John Krogstie • Haralambos Mouratidis • Jianwen Su (Eds.)



CAiSE Radar 2016

Proceedings



Held	at	CAiSE	2016,	Ljubljana,	Slovenia,	June	13–14,	2016





CAiSE Radar 2016



Copyright © 2016 Faculty of Computer and Information Science, University of Ljubljana.

All rights reserved.



Published by Faculty of Computer and Information Science, University of Ljubljana, Večna pot 113,

1000 Ljubljana, Slovenia.



Available online: http://caise2016.si/radar



Editors: John Krogstie • Haralambos Mouratidis • Jianwen Su



Ljubljana, 2016





CIP - Kataložni zapis o publikaciji

Narodna in univerzitetna knjižnica, Ljubljana



659.2:004(082)(0.034.2)



CAISE Radar 2016 : proceedings, held at CAiSE 2016, Ljubljana, Slovenia, June 13-14, 2016 / John Krogstie,

Haralambos Mouratidis, Jianwen Su (Eds.). - El. knjiga. - Ljubljana : Faculty of Computer and Information Science, 2016



Način dostopa (URL): http://caise2016.si/radar



ISBN 978-961-6209-88-5 (pdf)

1. Krogstie, John 2. Conference on Advanced Information Systems Engineering (2016 ; Ljubljana)

284925952





Preface



The Conference on Advanced Information Systems Engineering (CAiSE) has traditionally focused on aspects

that intersect our field – technological and human, theoretical and applied, organizational and societal. The theme for CAiSE 2016 of “Information Systems for Connecting People” emphasized the wish to satisfy the needs and requirements of people, both as individuals and as parts of organizations, which are socio-technical

systems. To further the research on these areas, it is important also to provide arenas where researchers can discuss new ideas in a supportive and exciting environment.

The CAiSE Radar is an experimental format, established for CAiSE 2016, to make CAiSE workshops livelier,

exciting, stimulate discussions, and attract additional active participants by establishing an environment where not only well established and validated research is reported but research in infancy, new ideas, and potentially interesting research projects can be presented and discussed. So similarly to a radar, the idea is to enable researchers to look into the future of the field and identify upcoming trends early. The aim of such effort is on one hand to contribute to the building of research communities and promote the integration of young researchers

into the community, and on the other hand to provide opportunities to discuss ideas early and to receive additional opinions on planned research.

In total, 7 papers were accepted as Radar papers and will be presented and discussed in the following workshops: ASDENCA, BUMDISE, COGNISE, and EMMSAD. As workshop chairs of CAiSE 2016, we

would like to express our gratitude to all the workshop organizers and authors for supporting this new idea and

for eliciting some very exciting papers.



May 2016

John Krogstie

Haralambos Mouratidis

Jianwen Su





EMMSAD 2016

On the Need for More Requirements Towards Visual Notation Design of BPMN Extension..........….1

Dirk van der Linden, Anna Zamansky, and Irit Hadar

Towards precise modeling of time constraints for processes with loops……………………………......5

Margareta Ciglic

Challenges for Assessing and Designing Business Continuity Processes……………………….…….....9

José Brás and Sérgio Guerreiro



ASDENCA/BUMDISE 2016

Identifying and Adressing Adaptability and Information System

Requirements for Tactical Management…………………………………………………………..….…13

Renata Petrevska Nechkoska, Geert Poels and Gjorgji Manceski

Enabling Corporate Innovation through the Middle…………………………………….……….….....25

Onur Saglam, Ecaterina Puricel, Fredrik Hacklin and Boris Battistini

COGNISE 2016

A Cognitive Study of Business Process and Business Rule Integration……………………….…..…...33

Wei Wang, Marta Indulska and Shazia Sadiq

The State of Research on Visualization in Information Systems Engineering ………...…………......38

Jens Gulden, Dirk van der Linden, Banu Aysolmaz, Irit Hadar,

Hajo A. Reijers and Erik H.A. Proper





On the Need for More Requirements Towards

Visual Notation Design of BPMN Extensions

Dirk van der Linden, Anna Zamansky, and Irit Hadar

Department of Information Systems, University of Haifa, Haifa, Israel

{djtlinden,annazam,hadari}@is.haifa.ac.il

Abstract. We present an initial exploration of how BPMN extensions

modify or add to the visual notation, and to what degree further clari-

fication on the requirements for the design of these extensions would be

warranted. We identify a number of concerns can that impact readability

and understandability of models using these modified visual notations.

We argue that these findings demonstrate a need for a more precise set

of requirements to be satisfied by modifications to the visual notation in

the BPMN standard.

Keywords: BPMN extension, visual notation, cognitive effectiveness

1

Introduction

Business Process Model and Notation (BPMN) is a de facto standard for pro-

cess modeling. It is extensively used for different domains, as evidenced by the

amount of extensions available for it. Besides adding to the abstract syntax,

these extensions often also make changes or additions to the visual notation of

BPMN itself. In the current BPMN 2.0 standard [5] there is little explicit re-

quirements posed towards such changes, primarily ensuring that extensions do

not modify the design of the BPMN core’s visual elements. Although it stresses

that a visual notation is desired that “all process modelers will recognize and

understand”, other potentially important requirements are left open, such as

ensuring graphic design or color use is cognitively effective or remains coherent

between extensions.

Using a literature review we investigated what additions and modifications

BPMN extensions have made to the core visual notation, and what understand-

ability and readability concerns with their design can be identified, possibly

stemming from a lack of explicit requirements towards that design. As a starting

point we use a recently published general review on BPMN extensions [1], fur-

ther complemented with an additional literature search. We used similar search

strings to [1], searching for “BPMN +extension”, “BPMN +domain”, and “BPMN

+domain-specific” in Google Scholar and Web of Science. However, our selec-

tion criteria were designed to only include those extensions which reported or

gave examples of changes made to the visual notation by the extension. This

resulted in a list of 27 papers for our analysis.

CAiSE 2016 Radar

1

2

Classification

A number of rules are given in [5, p. 8] to be followed when the visual notation

is extended, essentially stating that new visual markers can be added to other

elements, new shapes and line styles can be added if they do not conflict with

other existing elements, coloring can be used with specific semantics, and that

existing elements are not allowed to be changed. According to these criteria, we

classified the extensions discussed above according to the changes they made

to the concrete syntax (see Tab. 1). The addition of new markers to existing

elements is by far the most popular addition to the concrete syntax, followed on

almost equal footing by introducing new shapes and coloring. One would thus

expect readability concerns to be primarily related to these kinds of changes.

Table 1: Classification of changes to BPMN’s concrete syntax by investigated exten-

sions according to the BPMN standard’s criteria [5, sec. 2.1.3.]. For each kind of change

the total amount and percentage of papers doing so are listed in the final row.

Ref.

New Marker New Shape Coloring

Line

style

/ Indicator

Change

Altuhhov et al. (2013)

2





2





Brambilla et al. (2011)

2





Braun et al. (2014)

2



2



2





Braun & Esswein (2014)

2



2





Braun & Esswein (2015)

2





2





Brucker et al. (2012)

2





Charfi et al. (2010)



2



2





Friedenstab (2012)



2





2



Gagne & Trudel (2009)

2





2



Kopp et al. (2012)



2



2





Labda et al. (2014)

2





Lodhi et al. (2011)



2



2





Lohmann & Nyolt (2011)

2





2





Magnani & Montesi (2011)

2





Magnani & Montesi (2009)



2





2



Marcinkowski & Kuciapski (2012) 2





Martinho et al. (2015)

2





2





Rodriguez et al. (2007)

2





Roder et al. (2015)

2



2





Saeedi et al. (2010)

2





2





Saleem et al. (2012)

2





Salnitri et al. (2015)



2



2



2



Schultz & Radloff (2014)

2





Sungur et al. (2013)

2





Supulniece et al. (2010)

2





Yahya et al. (2015)

2





Zor et al. (2011)

2



2





Amount:

21 (77%)

10 (37%)

9 (33%)

5 (19%)

CAiSE 2016 Radar

2

3

Concerns Noted in BPMN Extensions

1. Suboptimal use of color. Extensions that use color often choose colors

based on their intuitive connotations. For example, green and red are often used

as the positive and negative sides to a bipolar aspect, in particular for security

or access related topics, whereas orange is often used for intermediate positions.

This poses a problem from a cognitive point of view if the notation is to be

generally usable by the wider public, as such design considerations do not take

into account color blindness. Furthermore, some extensions use color schemes

that are hard to distinguish when they lose their actual color in a conversion to

greyscale or even pure black and white, which is an expectable situation when

models are printed or written by hand. A final concern is that of readability with

for example color combinations like blue text on an orange background, which

make it difficult to distinguish and read all elements in a model.

2. Use of overly complicated symbols as markers. The use of markers or

indicators in models is important, as they are used to highlight specific attributes

or show that a particular element is a specific subtype. Thus, it is important

that such markers can be immediately read and understood. The extensions we

analyzed give rise to two challenges: (1) Markers are not semantically transparent

(i.e., have intuitive meaning) (2) Markers consist of too complicated graphics,

especially when scaled down. In a best case scenario, this leads to people needing

more time to read and understand a model, which slows down communication

in the modeling process. In a worse case they will misinterpret the meaning of

markers and interpret the model in an unintended way.

3. Homonymous use of symbols as markers. In some cases we identified the

same or strongly similar symbols being used for different meanings in different

extensions (e.g., a lightning symbol used by three extensions [4, 3, 6] all with a

distinct meaning, lock symbols being used by many security related extensions).

While in isolation this does not have to lead to concerns, it makes it more

difficult to interchange models. Furthermore, it can lead to a situation where a

particular extension becomes dominant and its use of the symbol becomes the

de facto interpretation, requiring others to follow suit.

4. Increased graphical complexity. Some extensions introduce significant

amounts of new symbols. While these symbols on their own are easily distin-

guishable, the sheer amount of different graphical elements present in a model

will pose a threat to such factors as reading speed and the ability of people

to easily distinguish between semantically different elements. This problem is

opposite to the problem of symbol deficit of BPMN, that is, its lack of visual

elements for all semantic constructs [2, p. 383].

4

Towards Design Guidelines for BPMN Extensions?

The concerns we have discussed here could become the starting point of a set of

‘anti-patterns’ that can guide how BPMN extensions approach the design of their

concrete syntax, by showing what to avoid, instead of attempting to prescribe

CAiSE 2016 Radar

3

an exhaustive method that would theoretically lead to a good notation. Doing

so will require additional work on determining appropriate requirements, and

involvement of the BPMN community – both of researcher and practitioner.

References

1. Braun, R., Esswein, W.: Classification of domain-specific BPMN extensions. In: The

Practice of Enterprise Modeling, pp. 42–57. Springer (2014)

2. Genon, N., Heymans, P., Amyot, D.: Analysing the cognitive effectiveness of

the BPMN 2.0 visual notation. In: Software Language Engineering, pp. 377–396.

Springer (2010)

3. Labda, W., Mehandjiev, N., Sampaio, P.: Modeling of privacy-aware business pro-

cesses in BPMN to protect personal data. In: Proceedings of the 29th Annual ACM

Symposium on Applied Computing. pp. 1399–1405 (2014)

4. Lohmann, N., Nyolt, M.: Artifact-centric modeling using BPMN. In: Service-

Oriented Computing-ICSOC 2011 Workshops. pp. 54–65. Springer (2011)

5. (OMG), O.M.G.: Business process model and notation (BPMN) version 2.0. Tech.

rep. (jan 2011), http://taval.de/publications/BPMN20

6. Röder, N., Wiesche, M., Schermann, M., Krcmar, H.: Workaround aware business

process modeling. In: Wirtschaftsinformatik. pp. 482–496 (2015)

Due to space constraints the references of the papers selected in the literature

review can be found in an online appendix at www.dirkvanderlinden.eu/data

CAiSE 2016 Radar

4





Towards precise modeling of time constraints for

processes with loops

Margareta Ciglic

Alpen-Adria-Universität Klagenfurt, Klagenfurt, Austria

margareta.ciglic@aau.at

Abstract. Business processes may contain activities that must be exe-

cuted in compliance with specified time constraints, e.g. activity B must

be completed at most 2 days after activity A. If such activities appear in

a loop, there is a lack of time constraints modeling that is able to define

precise appearance(s) of activities executed in a loop. In this paper, we

want to offer an extended time constraint definition that allows precise

modeling of time constraints on activities that appear in loops.

1

Introduction

Compliance with deadlines and other time constraints is crucial in business pro-

cesses, therefore (proactive) process time management is a very important and

deeply investigated topic. An overview of this field is given in [3].

The aim of proactive time management is to predict and avoid violations

of time constraints. First step towards this goal is the extension of the process

definition with activity durations and other time constraints. An overview of time

constraint types deliver Lanz et al. in [6]. In our work, we focus on two types

of time constraints: the upper-bound (UBC) and the lower-bound constraint

(LBC). They define the longest (respectively, shortest) time interval between

the starting or ending points of two activities [2]. After this step, further time

information is calculated and constantly monitored.

