Towards Data Awareness by Socio-technological Knowledge

Management

Alexander Heußner, Moritz H

oser and Sven Ziemer

Knowledge Management, Bauhaus Luftfahrt e.V., Willy-Messerschmitt-Str. 1, Taufkirchen, Germany

Keywords:

Data Awareness Challenge, Diagrammatic, Multi-perspective Modelling, Multi-modal Knowledge.

Abstract:

In today’s data-centric world, the data-awareness challenge is a crucial touchstone to existing knowledge man-

agement technologies. Adaptive, stakeholder-centric knowledge modelling approaches provide a solid ground

to tackle this challenge and open the door to enrich knowledge management by a socio-technological perspec-

tive. This paper proposes the use of a socio-technological approach to overcome the data-awereness challenge

by treating knowledge on data as a crucial business asset. Here, a data awareness generating, iterative, incre-

mental knowledge elicitation technique based on a multi-perspective, multi-modal diagrammatic knowledge

representation language serves as proof of concept.

1 INTRODUCTION

Data-driven software systems and data-based analy-

ses are the engines of today’s innovative businesses.

The key to start and operate such an engine is the

awareness an organization and its individual stake-

holders have of their data, its context and meta-data as

well as the corresponding implicit background know-

ledge. Establishing this data-awareness and address-

ing the awarenss challenge requires knowledge man-

agement (KM) to treat knowledge on data – and not

only data itself – as crucial business asset.

Thesis. Data awareness within an organization –

in the sense described above – can only be estab-

lished when a socio-technological perspective on the

involved stakeholders is included. Thus, we need to

rethink knowledge technologies and methods to (self-

) adapt to the needs of the individual stakeholders and

their multiple views, and not the other way round.

Setting. Knowledge assessment in an enterprise set-

ting requires to make knowledge on existing data, data

sources, corresponding meta-data on data quality, and

the related context knowledge. The results are tangi-

ble knowledge shared by the whole organization for

all participating stakeholders, e.g. from management

over IT to physical, industry workers – thus, deriving

an overview over the “data landscape”. Each stake-

holder represents different domains and functions and

has an individual perspective on the data, including

its context, that is processed by a system. The ac-

ceptance of a system depends on how well the sys-

tem is capable of reﬂecting the different user prior-

ities and perspectives and of facilitating knowledge

sharing among the stakeholders.

Knowledge elicitation as part of the requirements

analysis phase when developing next-generation data-

(& knowledge-)based enterprise information systems,

e.g. in the context of Industry 4.0, should there-

fore extend its mindset as well as its applied meth-

ods and tools, to include a socio-technical perspec-

tive. Existing methods and tools – such as UML or

BPMN – cover mostly the speciﬁcation and documen-

tation of software systems and processes. What is

needed, is thus the conception and speciﬁcation of a

method that ﬁt precisly into the required spot: a socio-

technical enbabled method to establish data aware-

ness as part of the requirements analysis phase. Es-

tablishing data awareness is a multi-disciplined task,

where exchanged knowledge must be accessible to all

stakeholders. As data-(& knolwedge-)based informa-

tion systems are rarely developed on the green ﬁeld,

we need to underpin their architecture and design with

sufﬁcient knowledge on the structure of existing data,

thus making the implicit knowledge (also hidden in

an organization’s hierarchy) and legacy knowledge

accessible for system engineers and data analysts as

well as software developers.

Crucial for the success of knowledge management

techniques is the provision of access to implicit back-

ground knowledge on various hierarchical levels in

the business organization, and the possibility to easily

adapt to each stakeholders’ context and the peculiari-

Heußner, A., Höser, M. and Ziemer, S.

Towards Data Awareness by Socio-technological Knowledge Management.

DOI: 10.5220/0008349602910298

In Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2019), pages 291-298

ISBN: 978-989-758-382-7

291

ties of the domain. In our view, this holistic view on

data awareness is a critical contribution to ensure the

acceptance of data-driven systems to be developed.

