Enterprise Integration and Interoperability Improving Business

Analytics

Georg Weichhart

PROFACTOR GmbH, Steyr, Austria

Keywords:

Enterprise Integration, Enterprise Interoperability, Business Analytics, Smart Grids.

Abstract:

In applied research and industrial business analytics (BA) projects data preparation requires around 80% of the

total effort. Preparation tasks include establishing technical, semantic interoperability of data and processes

to generate value. Enterprise Integration and Interoperability (EI2) approaches address these challenges, but

these approaches are hardly taken into account in business analytics. In this position paper, we analyse ap-

proaches for their contribution to improving business analytics by supporting the interoperability of data,

services, processes and business in general. For more details, we focus on the application domain of smart

grids. Existing and missing tool and methodological support as a basis for data-access required for efﬁcient

and effective descriptive, predictive and prescriptive business analytics.

1 INTRODUCTION

Before being able to analyse the intersection and con-

tributions of research in Enterprise Integration and In-

teroperability to Business Analytics, we brieﬂy sketch

both ﬁelds.

Business Analytics (BA) is a research ﬁeld where

quantitative methods meet decision making and infor-

mation technology (Hindle et al., 2020). Overall goal

is to generate business value by supporting actionable

decision making using descriptive, predictive or pre-

scriptive methods. This relies on an appropriate data

infrastructure. In addition to this, adaptive organisa-

tional capabilities to react to data and results from

the methods are required (Dubey et al., 2021; Omar

et al., 2019). Business Analytics enables to realize

goals like efﬁciency, effectiveness, ﬂexibility, and re-

silience. Business and Data Analytics methods are

applied to many domains. Electricity Networks is the

domain which we will use as application domain be-

low.

Enterprise Integration and Interoperability (EI2)

(Weichhart et al., 2021a; Weichhart et al., 2021b)

aims to research approaches enabling seamless in-

formation exchange between information systems (in

the general sense). Multiple systems are enabled to

communicate, coordinate and collaborate (Vernadat,

2010) in order to work towards a joint goal. This

https://orcid.org/0000-0002-1405-5825

includes technical systems like software (services),

physical systems like networked (manufacturing) ma-

chines and social systems like departments. The dif-

ference between the two endpoints of a continuum be-

tween integration and interoperability, is the degree

of coupling. Interoperability aims at loose coupling

of systems (e.g. federated interoperability). Interop-

erability provides approaches for systems-of-systems

and Cyber-Physical Systems (CPS), where the sys-

tems are developed independently. Integration is fo-

cusing on tight coupling, which includes a common

information model.

With respect to the underlying assumptions, both,

BA and EI2, share some common ground. From

a business informatics (in German Wirtschaftsinfor-

matik) point of view, both use IT to support organi-

sational tasks in the context of the enterprise. In this

context multiple human actors use different software

(and hardware) tools. Efﬁcient and effective support

by software for organisational tasks is a key aspect.

In applied research and industrial analytics

projects data preparation consumes 80% of the to-

tal effort, before any analysis method can be ap-

plied. The challenges for data preparation include,

identiﬁcation of data sources, pre-structuring of data

sets, cleaning of data, harmonisation and integration

of heterogeneous data sets by establishing technical

and semantic interoperability. EI2 can ﬁll these gaps.

There is a huge potential for EI2 methods and ap-

proaches for enabling better analytics projects. In-

Weichhart, G.

Enterprise Integration and Interoperability Improving Business Analytics.

DOI: 10.5220/0010761600003062

In Proceedings of the 2nd International Conference on Innovative Intelligent Industrial Production and Logistics (IN4PL 2021), pages 227-235

ISBN: 978-989-758-535-7

227

teroperability is also needed on business layer with

respect to analytics processes to generate value from

the data.

In this paper, we discuss, on conceptual level, the

suitability of Enterprise Integration, Interoperability

approaches for business analytics.

In the following we ﬁrst discuss and deﬁne enter-

prise integration and interoperability. This is followed

by a discussion of data preparation and business ana-

lytics challenges. To provide a more concrete picture

smart grids (energy networks) have been chosen to

provide examples for challenges. In the last but one

section, we discuss contributions of EI2 to BA.

2 ENTERPRISE INTEGRATION

AND INTEROPERABILITY

We ﬁrst provide brief deﬁnitions of Enterprise In-

tegration and Enterprise Interoperability. We use

a model to classify approaches on different levels.

These approaches are then discussed.

