Digitalized Cross-organizational Interoperability in Industrial Business

Ecosystems: Implications and Models for Process Industry

Petri Kannisto

and David H

astbacka

P.O. Box 553, 33014 Tampere University, Finland

Keywords:

Plant Lifecycle Information, Asset Information, Digital Business Ecosystem, Industry 4.0.

Abstract:

Interoperability between industrial organizations is a persistent challenge, particularly as the goal is to enable

digitalized communications between information and communications technology (ICT) systems operated by

different parties. This paper studies how the current interoperability tools and models support the foundations

of digitalized business-to-business communication. While the focus area is plant lifecycle information in pro-

cess industry, covering investment projects as well as operations and maintenance (O&M), the problems can

be generalized to all manufacturing. The results of this study suggest that the current interoperability mod-

els and standards offer little support for building cross-organizational interoperability with digitalized tools.

Thus, there should be consortia that span the enterprises in process industry, aiming to develop the architec-

ture of collaborative business networks that fulﬁl sustainable interoperability and the related governance. To

accomplish this goal, this paper shows what elements exist and what are lacking, mapping these to European

Interoperability Framework (EIF).

1 INTRODUCTION

The everyday operation of industrial enterprises

should improve in sustainability, but this progress is

hindered by interoperability issues between organi-

zations and their systems of information and com-

munications technology (ICT). In their operation,

the enterprises execute collaborative business pro-

cesses with multiple business partners involved. The

processes are often complex, but they also evolve

and change over time, which necessitates adaptation

(Agostinho et al., 2016). The processes should exploit

ICT in information exchange to improve efﬁciency,

reliability, and automation. From the functional view-

point, the information exchange could occur via direct

ICT system integrations in the point-to-point manner,

but this is economical only if the number of business

partners is low and the data volumes are high (Kan-

nisto et al., 2020). Such integrations are rigid and

expensive. Instead, the integration should be loosely

coupled, which is what interoperability refers to (Ver-

nadat, 2010). In general, interoperability issues and

other technical factors were identiﬁed as an obstacle

to data sharing by 73 % of respondents in a recent

study (Scaria et al., 2018, p. 75).

https://orcid.org/0000-0002-0613-8639

https://orcid.org/0000-0001-8442-1248

Interoperability has been a goal in both generic

ICT and industrial production for long, resulting in

interoperability models and technologies, but the goal

remains unreachable. Despite accomplishments to a

certain extent, such as communication protocols and

some information models, industrial enterprises still

communicate largely manually. It can be argued that

the low-hanging fruits have been collected, i.e., the

easiest problems have been solved especially if these

are generalizable enough to provide a solid return of

investment. For example, invoicing is a relevant activ-

ity for any enterprise, and as invoices share a general

format regardless of the industry, e-invoicing has ex-

panded rapidly (Koch, 2019). However, the narrower

the ﬁeld of application and the greater the heterogene-

ity, the less tempting it is to seek for a solution.

Regarding interoperability, process industry pro-

vides an example about a great potential in business-

to-business ICT interoperability but modest results

this far. Process industry is asset intensive, and the

business processes that require the exchange of plant

lifecycle information necessitate masses of data to

be communicated regarding engineering and equip-

ment. Concretely, such lifecycle processes are related

to investment projects or operations and maintenance

(O&M). Unfortunately, the practitioners still rely on

data exchange with manual tools, such as spread-

Kannisto, P. and Hästbacka, D.

Digitalized Cross-organizational Interoperability in Industrial Business Ecosystems: Implications and Models for Process Industry.

DOI: 10.5220/0011543900003329

In Proceedings of the 3rd International Conference on Innovative Intelligent Industrial Production and Logistics (IN4PL 2022), pages 233-241

ISBN: 978-989-758-612-5; ISSN: 2184-9285

233

sheets, in the absence of a digitalized medium. This

study considers that ”digitalized” refers to machine

readable rather than digitally accessible as deﬁned by

(Hendler and Pardo, 2012).

To contribute to the development of ICT interoper-

ability between organizations, this paper outlines the

current obstacles, tools, and models, and discusses

what is still lacking. The research question is:

How do interoperability models and existing

standards support digital interoperability in

the exchange of lifecycle information in the

business ecosystems of process industry, and

what are the future directions?

Although the viewpoint is lifecycle information in

process industry, the problems are analogous with

manufacturing and the related business.

The research method is constructive, starting with

a background survey in Section 2. For the basis of

the construct, Section 3 considers the practical busi-

ness needs of the exchange of lifecycle information.

Then, Section 4 outlines interoperability models and

standards that could and should contribute to digi-

tal ecosystems regarding lifecycle information in pro-

cess industry. These provide the components to show

which aspects of interoperability already exist and

which are still missing, indicating a research gap. Fi-

nally, Section 5 concludes the paper.

2 BACKGROUND

The ecosystem view of lifecycle information has

rarely been studied. The closest related research ﬁelds

are organizational interoperability in general and the

technical view of lifecycle information.