Note on Awareness. We utilize the term “awareness”

close to its psychological understanding as “percep-

tion or knowledge of something” and achieving ac-

curate reportability of the perceived process (APA,

2019). We apply awareness both regarding the stake-

holders’ understanding and their self-perception in

the model itself and the underlying context.

This paper presents an approach using a socia-

technological perspective to establish data awareness.

This buttom up and self-adapting approach consists of

a three-step description for involving all participating

stakeholders and a notation for data modelling. The

reminder of this paper is organized as follows: The

awareness challenge is further outlined in Section 2.

In addition, a review of existing approaches to multi-

perspective modelling is given. The proposed method

is presented in Section 3 and conclusions are given in

Section 4.

2 AWARENESS CHALLENGE

We argue that the KM community must devise their

next generation of knowledge management tech-

niques to inherently by design tackle the awareness

challenge. A technique to be proposed consists at

least of a method to elicitate and maintain knowledge

based on a corresponding knowledge representation

language. The challenge can be narrowed to the fol-

lowing set of requirements to such a technique:

• It shall address and adapt to multiple stakeholders

with different perspectives on different organiza-

tional layers (viz. Fig. 1);

• the underlying representation language shall be

both formal & sound, and thus open to algorith-

mic treatment, but also

• shall be easily accessible to the stakeholders, as

accessibility is key for generating awareness;

• the formalism shall interweave data, meta-data,

infrastructure and process models;

• the elicitation method shall be “minimal invasive”

and not disturb the business and disquiet the stake-

holders in their natural habitat;

• hence, a technique shall embed itself into organi-

zation and shall be suitable for everyday use;

• but shall also focus awareness from the point of

view of each of the individual stakeholders, i.e.

adapt to each stakeholders individual desidera-

tum;

• existing knowledge frameworks (e.g. process

maps, descriptions of IT infrastructure, etc.) shall

be sufﬁciently included or interfaced as far as ap-

propriate;

• the formalism shall bridge the gap between (pos-

sibly existing) formally described processes and

the way it is instantiated, i.e. lived, adapted and

adopted, always anew in everyday business;

Measuring awareness itself and thus evaluating a

solution to the awareness challenge is a scientiﬁc bar-

rier. We assume in the following, that a higher aware-

ness leads to a better quality of the underlying busi-

ness process, which is slightly easier to measure.

Regarding Existing Solutions. Existing approaches

on multi-perspective modelling, e.g. (Wood-Harper

et al., 1985) (Curtis et al., 1992) (Stohr and Zhao,

2001), ﬁx to certain set of perspectives regarding a

special application domain and often neglect acces-

sibility by all participating stakeholders. Putting the

stakeholder in the center of modelling, e.g. as in

(Jablonski, 2009), often leads to restrictions in expres-

sivity and readability.

The UML standard offers a range of diagrams as

notation for multi-view artefacts, which can be further

formalized with proﬁles and viewpoints (Object Man-

agement Group, 2017). However, these extensions to

UML are rather used to formally capture information

from spezialized domains, with current research on

multi-view notation and their consistency (Cicchetti

et al., 2019). Instead, a new formalism is needed that

focuses on accessible views for stakeholders with dif-

ferent background and level of operation.

More business artefact targeted approaches

like artifact-centric process modelling (Nigam and

Caswell, 2003) (Bhattacharya et al., 2007) (Kumaran

et al., 2008) (Cohn and Hull, 2009) miss multi-

ple perspectives and often focus solely high-level

management stakeholders.

Merging process and data modelling pari passu

is still extremely rare, e.g. (K

unzle and Reichert,

2011). Several approaches focus on a data-centric

view on business processes (Sun et al., 2006) (Aalst

et al., 2015), (Deutsch et al., 2009) or anticipate

an object-aware process support (Reichert and We-

ber, 2012), (K

unzle and Reichert, 2011). Other

data-centric modelling approaches research knowl-

edge awareness in process-aware information sys-

tems (Bhattacharya et al., 2009) (Reichert and We-

ber, 2012). Although both UML and BPMN (Object

Management Group, 2013) offer the coexistence of

process and data-related elements, the two concep-

tual models are rarely formally related and used sep-

arately by different teams (De Giacomo et al., 2017).