2.1 The Integration-interoperability

Continuum

Enterprise Integration is a research approach, that fo-

cuses on the interaction of information systems (in

the general sense). Integration of involved systems

ensures that an overarching objective is reached by

combining the functions of the systems (Morel et al.,

2007; Panetto et al., 2012). Systems are analysed on

multiple levels. Seamless data and information ﬂows

implies that data is exchanged, using a common tech-

nical format and semantic information models are de-

ﬁned in a common ontology. All systems and inter-

faces are aligned to enable this exchange. Services

and business processes of organisational systems are

aligned to align work-ﬂows to effectively meet busi-

ness goals.

The above described full and tight integration is

one end of the EI2 continuum. In this approach a

common model is used by all involved system. This

makes it easy to communicate and exchange informa-

tion, because it works seamlessly and is understood

in the same way by all. In short, syntax is standard-

ised, semantics is well deﬁned. Enterprise Integration

can be deﬁned as follows: “provide the right informa-

tion at the right place and at the right time and thereby

enable communication between people, machines and

computers and their efﬁcient cooperation and coordi-

nation” (Kosanke et al., 1999, p.85)

Opposite of tight and full integration, interop-

erability is found. Enterprise Interoperability is

grounded in Enterprise Integration and there is an

overlap between approaches. Enterprise Interoper-

ability is extending Enterprise Integration so that it

will be possible to meet the requirement of having

loose coupled systems (Weichhart et al., 2021a). In-

teroperability does not impose uniﬁed models. Inter-

operability assumes independent systems that follow

their own goals. As such this supports systems-of-

systems where systems are heterogeneous and inde-

pendent. The independence of systems (in a system-

of-systems) supports the evolution of sub-systems.

The loose coupling of systems-of-systems (as in con-

trast to integrated systems), makes them fundamen-

tally different. This is not only true for the struc-

ture but also for engineering processes (BKCASE Ed-

itorial Board, 2019; Morel et al., 2007). An inte-

grated system may be engineered in a process where

parts, which are under full control, are combined. In

contrast to that, interoperability is not a static state,

but a dynamic (and sometime continuous) process.

Individual systems in a system-of-systems maintain

their autonomy and evolve. In order to maintain in-

teroperability, the interactions in a system-of-system

need to be monitored and interoperability needs to be

re-established (by systems-engineers) when it is lost

(Ducq et al., 2012; Naudet et al., 2009; Weichhart

et al., 2021c).

2.2 Enterprise

Integration/Interoperability

Frameworks

In the following we are focusing on the problem space

described in enterprise integration research (Ducq

et al., 2012; Weichhart et al., 2021a). This problem

space is very well suited for enterprise integration as

well.

The ﬁrst dimension discussed in EI2 research ad-

dresses semiotic levels in the organisation (Stamper,

1993; Stamper et al., 2000). It addresses gaps and

barriers between multiple systems. The technological

level addresses the encoding of information (syntax).

The semantic level addresses the meaning of informa-

tion. This includes the conceptualisation of informa-

tion. The organisational level addresses the pragmat-

ics of an organisation. In some frameworks there is

also a societal level (legal level). This level includes

cultural aspects, encoded as laws, within which the

organisation has be behave (Panetto et al., 2019; We-

ichhart et al., 2021c).

In some approaches a second dimension is used

to specify a level of granularity (Chen et al., 2008).

EI2N 2021 - IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

228

It ranges from data over (technical) services and pro-

cesses to the business. The service addresses the pro-

vision of a single function. The process addresses

multiple, coordinated functions. The business the

overall enterprise.

The second or third dimension is the solution di-

mension. It ranges from federated interoperability

(loose coupling) to tight integration. In the next sec-

tion, we discuss approaches that provide interoper-

ability solutions. The approaches are discussed on

one level of the problem space, but often address mul-

tiple levels.

2.3 Technical Approaches

On technical level, approaches exist that exchange

data using messages or service APIs. With service-

based approaches both the used API (application pro-

gramming interface) and the data structures need to

be known. Some generic interaction protocols like

http(s) provide a more generic standard for coding

of web-services building on well known underlying

mechanisms like post and get methods to implement

a transparent interaction protocol like REST (REp-

resentational State Transfer) (Fielding and Taylor,

2002). This particular architectural approach has been

designed to support internet scale interoperability.