Enterprise modeling is a potential approach for

organizational interoperability. Model-driven meth-

ods would enable enterprises that adapt to changes in

a dynamic way, leading to sustainable interoperabil-

ity (Agostinho et al., 2016). Organizations should

constantly manage their enterprise models, making

these models a top priority and enabling the sens-

ing, smart, and sustainable (Sˆ3) enterprise (Weich-

hart et al., 2016). Furthermore, enterprise modeling

can, combined with knowledge representation, con-

tribute to organizational interoperability by enabling

reusable enterprise models (Weichhart et al., 2018).

Regarding the interaction of enterprises, the con-

cept ”digital business ecosystem” refers to a commu-

nity of business actors with digital information ex-

change. The concept is analogous to the biological

ecosystem, including the ”co-evolution” of the busi-

ness and the digital representations (Nachira et al.,

2007, p. 5). A business ecosystem involves a high

modularity and necessitates coordination, which is

more demanding than an open market or hierarchi-

cal supply chain (Pidun et al., 2019). The forma-

tion of the ecosystem can contain both bottom-up and

top-down elements, as the former contributes to co-

evolution and the latter to interoperability (Lenken-

hoff et al., 2018). To transform the production equip-

ment business from an open market to an ecosystem,

an ecosystem architecture has been proposed, but this

remains abstract and focuses on technology rather

than interoperability models (Kannisto et al., 2020).

Another study argues that systematically applied on-

tologies can lead to an ecosystem, but this requires

coordination (Ameri et al., 2022). The ecosystem can

even span multiple domains. These can, as assumed

in Industry Commons Ecosystem (ICE), co-operate in

a cross-domain manner by exploiting ”breakthrough

innovations”, which is enabled by seven layers that

include factors, such as societal values, ethics, envi-

ronment, contracts and legislation, intellectual prop-

erty, ﬁnance and payment systems, and data interop-

erability (Magas and Kiritsis, 2022).

Data-driven solutions have arisen in industrial sys-

tems recently, leading to goals for data sovereignty

and data autonomy. Data sovereignty refers to acting

according to the laws of the data origin, whereas data

autonomy means that the owner can determine who

can access its data and how to use it. Europe is lead-

ing this development, as European Commission has

announced ”a European strategy for data” that cov-

ers multiple data spaces, including industrial manu-

facturing, agriculture, green deal, and health, among

others (COM(2020) 66, 2020). The development has

resulted in a pursuit for platforms that aim to facili-

tate data sharing. In this effort, the initiative Gaia-X

aims to provide an appropriate infrastructure, whereas

International Data Spaces (IDS) target at controlling

data usage (Braud et al., 2021). Software is already

being developed for IDS, such as (Nast et al., 2020).

Multiple authors have studied semantic interoper-

ability. In data exchange, standardized properties for

the items can form the basis (Epple et al., 2017). Re-

garding the exchange of engineering data, there is a

need for consortia to reach and govern interoperabil-

ity (Fillinger et al., 2019). The application of the stan-

dard ISO 15926 has been studied for semantic inter-

operability (Kim et al., 2020). One study showed that

a fast data exchange solution may be tempting over a

slower standards-based one (Papakonstantinou et al.,

2019). The standards related to semantic interoper-

ability are referred to in Section 4.2.

To summarize, it appears that there is a research

gap between the domain-speciﬁc technology and gen-

EI2N 2022 - 16th IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

234

eral interoperability research. This leaves the organi-

zational aspects of process industry neglected.

3 BUSINESS NEEDS

3.1 Information Exchange

The lifecycle-related collaboration of process indus-

try has multiple challenges, resulting from the need

for interoperability. Earlier, these have been discussed

in (H

astbacka and M

asniemi, 2009) and (Kannisto

et al., 2020). Furthermore, this outline has been con-

tributed to by discussions with practitioners in Finnish

process industry although the issues are generally

similar all over the world.

The lifecycle processes in process industry, in-

cluding investment projects and O&M, are charac-

terized by a large number of heterogeneous business

actors. This complexity has led to a situation where

many aspects remain insufﬁciently covered by stan-

dardization despite a number of standards and the

large capital investment. Figure 1 illustrates the ac-

tors involved in the lifecycle of a plant, including

plant operators, equipment suppliers and manufactur-

ers, engineering agencies (investment projects only),

and maintance service providers (maintenance only).

For each actor but the operator, the ﬁgure presents the

most important types of lifecycle information deliv-

ered. The operator, on the other hand, needs data of

all of these types. Each actor and their tasks regarding

data delivery are explained in the following sections.

Plant operators

Maintenance service

providers

Engineering agencies

Equipment

manufacturers

Equipment suppliers

Investment

projects only

Maintenance

only

Product data and

documents

Product data and

documents,

catalogs

Product data and

documents,

catalogs

Process

requirements

Engineering

data

Figure 1: The stakeholders generally involved in investment

projects and maintenance service as well as the data and

documents exchanged with plant operators.

The information exchanged between the actors is

inherently complex. This results from the complexity

of the processes and equipment. There are standards

for the data, but these have a modest rate of adoption

and often require customization. The following sub-

sections explain the role of each actor, whereas Sec-

tion 3.2 explains the issues of data handover.