More recent approaches formalize the connection be-

KMIS 2019 - 11th International Conference on Knowledge Management and Information Systems

292

tween process and data, especially between BPMN

and UML activity diagrams with a “conceptual view”

(Combi et al., 2018a) (Combi et al., 2018b). Never-

theless, most approaches only seem to provide a com-

plicated ﬁx to existing modelling languages by reduc-

ing their readability, and are, thus, rather far from be-

ing applicable in an everyday setting.

Modelling frameworks often provide a toolbox of

different approaches and strategies to adapt to rather

different application domains and groups of stake-

holders. However, they require to be introduced top-

down with support from an organizations higher level

hierarchy, substantially inﬂuence the daily business

(thus are rarely minimal invasive), and do not focus

individual stakeholders and their demands. Though,

these frameworks often address a lot of the other

requirements of the awareness modelling challenge.

For example, ArchiMate (Lankhorst et al., 2009) em-

phasizes communication with the stakeholders and a

multi-viewpoint modelling approach; however, view-

points are restricted to a stakeholder group and the

target is not a holistic model. Also ArchiMate trades

ﬂexibility for formalization and ease of use. Further,

these frameworks need to be established top-down in

an organization – often not in a minimal invasive man-

ner. This is another hurdle for assuring the participa-

tion of all stakeholders and their awareness. Similar

objections would hold for other well-established ma-

jor frameworks like IDEF (Bruce, 1992), TOGAF or

IAF (Wang et al., 2016), or ZF (Zachman, 1987).

Supposing accessibility as key to awareness, most

notions of accessibility in practice are boiled down

to cognitive effectiveness or even only readability.

However, accessibility rarely played an initial prin-

cipal role in the design of state-of-the-art established

modelling languages like UML or BPMN (Reijers

and Mendling, 2011) (Moody and van Hillegers-

berg, 2008) (St

orrle and Fish, 2013). Moody’s “The

Physics of Notations” (Moody, 2009) has proposed a

catalogue principles that target this shortcoming and

includes several ideas how to devise a syntactically

based and measurable notion of accessibility. This

catalogue should be taken seriously from the start

when devising a knowledge representation language.

Thus, resolving the complete data awareness chal-

lenge with existing approaches is not straightfor-

wardly possible; especially, as existing approaches

in their practical implementation often mainly focus

the business-technological side that is established in a

top-down manner, and not the social embedding of the

stakeholders, i.e. their self-awareness and long-term

established accountability and responsibility regard-

ing a concrete business process and the correspond-

ing data. Existing frameworks often provide a bou-

Landscape

Process

Organization

Application

Utilisation

Applied

Process

Organizational

Level

Implementation

Level

instance of

Data View Process View

Figure 1: Combining a data and process view in one dimen-

sion with hierarchically different granularity due to differ-

ent systemic levels, we target a sweet spot (“ ”) for multi-

perspective modelling.

quet of different techniques, diagrams, methods, etc.,

to be combined into a set of knowledge tools that are

tailored to a concrete domain. However, these frame-

works and their a priori chosen set of tools not easily

adapt on-the-ﬂy to the stakeholders’ needs and in gen-

eral do not focus data awareness as central concern.

Knowledge Management 4.0. Recently, there are

several tendencies to re-invent knowledge manage-

ment in the context of current technological trends

and in the setting of Industry 4.0, e.g. (North and

Maier, 2019), (North et al., 2018), or (Peinl, 2017).

These are similar in spirit to our approach but focus

on generic, top-down business perspective, whereas

we primarily take a socio-technological, awareness-

focussed angle.