More recently message-based architectures gained

importance. A message broker architecture like

MQTT

and AMQP

provide standardised means to

decouple systems. Messages, well known to all par-

ticipants (on syntax and semantics level), are sent to

a broker which provides multiple topics. That com-

ponent then informs other systems that expressed the

interest in the same topic. This decouples processes

and supports concurrency.

An older approach to message-based interaction

are multi agent systems (MAS) (Wooldridge, 2009).

In contrast to broker-based systems there the systems

(agents) are communicating directly in a peer-to-peer

fashion. The syntax is provided by agent standards

like FIPA (Foundation of Intelligent Physical Agents)

(FIPA - Foundation for Intelligent Physical Agents,

2005). In that standard (as an example) the semantics

of a message is deﬁned through ontologies where in

every message the used ontology is speciﬁed.

2.4 Semantic Approaches

As mentioned above, ontologies are one way to de-

ﬁne the meaning of words and symbols (Gu

edria and

Naudet, 2014; Haller and Polleres, 2020). Ontologies

http://mqtt.org/

https://www.amqp.org/

also allow to specify relationships between words and

have mechanisms for providing rules. Depending on

the goal of an ontology different approaches can be

identiﬁed. An upper ontology like Cyc (Lenat et al.,

2004; Lenat, 2004) deﬁnes the meaning of central but

often abstract words and symbols to link the seman-

tics of different domains. Middle ontology are more

concrete than upper ontology and provide more con-

crete concepts, but are focused on a domain (of dis-

course) (Haller and Polleres, 2020). Domain-specifc

ontologies (e.g. the Ontology of Enterprise Interoper-

ability, OoEI (Naudet et al., 2010)) provide concepts

and rules with a particular focus on a domain.

Research in linked data is aiming to publish struc-

tured data in a decentralized and bottom-up manner.

Linking middle and domain ontologies in a top level

ontology an integrated ontology enabling automatic

retrieval should be possible and the (Linked Open

Data) LOD-Cloud

was created (McCrae et al., 2019;

Polleres et al., 2020).

This OoEI is one example of a domain speciﬁc

ontology (deﬁning concepts for enterprise interoper-

ability). Over the years it has evolved, and has been

extended using concepts from general systems theory

(von Bertalanffy, 1969; von Bertalanffy, 1950). This

allows to deﬁnes the enterprise as a system interact-

ing with other systems in an environment (Gu

edria

and Naudet, 2014).

In another approach, this OoEI has been the basis

where the ontology has been replaced by a Domain

Speciﬁc Language (DSL) (Weichhart et al., 2016a).

The goal of that project (OoEI

CAS

) was to re-use

the ontological concepts and extend it with an inte-

grated agent model, to support the description of dy-

namic and complex adaptive systems and their inter-

actions (Holland, 1998). The Domain Speciﬁc Lan-

guage (DSL) in OoEI

CAS

is implemented in the func-

tional programming language SCALA (Wampler and

Payne, 2014).

These approaches provide meaning to words and

symbols and also provide a conceptualisation of the

domain. The used examples above, are from the do-

main of enterprise interoperability, but several ontolo-

gies exist for different domains (Haller and Polleres,

2020).

2.5 Organisational Approaches

Enterprise Interoperability on organisational level is

researched by a few approaches. Building on the

view that the enterprise is a complex adaptive system

(Weichhart et al., 2016b) the S

ˆ3

-Enterprise (Sensing,

https://lod-cloud.net/

Enterprise Integration and Interoperability Improving Business Analytics

229

Smart and Sustainable) provides different view-

points on enterprise systems to capture different

aspects. The approach follows ISO/IEC 10746

ODP-RM-Open Distributed Processing - Refer-

ence Model and deﬁnes the following viewpoints

(ISO/IEC/JTC1/SC7, 2009):

enterprise viewpoint: the enterprise system and its

environment

information viewpoint: semantics of information

and its processing

computational viewpoint: functional decomposi-

tion and the distribution of (data) objects

engineering viewpoint: mechanisms and functions

supporting distributed interaction between objects

in the system

technology viewpoint: choice of technology in that

system

Overall goal is an architecture that works like an

enterprise operating system (EOS) and services are

provided through the EOS abstraction. Data from in-

telligent sensors is transferred to artiﬁcial and human

agents for smart decision making (Weichhart et al.,

2021c).

The MISA approach is focusing on a methodol-

ogy for collaborative organisations (B

enaben et al.,

2013; B

enaben et al., 2015). It provides a method

and supporting tools for interoperability in collabora-

tive, organisational networks. Focus here is also the

dynamics in networks and the independence of organ-

isations. This approach also conceptualizes the enter-

prise as a complex adaptive system.