Plant Operators. Plant operators utilize the pro-

duction equipment to make the end product. Regard-

less of the end product (e.g., food, chemicals, or met-

als), the industries process materials and must there-

fore measures, such as temperature, chemical concen-

tration, pressure, and mass. The processing requires

equipment, such as storages, pipes, valves, pumps,

and measurement equipment. For operation, the oper-

ators need to manage various types of information that

should become available during the lifecycle of the

plant. The operators must manage the requirements

of their production processes so that appropriate sys-

tems can be designed and built. During operation, the

operator needs information about the products so that

appropriate spare parts can be found and any updates

in the engineering design can be performed.

Equipment Manufacturers and Suppliers.

Equipment manufacturers and suppliers provide

the production devices. The manufacturers build

the equipment, providing a selection with varying

characteristics and capabilities. Respectively, the

suppliers sell the equipment, but even the manufac-

turer can sell directly without a dedicated supplier.

The supplier provides a catalog to potential buyers

about the selection.

Upon selling equipment, the manufacturer or sup-

plier should provide the related data and documents.

The operator needs these for the subsequent main-

tenance and engineering activities. Unfortunately,

the structure of the equipment data is heterogeneous

and speciﬁc to each equipment type, such as pumps,

valves, and measurement devices. Furthermore, a set

of attachment documents must follow, including but

not limited to bills of materials and certiﬁcates.

Engineering Agencies. Engineering agencies sell

services to the operator as the operator has specialized

in process control rather engineering. The agency

must receive process requirements from the operator.

Based on these, the engineers design an implementa-

tion, choosing equipment from a supplier. Once the

design is complete, the agency should hand over its

design data to the operator. The complexity of this

data is similar to the equipment data as engineering

determines the equipment properties.

Maintenance Service Provider. Maintenance ser-

vice providers take care of equipment replacements

Digitalized Cross-organizational Interoperability in Industrial Business Ecosystems: Implications and Models for Process Industry

235

Media

Formats

Receiving organizationSending organization

Generation of

information

Storage into

backend

Export into format

agreed with recipient

Delivery

Inspection; storage

into backend

Exploitation

Spreadsheet, PDF

Email, physical media, non-

standard cloud storages

Figure 2: Within the scope of lifecycle information, the stages required to deliver data between organizations, along with the

current delivery with a low automation degree. The stages are loosely based on (CFIHOS, 2021).

and the installation of spare parts as appropriate. The

service provider can have a storage to provide the

most commonly used products. The data delivery dif-

fers from an investment project due to its low volume

as the maintenance usally applies only to one piece

of equipment. Thus, to reduce the burden of data ex-

change, the service provider typically collects a data

batch to be delivered only every few months.

3.2 Issues in Data Delivery

Figure 2 illustrates the various stages required to de-

liver information from one organization to another

along with the manual delivery media. These stages

involve the generation of information, its storage lo-

cally, its export to a format suitable for communi-

cation, the actual delivery, inspection and storage in

the receiving end, and exploitation. These stages are

loosely based on (CFIHOS, 2021).

The state-of-art delivery occurs with tools that are

digitally accessible (spreadsheets and Portable Docu-

ment Format PDF) and use a medium with a low au-

tomation degree, i.e., email, cloud storages with no

support for standard data structures, or even physical

media (Kannisto et al., 2020). This is because there

are no tools to deliver the data in a format agreed by

all parties in the ecosystem. There are standards, but

their adoption rate is low and local customizations are

often necessary. Usually, the plant operator requests

to receive the technical data, such as equipment prop-

erties and engineering design, in its preferred spread-

sheet format. PDFs are applied for human-readable

documents, such as certiﬁcates and bills of material.

The manual spreadsheet-based process causes is-

sues in data quality and availability. Although spread-

sheets enable some automation, they are error-prone

and inefﬁcient by leaving too much freedom to the

user. The actual content of the data ﬁelds often varies

after the preferences of the creator. Besides, the man-

ual process can cause a delay of months before the

equipment supplier or maintenance service provider

delivers the up-to-date equipment data after collecting

a batch after one-by-one replacements. Meanwhile,

the updates remain inaccessible to the operator.

The manual practices persist because there is no

ecosystem-wide governance. The plant operators,

as the ultimate customer, have the power regarding

which data formats are used, but these are operator-

speciﬁc. Even these processes could use a machine-

readable format, but the volume of data delivery has

been too low for a proper incentive. Besides, the op-

erators lack knowledge about ICT solutions and stan-

dards. A sufﬁcient governance could change this.