Proposed Path to Tackle the Challenge. Inspired

by the well-established design principles of “keeping

it simple” and leaving out parts that “you aren’t gonna

need” (KISS & YAGNI), we tackle the data aware-

ness challenge by a “bottom-up” approach, starting

from the concrete participating stakeholders and their

embedding in a business process, as follows:

• A diagrammatic, multi-perspective modelling lan-

guage that is implicitly accessible to all stakehold-

ers, can be represented sufﬁciently formal, and is

embedded in

• an iterative, incremental elicitation method, sup-

porting the generation of (data) awareness, com-

bined with

• a self-adaption along these iterations of the both

language and method itself in a concrete knowl-

edge management situation.

This three step approach helped us to bottom-up

derive a diagrammatic modelling language that, em-

bedded in a suitable knowledge elicitation method,

helped us to establish a data awareness in a practi-

cal setting – thus, providing a proof of concept that

the awareness challenge can be tackled by taking a

socio-technological perspective on knowledge mod-

elling, elicitating and maintaining.

Towards Data Awareness by Socio-technological Knowledge Management

293

BusinessProductionInformation

Flow

Machine 1 Machine 2

Process Step 1 Process Step 2

Data Storage Information System

x xx

Workers

(intermediary)

Product

Figure 2: Running Example’s simpliﬁed Industry 4.0 pro-

duction context: different layers & perspectives.

3 PROOF OF CONCEPT:

DATA-AWARE MODELLING

We brieﬂy present the diagrammatic knowledge rep-

resentation language devised by following our path

to tackle a concrete practical instance of the aware-

ness challenge. We applied this multi-perspective lan-

guage in a series of elicitation interviews to success-

fully establish a notion of data awareness among all

participating stakeholders.

An Exemplary KM Application. In a recent infor-

mation systems engineering project, two of the au-

thors faced the task of creating a process-spanning

data-inventory in a manufacturing facility in order to

be able to design data analysis applications in the con-

text of enterprise systems. We experienced, however,

the difﬁculty to harmonize the speciﬁc information

from the perspectives of different stakeholder groups,

e.g. workers, process managers and data analysts.

This problem matches the information gap in the ex-

isting perspectives, as we could analyze the informa-

tion provided by the different stakeholder groups sep-

arately but we could not make them accessible for a

joint understanding and could not integrate them into

a holistic model. At its core, this situation is a pro-

totypical instance of the awareness challenge, with

additional intricacies belonging to the manufacturing

domain (e.g. only semi-digitalized data, strong hier-

archical dependencies between the stakeholders,. . . )

– these will be left out in the following discussion.

Exemplarily, a generic Industry 4.0 production

context is represented by Figure 2: A two-step pro-

duction in the physical world is based on preparing a

physical product on Machine 1, which is then moved

to and further processes on Machine 2. Both ma-

chines are operated by Workers and are embedded

into an IT landscape, e.g. reading machine parame-

ters from a database and exchanging information via

the ERP system. Without loss of generality, we as-

sume in the following that each process step runs on

one corresponding machine and that each machine is

operated by one worker. Further, we restrict ourselves

to processes that are chain-like and step-wise (e.g.

no forks, alternatives and joins). However, the pre-

sented approach can be straightforwardly generalized

to cover these more complex process models.

We refer to concrete data objects that are accessed

in a single instantiation of the process, i.e. producing

one concrete product, as data instances. Examples

here are a paper job card that is tracking what takes

place in the production process or machine parame-

ters that are taken from a data base.

Data Awareness Analysis Diagrams. The core of

our approach is a diagrammatic modelling language,

called Data Awareness Analysis Diagrams (DAAD),

that served as a medium for discussions with the dif-

ferent stakeholders. DAAD was derived pragmati-

cally in practice and it is based on a combination of

distinct (formal) diagrams. We provide sufﬁcient for-

mality for this notation derived from practical needs

to see its connection with existing, competing for-

malisms. A rigid formalization of the diagrammatic

formal language is currently in preparation.