3 DATA PREPARATION IN

BUSINESS ANALYTICS

Business Intelligence is a term for the processes and

tools that allows to discover valuable information and

knowledge in the companies’ databases (Zhang et al.,

2018). Big Data research is driving the ability of

systems to handle large and dynamic data (streams).

These two research ﬁelds provide a technical basis

for Business Analytics (BA) (Holsapple et al., 2014).

However, BA methods include (quantitative) methods

rooted in Operations Research (OR), Machine Learn-

ing (ML), and Decision Making (Hindle et al., 2020).

Overall the rationales for BA is to gain compet-

itive advantage and improved business performance

by generating value from data for informed decision

making and actionable insight (Holsapple et al., 2014;

Hindle et al., 2020).

The current increase in networked information

systems and processing power (in the edge and the

cloud) reduce the costs and effort for gathering big

amounts of data, analysing it with quantitative meth-

ods. New insights for business users can be gener-

ated in a variety of domains like software, marketing,

customer, supply chain, electricity (Holsapple et al.,

2014).

Business Analytics methods can be divided in

three approaches: (a) descriptive analytics, (b) pre-

dictive analytics, (c) prescriptive analytics. The ﬁrst

kind of methods make a situation transparent. The

second allow a decision maker (typically using statis-

tical methods) to predict a situation. The last kinds

of methods often stem from operations research and

quantitative models that allow to analyse multiple

courses of action and decision makers are able to

make optimized decisions.

“Success in business analytics is a complex mat-

ter, depending on a ﬁrm’s ability to harness ’simul-

taneously’ multiple resources and capabilities (peo-

ple, process, technology and organization) within a

business context, including the data itself (the input

and raw material), and deploy these synergistically

(key actions and decisions) to deliver a valued output”

(Vidgen et al., 2017, p. 634).

According to a study in the smart grid (SG) do-

main, the most signiﬁcant barrier in analytics projects

is the high data storage and manipulation costs. The

second most signiﬁcant barrier is data complexity and

the third data access issues (Bhattarai et al., 2019).

Data management, preparation, manipulation for

(business) analytics includes (Vidgen et al., 2017;

Zhang et al., 2018):

• integrating heterogeneous data sources

• improving and maintaining data quality

• cleansing and transforming data into common

meta-data structures

• governing data and meta-data

These issues are found in many, if not all, analyt-

ics project.

4 APPLICATION DOMAIN:

SMART GRIDS

The Smart Grid (SG) as application domain has many

data sources and generates Big Data. Some examples

are shown in ﬁg. 1. This ﬁgure shows how different

data sources in SG are.

For the management of small energy producing

units, found typically with sustainable energy genera-

tors (wind mills, water turbines, solar energy panels),

EI2N 2021 - IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

230

Figure 1: Data Sources in Smart Grid Environments (Zhang et al., 2018).

many more systems are generating relevant data. In

addition to the different types of data sources, these

are distributed in a large geographical area.

These sustainable energy systems, dependent on

weather conditions are less controllable. Analytics of

such energy networks for local micro grids is getting

more essential to maintain working energy networks

and avoid failures.

Taking the analytics approaches described above

we can identify the following services for analytic in

the energy domain.

Possible analytics services for SG with special at-

tention to sustainable energy systems includes (May-

ilvaganan and Sabitha, 2013; Veerlapati and Thota,

2021; Zhang et al., 2018):

1. Descriptive Analytics

(a) Asset health monitoring

(b) Fault detection

(d) Detection of energy loss

(e) Visualization of outage management

(f) Load disaggregation for reducing energy foot-

2. Predictive Analytics

(a) Electric device state estimation / health moni-

toring

(b) Predictive maintenance

(d) Renewable energy forecasting

(e) Load forecasting and proﬁling

3. Prescriptive Analytics

(a) Integrated resource allocation

(b) Transient stability analysis (resilience analysis)

(d) Balance the (predicted) load with energy pro-

ducers and consumers

The clustering into descriptive, predictive and pre-

scriptive services is not deterministic as most can be

placed in multiple categories. For example, load dis-

aggregation is the process of understanding the load

different devices generate at customer sites. This sup-

ports understanding where energy is consumed the

most. However, ultimate goal is to reduce the energy

consumption (by prescribing when to turn consuming

devices off).

