information objects and their respective individual

“variants“. Since all EMS functions are based on

typed information objects and have not been

implemented on specific object contents, this

approach exhibits a very high flexibility for

designing a generic database for an enterprise

management system.

3 CONTEXT RESOLUTION

ALGORITHM

In what follows we give a precise description of the

context resolution procedure. If an EMS information

object is accessed by a user, the algorithm

determines the most appropriate object content

within the object wrapper. For this purpose, it uses

the following criteria:

 user role, e.g. reader or author;

 user rights, e.g. read or create or modify or

delete or translate;

 language, e.g., English (EN), French (FR) or

German (GER);

 location, e.g. corporate headquarters or

location London, Paris, Munich etc. This

concept is in the style of the SAP client

concept.

Location, language and user role can be derived

from the login profile. User rights are granted

individually for each EMS information object. Each

location has its associated original language, in

which the information is maintained (see figure 3).

Figure 3: EMS Information Object Model (more

complex).

The most appropriate content of an EMS

information object is determined according to the

following rules:

Rule 1: Only One Location and only One Language

 A user with the role “reader” and the

corresponding access rights gets the latest

version with status published.

 A user with the role “author”, who only has

read access rights on that object, gets the latest

version with status published.

 A user with the role “author” and the

corresponding rights for modification gets the

latest version with status “in progress”, if such

a version exists. Otherwise, the latest version

with status published is opened.

Rule 2: Several Languages

 If an actual version of the user’s login

language (e.g. French) exists, this version is

displayed; otherwise, the latest version in

English is displayed.

 If that version does not exist, the information

is displayed in its original language (e.g.

German).

Rule 3: Several Locations (SAP-like Client Concept)

 If an actual version in the “login-location”

exists, this version is displayed.

 Otherwise, an actual version is searched for in

the “predecessor location”. The predecessor is

determined by an algorithm following the

information object link backward to the EMS

object from which it was created. This

backward step is repeated until the appropriate

content is found in the initial EMS object at

the latest.

Rule 4: Several Locations and Several Languages

 In general, the following rule applies: actual

version dominates language version.

To clarify the context resolution algorithm,

consider the following examples (see figure 4).

Figure 4: Context Resolution.

Example 1: A German-speaking employee

located in the German Headquarters gets the actual

version of the information object (Version 2 with

status “published”), if she has read access; as an

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems

254

Figure 5: System Architecture.

author with write access rights she gets Version 3

with status “in progress (draft)”. Given the history of

an information object by a version list, an author can

also open e.g. the latest published version.

Example 2: A Spanish-speaking employee

located in the German Headquarters gets the actual

version 2 with language English and status

“published”, because there is no Spanish version of

the information object.

Example 3: A French-speaking employee located

in Paris gets actual version 2 with language French

and status “published”. This information object was

derived as a location-specific variant of the

Headquarters information object.

4 SYSTEM ARCHITECTURE

On the basis of the described approach the EMS

DHC VISION was designed and implemented.

Technically, it is based on a scalable 4-tier archi-

tecture as presented in figure 5.

All data are stored in a central repository

implemented on a relational database system from

Oracle or Microsoft SQL Server. Communication

with the system server is realized via SQL and

ODBC.

The system server, which contains the

application logic, consists of four separate servers,

which communicate with each other via a self-

implemented protocol based on XML. Customers

can use that protocol to integrate legacy applications

and exchange data with the DHC VISION system.

The application module contains the business logic

of the EMS. The service module transforms

documents from their source format to their defined

display, print or download format. The scorecard

module is responsible for the management,

calculation and aggregation of performance indica-

tors. The information module handles modification

of data and search queries and triggers actions

according to data modification. For reasons of

scalability, each module can be instantiated a

number of times and can be run on the same or on

different physical machines.

The web server is used to provide the

information for presentation. Here, a data cache is

realized for quicker data access.

The workplace layer contains the graphical user

interface of the software (Tidwell, 2011). Figure 6

presents a rough overall impression of the interface.

A sales process is presented both graphically as a

process structure diagram and textually with its

corresponding information stored in different tabs.

The user interface is web-based and runs with

established browsers like Internet Explorer, Firefox

etc.

TowardsEnterpriseManagementSystems-AGenericandFlexibleInformationRepresentationApproach

255

Figure 6: User Interface, here Process Visualization and Process Cockpit.

MS Office products, MS Visio and MS Project

are integrated in DHC VISION, so that a user can

create, read and model the information objects with

his/her familiar working tools. The user interface

also provides functionalities for handicapped

persons (e.g. screen reader support, optical

enlargement).

5 RELATED WORK

Most of the work described in this paper is

influenced by graph theory (Chein and Mugnier,

2009), knowledge representation approaches,

especially semantic networks (Sowa, 1991), and

effective techniques for retrieving and processing

semi-structured and XML-based data via the

evaluation of path expressions (Abiteboul, Buneman

and Suciu, 1999). Nevertheless, the implemented

DHC VISION core functionality to manage a

network of information objects is self-conceived and

unique.