Time management becomes challenging if processes contain loops. In the

literature, loops are a) not handled at all [7], b) handled as a complex activity

[8], c) rolled out into a sequence [9] or rolled out into conditional blocks [5].

Representation of time constraints in processes with loops is, as well as the

loop handling itself, mostly left out of scope. Currently, there are no adequate

representations of time constraints, which allow to specify which of the appear-

ances of an activity are constrained. In [6], the proposed pattern solution for

iteratively performed processes introduces a special time constraint between two

process elements where the second one lies in the succeeding iteration. Combi

et al. [1] propose TNest, a new workflow modeling language for time constraints

definition, that can be used to express time constraints between two activities

in different cycles of a loop, however the notation has a limited scope.

In this paper we want to extend the expressiveness of time constraints. We

propose an extended time constraints definition for precise modeling of time

constraints in processes with loops.

CAiSE 2016 Radar

5



2

Time Constraints

A time constraint is a temporal relation between source and destination nodes

in a process graph P = (NP , EP ), where NP are process nodes and EP process

edges. If a time constraint is specified on a node that appears in a loop in process

P , the source and/or destination specification are more complex and until now

not possible to express, since it is not clear which node appearance is meant by

the node itself.

This problem is illustrated in a simple software development process shown

in figure 1. It contains 2 loops, denoted with LS (loop split) and LJ (loop join),

that may iterate arbitrarily often. There are 3 time constraints for this process,

defined by the customer and/or project manager:

TC1: (1, UBC, 180, A, D) Customer and the company agreed on customer

release 180 days after requirements elicitation.

TC2: (2, UBC, 90, B, D) The customer wants to see the first prototype at

least 90 days before customer release.

TC3: (3, UBC, 5, B, C) Project manager requires that after each software

development cycle the last test is completed within 5 days.

Since each activity that is placed in a loop might appear many times at

runtime, it is not clear which appearance of activities B and C must comply to

time constraints TC2 and TC3.

Fig. 1. Simple software development process

We introduce an extended time constraint definition based on time constraint

definition from [2], such that source and destination are not only a single activity

node but an expression that defines a set of source and destination nodes. With

such expressions we are able to define the exact appearance of an activity placed

in a loop. Extended time constraints, as defined below, can express all time

constraints from the example precisely according to their intended meaning:

TC1: (1, UBC, 180, A, D)

TC2: (2, UBC, 90, FIRST B, D)

TC3: (3, UBC, 5, EACH B, LAST RELATIVE C WITHIN LS2

SAME ITERATION LS1)

CAiSE 2016 Radar

6

Definition 1. (Time Constraint) A time constraint (ID, type, δ, source,

destination) ∈ T CP is a temporal relation between a source node set S ⊆ NP

and a destination node set D ⊆ NP from a process graph P , where ID is the ID

of the time constraint, type declares whether the time constraint is an LBC or

an U BC, δ is the required min. or max. time span between source and destina-

tion, and source and destination are specifications of the source node set S and

destination node set D, respectively.

The source specification is defined as (quantif ier, node label, loop ref erence)

and destination specification as (quantif ier, relation, node label, loop ref erence,

iteration ref erence)

In the syntax of source/destination specification defined below, a quantifier

specifies the topological node position, the relation whether a destination node

must follow the source node (relative) or not (absolute), loop reference binds the

quantifier to a particular loop and the iteration reference binds the destination

to the same or next iteration as the source node, regarding a particular loop.

source := [<quantifier>] node label [<loop reference>]

destination := [<quantifier> [<relation>]] node label [<loop reference>

[<iteration reference>]]

<quantifier> ::= FIRST | LAST | EACH

<relation> ::= RELATIVE | ABSOLUTE

<loop reference> ::= WITHIN loop label

<iteration reference> ::= <iteration> loop label

<iteration> ::= SAME ITERATION | NEXT ITERATION

A time constraint induces a set of time constraints with only one node as

source and one node as destination that apply to a particular appearance of

a source and/or destination node in a loop. The induction rule for such time

constraints is defined by the semantic of source and destination expressions and

the function atomize(tc) that creates source-destination node pairs from the

source and destination node sets.

We define the semantic of source and destination expressions with help of

instance types [4]. An instance type is a subgraph of a process graph P with an

arbitrary number of loop iterations where each XOR-split (condition) and each

LOOP-split have only one successor node. We define the function ξ that takes

any possible source or destination expression (specif ication) of a time constraint

tc as the input and returns a set of source or destination nodes n ∈ NI

from

P

the instance type IP of the process graph P as output:

ξ(specif ication, tc, IP ) : specif ication, tc, IP 7→ R ⊆ NIP

For one possible source or destination expression F IRST X the function ξ

is defined as ξ((F IRST X), tc, IP ) := {n ∈ NI |n.Label = X ∧

P

@m(m ∈ NI

∧ m.Label = X ∧ m < n)}

P

CAiSE 2016 Radar

7

3

Conclusion

So far, there is no possibility to model time constraints on activities that appear

in loops. In this paper we introduced an extension of time constraints definition

that enables modeling of time constraints on processes with loops. We proposed

a set of expressions for precise specifying of source/destination activity appear-

ances in a process with loops and showed the principle for defining the semantics

of such expressions.

Acknowledgments. The author would like to acknowledge and thank Prof. Jo-

hann Eder for his constant support, suggestions and comments on this research.

References

1. C. Combi, M. Gambini, S. Migliorini, and R. Posenato. Representing business

processes through a temporal data-centric workflow modeling language: An appli-

cation to the management of clinical pathways. Systems, Man, and Cybernetics:

Systems, IEEE Transactions on, 44(9):1182–1203, 2014.

2. J. Eder, E. Panagos, and M. Rabinovich. Time constraints in workflow systems.

In CAiSE’99 Proc. of the 11th International Conference on Advanced Information

Systems Engineering, pages 286–300. Springer, 1999.

3. J. Eder, E. Panagos, and M. Rabinovich. Workflow time management revisited.

In Seminal Contributions to Information Systems Engineering, pages 207–213.

Springer, 2013.

4. J. Eder, W. Gruber, and H. Pichler. Transforming workflow graphs. In Interoper-

ability of Enterprise Software and Applications, pages 203–214. Springer, 2006.

5. A. Lanz, R. Posenato, C. Combi, and M. Reichert. Controllability of time-aware

processes at run time. In OTM 2013 Proc. of the On the Move to Meaningful

Internet Systems Conferences, pages 39–56. Springer, 2013.

6. A. Lanz, B. Weber, and M. Reichert. Time patterns for process-aware information

systems. Requirements Engineering, 19(2):113–141, 2014.

7. R. Lu, S. W. Sadiq, V. Padmanabhan, and G. Governatori. Using a temporal

constraint network for business process execution. In ADC ’06 Proc. of the 17th

Australasian Database Conference, volume 49 of CRPIT, pages 157–166. Australian

Computer Society, 2006.

8. O. Marjanovic. Dynamic verification of temporal constraints in production work-

flows. In 11th Australasian Database Conference (ADC 2000), pages 74–81. IEEE,

2000.

9. J. H. Son, J. S. Kim, and M. H. Kim. Extracting the workflow critical path from the

extended well-formed workflow schema. Journal of Computer and System Sciences,

70(1):86–106, 2005.

CAiSE 2016 Radar

8





Challenges for Assessing and Designing

Business Continuity Processes

José Brás and Sérgio Guerreiro

a21400334@alunos.ulusofona.pt, sergio.guerreiro@ulusofona.pt

Lusófona University, Campo Grande 376, 1749-024 Lisbon, Portugal

Abstract. Currently organizations face greater resilience challenges due

to unpredictable and constant threat scenarios. Now these scenarios in-

clude, not only Severe Business Disruptions (SBD), but also the loss

of human lives, critical to maintain daily operations. The consciousness

of the need to address any weaknesses related to the resilience of or-

ganizations is now a topic of higher importance. From now on, organi-

zations need to be able to deal with unpredictable changes within the

IT and the business ecosystems, and for that, need to have a consistent

and well-structured Business Continuity Plan (BCP). In line with this

demand, the challenge to capture all business processes and document

them, sometimes from insufficient, fragmented and inconsistent informa-

tion, persists. The need to reduce the complexity of business processes

with data collection and restoration of a holistic picture of Business Con-

tinuity Process is mandatory and the Business Impact Analysis (BIA),

must be well addressed and support all business continuity activities in

order to build good recovery strategies.

Keywords: Business continuity, DEMO, business impact analysis

1

INTRODUCTION

What is Business Continuity? As a concept, business continuity (BC) was in-

troduced in the sixties as IT ”disaster recovery” and was incorporated in the

business environment with the motivation to ensure the high investments made

in computer systems. A BCP grounds the strategies, procedures and critical ac-

tions needed to comply with and manage a crisis situation [10] and expresses in

what state an organization is to deal with unexpected situations (disasters, forced

outages, reorganizations and sudden changes in the business paradigm) [5].

The British Continuity Institute [4] points that the objective of BC is to

provide a documented framework of processes in order to allow the organization

to resume all of its business processes within its recovery time objective (RTO)

and recovery point objectives (RPO) after a Disruptive Incident (DI). This allows

organizations to create resilience to manage unpredictable changes within the IT

and business ecosystems.

Recent disruptive events showed the vulnerability that the actual business

environment faces and all companies that are technology dependent, with higher

relevance for those in the finance, bank & insurance and telecommunications

industry have. Due to this leverage risk, this theme gained higher importance

for the global corporate landscape as BC preparedness can mean the difference

CAiSE 2016 Radar

9

between a company being able to recover from a DI and continue to operate or

disappear.

Moreover, it is important to have one or more methodologies that allows

an organization to change / adapt their processes, allowing the operation to

continue working with the resources available if a SBD befall. Happening, an

organization quickly needs to adapt its business to new realities and for that it

needs to redesign and re-engineering their processes.

The aim of this research should not only be to draw attention to the con-

stant need to work in business continuity processes, but also provide insight into

the development of the BC function and its processes, finding links to other

methodologies to leverage it.

Discuss around BC being part of the Governance, Risk & Compliance (GRC) [8]

or be a separate and independent discipline exists. The fact is that as a result

of a natural evolution along with the need to integrate similar disciplines to

provide new solutions [2] [3] is now a demand. BC needs new arguments to face

the unexpected and some buzz now about BC shift or adopt new paradigms is

emerging [1].

There are examples from several other disciplines that have shifted their

paradigms or adopted new methodologies, with major improvements and effi-

ciency: Enterprise Engineering with DEMO [6]; Project Management with Lean

and Agile; Enterprise Architecture with TOGAF [9], Quality control with Six

Sigma and so on. Some other hybrid methodologies were also found during this

research: Lean Six Sigma or Enterprise Operational Analysis Using DEMO and

the Enterprise Operating System [7], are also important examples.

1.1

Problem

Nowadays, the majority of companies rely on technology to support its business

and started to deal with more complex business processes that cross multiple

departments and disciplines. The challenge to create and maintain a BC Plan by

identifying, classifying and correctly document all the processes that compose

the business structure, is becoming a very demanding and complex task.

A business continuity plan relies and depends on the correct assessment of all

business processes to determine the cost impact of a sudden loss of key business

functions. Such analysis is done along with a Business Impact Analysis to help

evaluating which organization processes are most important and determining the

best recovery strategies.

The problem is that these processes are often modelled using transforma-

tional techniques and don’t define mechanisms to assess the consistency and

completeness of a business process. On the other hand, the documentation used

is often insufficient, fragmented and inconsistent, leading to an incorrect assess-

ment and to misinterpretations. Many process models are hard to understand

by other people and to keep up-to-date, mainly because don’t have guidelines or

pre-defined naming conventions. This can lead to wrong ideas of the elements of

the adopted notation. Furthermore, when the natural evolution of the process

takes place, these models often become inconsistent and difficult to maintain.

CAiSE 2016 Radar

10



Moreover, it is necessary to provide to management a comprehensive view

of the entire company and at the same time be deep enough, simplifying the

representation of complex processes.

1.2

Approach

The BIA process description (found on ISO 22301, point 8.2.2) was used and

analysed from the point of view of DEMO, in order to find contact points.

A business process involving different departments and external providers was

used for the case study. All business flows, dependencies and activities among

departments were captured and documented using DEMO.

The business process and the existent documentation (based on RTF), was

analysed with all key stakeholders involved and a conversation based technique

(DEMO) was used to assess, complete and validate the process, where applicable.

1.3

Integrating knowledge: DEMO & Business Impact Analysis

Design & Engineering Methodology for Organizations is a conversation-based

technique for the design, engineering, and implementation of organizations and

it is grounded in a theory named as Ψ -theory (PSI), where the standard pattern

of a transaction includes two distinct actor roles: the Initiator and the Executor.

The objective of the research on DEMO along with BCP is to empower the BIA

to reflect a more accurate calculation of the impact of a DI at a business by

giving emphasis to all critical aspect of the business process. Figure 1 points

where DEMO can be used during the BIA process to identify which business

units/departments and processes are essential to the survival of an organization.

Fig. 1. BIA steps adapted from ISO 22301:2012 to show where DEMO is relevant

CAiSE 2016 Radar

11

Observing the figure, step (A) of the BIA process, where is required to iden-

tify all activities that support the provision of products and services, DEMO is

applicable. This is well explained and visible using an ATD model and a trans-

action result table.