Analysis of stability and the resilience requires to

simulate disturbances and analyse a systems possible

responses, in order to identify the best response. So it

includes a predictive and a descriptive component.

A larger set of standards exists covering many as-

pects of the smart grid and approaches towards a gen-

eral architecture have been researched (Uslar et al.,

2019), but there are still challenges in the context

Business Analytics in the Smart Grid domain. “With

the fast deployment of smart meters and advanced

sensors, huge amount of data with multiple types and

structures from deference sources with a variety of

protocols are generated every second. However, the

Enterprise Integration and Interoperability Improving Business Analytics

231

lack of standard data format for the information soft-

ware and database structures, as well as the issue of

interoperability of different information and commu-

nication systems deployed in the smart grids, make it

complicated and difﬁcult to obtain data for real ap-

plication. The traditional way of isolated storage of

the data in various systems also increases the barrier

for data sharing among applications.” (Zhang et al.,

2018, p.19).

Enterprise Integration and Interoperability can

provide several approaches that meet the need of BA

in general and the SG domain in particular.

5 CONTRIBUTIONS OF EI2 TO

Enterprise integration and interoperability (EI2) ap-

proaches enable, by deﬁnition, a more ﬂuent way

of data exchange. Heterogeneous data sources are

brought together and the data-structures are homog-

enized. This capability to make heterogeneous data

sources and data models interoperable is a pre-

condition for all analytics approaches. Some EI2 ap-

proaches provide direct or indirect support for prepar-

ing data for the different types of analytics.

5.1 Descriptive Analytics

EI2 (in general) supports online access to different

parts of the system. A traditional data preparation

process for business intelligence (BI) is to establish

an automated Extraction-Transformation-Load (ETL)

process. This process is engineered and leads to a

solution where the data source schemata and the tar-

geted business intelligence tool needs to be stable.

The ﬁxed schemata allows an automated extraction.

To provide an interactive user interface the BI tools

need the data in their own - proprietary format. This

addresses interoperability on business process, data

semantics, and syntax level.

EI2 tools supporting online analytics processes

make data sources more transparent and allow ad-

hoc queries. This is relevant for distributed systems,

where for example supplier speciﬁc data is stored

localy with the supplier, but access to that data is

granted for customers. Also low-code environments

for the ETL process support more advanced end-users

in getting the right data to the BI and BA tools. An

example of such an environment is apache NIFI

In Smart Grids a number of examples for analytics

support that require detailed know-how on the current

https://niﬁ.apache.org/

network state exists. Here EI2 can support BA by pro-

viding online access and uniﬁcation in a common data

model for analytics.

5.2 Predictive Analytics

Predictive analytics is one of three general analytics

methodological approaches. To cover all three, we

mention it here as well, but currently there are no

EI2 approaches that speciﬁcally support prediction,

beyond the basic need for data exchange. However,

access to large amount of clean and consistent data

is an important precondition for predictive analytics.

This includes predictive maintenance for energy as-

sets.

5.3 Prescriptive Analytics

The S

ˆ3

-Enterprise (Sensing, Smart and Sustainable)

provides multiple models and points of view to enable

an Enterprise Operating System (Weichhart et al.,

2016b; Weichhart et al., 2021c). The different model

types are used to make processes and data structures

in the enterprise transparent. Overall goal is to sup-

port smart decision making based on a solid and trans-

parent data basis (Weichhart et al., 2018).

Using process-models for prescription of busi-

ness behaviour supports interoperability on business

level. Modular process approaches like the Subject-

oriented Business Process Management (S-BPM) ap-

proach (Fleischmann et al., 2012) supports on the

one-hand interoperability in general (Weichhart and

Wachholder, 2014) and on the other hand can be used

to unify the behaviour of agents so that it becomes in-

teroperable. An example for the later is research in the

ROBxTASK project where processes for human and

robotic agents have to be aligned (Weichhart et al.,

2021d). Such process models can be the result of

prescriptive analytics. In particular the ROBxTASK

process models will be modular and as such provide

task descriptions on a higher level of abstraction that

allows decision makers process planning and optimi-

sation without the need for detailed robotic program-

ming know-how (Weichhart et al., 2021d).

To capture dynamic energy processes for manage-

ment EI2 approaches like the OoEI

CAS

Domain Spe-

ciﬁc Language (DSL) can be a basis (Weichhart et al.,

2016a). The agent based approach can (by its nature)

be easily extended to not only include the systemic as-

pects of enterprise systems but also of energy systems.