4 MODELS AND STANDARDS

4.1 Interoperability Models

Interoperability models enable practitioners to form

common concepts and structures to facilitate discus-

sion and the development of concrete solutions. This

section outlines models relevant in ICT and produc-

tion systems, the selection criteria being: either in-

volved in industrial research (1), designed for pro-

duction systems (2), or stems from a technological

background (3). Thus, the selected models are Euro-

pean Interoperability Framework (EIF), the data fed-

eration pyramid, and Reference Architectural Model

Industrie 4.0 (RAMI 4.0). EIF is domain agnostic but

included as there are studies for production systems

(criterion 1), e.g., (Panetto et al., 2019). The data fed-

eration pyramid is domain agnostic as well but stems

from a technical motivation similar to this study (cri-

terion 3). Finally, RAMI 4.0 was created explicitly

for industrial systems (criteria 1–3). It remains future

work to include more of interoperability models espe-

cially from other domains for more of material.

4.1.1 EIF

EIF is a conceptual model about the elements of inter-

operability, published by European Commission (EIF,

2017). Figure 3 illustrates the elements. In the core,

there are interoperability layers called technical, se-

mantic, organizational, and legal. The element inter-

operability governance spans over each layer, refer-

ring to the decisions, structures, and arrangements re-

EI2N 2022 - 16th IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

236

quired to reach and maintain interoperability. Another

governance element, integrated public service gover-

nance, refers to involving or introducing the public

services necessary for persistent interoperability. This

covers security and privacy, information services, and

catalogs. Although EIF has been meant for public ser-

vices, its abstract, generic nature provides an analogy

with the industrial production operated by enterprises.

Interoperability governance

Legal

Organizational

Semantic

Technical

Integrated public

service governance

Figure 3: The elements of EIF. Modiﬁed, re-drawn based

on (EIF, 2017).

EIF is domain agnostic, but there is an extension

modeled for cyber-physical manufacturing systems.

This extends the layers of EIF with related elements.

The two lowest levels, technical and semantic inter-

operability, form the cyber world, whereas organiza-

tional and legal interoperability are the physical world

including people. The cyber world must enable the in-

teroperability of models, whereas the physical world

must enable knowledge transfer, resilience, and sus-

tainability. (Panetto et al., 2019)

4.1.2 Data Federation Pyramid

The data federation pyramid provides a layered inter-

operability model that builds upon a technical view-

point but places business needs as well as trust on

top of technology (see Figure 4). It shows how the

ICT domain initially struggled with interoperability

problems even regarding hardware and operating sys-

tems. Later however, the interoperability of software,

networks, and data representation has been accom-

plished. In the present days, the community works

on semantics and pragmatics. Pragmatics refers to

understanding what kind of requirements arise from

the interoperability effort, whereas trust refers to the

reliability of the data. (Bergman, 2018, pp. 69-71)

Software and database;

operating system; hardware

Network

Semantics

Trust

Pragmatics

Data representation

Figure 4: The data federation pyramid. Modiﬁed, re-drawn

based on (Bergman, 2018, p. 71).

In the data federation pyramid, the two topmost

layers, pragmatics and trust, involve organizational

complexity whereas the others have a clear techni-

cal emphasis. Pragmatics is not to be confused with

pragmatic interoperability, which has no commonly

agreed deﬁnition (Asuncion and van Sinderen, 2010).

However, the concept resembles what is called or-

ganizational interoperability in EIF, although organi-

zational interoperability must be wider as it includes

more of aspects than only the pragmatic consequences

of aiming at interoperability. In the context of cross-

organizational interactions, pragmatics will result in

all of the collaborative effort required to reach inter-

operability. Bergman’s deﬁnition of trust is restricted

as it considers only reliability and thus lacks trust in

the data usage of collaborators in the spirit of data au-

tonomy.

4.1.3 RAMI 4.0

RAMI 4.0 provides a three-dimensional model for in-

teroperability as shown in Figure 5. The three di-

mensions are layers, hierarchy levels, and lifecycle

and value stream. From interoperability viewpoint,

the layers are most relevant, covering the physical as-

set, its connectivity with a communication protocol,

the information, functionality, and business. Com-

pared to EIF, this focuses more on devices and less

on business, containing no explicit element for gov-

ernance. The lifecycle dimension focuses on devices

and is therefore more restricted than the lifecycle of

entire plants. (Adolphs et al., 2015)

Hierarchy

levels

Life cycle and

value stream

Layers

Integration

Communication

Asset

Information

Functional

Business

Figure 5: RAMI 4.0 reference architecture. Modiﬁed, re-

drawn from (Kannisto et al., 2022), based on (Adolphs

et al., 2015).

RAMI 4.0 appears to be inspired by Smart Grid

Reference Architecture (SGAM). SGAM provides a

three-dimensional reference architecture for power

systems (SGAM, 2012). The most remarkable differ-

ence is that RAMI 4.0 replaced the power distribution

hierarchy with the lifecycle dimension.

Digitalized Cross-organizational Interoperability in Industrial Business Ecosystems: Implications and Models for Process Industry

237

4.1.4 Models Compared

Interoperability models are often layered and vary in

the level of detail depending on the domain. Com-

monly, the layers include technology or infrastructure

on the bottom, followed by semantics, business, and

legal, although not all models cover these aspects, as

indicated by a comparison of EIF, SGAM, and two

other interoperability models GridWise and eHealth

(Reif and Meeus, 2020). Still, some models intro-

duce more of aspects, as RAMI 4.0 has two additional

domain-speciﬁc dimensions and EIF provides an ab-

stract model for governance.