Inspiration for DAAD was certainly drawn from

the broad background of diagrams applied in soft-

ware engineering and in the stakeholders’s daily work

practice. Hence, reminiscences to UML class, object,

package and use case diagrams as well as business

process modelling languages are implicit but were

initially not explicitly intended, as the notation grew

from a pragmatic need and was not a priori designed

in a purely theoretic setting. Further, to cover the

needs of a multi-perspective, i.e. holistic, view, we

adapted the 4 + 1, or better n + 1 architectural view

approach of (Kruchten, 1995) to our needs, by orbit-

ing a central diagram with a set of additional “per-

spective” diagrams related to the different views of

distinct stakeholder groups.

To simplify the presentation, we will not draw a

distinction between the view or perspective itself and

its representation in the corresponding diagram in the

following and will not distinguish between the ab-

stract and concrete syntax of the diagrams.

Principal View. In the following, we present these

views inside-out by starting with the central Principal

View. The Principal View combines a data-oriented

with a process-oriented view, and “ﬂoats” between the

organisational and the implementational level, as in-

troduced in Figure 1. Thus, this view can be seen as

the nexus of the knowledge elicitation interview tech-

nique. While the Principal View serves as the main

communication anchor for operational users, the sur-

KMIS 2019 - 11th International Conference on Knowledge Management and Information Systems

294

Material

Cutting Drilling Inspection

:Instruction

:Parts

:Instruction

:Design

:Standard

doc

Input

Output

design

:WorkRecord

:MachineData

table

:Instruction

:Standard

:WorkRecord

:MachineData

:QualityControl

:QualityFeature

:Equipment

:WorkRecord

:Inspection

:Measurement

(a) Principal View: Tracing data instances along the process steps.

OrderStore

AssemblyDB

QualityCtrlSystem

(b) Data Architecture View: Relating data instances to the

storage architecture.

QualityControl

QualityFeature

-setPoint

-location

TestEquipment

-instrument

Measurement

-value

-unit

their relations.

abstract

& get context

locate

Figure 3: Example DAAD. We only present an excerpt containing three prominent perspectives. Color is applied to connect

entities between the different views, e.g. the QualityFeature data instance is subsumed by the QualityFeature class that can

be found in the Quality Control System in the IT landscape. Different stakeholders can insert their background knowledge in

their “closest” perspective. Then, the other perspectives are updated accordingly. Taking the view from a different perspective

supports any stakeholder to reconsider their position in the underlying business operation and, thus, to gain additional self-

awareness besides the data awareness.

rounding views are pivotal for bridging adjacent do-

main areas.

Supposing a step-wise representation of the under-

lying production process, the basic unit of the princi-

pal view represents one of these steps by the symbol

and relates the data instances accessed in this step.

Each can include additional information, e.g. an

identiﬁer, or description of the machinery used. We

differentiate data instances that are read by actions

in this step, i.e. its input, and those that are writ-

ten, i.e. its output. We depict the input instances

above and the output instances below. See for

example the “Material” step in Figure 3a: Instruc-

tions feed into the Material- and we output (a col-

lection of) parts. Data instance that are updated, i.e.

both read and written, appear both above and below of

, whereas the identity is assured by assigning the

same instance identiﬁer. We resort to depict these data

Towards Data Awareness by Socio-technological Knowledge Management

295

instances similar to object in UML object diagrams.

Thus, we can also add additional information, e.g. ex-

ample documents or screenshots of the user interface

to the corresponding data instances. These data an-

notations build a visual bridge between process and

entity elements for stakeholders from the process do-

main.