This allows to dynamically reason over the current

state and possible actions in the network. In addition

to this general possibility, does the agent-based ap-

proach of OoEI

CAS

allow to do the analytics tasks in

EI2N 2021 - IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

232

a distributed manner. That allows immediate reaction

to local events. The communication of the agents al-

lows a balancing of energy production and consump-

tion across the network, given that agents have inﬂu-

ence over the energy assets they represent.

6 CONCLUSIONS

In this work we have made an initial proposal to align

research in Enterprise Integration and Interoperability

(EI2) and Business Analytics (BA). The work is mo-

tivated by observations that a great share of the effort

in analytics projects is for data preparation. EI2 of-

fers existing tools and methods to support this. There

is great beneﬁt for BA research to better address the

initial steps of the process.

Further work is required to better align tools and

methods of both ﬁelds. The overall vision is that de-

cision makers, using descriptive, predictive and pre-

scriptive analytics, are enabled to immediately access

relevant data also in the context of changing ques-

tions. To enable this vision, a speciﬁc EI2 approach

for meeting business analytics speciﬁc needs is to be

researched.

Taking a look at all three methodological ap-

proaches, initial support for descriptive analytics is

already provided by existing approaches. However,

end-user tools for immediate change of data sources

(e.g. inclusion of CPS and IoT devices) is still a chal-

lenge, in particular in the context of distributed sys-

tems. We where not able to identify EI2 approaches

for prescriptive analytics; more work is needed in this

area. While some approaches can be used for pre-

dictive analytics, more needs to be done to support

end-users in modeling and analysing their current or

future system based on interoperable data.

From a more general view, BA currently relies on

centralized integrated data sets. This stands in con-

trast to interoperability and decentralized but interop-

erable systems. Here, we see an interesting ﬁeld for

further research. How to enable ad-hoc analytics (i.e.

reducing setup times and tasks for data preparation)

in decentralized systems. For example, smart grids

where multiple factories and many households are in-

volved in exchanging energy from sustainable energy

sources. Analytics tools are needed for balancing the

generation and consumption of power. But all par-

ticipants have their own point of view and should not

have full access to all details in terms of energy needs.

In this initial work we described existing and

missing support for business analytics with respect to

access to interoperable data sets. More work that sup-

ports decision makers in seamless access to data from

heterogeneous sources and data processing services is

needed.

ACKNOWLEDGEMENTS

The research has been funded by the European

Commission and the Government of Upper Austria

through the program EFRE / IWB 2020, the prior-

ity REACT-EU and the project RESINET (RESIlien-

zsteigerung In energieNETzen – RESilience increase

In energy NETworks; Nr.:WI-2020-701900/12).

REFERENCES

enaben, F., Boissel-Dallier, N., Pingaud, H., and Lorre,

J.-P. (2013). Semantic issues in model-driven man-

agement of information system interoperability. In-

ternational Journal of Computer Integrated Manufac-

turing, 26:1042–1053.

enaben, F., Mu, W., Boissel-Dallier, N., Barthe-Delanoe,

A.-M., Zribi, S., and Pingaud, H. (2015). Support-

ing interoperability of collaborative networks through

engineering of a service-based mediation information

system (mise 2.0). Enterprise Information Systems,

9:556–582.

Bhattarai, B. P., Paudyal, S., Luo, Y., Mohanpurkar, M.,

Cheung, K., Tonkoski, R., Hovsapian, R., Myers,

K. S., Zhang, R., Zhao, P., et al. (2019). Big data

analytics in smart grids: state-of-the-art, challenges,

opportunities, and future directions. IET Smart Grid,

2(2):141–154.

BKCASE Editorial Board (2019). Guide to the systems en-

gineering body of knowledge (SEBoK). techreport,

INCOSE - International Council on Systems Engi-

neering.

Chen, D., Doumeingts, G., and Vernadat, F. B. (2008).

Architectures for enterprise integration and interoper-

ability: Past, present and future. Computers in In-

dustry, 59(7):647 – 659. Enterprise Integration and

Interoperability in Manufacturing Systems.

Dubey, R., Gunasekaran, A., Childe, S. J., Wamba, S. F.,

Roubaud, D., and Foropon, C. (2021). Empirical

investigation of data analytics capability and organi-

zational ﬂexibility as complements to supply chain

resilience. International Journal of Production Re-

search, 59(1):110–128.