Although RAMI 4.0 stems from industrial do-

main, EIF appears best from the viewpoint of orga-

nizational coverage. EIF stresses the importance of

governance, whereas RAMI 4.0 focuses on what ap-

pears solely technical from EIF viewpoint. The data

federation pyramid is even more technical, providing

little elaboration about the two topmost layers that

cover organizational issues.

4.2 Standards

Several standards contribute to exchanging lifecycle

information in process industry. ISO 15926 con-

sists of multiple parts that aim for interoperability

for lifecycle information (ISO 15926, 2004). The

parts include, for instance, a data model and refer-

ence data library, providing generic concepts that en-

able application-speciﬁc extensions. From this foun-

dation, Data Exchange in Process Industry (DEXPI,

nd) has been created for the exchange of piping and

instrumentation diagrams. Capital Facilities Informa-

tion Handover Speciﬁcation (CFIHOS, 2021) aims to

provide a ”common language” for the delivery of life-

cycle information. CFIHOS speciﬁes properties and

equipment classes as the basis of common data in-

formation modeling. IEC 61987 series, such as (IEC

61987-10, 2009) for valve data, speciﬁes structures to

deliver both equipment and engineering information,

among others. This list of standards is not exhaustive

but provides an overview with examples.

Asset Administration Shell (AAS) is an initiative

and standard to provide interoperability within the

value chain. AAS provides abstract information mod-

els, associating these to data models and communi-

cation protocols. Introduced along with RAMI 4.0,

AAS enables a standardized interface for assets to

provide their data (Adolphs et al., 2015). AAS is be-

ing standardized (AAS Pt. 1, 2022). Regarding AAS,

Digital Twin is a related concept, referring to the digi-

tal representation of a concrete object or process. One

aim of the concept is to facilitate the exchange of en-

gineering information (Sierla et al., 2020). Still, this

communication requires a concrete medium, which is

lacking from process industry at least. This study re-

gards Digital Twin as a container of data and models

that still requires concrete standards for the interfaces

that enable interoperability, thus providing no actual

interoperability tools.

Additionally, there is a group of competing

domain-agnostic standards to deliver business docu-

ments or messages in a standardized format as re-

viewed by (Chituc, 2017). This ﬁeld is, therefore,

covered better than most of process-industry-speciﬁc

communication. Such business documents include,

for example, request for quotation, order, and in-

voice. The standards reviewed by Chituc include,

e.g., Universal Business Language (UBL), Open Ap-

plication Group Integration Speciﬁcation (OAGIS),

RosettaNet, and Electronic Business using Extensible

Markup Language (ebXML). However, the standards

provide no help in exchanging lifecycle information.

Open Platform Communications Uniﬁed Archi-

tecture (OPC UA, 2017) provides standards for com-

munication in industrial plants but lacks a focus

on both cross-organizational communication schemes

and lifecycle information. Although the newer OPC

UA ”PubSub” speciﬁcation (part 14) introduces data

models enabling cloud storage, there is no standard

medium for multi-actor schemes. Besides, OPC UA

lacks an information model for process requirements,

product and engineering data, and catalogs.

Despite a number of existing standards, their

adoption and inﬂuence on plant-lifecycle-related

business processes is limited. The reasons not to

adopt standards stem from the business environ-

ment, the organizations, and the standards themselves

(Braaksma et al., 2011). Clearly, the complexity

of the environment and information structures ham-

per digital interoperability efforts. Consequently, the

business practices remain heterogeneous, and there

are neither common information models nor appro-

priate data platforms.

4.3 Future Directions

Despite the challenges, there are initiatives to change

the situation. In Sweden and Finland, research in-

stitutes and companies organized an initiative called

Nordic Interoperability Cooperation (NIC) to recog-

nize new potential business models, Norway has been

involved in the discussions, and there are Europe-

wide efforts as well (SEIIA, 2022).

AAS aims to enable interoperability within the

value chain. It has gained the attention of scholars,

inspiring an entire workshop in 2022, e.g., (Jacoby

EI2N 2022 - 16th IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

238

et al., 2022). Currently, AAS has a scope differ-

ent from lifecycle information where information and

data models as well as semantics are essential. To be

applicable for lifecycle data, AAS still necessitates a

related information model as well as some medium

and organization to maintain interoperability. It re-

mains to be seen if AAS will either extend its appli-

cability or at least inspire the development of lifecycle

information processing.

The themes data sovereignty and data autonomy

have gained interest, which can contribute to lifecycle

information as well. As these are domain-agnostic

topics, they again lack a direct contribution to se-

mantic interoperability and the related governance for

lifecycle information. However, the dataspaces Gaia-

X and IDS are penetrating even to industrial setups,

contributing to communication and data autonomy

(Usl

ander et al., 2022).