Within the last years, more and more software

vendors, originating mainly from the area of

business process management systems, pursue

comparable approaches to develop integrated

enterprise management systems.

Companies like e.g. BOC Group (BOC Group,

2013), BWise (BWise, 2013), EMC Corporation

(EMC Corporation, 2013) or Software AG (Software

AG, 2013) provide tools with similar approaches

and features. Because all these systems are

proprietary ones, we do not know their internal

information representation structures and the

underlying information processing algorithms.

Nevertheless, confirmed by feedback of

interested parties and customers, our approach is

rated as a sound and sophisticated one with regard to

flexibility and extensibility. Several awards won by

the software DHC VISION within the last two years

emphasize this statement (DHC Business Solutions

GmbH, 2013).

6 CONCLUSIONS

The technical approach of DHC VISION is a state-

of-the-art 4-tier client/server architecture for an

enterprise management system. All the underlying

information can be addressed indirectly and

dynamically on the basis of a complex information

network with typed links and a sophisticated

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems

256

versioning mechanism. This generic approach is

used to describe and specify well-known

management methods like BPM, QM, ICS, EAM by

defining their semantic data as generic EMS

information objects; therefore, also customer-

specific management methods can be easily

integrated as add-on module to the core system. New

information objects for new management solutions

can be specified by a customizing component and do

not have to be implemented in a hard-coded form.

Within the last three years, the DHC VISION

system was completely re-designed and

implemented towards an integrated EMS with a

huge set of standard management solutions. The

concept of EMS information objects guarantees the

re-usability of important enterprise information like

processes, process steps, organizational charts

(organizational units, roles), IT systems and general

objects like resources, documents, training material,

measures, risks, audits, controls, KPIs etc. All these

objects can consistently be used for the definition of

an integrated enterprise management solution with

focus on business process management, quality

management, audit management, internal risk and

control management, enterprise architecture

management, compliance management.

ACKNOWLEDGEMENTS

We want to thank the people that have directly or

indirectly contributed to the creation of this paper.

Special thanks go to C. Houy (DHC Business

Solutions GmbH & Co. KG) and G. Rock (Trier

University of Applied Sciences) for fruitful

discussions and very helpful comments on drafts of

this paper.

REFERENCES

Abiteboul S., Buneman, P., Suciu, D., 1999. Data on the

Web: From Relations to Semistructured Data and

XML. Morgan Kaufmann. 1

edition.

BOC Group, 2013. Business Process Management with

ADONIS, Online, available from Internet:

http://www.boc-group.com/products/adonis, last

access: 2013/01/30.

BWise, 2013. GRC Platform, Online, available from

Internet: http://www.bwise.com/solutions-services/-

grc-platform, last access: 2013/01/30.

Chein, M., Mugnier, M., 2009. Graph-based Knowledge

Representation, Computational Foundations of

Conceptual Graphs. Springer.

DHC Business Solutions GmbH, 2013. Software DHC

VISION, Online, available from Internet:

http://www.dhc-gmbh.com/software_dhcvision and

http://www.dhc-gmbh.com/de/software-dhc-vision/

awards, last access: 2013/04/10.

EMC Corporation, 2013. RSA Archer eGRC, Online,

available from Internet: http://www.emc.com/-

security/rsa-archer.htm, last access: 2013/01/30.

ISO, 2008. The integrated use of management system

standards. ISBN 978-92-67-10473-7. ISO Online

Store.

Laudon, K., Laudon, J., 2010. Management information

systems: Managing the digital firm. Pearson Prentice

Hall, 11

edition.

Lux A., Hess J., Herterich R., 2013. Business Process

Management as Basis for Enterprise Management

Systems. Submitted for publication.

Pardy, W., Andrews T., 2009. Integrated Management

Systems: Leading Strategies and Solutions.

Government Institutes.

Software AG, 2013. ARIS Platform, Online, available

from Internet: http://www.softwareag.com/corporate/-

products/aris_platform/default.asp, last access:

2013/01/30.

Sowa, John F., 1991. Principles of semantic networks.

Explorations in the representation of knowledge,

Morgan Kaufmann. San Mateo, Cal.

Tarantino, A., 2008. Governance, Risk, and Compliance

Handbook: Technology, Finance, Environmental, and

International Guidance and Best Practices. John Wiley

& Sons.

Tidwell, J., 2011. Designing Interfaces, O'Reilly Media,

edition.

vom Brocke, J., Rosemann, M. (2010), Handbook on

Business Process Management: Strategic Alignment,

Governance, People and Culture (International

Handbooks on Information Systems). Springer, Berlin.

TowardsEnterpriseManagementSystems-AGenericandFlexibleInformationRepresentationApproach

257