Ducq, Y., Chen, D., and Doumeingts, G. (2012). A con-

tribution of system theory to sustainable enterprise in-

teroperability science base. Computers in Industry,

63(8):844 – 857. Special Issue on Sustainable Inter-

operability: The Future of Internet Based Industrial

Enterprises.

Fielding, R. T. and Taylor, R. N. (2002). Principled design

of the modern web architecture. ACM Transactions on

Internet Technology (TOIT), 2(2):115–150.

Enterprise Integration and Interoperability Improving Business Analytics

233

FIPA - Foundation for Intelligent Physical

Agents (2005). Standard ﬁpa speciﬁcations.

http://ﬁpa.org/repository/standardspecs.html.

Fleischmann, A., Schmidt, W., Stary, C., Obermeier, S., and

orger, E. (2012). Subject-Oriented Business Process

Management. Springer, Berlin Heidelberg.

edria, W. and Naudet, Y. (2014). Extending the ontology

of enterprise interoperability (ooei) using enterprise-

as-system concepts. In Mertins, K., B

enaben, F.,

Poler, R., and Bourri

eres, J.-P., editors, Enterprise In-

teroperability VI, volume 7 of Proceedings of the I-

ESA Conferences, pages 393–403. Springer Interna-

tional Publishing Switzerland.

Haller, A. and Polleres, A. (2020). Are we better off

with just one ontology on the web? Semantic Web,

11(1):87–99.

Hindle, G., Kunc, M., Mortensen, M., Oztekin, A., and Vid-

gen, R. (2020). Business analytics: Deﬁning the ﬁeld

and identifying a research agenda. European Journal

of Operational Research, 281(3):483–490. Featured

Cluster: Business Analytics: Deﬁning the ﬁeld and

identifying a research agenda.

Holland, J. H. (1998). Emergence: From Chaos to Order.

Basic Books.

Holsapple, C., Lee-Post, A., and Pakath, R. (2014). A uni-

ﬁed foundation for business analytics. Decision Sup-

port Systems, 64:130–141.

ISO/IEC/JTC1/SC7 (2009). Iso/iec 10746-3:2009 informa-

tion technology – open distributed processing – ref-

erence model: Architecture. Technical report, Inter-

national Standards Organization / International Elec-

trotechnical Commission, Joint Technical Committee

1, Subcommittee 7: Software and systems engineer-

ing, Geneva, Switzerland.

Kosanke, K., Vernadat, F. B., and Zelm, M. (1999). Cimosa:

enterprise engineering and integration. Computers in

Industry, 40(2):83 – 97.

Lenat, D. (2004). cyc common sense ontology.

Lenat, D. et al. (2004). Cyc upper ontology open version.

Mayilvaganan, M. and Sabitha, M. (2013). A cloud-based

architecture for big-data analytics in smart grid: A

proposal. In 2013 IEEE International Conference on

Computational Intelligence and Computing Research,

pages 1–4.

McCrae, J. P., Abele, A., Buitelaar, P., Cyganiak, R.,

Jentzsch, A., Andryushechkin, V., and Debattista, J.

(2019). The linked open data cloud.

Morel, G., Panetto, H., Mayer, F., and Auzelle, J.-P.

(2007). System of enterprise-systems integration is-

sues: an engineering perspective. In IFAC, editor,

IFAC Conference on Cost Effective Automation in

Networked Product Development and Manufacturing,

IFAC-CEA’07, page CDROM, Monterrey, Mexique.

Elsevier. Invited plenary survey.

Naudet, Y., Gu

edria, W., and Chen, D. (2009). Systems sci-

ence for enterprise interoperability. In Interoperabil-

ity for Enterprise Software and Applications, China,

2009. IESA’09. International Conference on, pages

107–113.

Naudet, Y., Latour, T., Gu

edria, W., and Chen, D. (2010).

Towards a systemic formalisation of interoperability.

Computers in Industry, 61:176–185.

Omar, Y. M., Minoufekr, M., and Plapper, P. (2019). Busi-

ness analytics in manufacturing: Current trends, chal-

lenges and pathway to market leadership. Operations

Research Perspectives, 6:100127.

Panetto, H., Iung, B., Ivanov, D., Weichhart, G., and Wang,

X. (2019). Challenges for the cyber-physical manu-

facturing enterprises of the future. Annual Reviews in

Control, 47:200 – 213.

Panetto, H., Jardim-Goncalves, R., and Molina, A. (2012).

Enterprise integration and networking: Theory and

practice. Annual Reviews in Control, 36:284–290.