Inspired from existing standards and frameworks,

Figure 6 illustrates how the current and future contri-

butions can enable interoperability for lifecycle infor-

mation. EIF is was chosen for the foundation due to

its emphasis on interoperability governance. The ﬁg-

ure illustrates the four layers as well as the pervasive

governance. Each layer is explained in the following

paragraphs.

Interop.

governance

Ecosystem-wide

organizations and

bodies

Commonly agreed

information models

Data mapping with

ontologies

Commonly agreed

business processes

Choreographies

International

collaboration

Clouds with Gaia-X

and IDS

Gaia-X

sovereignty

ISO 15926

IEC 61987

RAMI 4.0

AAS

DEXPI

CFIHOS

Organizational interoperability

Semantic interoperability

Technical interoperability

Legal interoperability

New contributions

from consortia

Figure 6: Standards and contributions regarding lifecycle

information in process industry; built upon (EIF, 2017).

The bottom layer covers technology, including

platforms, that enables data exchange in an au-

tonomous manner. Gaia-X and IDS provide the in-

frastructure and dataspaces. Furthermore, RAMI 4.0

and AAS provide communication interfaces.

The second layer is semantics, which is a per-

sistently challenging issue in lifecycle information.

Multiple standards contribute to this layer, at least

RAMI 4.0, AAS, ISO 15926, IEC 61987, CFIHOS,

and DEXPI. Of these, some provide static informa-

tion models, whereas others map these into ontolo-

gies. Ontologies help in various production-system-

related information management tasks (Batres, 2017).

The third layer refers to business processes, in-

cluding any common agreements for business-to-

business interaction. This is where new consortia are

necessary as there is currently no sufﬁcient coordina-

tion regarding the application of information models

and the required technical platforms. Although RAMI

4.0 covers this area, it provides no concrete tools to

manage organizational interoperability. The consor-

tia can arise from existing national organizations or

the ones already developing the standards. In forming

such consortia, a rulebook from Sitra helps to create

a fair ecosystem (Rulebook for a Fair Data Economy,

2021).

The topmost layer, legal interoperability, is im-

plemented with Gaia-X that has data sovereignty as

a core goal. The layer still requires more of consid-

eration in an open consortium to guarantee suitability

for the speciﬁcs of process industry. The earlier men-

tioned Sitra rulebook helps here as well (Rulebook for

a Fair Data Economy, 2021).

4.4 Discussion

The proposed model, mapping standards and activi-

ties to EIF, is a potential approach for interoperabil-

ity regarding lifecycle information exchange in pro-

cess industry. Because the business is currently an

open market, the way for an interoperable ecosystem

is to increase coordination (Pidun et al., 2019). For

the concrete data exchange platform, there is no guar-

antee if Gaia-X is the solution, but at least its ideas

should be followed to reach data autonomy, which is

desirable to industrial actors to protect their property.

The model could still be more concrete requiring

the organizations of process industry. It could map

the concrete consortia of process industry into the ﬁg-

ure and show their relationships with the non-domain-

speciﬁc items, such as Gaia-X and AAS. Clearly, suit-

ability to process industry can realize only if the do-

main demands a recognition for its requirements. Ad-

ditionally, the model could be seek for acceptance

within the business actors, such as operators and

equipment suppliers. This would increase credibil-

ity and potentially introduce new elements as well as

encourage discussion within the domain.

As a limitation, this study only considers lifecy-

cle information in process industry, excluding the ex-

plicit needs of other domains and activities. Still, en-

gineering, equipment, and maintenance are equally

important in manufacturing, and the respective in-

teroperability governance is necessary in any cross-

Digitalized Cross-organizational Interoperability in Industrial Business Ecosystems: Implications and Models for Process Industry

239

organizational communication that faces evolution re-

gardless of the domain.

5 CONCLUSIONS

This paper studied the support from models and stan-

dards for organizational interoperability in process in-

dustry, ﬁnally suggesting future directions. The au-

thors argue that the complex, heterogeneous nature of

enterprises and collaborative tasks effectively slows

down the progress towards common practices. Al-

though such practices would increase efﬁciency, it

will take a considerable effort from any enterprise.

Regarding interoperability models, it was discovered

that while some models mention organizational is-

sues, there is little concrete support, and others ex-

clude organizational factors altogether. Of the exam-

ined interoperability models, EIF provides the best

foundation by stressing interoperability governance,

a core element in sustainable interoperability. EIF

is, however, abstract and oriented to public services

rather than industrial production, lacking any con-

crete concepts for an ecosystem. Fortunately, ongoing

projects and proposal can relieve the interoperability

problems but only if the enterprises are ready to col-

laborate and participate in the bodies that govern stan-

dardization. Additionally, more work is necessary for

the data infrastructures, such as Gaia-X and IDS.

Considering the data federation pyramid sug-

gested by (Bergman, 2018, pp. 69-71), the challenges

of semantics and pragmatics still dominate the ﬁeld

of interoperability. Semantic interoperability is cur-

rently partially reached as some information mod-

els are available and others not. Pragmatics is what

causes the most tangible interoperability issues as this

covers the actual realization of interoperability in the

everyday tasks.