In the Principal View, we now combine differ-

ent steps along a temporal axis to sequentially repre-

sent the underlying production process (see Figure 3a:

step sequence from “Material” over “Cutting” and

“Drilling” to “Inspection”). As before, anonymous

objects depict different data instances of the same

type along the process chain, while data instances us-

ing the same identiﬁer (not shown here) represent the

same data object. For example in Figure 3a: The “In-

struction” and “WorkRecord” data from the ﬁrst steps

are from the same IT system. Recall that without loss

of generality we assumed these processes to be linear,

i.e. chains, only. Hence, we put the linear sequence

of in the center of the diagram from left-to-right.

Adding Additional Perspectives. Covering the prin-

cipal information regarding the interaction between

data and the (business) process and, thus, aiming at

the heart of the information gap, we enrich the Princi-

pal View by different additional, stakeholder-driven

perspectives. Thus, we provide access to the sys-

tem parts mentioned in Figure 1 through intermediary

views that refer to elements of both the Principal View

and the existing enterprise information system. For

simplicity, we rely on the (UML-inspired) diagrams

already established and applied in each stakeholder’s

domain as far as possible to assure readability and ac-

cessibility:

Data Architecture View. We provide an UML pack-

age diagram with additional meta-information to rep-

resent data context groups on an abstract level. The

groups can be UML-ish packages or class stereotypes

and are associated with a simple coloring scheme.

This view is presented to IT and relates the con-

text groups to their concrete location in the underly-

ing IT landscape, e.g. servers, information systems,

databases. The coloring scheme is used throughout

all views to re-identify the context groups.

Data Entity View. An UML class diagram is used

to detail the data context groups. The Data Entity

View provides the data analyst with a detailed view

on the processed data from a abstract conceptual per-

spective. See Figure 3 for an example.

Data Characteristics View. Additionally we pro-

vide the data analyst with data characteristics, de-

rived from the V’s” of Big Data (Gandomi and Haider,

2015; Sagiroglu and Sinanc, 2013), represented in an

appropriate formalism. The choice of this formalism

heavily depends on the analysis framework to be ap-

plied later on in the engineering process. In our case,

we applied a simple table assigning “V”-qualities to

types of data instances with the help of a table.

Process View. The Process View provides a busi-

ness process language representation of the underly-

ing process. However, in our context, the representa-

tion by the chain of already included in the Prin-

cipal View was sufﬁcient to represent the perspective

from process management.

Business View. The Business View embeds the in-

formation contained in all other perspectives, and es-

pecially from both the Process View and the Data

Architecture View, into high level business perspec-

tive in form of an UML Use Case diagram. This

permitted us to embed the process and data/architec-

ture perspective into the organizations management

context and to highlight the participating stakeholders

and their connection in the business case underlying

the production process; thus, providing a general and

very abstract anchor point for all gathered informa-

tion. The Business View also embeds the whole en-

deavor in the underlying information system develop-

ment’s requirement analysis model and provides ref-

erence points for traceability.

Additionally, we harmonized the different per-

spectives by making the connection of elements be-

tween them visually explicit, e.g. by using the same

color representation for instances and the correspond-

ing class. This allowed us to provide a straightforward

traceability of concepts between the different abstrac-

tion layers. For example, one could easily match

a data instance :QualityControl in the Principal View

with its class’s data context group in the Data Entity

View and thus the corresponding IT system where this

instance is maintained/stored in the Data Architecture

View by tracing the instance’s color.

Thus, the diagrammatic modelling language

DAAD is the spearhead of our approach that targets

the data awareness challenge via its embedding in an

iterative, incremental knowledge elicitation method.

4 CONCLUSION

Posing data awareness as central concern of knowl-

edge management proves to be a challenge to

both existing formalisms and techniques. Taking a

socio-technological perspective and a bottom-up self-

adapting approach supplies sufﬁcient tools to tackle

this challenge in concrete practical situations. Here,

KMIS 2019 - 11th International Conference on Knowledge Management and Information Systems

296

we provided a proof of concept regarding a newly de-

vised modelling language in an industry 4.0 domain

that embraces data awareness as central concern.