Moreover, in step (D) the identification of dependencies that support the

resources for these activities for all interested parties, is one of the best attributes

that DEMO has by providing a complete and unified view of the process.

To make a BIA, all accountable aspects, possible interactions and depen-

dencies are necessary to correctly calculate the impact of a DI on a business.

DEMO uses business processes as their main focus, this characteristic become

an essential tool to leverage the BIA.

1.4

Conclusions

The DEMO model used permitted to represent more realistically all accountable

aspects of a business process and empowered the BIA to build a more accurate

calculation of the impact of a SBD at the business. A BIA is prone to having

flaws that need to be mitigated in advance (misinterpretations, sometimes unable

to clearly identify business processes and by not having a clear and wide view of

them), areas where DEMO can be used to mitigate these issues. DEMO models

also can serve as a chance to discuss design issues and optimization opportunities

with the stakeholders for the processes that are being assessed.

References

1. Armour, M.: Talking about a (business continuity) revolution: Why best prac-

tices are wrong and possible solutions for getting them right. Journal of business

continuity & emergency planning 9(2), 103–111 (2015)

2. Asnar, Y., Massacci, F.: A method for security governance, risk, and compliance

(grc): a goal-process approach. In: Foundations of security analysis and design VI,

pp. 152–184. Springer (2011)

3. Balaouras, S.: The state of business continuity preparedness. Disaster Recovery

Journal 25, 14–16 (2012)

4. BCI, C.: Risk and business continuity management guide. Business Continuity

Institute (2009)

5. COBIT: Cobit 5 for Assurance. ISACA (2013), https://books.google.pt/

books?id=FDdbAwAAQBAJ&lpg=PA1&dq=cobit%205&hl=pt-PT&pg=PA2#v=

onepage&q=cobit%205&f=false

6. Dietz, J.: Enterprise ontology: theory and methodology. 2006. Berling: Springer

(2006)

7. Dudok, E., Guerreiro, S., Babkin, E., Pergl, R., van Kervel, S.J.: Enterprise oper-

ational analysis using demo and the enterprise operating system. In: Advances in

Enterprise Engineering IX, pp. 3–18. Springer (2015)

8. Potter, P.: What does governance, risk & compliance have to do with business con-

tinuity planning anyway? (2013), http://www.continuityinsights.com/article/

2013/06/what-does-governance-risk-compliance-have-do-business-

continuity-planning-anyway

9. TOGAF: TOGAF Version 9.1. Open Group Standard (2011)

10. Tucker, E.: Business Continuity from Preparedness to Recovery: A Standards-

Based Approach. Elsevier Science (2014), https://books.google.pt/books?id=

v95FBAAAQBAJ

CAiSE 2016 Radar

12





Identifying and Addressing

Adaptability and Information System Requirements

for Tactical Management

Renata Petrevska Nechkoska1,2, Geert Poels1, Gjorgji Manceski2

1Ghent University, Ghent, Belgium

{renata.petrevskanechkoska,geert.poels}@ugent.be

2St. Clement Ohridski, Bitola, Macedonia

gmanceski@t-home.mk

Abstract. Making tactical decisions is a daily reality for many people in

different kinds of organizations. Tactics denotes dynamically choosing between

alternative pathways, requiring continuous adaptation while realizing the

organizational strategy within the current context. We are pointing out that

tactical management is a dynamic capability with inherent adaptability that

should be supported by a properly designed information system. Based on

expert interviews and literature study, we identified adaptability needs for

tactical management and derived the specific requirements for a tactical

management information system. Through an Action Design Research

involving four different companies in two countries, and by theoretically

grounding the solution in Haeckel’s Sense-and-Respond Framework for

adaptability, we designed a method that uses Social Network Analysis

techniques and tools and is aimed to support the person responsible for tactical

management in the design and continuous revision of a personalized

information system.

Keywords: Tactical Management, Information Systems, Sense-and-Respond,

Adaptability, Capability, Complex Adaptive Systems, Social Network Analysis

1

Introduction

Tactical management is a managerial function aiming to achieve given goals, with

given resources, in given circumstances, with respect to given rules, preconditions

and strategic guidelines in a dynamic environment. We perceive tactical management

as a capability that needs proper (1) common definition and (2) support with IS. There

is a prerequisite for this function to be adaptable to dynamic changes in each of the

‘givens’ (strategy, goals and KPIs, processes, resources) – As a consequence, the

tactical manager needs to continuously probe the context and get the necessary and

complete information for successful facilitation and steering of the socio-technical

system towards given goals. The ability to deliver a certain business value

continuously, while the circumstances are dynamically changing is denoted by the

concept of business capability [28]. A dynamic capability is defined as the ability to

integrate, build and reconfigure internal and external competences to address the

CAiSE 2016 Radar

13

changes in the environment [29]. Tactical management facilitates continuous

adaptation of an open system of interrelated entities achieve a goal through

dynamically changing expectations, resources, circumstances – by maneuvering with

what is given. To clarify tactical management challenges in adaptability, we are

adding the notion of complex adaptive systems. The company, the team [19] the

system of stakeholders directed towards a purpose that tactical manager needs to steer

towards given goals can be characterized as a complex adaptive system (CAS). It is an

open system with autonomous agents networked together, complex and non-

determined processing of inputs into outputs and emergent behavior [10]. Managing a

CAS, along with the numerous ‘givens’ discussed in the beginning, enhances the need

for tactical management to be adaptable.

The current information system models and solutions are addressing tactical

management in a rudimentary form – mostly because they are not considering its

specificity and needs. The information system supporting tactical management should

provide continuous context capture, broad scope of information entities and diverse

type of information attributes. At the same time, it needs to support system view and

handle the mismatch of the incoming data with expected outcomes. By addressing

tactical management as static, rigid, mid-term planning oriented and process-

prescribed managerial function that is similar to strategic or to operational

management, the information systems are not supporting in an effective manner this

uncovered source of competitive advantage [18].

Our work follows the direction of the dynamic alignment modeling discourse of [6]

who provide a framework to offer systematic methods and tools for capturing,

representing, and reasoning about enterprise and IT capabilities when co-designing

organizational and IT architectures; the runtime adjustments of [33]; along with the

dynamic capability modeling for strategic management [29]. Our work is

complementing these approaches with the goal for shaping and addressing the

dynamic capability of tactical management, its adaptability and information systems.

Our specificity is in the way how to achieve it by placing a focus on the person – the

manager. We are situating the research in the domains of Management and

Information Systems. In order to emphasize the adaptability of this capability, we

provided a managerial method that endorses strategic adaptability (based on the

Sense-and-Respond framework [8]) and developed and applied it for tactics. To map

its information requirements we incorporated components in the method which enable

information system continuous self-design and revision. We are arguing that by

enabling design of a personalized information system by the manager we are

contributing an important component in realization of the adaptability of the dynamic

capability of tactical management towards effective business-IT alignment.

One of the most vital questions we pursued in the research is how to model and

visualize a CAS and its behavior for the purpose of mutual understanding and

orchestrated action of all involved parties? Will this model and visualization help the

manager convey the adaptability of his/her system and map his/her information

needs? CAS can be modeled with fractals, differential equations, agent based models,

cellular automata and networks. Graphs and matrices of the Social Network Analysis

(SNA) [30] [13], even though with very high potential, have been used in

organizations very little. SNA metrics have been used in: construction [21] and in

supply chain management [23]. We will draw attention to the applicability and the

benefits of the SNA visualizations and metrics for managerial purposes of adaptability

– as well as for information system requirements elicitation.

CAiSE 2016 Radar

14

The research has followed Design Science Methodology [9] respecting the

guidelines of the organizational design and information systems design. The relevance

of the research has originated in 30 expert interviews for positioning of the problem,

followed by literature study of current contributions and knowledge gaps. The

knowledge base has been repeatedly consulted for foundations, methodologies and

tools on how to design the artifact. This resulted with selection of the Sense-and-

Respond framework, along with the concepts of dynamic capabilities, Complex

Adaptive Systems and the techniques of Social Network Analysis used in the

research. An Action Design Research (ADR) [25] has been conducted with

practitioners in 4 international companies in two countries – Belgium and Macedonia.

Our research effort resulted with design of an artifact – a method for tactical

management adaptability and information systems self-design. The research has been

communicated through conferences and publications.

We proceed as follows: The next section provides a background for tactical

management, its managerial and information system support. Next, we describe the

research methodology employed in the research. We then discuss the delivery of the

artifact; by elaborating our theoretical starting point, the findings of the ADR and our

idea on addressing them. The final section concludes the implications, limitations and

contributions of the research.

2

Problem investigation

We are investigating two interdependent problems – the tactical management’s need

for modern definition and recognition; and the tactical management’s information

systems design. It is necessary to derive the second from the first, while emphasizing

the emergence of the adaptability as a most significant feature of the tactical

management capability.

Strategy is determining the goals of the organization along with the set of coherent

choices concerning the allocation of resources, activities and approaches to realize

those goals. The main concerns of strategic management are effectiveness and

organizational alignment. Strategic management involves strategy formulation,

implementation, and measurement of strategic benefits realization. Support for these

activities is available in the form of a rich and diverse set of conceptual tools and

management instruments (e.g., Balanced Scorecard, Strategy Maps, VMOST analysis,

SWOT analysis, the Value Chain concept, 5 Forces Analysis, the Performance Prism).

Business Informatics research has integrated such techniques in the design of several

modeling techniques providing understanding, analysis and design support for

strategic management (e.g., the Business Motivation Model [3], the Business

Intelligence Model [14], the Business Model Canvas [16], the Component Business

Model [5]). Furthermore, strategic management information in the form of scorecards

and dashboards with KGIs and KPIs is offered by different types of enterprise

information system [15].

The key element in the contemporary view on operations is the process (e.g.,

production process, service process, or business process in general). The main

concern of operations management is process efficiency in terms of cost, time and

quality. Appropriate managerial methods and techniques include Six Sigma, Theory

of Constraints (TOC), Total Quality Management (TQM), (Lean) Six Sigma,

CAiSE 2016 Radar

15

Statistical Process Control (SPC), Agile and others. Operations are nowadays

characterized as a “high frequency – low latency environment” [12].

Compared to operational and strategic management, relatively few managerial

methods and techniques relate to tactics. The managerial function most closely

practicing tactical management is project management – addressed with PMBOK,

Prince2, Scrum, MS Project. However, project is “a temporary endeavor undertaken

to create a unique product, service, or result” [20] and “must be completed by a

specific time, within budget, and according to specification” [31] – tactical

management continues for an undetermined period of time and requires a ‘systems’

approach and capability for adaptability rather than a ‘projects’ approach and

predictability. Tactics is a concept that is much harder to characterize than strategy

and operations. Abstracting from its originally military context, we can describe it

loosely as employing available means to accomplish an end. More specifically, tactics

refers to the residual choices open to a firm by virtue of the business model that it

employs [4]. As working definition for our research we define tactical management as

the managerial function that addresses the following question: How to achieve what is

expected by utilizing what is given and following certain governing principles in the

current context of the organization and environment? [17]

The IS requirements elicitation and analysis, and to certain extent specification and

validation [32] have been achieved through the following strategies. We explored the

literature, for the generic IS requirements for all managerial functions; we supported

and complemented it with the initial and the secondary set of interviews; after which

the notion was completed with analysis of the behavior of end-user 1 and end-user 11

when they were using our artifact. In the latter, the enhanced adaptability enabled

genuine authentication of the tactical management IS needs.

In [18] based on an in-depth review of the literature we observed that when

examining the support of information systems for different management levels, there

is significantly less coverage of tactical management in general, while operational

management is in hive of solutions, followed by strategic management. There have

been attempts at interconnecting business intelligence and performance management

in a closed-loop approach [11]. For instance, the Business Activity Monitoring

(BAM) approach integrates strategic and operational management levels through

closed loops, providing for tactical management informational input for an event-

driven complement of traditional monitoring [26]. The diffusion of BI into

operational and tactical management layers has been coined Operational BI [11].

Another example is the Corporate Performance Management (CPM) Integration Grid

[22] attempts to provide a multidimensional approach where the tactical management

level is cohered to the strategic level. In general, we concluded that there is a

significant ‘ingestion’ of tactics by operations or strategy resulting in a scarcity of

information systems and business informatics modeling and analysis tools that

support the specificities of tactical management. These specificities relate mainly to

taking systems view of the organization (rather than processes or projects view) and

the need for right-time information on contextual changes (rather than real-time

information). More than anything else, tactical management information systems

should help realizing tactical management’s essential feature of adaptability, as much

as it was associated with mid-range planning, in the past.

The main threads of answers with regards to how appropriate the IS in the

company is for the manager dealing with tactical issues – have been that strategic

dashboards don’t capture the current context, while the operational real-time data is

CAiSE 2016 Radar

16

too overwhelming and not needed for tactical management (with some exceptions).