Polleres, A., Kamdar, M. R., Fern

andez, J. D., Tudorache,

T., and Musen, M. A. (2020). A more decentralized

vision for linked data. Semantic Web, 11(1):101–113.

Stamper, R. (1993). A semiotic theory of information

and information systems. In Invited papers for the

ICL/University of Newcastle Seminar on Information.

Stamper, R., Liu, K., Hafkamp, M., and Ades, Y. (2000).

Understanding the roles of signs and norms in orga-

nizations - a semiotic approach to information sys-

tems design. Behaviour & Information Technology,

19(1):15–27.

Uslar, M., Rohjans, S., Neureiter, C., Pr

ostl Andr

en, F.,

Velasquez, J., Steinbrink, C., Efthymiou, V., Migli-

avacca, G., Horsmanheimo, S., Brunner, H., and

Strasser, T. I. (2019). Applying the smart grid archi-

tecture model for designing and validating system-of-

systems in the power and energy domain: A european

perspective. Energies, 12(2).

Veerlapati, R. and Thota, R. (2021). Data analytics

functions and practices in electric distribution util-

ities management-a new perspective. In 2021 In-

ternational Conference on Intelligent Technologies

(CONIT), pages 1–6.

Vernadat, F. B. (2010). Technical, semantic and organi-

zational issues of enterprise interoperability and net-

working. Annual Reviews in Control, 34(1):139 – 144.

Vidgen, R., Shaw, S., and Grant, D. B. (2017). Manage-

ment challenges in creating value from business an-

alytics. European Journal of Operational Research,

261(2):626–639.

von Bertalanffy, L. (1950). An outline of general system

theory. The British Journal for the Philosophy of Sci-

ence, 1(2):134–165.

von Bertalanffy, L. (1969). General System Theory - Foun-

dations, Development, Applications. George Braziller,

New York, 17th edition edition.

Wampler, D. and Payne, A. (2014). Programming Scala:

Scalability = Functional Programming + Objects.

O’Reilly and Associates.

Weichhart, G., Ducq, Y., and Doumeingts, G. (2021a).

An IFIP WG5.8 State-of-the-Art View on Methods

and Approaches for Interoperable Enterprise Systems,

pages 222–244. (C) IFIP International Federation for

Information Processing, published by Springer Inter-

national Publishing, Cham.

EI2N 2021 - IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

234

Weichhart, G., Gu

edria, W., and Naudet, Y. (2016a). Sup-

porting interoperability in complex adaptive enter-

prise systems: A domain speciﬁc language approach.

Data and Knowledge Engineering, 105:90–106.

Weichhart, G., Mangler, J., Raschendorfer, A., Mayr-

Dorn, C., Huemer, C., H

ammerle, A., and Pichler, A.

(2021b). An adaptive system-of-systems approach for

resilient manufacturing. e & i Elektrotechnik und In-

formationstechnik.

Weichhart, G., Molina, A., Chen, D., Whitman, L., and

Vernadat, F. (2016b). Challenges and current devel-

opments for sensing, smart and sustainable enterprise

systems. Computers in Industry, 79:34–46.

Weichhart, G., Panetto, H., and Molina, A. (2021c). Inter-

operability in the cyber-physical manufacturing enter-

prise. Annual Reviews in Control, 51:346–356.

Weichhart, G., Pichler, A., Strohmeier, F., Schmoigl, M.,

and Z

orrer, H. (2021d). The ROBxTASK architecture

for interoperability of robotic systems. In 2021 IEEE

International Workshop on Metrology for Industry 4.0

and IoT, Proceedings of. IEEE.

Weichhart, G., Stary, C., and Vernadat, F. (2018). En-

terprise modelling for interoperable and knowledge-

based enterprises. International Journal of Production

Research, 56(8):2818–2840.

Weichhart, G. and Wachholder, D. (2014). On the interop-

erability contributions of s-bpm. In Nanopoulos, A.

and Schmidt, W., editors, S-BPM ONE - Scientiﬁc Re-

search, volume 170 of LNBIP - Lecture Notes in Busi-

ness Information Processing, pages 3–19. Springer In-

ternational Publishing.

Wooldridge, M. (2009). An Introduction to MultiAgent Sys-

tems. John Wiley & Sons, second edition edition.

Zhang, Y., Huang, T., and Bompard, E. F. (2018). Big data

analytics in smart grids: a review. Energy Informatics,

1(1).

Enterprise Integration and Interoperability Improving Business Analytics

235