Beyond the presentation here, we supplied the

modelling language with a proper formal founda-

tion and were able to show, that it shares seman-

tic similarities with UML but gains better accessi-

bility in general. A formalization of updating and

harmonizing multi-perspective models in DAAD and

tool support is planned as future work. Further,

we are still augmenting the language to cover prac-

tical demands in data-analytical business processes,

e.g. including context information (as crucial for sus-

tainable data-analyses) and classical organizational

knowledge models in addition to data and process

models by additionally embracing knowledge maps

(Eppler, 2001) as additional perspective. Seeing the

success of the bottom-up derivation of a self-adapting

knowledge representation formalism, we want to in-

vestigate further the direction of modelling frame-

works that embrace at their core a notion of on-the-ﬂy

self-adaption to concrete stakeholder demands.

Future work will also include the transfer of

DAAD to other domains and apply the three-step

approach to tackle the awareness challenge in other

practical knowledge management application situa-

tions. This will serve the long term sustainability and

development of DAAD.

ACKNOWLEDGEMENTS

The project underlying this

research (EFFPRO 4.0 – In-

tegration and Analysis of De-

sign and Production Data for

a more efficient Development

Process Chain) has received

funding from the German

Federal Ministry for Economic Affairs and Energy

under grant agreement no. 20Y1509E.

REFERENCES

APA (2019) Awareness. In Online Dict. of the Am. Psych.

Assoc., online: https://dictionary.apa.org/awareness.

Aalst, W. V. D., Zhao, J. L., and Wang, H. J. (2015). Busi-

ness process intelligence: Connecting data and pro-

cesses. ACM TMIS, 5(4):18e.

Bhattacharya, K., Gerede, C., Hull, R., Liu, R., and Su,

J. (2007). Towards formal analysis of artifact-centric

business process models. In Int. Conf. on BPM, pages

288–304. Springer.

Bhattacharya, K., Hull, R., and Su, J. (2009). A data-

centric design methodology for business processes. In

Handbook of Research on Business Process Modeling,

pages 503–531. IGI Global.

Bruce, T. A. (1992). Designing quality databases with

IDEF1X information models. Dorset House.

Cicchetti, A., Ciccozzi, F. and Pierantonio, A. (2019).

Multi-view approaches for software and system mod-

elling: a systematic literature review. Software & Sys-

tems Modeling, pages 1–27. Springer.

Cohn, D. and Hull, R. (2009). Business artifacts: A data-

centric approach to modeling business operations and

processes. IEEE Data Eng. Bull., 32(3):3–9.

Combi, C., Oliboni, B., Weske, M., and Zerbato, F. (2018a).

Conceptual modeling of inter-dependencies between

processes and data. In Proc. of ACM Symposium on

Applied Computing, pages 110–119. ACM.

Combi, C., Oliboni, B., Weske, M., and Zerbato, F. (2018b).

Conceptual modeling of processes and data: Connect-

ing different perspectives. In Int. Conf. on Conceptual

Modeling, pages 236–250. Springer.

Curtis, B., Kellner, M. I., and Over, J. (1992). Process mod-

eling. Communications of the ACM, 35(9):75–90.

De Giacomo, G., Oriol, X., Estanol, M. and Teniente, E.

(2017). Linking data and BPMN processes to achieve

executable models. In Int. Conf. on Advanced Infor-

mation Systems Engineering. Springer.

Deutsch, A., Hull, R., Patrizi, F., and Vianu, V. (2009).

Automatic veriﬁcation of data-centric business pro-

cesses. In Proc. of ICDT 2009, pages 252–267. ACM.

Eppler, M. J. (2001). Making knowledge visible through

intranet knowledge maps: Concepts, elements, cases.

In Proc. of HICSS-34. IEEE Computer Society.

Gandomi, A. and Haider, M. (2015). Beyond the hype: Big

data concepts, methods, and analytics. Int. J. of Infor-

mation Management, 35(2):137–144.

Jablonski, S. (2009). Process modeling for holistic process

management. In Handbook of Research on Business

Process Modeling, pages 49–68. IGI Global.