The interviewees had all different interfaces (paper, electronic, combined) for

organizing the wide variety of obligations deriving from their tactical management

function. In terms of reporting for tactical management needs – the users addressed a

struggle between daily, detailed operational reports and periodical (monthly,

quarterly, annually) reports, usually being too late for something to be effectively

improved. The interviews shed light on the notion that numerous entities (stakeholder,

other department, external collaborator) or events (developments) are not captured in

the information flows for tactical management. With regards to the managerial

methods, the users practiced Agile, Scrum, Microsoft SureStep, Waterfall models or

any method or tool implicitly incorporated in the IS. The finding of unaddressed

mismatch of incoming data and expected KPIs has been consensual for all.

The offer for systematic methods and tools for capturing, representing, and

reasoning about the enterprise, its subsystems and the IT capabilities when co-

designing organizational and IT architectures is scarce. Danesh et al. recognize the

need to “(i) represent and monitor the environment in which the enterprise is situated,

(ii) represent and analyze the strategic objectives and positioning of the enterprise

[29], (iii) design flexible and reconfigurable enterprises that enable transformation

(Combs, 2011), and (iv) specify and build adaptable services that can adhere to

changes in their context and deliver value to consumers [33]” [6]. Complementary to

these needs, we are investigating whether for tactical management there can be

person- not organization-oriented support for IS design; handling the mismatch of

incoming data and expected goals; incorporating risk management; using visuali-

zations for communication and orchestration purposes; supporting systems design.

3

Research Methodology

As the goal of our research is a method (i.e., a designed artifact) to be used by tactical

managers for designing and continuously revising (i.e., contextual awareness and

adaptability) personalized information systems that support their tactical decision

making, Design Science Research (DSR) [9] provides an overall guiding framework

for our research [17].

The identification of Tactical Management adaptability needs and information

systems requirements was performed to further characterize the adaptability needs of

tactical management and derive from these requirements for tactical management

information systems. We performed thirty semi-structured expert interviews with

senior, middle, operational and project managers as well as SME owners, in national

and international companies in the authors’ countries Belgium and Macedonia, to

assess their perception of the role of tactical management in organizations and how

this role is currently supported by information system artifacts, with a special

emphasis on the need for adaptability based on context capture and approaches

tailored to individual needs. A parallel effort has been placed in literature review on

current managerial methods and information systems support for operational, tactical,

strategic and project management, focusing on the identification of specific needs of

tactical management for adaptability to changes and information systems support.

A separate search has been performed to find appropriate theoretical foundation

for the envisioned method. To provide rigor to our research, we wished to ground the

design of the envisioned method in existing theoretical frameworks and concepts that

CAiSE 2016 Radar

17

we deemed appropriate for addressing the tactical management adaptability needs and

information system requirements identified in the first stage of the research. We

therefore looked at various contributions in different fields like Strategic and

Operational

Management,

Leadership,

Information

Systems,

Knowledge

Management, Complexity Theory, Behavioral science, Systems Theory, Network

Theory, Social Network Analysis, and Social Systems Design.

To strengthen the relevance of the research, and to develop, build, justify and

evaluate an artifact that has been immediately proven to work in at least one real

environment, we performed Action Design Research (ADR) [27]. The Building-

Intervention-Evaluation cycles (BIE) of the ADR took place in 4 companies in

Belgium and Macedonia – 2 small and 2 big ones (Company 1, 2, 3, 4) with 11

managers as end-users. In Company 2 and 3 we have investigated tactical

management issues of optimizing staff utilization across projects and shifting the

customer perception of the company – and proposed S&R framework-based solution

for the management to follow. We consider this to be the Alpha-version in the artifact

design. In Company 1 we have investigated the tactical management issues of

enabling customer’s management to spend least time possible on remote

communication with geographically scattered staff members. In Company 4 the issue

was to provide earliest possible information status and discrepancies to management

in a new factory and equipment alignment project. The solution design for the last two

companies is the Beta-version in the artifact design. The research encompassed group

sessions and individual conversations.

4

Design and Development

In the following section we will elaborate how the systems design concepts are used

to design personalized tactical management information systems and how the SIDA

loop is used to continuously reform the structure of the tactical management system

towards its purpose and the personalized information system towards its runtime

adjustment.

The Sense-and-Respond (S&R) Framework [8] has been selected as most

appropriate for supporting tactical management adaptability [17]. The main tenets of

the S&R Framework are outcome instead of output, accountability instead of

traditional job description responsibilities, effectiveness before efficiency, and system

design before process design. These ideas are fundamental to creating adaptability in

environments where there is high unpredictability. The S&R Framework centers both

on Systems Design (SD) – using concepts of purpose, governing principles, roles and

accountabilities, conditions of satisfaction, negotiations – and on the Sense-Interpret-

Decide-Act (SIDA) loop for the continuous discovery of early signals, reasoning upon

them, and introducing changes and reconfiguring the system accordingly. With the

S&R Framework, systems should be structured around a mutually agreed purpose,

which is always defined from the outside-in, or customer-back, not firm-forward.

The solution artifact is a method to be used by a tactical manager that embodies

principles, guidelines and prescriptions on how to achieve adaptability for the tactical

management function by means of reasoning how to act and proper information

system self-design. What we have conceptualized in the Alpha-version of the artifact

has been implemented and evaluated in the four organizational contexts in Belgium

and Macedonia, and with all end-users 1-11. We here present the Beta-version of the

CAiSE 2016 Radar

18



method, which incorporated improvements and was given for implementation and

use, to the end-user 1 in Belgium and end-user 11 in Macedonia:

1. Designing a System, according the S&R Framework principles

o Starting from the Purpose (i.e., the end, the reason for being)

o Visualizing the Role-and-Accountability Diagram

o Specifying Conditions of Satisfaction for every negotiated outcome

2. Designing Information Sensors – what the tactical manager would need to

have as information (regardless of the current supply with reports) in order to

have overview of his system

3. Designing the Information Emitters – what the tactical manager would like

to have been told by the other roles in order to be aware on time for possible

issues disturbing the agreed outcomes

4. Designing the Risk Management

o Visualizing the Information Sensors, Information Emitters, and

Risks per role, around the role of the system designer

o Stating the necessary attributes and their indicators

5. Continuous Revision of the System, Accountabilities, Roles, Information

Sensors, Information Emitters and Risks by performing the Sense-Interpret-

Decide-Act (SIDA) loop and deciding on next steps



Fig. 1. Snapshot 1 with Bi-Partite directed graph using Social Network Analysis for showing a

Role-and-Accountability Diagram, for the ‘role’ of ‘Advisor’



CAiSE 2016 Radar

19

To support the application of the method we have developed an Excel Workbook

with 4 separate sheets for points 1 – 4, while the changes initiated by the SIDA

revisions (i.e., point 5) are registered as monthly changes of the information system

elements, their attributes and indicators. The end-users had the responsibility to

manage their system upon instructions and revise the content of the worksheets

according to everyday business activities. They had to note needs, changes, addressed

and un-addressed issues of adaptability and information needs, and neatly record any

change to the system during the research.

For visualizing the result of the design and revision of the tactical management

information system, we use Social Network Analysis (SNA) and the software tools

Gephi and NodeXL. Fig. 1 presents an example snapshot of the artifact-in-use of end-

user 1 (Company 1). This SNA visualization shows the Role-and-Accountability

diagram onto which the tactical information needs for the role of the ‘Advisor’ are

mapped (i.e., information sensors, information emitters and risks).

The primary outcome ‘ least time spent on remote communications’, highlighted on

the figure with the red oval shape, is the primary purpose of the system around which

the initial system of roles and accountabilities is designed. The roles are nodes

represented with big filled circles colored blue and orange (for the role fulfilled by the

tactical manager of interest). The blue edges represent accountabilities between roles.

They are directed lines with an arrow from the provider role – the role that is

accountable for an outcome (in orange text) – towards the client role – the role that is

receiving the effect of the specific interaction. The orange edges denote how each role

is connected with information system entities (i.e., information sensors, information

emitters and risks). The information sensors, emitters and risks are visualized with

blue, yellow and red circles respectively.

After the initial design of the tactical information system by the ADR researcher

and practitioner (as in Fig. 1), the end-user used this SNA visualization during the

SIDA loop – the perpetual engine for scanning of the context and providing

adaptability to the system of roles and accountabilities – along with the Excel

workbook with the four sheets of details (attributes and indicators) for the role-and-

accountability diagram, the information sensors, information emitters and risks. A

view of the Role-and-Accountabilities diagram (Fig. 2) when, after changes in the

context of work, there was a need to introduce a new node – a ‘role’ – absence

coverage – due to a high level of staff turnover in the developers role. The new role

has

accountability

relationships

with

three

existing

roles.

Such

new

roles/accountabilities get introduced or get extinguished as part of the SIDA loop

mechanism for adaptability to the changing context. Also, some of the information

sensors, information emitters and risks have been marked with red, black or blue

squares (colored to denote at least three types of changes in attribute measurements) –

to visualize changes in the content of the attributes for the respective information

flows. The need for a complete re-designed system of roles and accountabilities

occurs after a change of the primary purpose of the system. This situation occurred

with end-user 1 after 10 months of use and revision of the initial system.

CAiSE 2016 Radar

20





Fig. 2. Snapshot 2 – Revised R&A Diagram with changes in nodes, edges and IS attributes

5

Discussion

We have supported the manager performing the tactical management function with an

artifact for providing adaptability (Sense-and-Respond framework implemented and

enhanced for tactical management) and visualized the behavior of the system being

managed (with Social Network Analysis), through a period of two years. The artifact

is a managerial method that is consisted of components how to manage a Complex

Adaptive System and components how to self-design own information system,

continuously, responding to changes while aiming to reach the given goals.

The discussion of the experiences and outcomes can go in several directions – first

of which is the need for shifting the mindset of the business collaborators in the ADR,

as well as of the general audience in this research – from command-and-control to

sense-and-respond way of thinking and acting; from output to outcome; from resource

planning in terms of people towards roles and accountabilities; from process to

system thinking. When establishing the manager’s system it was necessary for his/her

hierarchy level to have a helicopter view and proper authority over the system.

The end-user 1 experience has been of biggest importance, for several reasons. One

was the answer to the question how did planning and adaptability get along in the

tactical management challenge? Important learning from this interaction was the

user’s addition of a Plan-Do-Check-Act (PDCA) loop for each of the attributes in the

four sheets. While our research focused on providing an adaptability capability for

handling uncertainty by means of the SIDA loop, the end-user felt the need of better

support of planning. In fact, for each of the information sensors, emitters and risks,

the user had idea how something in the specific roles or accountabilities can be

improved – and followed it up until actual implementation. This behavior finds

CAiSE 2016 Radar

21

support in the single- and double-loop organizational learning theory[1] The single-

loop denotes operationalization of given or chosen goals, values, plans and rules. But,

double-loop learning occurs the moment when there is critical questioning of the

given mechanisms and altering the ‘givens’ or deploying different tactics, or different

strategies to reach the goals. This SIDA + PDCA loop is included in the method.

The S&R framework offers components for adaptability such as: (1) negotiations

between roles concerning the conditions of satisfaction, (2) population and re-

population of roles with different employees/departments according to changes in

context, (3) introduction or re-introduction of roles and/or accountabilities in order to

respond to the changing context while aiming for the same primary purpose. But once

it is changed, then there is punctuation of the equilibrium [24] and new system needs

to be designed – which actually happened with end-user 1, denoting continuous

dynamic reconfiguration of roles and capabilities, and flexibility and re-

configurability in general for the purpose of continuously providing value to the

customers. The information flows paralleled the adaptability of the system – the

indicators showed that attributes record continuously changing values. This behavior

corresponds with the needs of tactical managers to know and dynamically self-design

their own personal information flows. For example, out of all the information sensors,

emitters and risk issues, 48% changed their attribute frequency (daily, weekly, bi-

weekly, monthly, quarterly), more specifically 26% increased and 22% decreased

their frequency. There was a shift of 13% in the attribute manner of obtaining, from

an on-demand to daily or weekly, which denotes an important change in the manner

of how the tactical manager wished to obtain the information. Overall, 61% of

incoming information is on-demand unstructured information, which means that only

39% of the information is provided through event-driven reports. It is further

noteworthy that 87% of all sensors and emitters contained qualitative information.

The system sustained by our method, lasts until there is no change in the primary

purpose. In the 10th month, this development occurred with end-user 1, so a

completely new design was needed – and completely new information content. The

scope of entities translated into information entities represented on the R&A diagram

is widest possible – the employees of the Advisor are populating only two other roles,

while the roles of clients, clients of clients, technology provider, developers (Fig. 1)

are populated by persons/departments/companies out of the Company 1.

The Social Network Analysis (SNA) visualizations and metrics have proven to be

the motivating force of distinguishable importance for the end-users, in visualizing

their system, needs, changes, and communicating it with the management,

stakeholders, clients. When presented on a timeline, the SNA visualization gives

precious insight on the adaptability of the system (with roles (nodes) popping-up or

being removed, accountabilities re-negotiated, or complete system re-design). The

same occurs with the Information Sensors, Emitters, Risks changing every of their

attributes (frequency, scope). The method provided the managers insight in their own

responsibilities but also of the people they were in contact with, by combining several

points of view into a 360° overview of the workspace they were functioning in.