For future work, there are multiple topics. First,

there could be contributions towards a concrete

ecosystem for lifecycle information, along with the

required coordination and governance, with both

bottom-up and top-down elements for both co-

evolution and interoperability (Lenkenhoff et al.,

2018). Second, the scope of the study could be ex-

tended to cross-domain scenarios. This could in-

clude, e.g., energy management and building infor-

mation modeling (BIM) as these should be consid-

ered in daily industrial operation to reach the green

transition. Third, the application of platforms, such

as Gaia-X, could be studied for interoperability and

data autonomy.

ACKNOWLEDGEMENTS

This work was supported by Business Finland [deci-

sion ID 45392/31/2020] via the project Nordic Inter-

operability Cooperation Finland (NIC FI). The funder

had no role in the actual research. The authors want

to express their sincere gratitude.

REFERENCES

AAS Pt. 1 (2022). Details of the asset administration shell

part 1 – the exchange of information between partners

in the value chain of Industrie 4.0 (3.0RC2). BMWK.

Adolphs, P., Bedenbender, H., Dirzus, D., Ehlich, M., Ep-

ple, U., Hankel, M., Heidel, R., Hoffmeister, M.,

Huhle, H., K

archer, B., Koziolek, H., Pichler, R.,

Pollmeier, S., Schewe, F., Walter, A., Waser, B., and

Wollschlaeger, M. (2015). Reference architecture

model Industrie 4.0 (RAMI4.0). VDI/ZVEI.

Agostinho, C., Ducq, Y., Zacharewicz, G., Sarraipa, J.,

Lampathaki, F., Poler, R., and Jardim-Goncalves, R.

(2016). Towards a sustainable interoperability in net-

worked enterprise information systems: Trends of

knowledge and model-driven technology. Comput.

Ind., 79:64–76.

Ameri, F., Sormaz, D., Psarommatis, F., and Kiritsis, D.

(2022). Industrial ontologies for interoperability in

agile and resilient manufacturing. Int. J. Prod. Res.,

60(2):420–441.

Asuncion, C. H. and van Sinderen, M. J. (2010). Prag-

matic interoperability: A systematic review of pub-

lished deﬁnitions. In EAI2N 2010, IFIP Advances in

Information and Communication Technology, pages

164–175, Berlin, Heidelberg. Springer.

Batres, R. (2017). Ontologies in process systems engineer-

ing. Chem. Ing. Tech., 89(11):1421–1431.

Bergman, M. K. (2018). A Knowledge Representation Prac-

tionary: Guidelines Based on Charles Sanders Peirce.

Springer International Publishing, Cham.

Braaksma, A. J., Klingenberg, W. W., and van Exel, P. P.

(2011). A review of the use of asset information stan-

dards for collaboration in the process industry. Com-

put. Ind., 62(3):337–350.

Braud, A., Fromentoux, G., Radier, B., and Le Grand, O.

(2021). The road to European digital sovereignty with

Gaia-X and IDSA. IEEE Netw., 35(2):4–5.

CFIHOS (2021). Scope and procedures V.1.5. Interna-

tional Association of Oil & Gas Producers. URL https:

//www.jip36-cﬁhos.org/cﬁhos-standards/ [Visited 28

Jun 2022].

Chituc, C.-M. (2017). XML interoperability standards for

seamless communication: An analysis of industry-

neutral and domain-speciﬁc initiatives. Comput. Ind.,

92-93:118–136.

COM(2020) 66 (2020). Communication from the commis-

sion to the European parliament, the council, the Eu-

ropean economic and social committee and the com-

mittee of the regions: A European strategy for data.

EI2N 2022 - 16th IFAC/IFIP International Workshop on Enterprise Integration, Interoperability and Networking

240

DEXPI (n.d.). Data exchange in the process industry. URL

https://dexpi.org [Visited 28 Jun 2022].

EIF (2017). New European interoperability framework.

European Commission. URL http://doi.org/10.2799/

78681.

Epple, U., Mertens, M., Palm, F., and Azarmipour, M.

(2017). Using properties as a semantic base for inter-

operability. IEEE Trans. Ind. Informat., 13(6):3411–

3419.

Fillinger, S., Esche, E., Tolksdorf, G., Welscher, W., Wozny,

G., and Repke, J.-U. (2019). Data exchange for

process engineering – challenges and opportunities.

Chem. Ing. Tech., 91(3):256–267.

Hendler, J. and Pardo, T. A. (2012). A primer on machine

readability for online documents and data. Data.gov.

URL https://www.data.gov/developers/blog/primer-

machine-readability-online-documents-and-data

[Visited 21 Jun 2022].

astbacka, D. and M

asniemi, T. (2009). Unifying pro-

cess design with automation and control application

development - an approach based on information inte-

gration and model-driven methods. IFAC Proceedings

Volumes, 42(4):1227–1232. 13th IFAC Symposium

on Information Control Problems in Manufacturing.

IEC 61987-10 (2009). Industrial-process measurement and

control — data structures and elements in process

equipment catalogues — part 10: Lists of properties

(LOPs) for industrial-process measurement and con-

trol for electronic data exchange. IEC.