Kruchten, P. B. (1995). The 4+ 1 view model of architec-

ture. IEEE software, 12(6):42–50.

Kumaran, S., Liu, R., and Wu, F. Y. (2008). On the duality

of information-centric and activity-centric models of

business processes. In Int. Conf. on Adv. Information

Systems Engineering, pages 32–47. Springer.

unzle, V. and Reichert, M. (2011). Striving for object-

aware process support: How existing approaches ﬁt

together. In Int. Symp. on Data-Driven Process Dis-

covery and Analysis, pages 169–188. Springer.

Lankhorst, M. et al. (2009). Enterprise architecture at work,

volume 352. Springer.

Moody, D. (2009). The “physics” of notations: toward a sci-

entiﬁc basis for constructing visual notations in soft-

ware engineering. IEEE Trans. on Software Engineer-

ing, 35(6):756–779.

Moody, D. and van Hillegersberg, J. (2008). Evaluating

the visual syntax of uml: An analysis of the cognitive

effectiveness of the uml family of diagrams. In Int.

Conf. on Software Language Engineering, pages 16–

34. Springer.

Towards Data Awareness by Socio-technological Knowledge Management

297

Nigam, A. and Caswell, N. S. (2003). Business artifacts: An

approach to operational speciﬁcation. IBM Systems

Journal, 42(3):428–445.

North, K. and Maier, R. (2019). Wissensmanagement f

industrie 4.0. herausforderungen und l

osungsans

atze.

Industrie 4.0 Management, pages 7–12.

North, K., Maier, R., and Haas, O., editors (2018). Knowl-

edge Management in Digital Change. Springer.

Object Management Group (2013). Business Process

Model and Notation (BPML). https://www.omg.org/

spec/BPMN/2.0.2/.

Object Management Group (2017). Uniﬁed Modeling

Language (OMG UML). https://www.omg.org/spec/

UML/2.5.1/.

Peinl, R. (2017). Knowledge management 4.0 - lessons

learned from IT trends. In Sure-Vetter, Y., Zander,

S., and Harth, A., editors, Proc. of Prof. Knowledge

Management 2017, pages 112–117. CEUR-WS.org.

Reichert, M. and Weber, B. (2012). Enabling ﬂexibil-

ity in process-aware information systems: challenges,

methods, technologies. Springer.

Reijers, H. A. and Mendling, J. (2011). A study into the

factors that inﬂuence the understandability of busi-

ness process models. IEEE Trans. on Sys., Man, and

Cybernetics-Part A, 41(3):449–462.

Sagiroglu, S. and Sinanc, D. (2013). Big data: A review. In

Proc. of CTS 2013, pages 42–47. IEEE.

Stohr, E. A. and Zhao, J. L. (2001). Workﬂow automation:

Overview and research issues. Information Systems

Frontiers, 3(3):281–296.

orrle, H. and Fish, A. (2013). Towards an operationaliza-

tion of the “physics of notations” for the analysis of vi-

sual languages. In Proc. of Int. Conf. on Model Driven

Engineering Languages and Systems, pages 104–120.

Springer.

Sun, S. X., Zhao, J. L., Nunamaker, J. F., and Sheng, O.

R. L. (2006). Formulating the data-ﬂow perspective

for business process management. Information Sys-

tems Research, 17(4):374–391.

Wang, J., Wang, H., Ding, J., Furuta, K., Kanno, T., Ip,

W., and Zhang, W. (2016). On domain modelling of

the service system with its application to enterprise

information systems. Enterpr. Inf. Sys., 10(1):1–16.

Wood-Harper, A. T., Antill, L., and Avison, D. E. (1985).

Information systems deﬁnition: the multiview ap-

proach.

Zachman, J. A. (1987). A framework for information sys-

tems architecture. IBM sys. journ., 26(3):276–292.

KMIS 2019 - 11th International Conference on Knowledge Management and Information Systems

298