6

Conclusion

We used the Sense-and-Respond framework [8] as a generic managerial method for

adaptability and introduced it for tactical management. This enabled the exposure of

CAiSE 2016 Radar

22

the real essence of the adaptability as authentic behavior of the manager and his/her

Complex Adaptive System. Our method has integral feature of information system

self-design and its continuous revision – which resulted with mapping the information

system distinctive requirements for tactical management. By matching the

characteristics of Complex Adaptive System, Sense and Respond framework and how

they can be visualized and measured by the Social Network Analysis – we have been

able to design a method for tactical management to be used by the person, supporting

adaptability and continuously designing own information system requirements.

In this study we contribute to management and its information systems by

emphasizing the notion that for the tactical management capability – adaptability is

essential. Likewise, the manager should be positioning his/her information sensors

and emitters according to his/her own context, and review them continuously. Our

study provides theoretical and empirical evidence that adaptability to changes,

especially for a function ‘in the middle’ such as tactical management, with many

‘givens’ needs to be addressed with managerial support of thinking and acting, and

appropriate information system support. Along with capturing of the multi-faceted

aspects of context, we are introducing the risk awareness and management to be

conscious in the reasons that may disable a role from fulfilling it accountability. Last

but not least, we focused on the person, as a source of adaptability and alignment.

In order to practice the method, the manager has to comprehend and apply the

principles of S&R framework and the method for tactical management. Also, the tool

support for Social Network Analysis has been in beta-versions, putting the

practitioners in a position to depend on researcher’s input of the visualizations and

metrics – suppressing greater creativity and independence of the current users. There

is an issue of getting every stakeholder on board in the system of Role and

Accountabilities, with the same (different) way of thinking and acting – which is not

always possible nor is the case – resulting with resistance and hardship in this sense.

Tactical management is dynamic, complex, very person-dependent function that is

hard to describe as well as support in terms of management and management

information systems. Very few artifacts try to address the person, not the

organization. Even fewer support the visualization of the system being managed. Our

artifact is generic enough to be applied for any business type, category, managerial

level or profile, environment, business or life in general. Hopefully, in near future,

this dynamic capability will become focused source of competitive advantage.

References

1. Argyris, C., & Schön, D. 1978. Organizational learning: A theory of action perspective,

Reading, Mass: Addison Wesley.

2. Bērziša, S., España, S., Grabis, J., Henkel, M., Jokste, L., Kampars, J., Koç, H., Sandkuhl, K.,

Stirna, J., Valverde, F., Zdravkovic J., 2015. Task 5.1 Result Report: State-of-the-Art in

relevant methodology areas, CaaS – Capability as a Service in Digital Enterprises

3. Berkem B. 2008, From The Business Motivation Model (BMM) To Service Oriented

Architecture (SOA), Journal of Object Technology Vol. 7 No. 8, ETH Zurich

4. Casadesus-Masanell R., Ricart J. E., 2010. From Strategy to Business Models and to Tactics,

Elsevier, Long Range Planning, 43.

5. Cherbakov L., Galambos G., Harishankar R., Kalayana S., Rackham G. 2005, Impact of the

service orientation at the business level, IBM Systems Journal, Vol.44, No. 4

6. Danesh, M. H., Loucopoulos, P., Yu, E. 2015. Dynamic Capabilities for Sustainable Enterprise

IT – A Modeling Framework, Conceptual Modeling, pp. 358-366 Springer CAiSE 2016 Radar

23

7. Davis, G.B. 1982. Strategies for information requirements determination. IBM Systems

Journal, 21(1): 4-31.

8. Haeckel, S., 1999. Adaptive Enterprise: Creating and Leading Sense-and-Respond

Organizations, Harvard Business Press.

9. Hevner, A., March, S.,Park, J., Sudha, R., 2004, Design Science in Information Systems

Research, MISQ, March

10. Holland, J. H., 1996. Hidden order. How adaptation creates complexity. Addison Wesley

11. Kemper, H-G., Rausch, P., Baars, H., 2013. Business Intelligence and Performance

Management: Introduction, Advanced Information and Knowledge Processing, Springer.

12. Khan, I., 2013. Reducing Data Latency leads to faster decisions, SAP Sybase IQ–Ver. 16.

13. Müller-Prothmann, T., 2007. Social Network Analysis: A Practical Method to Improve

Knowledge Sharing, Hands-on knowledge co-creation and sharing, Kazi, A. S., Wohlfahrt, P.

L., eds., pp. 219-233, Knowledge Board, Stuttgart

14. Nalchigar S., Yu E. 2013, From Business Intelligence Insights to Actions: A Methodology for

Closing the Sense-and Respond Loop in the Adaptive Enterprise, The Practice of Enterprise

Modeling, Springer Berlin Heidelberg

15. Neely A. D., Adams C., Kennerley M., 2002. The Performance Prism: The Scorecard for

Measuring and Managing Business Success, Financial Times Prentice Hall.

16. Osterwalder, A., Pigneur, Y., Tucci, C. L. 2005, Clarifying Business Models: Origins, Present,

and Future of the Concept. Communications of the Association for Information Systems 15, pp.

1-25.

17. Petrevska Nechkoska, R., Tactical Management in Focus: Adaptability and Information

Systems, 2015a, p. 19-27, Vol 1415, Proceedings of the CAiSE'2015 Doctoral Consortium at

the 27th CAiSE 2015, CEUR Proceedings,.

18. Petrevska Nechkoska, R., Poels, G., Manceski, G., 2015b. Bridging Strategic, Operational and

Project Management Information Systems for Tactical Management Information Provision.

Electronic Journal Information Systems Evaluation, 18(2) 146-158.

19. Plsek, P. E.: Working Paper: Some Emerging Principles for Managers of Complex Adaptive

Systems (CAS), Accessed March 2015.

20. Project Management Institute, 2013; A Guide to the Project Management Body of Knowledge

(PMBOK Guide), Fifth Edition.

21. Pryke, S. 2012, Social Network Analysis in Construction, Wiley-Blackwell

22. Rausch, P., Stumpf, M., 2013. Linking the Operational, Tactical and Strategic Levels by Means

of CPM: An Example in the Construction Industry, Advanced Information and Knowledge

Processing, Springer

23. Rodriguez-Rodriguez R., Leon R. 2016, Social network analysis and supply chain management,

International Journal of Production Management and Engineering, pp. 35-40

24. Romanelli, E., Tushman, M. L. 1994, Organizational transformation as punctuated equilibrium:

an empirical test, Academy of Management Journal, Vol. 37/5, pp 1141-1166.

25. Rossi M., Purao S., Sein M. K., 2012. “Generalizating from Design Research”, IT Artefact

Design & Workplace Intervention

26. Schmidt, W., 2013. Business Activity Monitoring (BAM). Advanced Information and

Knowledge Processing, Springer.

27. Sein M., Henfridsson O., Purao S., Rossi M., Lindgren R., 2011. Action Design Research, MIS

Quarterly Vol. 35

28. Stirna J., Grabis J., Henkel M., Zdravkovic J., 2012. Capability Driven Development – an

Approach toSupport Evolving Organizations, In: The Practice of Enterprise Modeling, pp. 117-

131, Springer Berlin Heidelberg

29. Teece, D. J., Pisano, G., Shuen, A., 1997. Dynamic Capabilities and Strategic Management,

Strategic Management Journal, Vol. 18, No. 7, pp. 509-533, John Wiley & Sons.

30. Tichy, N. M., Tushman, M. L., Fombrun, C. 1979. Social Network Analysis for Organizations,

The Academy of Management Review, Vol. 4/4, pp. 507-519

31. Wysocki, R., 2009. Effective Project Management: Traditional, Agile, Extreme, Wiley

32. Wiegers, K. E. 2003. Software requirements, 2nd Ed. Redmond, Wash.: Microsoft Press.

33. Zdravkovic, J., Stirna, J., Henkel, M., Grabis, J., 2013. Modeling Business Capabilities and

Context Dependent Delivery by Cloud Services, CAISE 2013, LNCS 7908, Springer-Verlag

Berlin Heidelberg

CAiSE 2016 Radar

24





Enabling Corporate Innovation through the Middle

Onur Saglam1, Ecaterina Puricel1, Fredrik Hacklin1, Boris Battistini1



1Department of Management, Technology and Economics, Weinbergstrasse 56/58, CH-

8092, Zurich, Switzerland

{osaglam, epuricel, bbattistini, fhacklin}@ethz.ch

Abstract. Recent research on innovation in large corporations has been

polarized to two extremes: top management decision on new initiatives or

innovations emerging from the bottom of the organization. The focus on these

extremes may partially blunt the crucial role of middle management for the

successful implementation of new innovation initiatives. Building on

illustrative cases of three innovation initiatives, we explore the distinctive

organizational capabilities that need to be developed in order to embed

corporate innovation programs into the organization’s middle layer, where

middle managers drive, adapt and implement these programs.

Keywords: middle management, corporate innovation, organizational

capabilities

1 Principal topic

Established firms need to continuously invest into innovation in order to stay on top

against upcoming transformations and disruptions in the market [1], [2], [3].

Traditionally, firms’ internal R&D capability, such as corporate R&D labs and centers

have served to fulfill this need. However, the model is reaching its limits as it has

proven not sufficient for capturing developments and trends taking place outside the

boundaries of the firm—or even beyond the industry or market. In this regard,

established corporations are increasingly finding themselves in the search for new and

unconventional sources of innovation potential to tap into [4], [5]. As a result, top

management decides to set-up new innovation programs to meet the challenges and

maintain competitiveness. While top managers give generic directives and initial

mandate for such novel programs—e.g., with the primary goal to open up a new

market or improve quality—, an efficient middle management is needed to design

systems, carry them out and re-direct the staff’s activities accordingly.

Recently, the literature on middle managers’ role and importance for strategy

formation and firm performance has regained attention [6]. Existing literature

describes middle managers’ role in areas of strategy [7], innovation [8] or corporate

entrepreneurship [9]. Several studies have described middle managers as the source of

novel innovation [10], key actors for implementing, driving and communicating novel

programs [11] and mediators between top management and the rest of the

organizations [12], [13]. However, only recently the literature looks at the

CAiSE 2016 Radar

25

mechanisms, which are used by middle managers, to achieve their goals. For

example, a recent study by Van de Oever and Martin [14] investigated the micro-

processes middle managers use to initiate and drive business model changes in large

organizations. While these findings are relevant to deepen our understanding on how

middle managers drive innovation (e.g. through business model change), we still lack

more fine-grained insights on the organizational capabilities needed to form a

stimulating environment—i.e. an enabling context—for middle managers to act in.

Indeed, existing literature points out that we need to better understand “how they [the

middle managers] fit these ideas into the broader strategic context, and the overall

effect of these behaviors on the firm’s innovative capacity” [15].

Against this background, we build on the illustrative cases of three innovation

initiatives in large organizations with the aim of exploring the distinctive

organizational capabilities firms need to develop in order to be able embed corporate

innovation programs into the organization’s middle layer, where middle managers

drive, adapt and implement these programs. Our findings suggest that, established

firms are able to innovate through the middle by enabling four distinct organizational

capabilities i.e., stakeholder commitment, experience continuity, mission longevity,

and openness to change.



2 Methodology & research setting

Unified by a shared disbelief of a corporate future where R&D labs take the same

center stage as they have done in the past, we brought together a group of senior

industry leaders with the mission to develop a new vision for corporate innovation.

The core team consisted of experienced middle managers from major European

industry corporates, Alpha, Beta and Gamma. This set of companies represents a

highly interesting and relevant sample out of several reasons. First, all of the three

companies represent—in one way or another—traditional European R&D-centric

industry firms, with significant track records in their respective industries (and

industrial histories), as well as with significant sizes. Second, all of the companies

shared the common concern about the limits of the current R&D-based approach to

innovation, and agreed on the upcoming challenges related to novel sources of

innovation, whatever these may be. Third, the companies were all traditionally

strongly relying on technology. While the underlying differences in technological

legacies may seem obvious, the companies where at different stages in evolution in

terms of freeing themselves up from their dependence on technology per se. Lastly, as

the companies at the time of the study where not active in mutually competing

businesses, this setting allowed an open and free-minded exchange of experiences,

concerns and ideas.



On the basis of this unique access to senior-level middle managers, we were not

only able to develop and document three distinct case studies, but moreover all firms

joined a 2-year series of regular workshop roundtables to discuss challenges related to

CAiSE 2016 Radar

26

the future of corporate innovation, where the participating companies sent senior

executives and had placed this topic on a senior agenda.

As an outcome of these two elements of data collection and generation of insights,

this study presents the shared collective knowledge from a joint learning journey. In

particular, data was collected through interviews, observations, workshops, informal

meetings, archival data and other, publicly-available documentation at various points

of time during an observation period of 36 months. We used structured approaches

and formal coding schemes [16], [17] to derive insights and concepts. Emerging

findings and constructs were iterated within our research team, but also with key

informants, revisiting the cases through additional interviews, allowing continuous

iteration between theory and observations [18].