ISO 15926 (2004). Industrial automation systems and in-

tegration – integration of life-cycle data for process

plants including oil and gas production facilities – part

1: Overview and fundamental principles. ISO.

Jacoby, M., Volz, F., Weißenbacher, C., and M

uller, J.

(2022). FA

ST service – an open source implemen-

tation of the reactive Asset Administration Shell. In

First Workshop on Implementing Asset Administration

Shells (ImplAAS). In press.

Kannisto, P., H

astbacka, D., Guti

errez, T., Suominen, O.,

Vilkko, M., and Craamer, P. (2022). Plant-wide inter-

operability and decoupled, data-driven process control

with message bus communication. J. Ind. Inf. Integr.,

26:100253.

Kannisto, P., H

astbacka, D., and Marttinen, A. (2020). In-

formation exchange architecture for collaborative in-

dustrial ecosystem. Inf. Syst. Front., 22(3):655–670.

Kim, B. C., Kim, B., Park, S., Teijgeler, H., and Mun, D.

(2020). ISO 15926–based integration of process plant

life-cycle information including maintenance activity.

Concurr. Eng., 28(1):58–71.

Koch, B. (2019). The e-invoicing journey 2019-2025.

Billentis. URL https://www.comarch.com/ﬁles-

com/ﬁle

441/report the e-invoicing journey 2019-

2025.pdf [Visited 27 Jun 2022].

Lenkenhoff, K., Wilkens, U., Zheng, M., S

uße, T., Kuh-

lenk

otter, B., and Ming, X. (2018). Key challenges

of digital business ecosystem development and how

to cope with them. Procedia CIRP, 73:167–172.

Magas, M. and Kiritsis, D. (2022). Industry commons: an

ecosystem approach to horizontal enablers for sustain-

able cross-domain industrial innovation (a positioning

paper). Int. J. Prod. Res., 60(2):479–492.

Nachira, F., Nicolai, A., and Dini, P. (2007). The digital

business ecosystems: Roots, processes and perspec-

tives. In Nachira, F., Nicolai, A., Dini, P., Le Louarn,

M., and Rivera Leon, L., editors, Digital Business

Ecosystems, pages 1–20. European Commission.

Nast, M., Rother, B., Golatowski, F., Timmermann, D.,

Leveling, J., Olms, C., and Nissen, C. (2020). Work-

in-progress: Towards an International Data Spaces

connector for the Internet of Things. In 2020 16th

IEEE International Conference on Factory Communi-

cation Systems (WFCS), pages 1–4.

OPC UA (2017). OPC uniﬁed architecture part 1, overview

and concepts, release 1.04. OPC Foundation.

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

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

facturing enterprises of the future. Annu. Rev. Control,

47:200–213.

Papakonstantinou, N., Karttunen, J., Sierla, S., and Vyatkin,

V. (2019). Design to automation continuum for indus-

trial processes: ISO 15926 – IEC 61131 versus an in-

dustrial case. In 2019 24th IEEE International Confer-

ence on Emerging Technologies and Factory Automa-

tion (ETFA), pages 1207–1212.

Pidun, U., Reeves, M., and Sch

ussler, M. (2019). Do you

need a business ecosystem. Boston Consulting Group.

Reif, V. and Meeus, L. (2020). Getting our act together on

the EU interoperability acts. Policy Briefs, Florence

School of Regulation, Energy, 2020/30.

Rulebook for a Fair Data Economy (2021). Rulebook tem-

plate for data networks, version 1.3 en. Sitra.

Scaria, E., Berghmans, A., Pont, M., Arnaut, C., and

Leconte, S. (2018). Study on data sharing between

companies in Europe. European Commission.

SEIIA (2022). P

aende och framtida projekt. URL https:

//seiia.se/projekt/ [Visited 14 Sep 2022].

SGAM (2012). Smart grid reference architecture 3.0. CEN-

CENELEC-ETSI Smart Grid Coordination Group.

Sierla, S., Azangoo, M., Fay, A., Vyatkin, V., and Papakon-

stantinou, N. (2020). Integrating 2D and 3D digi-

tal plant information towards automatic generation of

digital twins. In 2020 IEEE 29th International Sympo-

sium on Industrial Electronics (ISIE), pages 460–467.

Usl

ander, T., Baumann, M., Boschert, S., Rosen, R., Sauer,

O., Stojanovic, L., and Wehrstedt, J. C. (2022). Sym-

biotic evolution of digital twin systems and datas-

paces. Automation, 3(3):378–399.

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

zational issues of enterprise interoperability and net-

working. Annu. Rev. Control, 34(1):139–144.

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

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

opments for sensing, smart and sustainable enterprise

systems. Comput. Ind., 79:34–46.

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

modelling for interoperable and knowledge-based en-

terprises. Int. J. Prod. Res., 56(8):2818–2840.

Digitalized Cross-organizational Interoperability in Industrial Business Ecosystems: Implications and Models for Process Industry

241