Against the background of a plethora of insights generated around best practices,

methods and tools, the purpose of this article is not to jump into reporting on

company-specific details on how such where implemented and applied. Instead, the

abstraction, comparison and iteration across the three cases yielded a set of

generalizable findings with respect to the organizational capabilities these three

corporation developed in order to be able to inject innovation programs through their

middle management. In the following we will first give some insights on the distinct

companies and their innovation programs (see Table 1 for an overview).





Table 1. Researched firms and innovation programs

Company

Industry

Innovation program

Alpha

Manufacturing and energy

Corporate venture

distribution

capital unit

Beta

Special chemistry

New business

development unit

Gamma

Manufacturing and engineering

Advanced services unit





The case of Alpha

Alpha is international firm with a focus in manufacturing and energy distribution. To

complement its long-standing technological expertise and renowned R&D centers

around the world, Alpha sought after additional means to reach the next frontier or

corporate innovation. In this context, the group’s top management decided to

established corporate venture capital (CVC) unit in 2009. The CVC unit invests in

early and growth stage entrepreneurial ventures with technologies or business models

of strategic interest to Alpha. It was established specifically to execute upon the

investment opportunities, which would both complement the R&D and M&A

activities. Specifically, corporate venturing entails the origination, financing, and

development of entrepreneurial ventures that are introducing to the market new

technologies and solutions that increase performance, open up new business models

or commercial opportunities.

CAiSE 2016 Radar

27

While every venture investment that the CVC unit enters into passes typical

venture capital thresholds for financial returns, it must additionally meet the strategic

objectives of Alpha and help accelerate a business line or a potential area of business

interest. To ensure and maximize the strategic value from the venturing activities the

CVC unit is strategically aligned to the corporate parent, for example, with linkages

with Alpha’s business units and R&D/technical resources. Moreover, a thorough

understanding of the strategic interests of Alpha’s business units and R&D priorities

allows the CVC unit to work closely with them so that the venture investments are

carefully coordinated and corporate’s knowledge and assets can be leveraged both

early in the evaluation process and later in the post-investment activities.





The case of Beta

Beta is one of the world leading companies within the specialty chemicals industry.

Beta’s products provide innovative and sustainable solution for various industries.

Analysts describe Beta as a financial healthy and lean organization, which

successfully manage a turnaround.

Traditionally, companies within the specialty chemistry industry have been

growing mainly organically from inside or through acquisitions and mergers with

competitors. Though, in 2010, driven by tensions within the industry, Beta decided to

launch a new corporate innovation initiative to foster new business creation by

establishing a distinct corporate unit for new business development (NBD). In

particular, the aim of this corporate innovation initiative is to complement the

traditional forms of R&D and business development by looking into alternative

sources of innovation and value creation. The focus is to look at opportunities or

business models completely new to the organization -which differ from Beta’s core

technologies and markets- with a high growth potential.

From the perspective of organizational structure and embeddedness, the NBD unit

forms an autonomous, corporate unit within the organization with no significant own

budget to fund projects. It reports directly to Beta’s CTO, but direct links to the

executive committee are ensured as well. Decisions about projects to be executed and

opportunities to be pursued are taken by a steering committee, which is formed by

members of the executive committee, the CTO and other relevant company-internal

stakeholders. Furthermore, the unit collaborates for specific projects closely with

operating units and the traditional R&D centers, so it is well embedded and connect

within and to the entire organization.





The case of Gamma

Gamma is a globally operating, leading industrial engineering & manufacturing firm.

Gamma offers a wide range of innovative products and sustainable solutions in

rotating equipment and follows the typical business model of a manufacturing firm: it

sells products and life cycle services oriented around maintenance and repair.

Over the last decade, Gamma’s top management has put major effort to develop its

service business and in 2013, Gamma went through a major re-structuring to set-up

separate product and service divisions. Today services contribute to more than 40% of

Gamma’s revenues. Distinctively, besides traditional services such as repairs or

CAiSE 2016 Radar

28

maintenances, Gamma has also placed emphasis on advanced services, which is based

on professional skills and expertise, to create opportunities for exploiting the existing

service and product business through further leveraging customer relationships; on the

basis of existing technological capabilities. This top management emphasis on

services in general has turned a business under the radar –i.e. advanced services– into

a strategic one. Advanced services may not seem financially attractive at first sight;

however, the middle management realized and eventually championed the strategic

dimension of such services in gaining organizational recognition. For example, the

technology and knowledge intensity of advanced services enables Gamma to spot

latent customer needs and to create spillover business through sales of new equipment

and related lifecycle services. Moreover, as a company dedicated to keep its strong

product identity, advanced services help Gamma in optimizing the product vs.

services sales and managing the complementarities between the separate businesses

through mitigating value conflicts inherent to two distinct businesses.



3 Results and implications

A closer examination of the three diverse and illustrative cases reveals four

organizational capabilities— stakeholder commitment, experience continuity, mission

longevity, and openness to change—that enable middle managers’ action for

corporate innovation. In the following we will briefly explain the identified

capabilities and their importance.



Stakeholder commitment. As boundary spanners, middle managers perform a

coordinating role through mediating, negotiating and interpreting connections

between the different internal and external stakeholders of an organization [19].

Middle managers interface role with multiple stakeholders enables access to diverse

knowledge and creates a greater potential for knowledge creation. For example in the

case of Alpha, the established CVC unit perceived great commitment from the top

management, but also from other parts of the organization. This stakeholder

commitment enables the CVC unit’s middle managers to established important

relationships with the external VC environment as well as the start-up ecosystem,

which are crucial for the success of Alpha’s innovation program.

Experience continuity. While literature focuses on the role of middle managers in

organizational change [6], it has neglected the importance of middle manager

experience and political capital within the organization. Our study revealed the

capability of experience continuity to be necessary for the successful implementation

of a corporate innovation program as the developed and accumulated political capital

facilitates issue selling in ensuing change. For example, in the case of Beta two third

of members of the NBD unit had a long employment history in Beta, while the

remaining third started their employment with this innovation program. This was a

deliberate choice as personal development of employees and succession planning is

deeply anchored in Beta’s overall strategy. This combination of seniority with Beta

among the middle managers not only allowed a transfer of deep technical knowledge

CAiSE 2016 Radar

29

and political experience, but also supported the successful execution of the innovation

program.

Mission longevity. Corporate innovation programs have a long-term orientation

rather than fulfilling short-term operational objectives. As in any innovation activity,

although program success is expected be lucrative, the process bears unpredictable

contingencies and involve high probability of failure [20]. For example in the case of

Gamma, although advanced services as an independent business is not financially

attractive at first sight, its synergistic effects with core businesses have turned it into a

strategic unit, as it uncovers latent customer needs and opens the door for spillover

product and service sales. The effect was not immediately prominent to top

management but rather emerged in the long run through the observation of middle

management and operational levels. Hence, organizations willing to uncover hidden

middle management potential require a more failure-tolerant behavior regarding its

innovation initiatives. This is also in line with a recent study by Tian & Wang in 2014

[21], which find positive relationship between failure tolerance and innovation

productivity.

Openness to change. Recent literature suggests that middle managers play a

crucial role in organizational change [14]. While our insights confirm these findings,

they also suggest that the organization needs to have certain openness to change in

order to allow middle managers to implement the innovation programs. For example

in the case of Beta, the R&D centers realized that the NBD unit implemented an

efficient process by using a specifically designed database and decided to adapt this

process also for their activities. The Beta’s openness to change organizational

processes stimulated the middle managers to further refine and communicate their

process leading to a better implementation of the corporate innovation program.



In summary, our preliminary findings suggest that large firms need to develop

certain organizational capabilities in order to inject novel corporate innovation

programs into the middle layer. We contribute to the growing literature on middle

management perspective [6] by identifying distinct organization capabilities as

important elements for successful middle managers’ action. This has relevant

implication for academics and practitioners alike. Further studies, should examine

how an organization can deliberately develop such capabilities and how they

influence middle managers’ micro-processes [14]. From a practitioner’s perspective,

our findings should redirect managers’ attention from the ideation of the corporate

innovation program to the important elements of its implementation. Senior

executives should be aware of the organizations need to develop distinct capabilities

in order to harness middle management efficacy and allow them to achieve their full

potential to drive corporate innovation programs.





CAiSE 2016 Radar

30

References

1. Guth, W.D., Ginsberg, A.: Guest editor's introduction. Corporate entrepreneurship.

Strategic Management Journal. 11, 5--15 (1990).

2. Corbett, A., Covin, J.G., O'Connor, G.C., Tucci, C.L.: Corporate Entrepreneurship:

State of the Art Research and a Future Research Agenda. Journal of Product

Innovation Management. 30, 812--820 (2013)

3. Crockett, D.R., McGee, J.E., Payne, G.T.: Employing new business divisions to

exploit disruptive innovations: The interplay between characteristics of the

corporation and those of the venture management team. Journal of Product

Innovation Management. 30, 856--879 (2013)

4. Cassiman, B., Veugelers, R.: In search of complementarity in innovation strategy:

internal R&D and external knowledge acquisition. Management science. 52, 68—82

(2006)

5. Kuratko, D. F., & Audretsch, D. B. (2013). Clarifying the domains of corporate

entrepreneurship. International Entrepreneurship and Management Journal,9(3), 323-

335.

6. Wooldridge, B., Schmid, T., Floyd, S.W.: The middle management perspective on

strategy process: contributions, synthesis, and future research. Journal of

Management 34. 1190--1221 (2008)

7. Mintzberg, H., Waters, J.A.: Of strategies, deliberate and emergent. Strategic

Management Journal. 6, 257--272 (1985)

8. Kodama, M.: Knowledge creation through network strategic communities: Case

studies on new product development in Japanese companies. Long Range Planning.

38, 27--49 (2005)

9. Kuratko, D.F., Ireland, R.D., Covin, J.G., Hornsby, J.S.: A model of middle-level

managers’ entrepreneurial behavior. Entrepreneurship Theory and Practice. 29, 699--

716 (2005)

10. Chen, M.H., Chang Y.Y., Chang Y.C.: Entrepreneurial Orientation, Social Networks,

and Creative Performance: Middle Managers as Corporate Entrepreneurs. Creativity

and Innovation Management. 24, 493--507 (2015)

11. Floyd, S. W., & Wooldridge, B. Building strategy from the middle. Sage

Publications, Thousand Oaks, CA (2000)

12. Nonaka, I.: A dynamic theory of knowledge creation. Organization Science. 5, 14--37

(1994)

13. Ireland, R.D., Covin, J.G., Kuratko, D.F.: Conceptualizing corporate entrepreneurship

strategy. Entrepreneurship Theory and Practice. 33, 19--46 (2009)

14. Van den Oever, K., Martin, X.: Business Model Change: Managerial Roles and

Tactics in Decision-Making. In: Baden Fuller, C., Mangematin, V. (eds.) Business

Models and Modelling. Advnaces in Strategic Management, vol. 33, pp. 387--420,

(2015)

15. Wooldridge, B., Schmid, T., Floyd, S.W.: The middle management perspective on

strategy process: contributions, synthesis, and future research. Journal of

Management. 34, p.1212 (2008)

16. Gioia, D.A., Corley, K.G., Hamilton, A.L.: Seeking qualitative rigor in inductive

research notes on the Gioia Methodology. Organizational Research Methods. 16, 15--

31 (2013)

17. Langley, A.: Strategies for theorizing from process data. Academy of Management

Review. 24, 691--710 (1999)

18. Miles, M.B., Huberman, M. Qualitative Data Analysis. Sage Publications, Beverly

Hills, CA (1994)

CAiSE 2016 Radar

31

19. Floyd, S.W., Wooldridge, B.: Middle management’s strategic influence and

organizational performance. Journal of Management Studies. 34, 465--485 (1997)

20. Holmstrom, B.: Agency costs and innovation. Journal of Economic Behavior and

Organization. 12, 305--27 (1989)

21. Tian, X., Wang, T.Y.: Tolerance for Failure and Corporate Innovation. Review of

Financial Studies. 27, 211--255 (2014)





CAiSE 2016 Radar

32





A Cognitive Study of Business Process and Business Rule

Integration

Wei Wang1, Marta Indulska2, Shazia Sadiq1

1School of Information Technology and Electrical Engineering,

The University of Queensland, Brisbane, Australia

w.wang9@uq.edu.au, shazia@itee.uq.edu.au

2University of Queensland Business School,

The University of Queensland, Brisbane, Australia

m.indulska@business.uq.edu.au

Abstract. The conceptual and pragmatic overlap between business process mod-

els and business rules indicates a need to model the two related aspects together,

at least under some specific conditions. While a considerable amount of research

has focused on the development of integration methods for process and business

rule modeling, whether such integration improves (or diminishes) the under-

standing of business processes has not been investigated. Following our prior

work in which we proposed a cognitive process model to explore theoretical

foundations that underpin the understanding of process models, in this paper, we

propose a design of an eye tracking study to explore integrated and separated

modeling of processes and rules on the backdrop of the cognitive process.

Keywords: Business Process Modeling, Integrated Modeling, Cognition The-

ory, Eye-Tracking, Human Information Processing

1

Introduction

Conceptual models are widely used in organizations by information systems analysts

and designers to represent, understand and analyze complex business domains [1]. A

good understanding of a domain is a prerequisite to effective communication and de-

sign. Thus, how conceptual models improve human cognition of the domain repre-

sented is a very important research question. Such questions have also been explored

in the context of business process models.

Business process models mainly focus on the modeling of business activities of an

organization. In practice, business rules, which are extracted from laws, policies and

procedures, play an indispensable role in the design, specification and implementation

of process models. Business rules can be represented in a separated or an integrated

manner, and often by a mix of the two. By ‘separated manner’, we mean the rules con-

straining process activities are documented in separate documents or rule engines, and

the relations and connections of business process models and the rules are not explicitly

represented in the process models. By ‘integrated manner’, we mean graphically in a

process model. In such integrated models, business rules can be represented either as

CAiSE 2016 Radar

33

text annotations, as graphical links to external rules, or diagrammatically using a com-

bination of sequence flows, activities and gateways.

Despite arguments for integration and despite a variety of integration methods being

available, whether such integration improves user understanding of the process models

has not been investigated. In particular, while researchers have argued that integrated

modeling can improve the understanding of business processes, this proposition has

neither been theoretically analyzed, nor empirically evaluated. Yet, such understanding

is crucial for the advancement of process modeling methods. To bridge this gap, in our

prior work [2], we proposed a four-stage cognitive process model to explore the theo-

retical foundations that underpin the understanding of process models and different in-

tegration methods, i.e. rule linking, annotation, and diagrammatical integration. Fol-

lowing this contribution, three important questions remain to be answered: (1) what are

the differences in cognitive activities when rules are modeled in a separated manner

and in an integrated manner, (2) what are the cognitive differences of different integra-

tion methods, and (3) what is the best integration method for different types of rules in

terms of model understanding. In this paper, we focus on the first question. We use rule

linking as a representative integration method to compare with rules modeled in a sep-

arated method, with an innovative design to evaluate those arguments in our prior work

[2] which can be measured using an eye-tracking device.

2

Cognitive Process Model

In our prior work [2], we used cognitive load theory [3], information representation

theory [4], and information integration theory [6] to reason about the cognitive process

of a user using a process model with related business rules. We argue that to fully un-

derstand a business process, three components need to be studied: the process model,

the business rules, and the impact or implications the rules have on process activities.

Thus, our paper [2] proposed a four-stage model based on a human information search-

ing and processing cognitive model [7]. The four cognitive stages involve rule aware-

ness (being aware of the existence of rules that the business activities are required to be

in compliance with), rule locating (finding the relevant rules), rule comprehension (de-

velopment of understanding of an individual information element), and information in-

tegration (combining knowledge of the process and how the business rules constrain it

together).

In the next section, we will introduce our experimental design based on the cognitive

process to explore the effect of business rule integration.

3

Experiment Design

Our aim of the experimental design is to identify differences in cognitive activities

when rules are modeled in a separated manner and in a linked manner, thus a between-

groups design will be used. We aim to have 40 - 100 students participate in the study,

all of whom have a background in conceptual modeling (although not process model-

ing). All student participants will have knowledge of Entity-Relation diagram modeling

CAiSE 2016 Radar

34



and analysis. To facilitate the data collection in the eye-tracking stage, the process

model, rules, and the questions are presented on the screen at the same time. The layout

is illustrated in Fig. 1. In the following, we will use area Q, P, R to refer to the question

area, process model area, and rules area respectively. Each participant will be presented

one of the treatment randomly. In treatment A, a process model and rules will be dis-

played separately, and all the rules will be presented in area R. While in treatment B,

the same process and rules will be linked together, i.e. when a rule icon in area P is

clicked, only the corresponding rule will be displayed in area R. The order of questions

and rules will be random to control for learning effects.



Fig. 1. Screen layout illustration

Participants will first be briefly introduced to business processes, business process

models and BPMN, as well as business rules. Then in the experiment, a participant

will be asked to answer eight comprehension questions based on the process model

and related business rules. Finally, perception questions about the difficulty of the

model and rules will be asked, as well as the difficulty of the questions, and the confi-

dence of answers given.

Efficiency and effectiveness are two typical objective measurements in classical

conceptual model understanding experiments [8] . In this study, we are firstly motivated

to explore whether integrated process and rule representations have better efficiency

and effectiveness in model understanding. Thus, we use the time to complete model

comprehension questions as the measure of efficiency, and the correct number of an-

swers as the measure of effectiveness. In the following section, we will use micro-level

cognitive behavior data to evaluate our arguments.

4

Measuring Cognitive Activities

Using eye-tracking devices, we will be able to record micro-level cognitive behavior

data, which we would not be able to easily capture without such equipment. We will

rely on [9] to reuse common eye tracking metrics. We use Tobii X2-30 as the eye track-

ing hardware and Tobii Studio as the data collecting and analyzing software. We will

analyze the difference of eye fixation durations, fixation counts, the sequence of fixa-

tions, and visit counts between the control and the treatment. Fig. 2 illustrates the se-

quence of fixations, fixation heat map, and statistics of visit count and fixation durations

from a test of the equipment on control group experimental materials.

CAiSE 2016 Radar

35





a b

c

Fig. 2. Data capture illustration

Although the earlier outlined four cognitive activities are different, they cannot be

easily separated using eye tracking devices. Moreover, as the sequences of these cog-

nitive activities can be interlaced, the identification of each stage becomes even more

difficult. Thus, we may need to measure some stages together, which is a limitation of

this research. In the following discussion, we introduce our arguments and possible

measurements, which are still being refined.

Rule awareness. We argue that “awareness of business rules can be improved by

integrating the rules into relevant process model diagrams through any of the already

existing integration approaches” [2].

Measurement 1: Rule awareness can be captured by the time from start of reading of

a related question to the time the participant starts to look for a relevant rule. In Fig. 1,

this would refer to the first time the participant’s eyes leave area P and enter area R.

Measurement 2: Lack of rule awareness occurs when, after reading a question, the

participant starts to answer the question only after focusing on the process model but

not reading the rules, i.e. the eyes only focus in area Q and P but not area R. The number

of times this happens can be a measurement of rule awareness.

Rule locating. We argue that “by integrating business rules into business process

models the cognitive effort in searching for relevant rules can be reduced” [2]. Three

measurements can be used, depending on different scenarios:

Measurement 1: If a participant finds the right business rule at the first opportunity

and remains fixed on that rule, then rule locating can be the sum of time the eyes stay

in area R, from the point of reading the question until the initial time the target rule is

found.

Measurement 2: If a participant finds the target rule but continues to explore other

rules, this situation indicates that participant does not realize they have located the right

rule. The number of times this scenario happens is the measurement.

Measurement 3: If a participant starts to answer the question without reading any

rule, this is an indication of rule ignorance.

Rule comprehension and information integration. Although rule comprehension and

information integration are two different cognitive behaviors, it is hard to separate the

two using the eye-tracking device. I.e. when eyes are fixed on an activity or a rule, we

cannot tell if the participant is comprehending the information, or integrating the infor-

mation with other information. Accordingly, we will measure these two together. Given

that we argue that if information elements are not integrated physically then one has to

mentally integrate them, which involves dividing attention between the multiple

CAiSE 2016 Radar

36

sources of information, cross-referencing each source [2], our measurements are

planned as follows:

Measurement 1: Number of visits in area P and R respectively. A visit is defined as

the entering of eye fixations to an area from outside. A visit means an attention switch-

ing to this area from another area.

Measurement 2: Number of eye-fixations and fixation durations on the target activity

and the target business rule, which are indications of cognitive function [8].

5

Discussion

In this paper, we proposed our experiment to evaluate the differences of cognitive ac-

tivities between rules modeled in a separated manner and rules modeled in a linked

manner, based on our prior proposal of a cognitive process model. We expect that mod-

eling business process in an integrated manner can achieve better efficiency and effec-

tiveness than in a separated manner, and we proposed three arguments and several pos-

sible cognitive activity measurements that can indicate such difference. The result of

this study will evaluate our arguments and provide a variety of detailed cognitive be-

havior data that will provide insight on how users understand business process models.

References

1.

Recker, J., Rosemann, M., Green, P.F., Indulska, M.: Do ontological deficiencies in mod-

eling grammars matter? MIS Quarterly. 35, 57–79 (2011).

2.

Wang, W., Indulska, M., Sadiq, S.: A Theoretical Perspective on Integrated Modeling of

Business Processes and Rules. In: Proceedings of the 28th Internation Conference on Ad-

vanced Information Systems Engineering (Forum). Springer (2016).

3.

Paas, F.G., Van Merriënboer, J.J.: Instructional control of cognitive load in the training of

complex cognitive tasks. Educational psychology review. 6, 351–371 (1994).

4.

Larkin, J.H., Simon, H.A.: Why a diagram is (sometimes) worth ten thousand words. Cog-

nitive science. 11, 65–100 (1987).

5.

Chandler, P., Sweller, J.: The split-attention effect as a factor in the design of instruction.

British Journal of Educational Psychology. 62, 233–246 (1992).

6.

Chandler, P., Sweller, J.: Cognitive load theory and the format of instruction. Cognition and

instruction. 8, 293–332 (1991).

7.

Guthrie, J.T.: Locating Information in Documents: Examination of a Cognitive Model.

Reading Research Quarterly. 23, 178–199 (1988).

8. Duchowski, A.T.: Eye tracking methodology: theory and practice. Springer, London

(2007).

9.

Rayner, K.: Eye movements in reading and information processing: 20 years of research.

Psychological bulletin. 124, 372 (1998).

CAiSE 2016 Radar

37





The State of Research on Visualization in

Information Systems Engineering

Jens Gulden1, Dirk van der Linden2, Banu Aysolmaz3, Irit Hadar2,

Hajo A. Reijers3, and Erik H.A. Proper4

1 University of Duisburg-Essen, 45141 Essen, Germany

jens.gulden@uni-due.de

2 University of Haifa, Department of Information Systems, Haifa 31905, Israel

djtlinden@is.haifa.ac.il

3 VU University Amsterdam, 1081 HV Amsterdam, The Netherlands

b.e.aysolmaz@vu.nl

4 Luxembourg Institute of Science and Technology, L-4362 Esch-sur-Alzette,

Luxembourg

e.proper@acm.org

Abstract. The relevance of information visualization and visual com-

munication is continuously growing on a large socio-economical scale.

This implies an increasing variety of applications and research questions

related to visualizations that are used as part of information systems,

and for the design thereof. Several overview publications have pointed

out the relevance of scientific reflection on the use and the effects of visu-

alizations in information systems engineering. This paper aims at taking

stock of recent developments in the field and systematizing some recent

contributions dealing with cognitive, formal and empirical challenges.

Keywords: information visualization, data visualization, modeling, di-

agrams, information systems, systems engineering, cognitive factors

1

Visualization in Information Systems Engineering

There is a growing need for systematic and scientific underpinnings of visualiza-

tions, as the visualizations incorporated in today’s system engineering methods

will be implemented in tomorrow’s infrastructures and system environments, and

thus will shape the way humans interoperate with and benefit from a technolog-

ically engineered infrastructure in the future. Information systems engineering

is concerned with the processing and usage of information of diverse kinds. To

understand and communicate the settings in which systems operate, to interact

with them, as well as to conceptualize requirements and design specifications,

visualizations have become an important cultural technique for designing and

operating systems.

Not despite, but because of its cross-cutting relationships to diverse fields of

systems engineering and its fundamental role as an ubiquitous cultural technique,

the use of visualizations has not yet systematically been put into the focus of

CAiSE 2016 Radar

38

research activities in the engineering discipline. This seems to be caused by the

fact that from the individual perspective of a specialized engineering discipline,

the use of visualizations is naturally perceived as merely imported from other

disciplines such as graphic design, visual communication, and semiotics. Even

in closely related fields of a single discipline such as information systems, there

are groups working in, e. g., business process modelling, enterprise modelling,

software engineering, and requirements engineering that would share similar, if

not outright the same, underlying fundamental challenges, yet do reinvent the

wheel by spending time on fundamentals that could be investigated in joint

efforts.

The Visualization in Information Systems Engineering Network (VISE-Net-

work) is a recently launched pan-European international collaboration between

academic and industrial partners located in Germany, Israel, Luxembourg, the

Netherlands, and Turkey. Researchers involved in VISE have all come across

different visualization challenges at some point, and realized the use of stronger

collaboration to learn from each other’s work in dealing with such challenges.

This paper is meant to give a summarizing look on recent advances that

VISE members have made in visualization research with respect to information

systems engineering. These reflect the research interests of the different members

and institutes, as well as already ongoing collaborative efforts for particular

visualization challenges. We further hope to stimulate more discourse on these

topics, as well as involvement of others interested in visualization research.

2

What have we achieved so far?

The challenges investigated by different VISE members address various concerns,

but so far in particular center around those of a cognitive, formal or empiric and

methodical nature. The below sections give a brief overview of current ongoing

work we address.

2.1

Cognitive viewpoints on visualizations

One of the main purposes of using visualizations in information systems engineer-

ing is to communicate knowledge between different stakeholders with conceptual

models. The work of the VISE network has shown that the model factors, the

characteristics about how the model is structured using its notational elements,

have an important effect on the understandability of the models [17]. Even when

the model factors are the same, presentation of the models in different modu-

lar structures may affect the understandability [16]. Moreover, we have shown

that the concrete syntax; the design of symbols, colors and position of nodes;

and the design of labels and icons on model elements have an important impact

on model comprehension [18, 19, 16]. The guidelines on structuring the specific

types of models we have developed became a well-accepted practice [11].

In our studies, the comprehension of models is found to be dependent on

personal factors such as background, experience, motivation and learning style

CAiSE 2016 Radar

39

[20, 14]. Improving such personal factors for the specific case where a concep-

tual model is used even without changing the visual aspects of the conceptual

model may help enhance comprehension and communication between stakehold-

ers. Overall, we were able to categorize the factors affecting the comprehension

of models based on user characteristics, context and notational properties [24].

We have also focused on utilizing advanced highlighting and animation tech-

niques to enhance the model reading experience [2, 15, 1]. With respect to dif-

ferent modes of perception and their influence on cognitive processing of infor-

mation, [5] discusses fundamentals on how the use of audio as part of concrete

syntax elements of modeling languages can look like.

2.2

Formalizations of visual languages and visual artifacts

Next to ensuring that visualizations are appropriate for their users and their cog-

nitive make-up, ensuring that they can be systematically designed and used is of

significant importance. Work of the VISE Network has focused on several aspects

here, ranging from introducing meta-models and DSMLs to appropriately use

visualizations, to more fundamental work comparing modeling languages using

foundational ontologies and investigating the very feasibility itself of formalizing

the requirements we have for the design of good visualizations.

The state-of-the-art view on formal visualization modeling assumes that vi-

sualizations can sufficiently be described by a one-to-one mapping between con-

ceptual elements and visual representations [8]. To overcome these limitations,

[6] suggests a meta-model that introduces additional concepts specific to the

domain of visualizations which enrich the semantics of formalized visualization

descriptions beyond simple type-to-symbol mappings.

Work on incorporating existing philosophical models into the discourse on

visual modeling languages is presented in [13], which examines the Business

Process Modeling Notation (BPMN) language on the background of concepts

provided in the Bunde Wand Weber (BWW) ontology. Using philosophical work

in a responsible way to enrich the discourse in the Information Systems discipline

is one of the directions the VISE Network is picking up in ongoing research.

Formalization is used as a way to operationalize aspects of conceptual mod-

eling languages, that is, to ensure their application is systematic and leads to

replicable results. The call for formalization has led to efforts to formally ver-

ify whether visual notations are cognitively effective or not, by operationalizing

(parts of) a widely used theory: the Physics of Notations (PoN) [12]. However,

in recent work we examined whether all parts of the PoN theory are equally,

or at all, suitable for formalization. We found [23] that not all principles of the

PoN can be represented in a formal manner (e. g. using set theory), because

they require additional external information which in itself might be difficult to

formalize, or in the most difficult cases require direct user involvement.

2.3

Empirical user studies

Empirical work, in particular work involving users of visualization artifacts, is

of vital importance to evaluating its benefit in practice. Research done by VISE

CAiSE 2016 Radar

40

members ranges from experiments comparing what kind of diagrams contribute

most to comprehension of a domain (e. g., [9]) to studies evaluating the degree

to which visualization design involves users in the first place, to the design of

platforms fostering collaboration between users of visualizations themselves.

Our network has focused on empirically evaluating model comprehension us-

ing different techniques. Retention and transfer tasks have been used heavily

to compare the model comprehension under different personal and notational

factors [15, 19, 16, 10]. By means of these experiments, we have understood the

effect of cognitive viewpoints as discussed in Subsection 2.1 on model compre-

hension better. The survey studies also support us to understand the viewpoint

of model readers [17, 18]. Based on the findings, we deduct that to enhance

cognitive aspects in modeling domain, new visualization studies shall consider

notational semantics, concrete syntax and personal factors equally to provide

better understanding for the users.

While the importance of user involvement in the design and development of

visual artifacts is widely accepted, many efforts do not manage to reach signifi-

cant levels of user involvement. In a recent work [22, 21] we explored to what de-

gree applications of the PoN involved users. We found that a significant amount

of applications did not involve users at all, whether for the critical analysis of

existing visual notations, or the design and implementation of new ones. This

means that little requirements elicitation for the visual notation is actually done

with the very people using it.

In an ongoing project we are designing a platform, or perhaps, marketplace,

for the collaborative design and evaluation of visual elements. Initially this is

targeted at elements of visual notations for modeling languages, but it can be

used for other visualization efforts as well. This could lead to a kind of ‘cer-

tification’ for visual elements: that particular core elements are known to be

interpreted correctly and efficiently by users, in which contexts it is so, and for

what purposes it is already used.

2.4

Applications of visualizations

Besides knowing how to create good visualizations, it is important to know how

to put them to good use. This ranges from being able to select the appropriate

kind of visualization and its parameters, to setting up interaction strategies.

VISE members are working on topics like the animation of process models and

other kinds of annotations, as well as on methods for applying visual models in

software development processes.

We have used highlighting of process model constructs to focus the readers

to the relevant parts of the model [15]. To include the readers better in the

process of reading, animation support based on different scenarios, textual and

audio annotations, and active involvement of users via controls are essential [2].

Currently we are developing process model animation techniques to guide the

focus of the users as they read the models and enhance comprehension [1].

Utilizing visual models as construction tools in model-driven software devel-

opment processes requires to embed them into a tool chain where their formal

CAiSE 2016 Radar

41

content is further analyzed and transformed by code generation and interpreta-

tion techniques. While traditional approaches mainly focus on general purpose

modeling languages with few semantic expressiveness, [4, 3] put semantically rich

domain-specific visual enterprise modeling languages into focus for this.

An approach for automatically suggesting visualizations for data analyses in

specific domains is introduced in [7]. It is based on domain-specific semantic

characteristics of the underlying data, which allows to narrow down the range

of useful analytic views compared to merely syntactic approaches.

3

Future outlook

In this paper we have given an overview of recent work by VISE members on

different visualization challenges in information systems engineering, from deal-

ing with the cognitive point of view to formalization and empirical work. We

hope to stimulate more researchers interested in visualization to collaborate on

these and other challenges and exchange their insights on research strategies to

best deal with these and other emerging challenges in the field of visualization

in Information Systems.

References

1. Banu

Aysolmaz.

PRIME

Process

Model

Animation,

2016.

http://prime.cs.vu.nl/prime.html (2016-05-06).

2. Banu Aysolmaz and Hajo Reijers. Towards an Integrated Framework for Invigorat-

ing Process Models: A Research Agenda. In Ross Brown, Simone Kriglstein, and

Stefanie Rinderle-Ma, editors, Proc. Int. Work. Theory Appl. Process Vis. Conf.

Bus. Process Model. (BPM 2015), Innsbruck, 2015. Springer.

3. Jens Gulden. Meta-model patterns for expressing relationships between organiza-

tion model concepts and software implementation concepts. In B. Barn et al., edi-

tors, Proceedings of the workshop Towards the Model Driven Organization (AMINO

2013), volume Vol. 1102 of CEUR Workshop Proceedings. IEEE, 2013.

4. Jens Gulden. Methodical Support for Model-Driven Software Engineering with En-

terprise Models. PhD thesis, Universität Duisburg-Essen, Berlin, 2013.

5. Jens Gulden. Towards a generalized notion of audio as part of the concrete syntax

of business process modeling languages. In Ross Brown, Simone Kriglstein, and

Stefanie Rinderle-Ma, editors, Proc. Int. Work. Theory Appl. Process Vis. Conf.

Bus. Process Model. (BPM 2014), Eindhoven, 2014. Springer.

6. Jens Gulden. A description framework for data visualizations in enterprise informa-

tion systems. In Proceedings of the 19th IEEE International Enterprise Distributed

Object Computing Conference (EDOC 2015) in Adelaide, Australia, 2015-09-22 –

2015-09-25. IEEE Xplore, 2015.

7. Jens Gulden. Semantic support for visual analyses in electronic commerce set-

tings. In Proceedings of the 18th International Conference on Human-Computer

Interaction, HCI International 2016, New York, 2016. Springer. HCI International,

Toronto, Canada, 17-22 July 2016.

8. Jens Gulden and Hajo A. Reijers. Toward advanced visualization techniques for

conceptual modeling. In Janis Grabis and Kurt Sandkuhl, editors, Proceedings of

CAiSE 2016 Radar

42

the CAiSE Forum 2015 Stockholm, Sweden, June 8-12, 2015, CEUR Workshop

Proceedings. CEUR, 2015.

9. Irit Hadar and Orit Hazzan. On the contribution of uml diagrams to software

system comprehension. Journal of Object Technology, 3(1):143–156, 2004.

10. Irit Hadar, Iris Reinhartz-Berger, Tsvi Kuflik, Anna Perini, Filippo Ricca, and

Angelo Susi. Comparing the comprehensibility of requirements models expressed

in Use Case and Tropos: Results from a family of experiments. Inf. Softw. Technol.,

55(10):1823–1843, oct 2013.

11. J. Mendling, Hajo Reijers, and W.M.P. van der Aalst. Seven process modeling

guidelines (7PMG). Inf. Softw. Technol., 52(2):127–136, feb 2010.

12. Daniel L. Moody. The “physics” of notations: Toward a scientific basis for con-

structing visual notations in software engineering. IEEE Transactions on Software

Engineering, 35(6):756–779, 11/12 2009.

13. Jan Recker, Marta Indulska, and Peter Green. Extending representational analysis:

Bpmn user and developer perspectives. In Proceedings of the 5th International

Conference on Business Process Management, BPM’07, pages 384–399, Berlin,

Heidelberg, 2007. Springer-Verlag.

14. Jan Recker, Hajo Reijers, and Sander G. van de Wouw. Process Model Compre-

hension: The Effects of Cognitive Abilities, Learning Style, and Strategy. Commun.

Assoc. Inf. Syst. Vol., 34(9):199–222, 2014.

15. Hajo Reijers, Thomas Freytag, Jan Mendling, and Andreas Eckleder. Syntax high-

lighting in business process models. Decis. Support Syst., 51(3):339–349, 2011.

16. Hajo A. Reijers and Jan Mendling. Modularity in Process Models: Review and

Effects. Bus. Process Manag. - 6th Int. Conf. BPM 2008, Milan, Italy, Sept. 2-4,

2008. Proc., pages 20–35, 2008.

17. Hajo A. Reijers and Jan Mendling. A Study Into the Factors That Influence the

Understandability of Business Process Models. IEEE Trans. Syst. Man, Cybern. -

Part A Syst. Humans, 41(3):449–462, may 2011.

18. Marcello La Rosa, Arthur H. M. ter Hofstede, Petia Wohed, Hajo A. Reijers, Jan

Mendling, and Wil M. P. van der Aalst. Managing Process Model Complexity via

Concrete Syntax Modifications. IEEE Trans. Ind. Informatics, 7(2):255–265, 2011.

19. Matthias Schrepfer, Johannes Wolf, Jan Mendling, and Hajo Reijers. The Impact

of Secondary Notation on Process Model Understanding. In Anne Persson and

Janis Stirna, editors, Pract. Enterp. Model. SE - 13, volume 39 of Lecture Notes in

Business Information Processing, pages 161–175. Springer Berlin Heidelberg, 2009.

20. Dirk van der Linden. Personal semantics of meta-concepts in conceptual modeling

languages. PhD thesis, Radboud Universiteit Nijmegen, 2014.

21. Dirk van der Linden. An argument for more user-centric analysis of modeling

languages visual notation quality. In Advanced Information Systems Engineering

Workshops, pages 114–120. Springer, 2015.

22. Dirk van der Linden and Irit Hadar. User Involvement in Applications of the

PoN. In Proceedings of the 4th International Workshop on Cognitive Aspects of

Information Systems Engineering (COGNISE). Springer, 2016.

23. Dirk van der Linden, Anna Zamansky, and Irit Hadar. How cognitively effective is

a visual notation? On the inherent difficulty of operationalizing the physics of no-

tations. In Proc. 21st Intl. Conf. Expl. Mod. Meth. f. Sys. Anal. Des. (EMMSAD).

Springer, 2016.

24. Sander G Van De Wouw, Hajo Reijers, and Jan Recker. An Integrative Framework

of the Factors Affecting Process Model Understanding: A Learning Perspective. In

AMCIS 2010 Proc., pages 1–10, 2010.

CAiSE 2016 Radar

43





Document Outline


RadarprefaceTOC

RadarProceedings_content vanderLinden-EMMSAD-radar-abstract

Ciglic_EMMSAD2016_Poster_final

Challenges for Assessing and Designing Business Continuity Processes_final

CrC Renata Petrevska Nechkoska CAiSE Radar v2

Extended abstract_Saglam, Puricel, Hacklin, Battistini 2016

COGNISE_2016_paper_12

The State of Research on Visualization in Information Systems Engineering