An e-Government Project Case Study
Interview based DEMO Axioms' Benefits Validation
Duarte Pinto
1
and David Aveiro
1,2,3
1
Exact Sciences and Engineering Centre, University of Madeira, Caminho da Penteada 9020-105 Funchal, Portugal
2
Madeira Interactive Technologies Institute, Caminho da Penteada, 9020-105 Funchal, Portugal
3
Center for Organizational Design and Engineering, INESC-INOV, Rua Alves Redol 9, 1000-029 Lisboa, Portugal
Keywords: Enterprise Engineering, Enterprise Change, DEMO, Case Study, Validation.
Abstract: This paper has as its background, a practical enterprise change project where the Design and Engineering
Methodology for Organizations (DEMO) was used in the initial stage as to give a neutral and concise but
comprehensive view of the organization of a local government administration in the process of
implementing an e-government project. The main contribution presented in this paper is an interview based
qualitative validation of some of DEMO's axioms and claimed benefits – something that, to our knowledge
has never been done up to now. Namely, we were able to validate DEMO's qualities of conciseness and
comprehensiveness brought about by the transaction and distinction axioms and also the stability of its
ontological models which are, by nature, highly abstracted from the human and technological means that
implement and operate an organization.
1 INTRODUCTION
This paper has as its background a practical
enterprise change project where the Design and
Engineering Methodology for Organizations
(DEMO) was used in the initial stage with the
purpose to give a neutral and concise but
comprehensive view of the organization of a local
government administration having, itself, the
purpose to implement an e-government project. This
administration is present in a small island of a
European archipelago that is dependent on a main
island that has its own autonomous regional
government. We chose to apply DEMO in the
project, due to its growing use in projects in Europe
and purported qualities and benefits given by the
method. Such qualities highly fitted our work
context which had the need to harness the huge
complexity of the government administration target
of the project. The fact that, as far as we are aware
of, no academic study (qualitative or quantitative)
has ever been made to validate DEMO's qualities
gave rise to the idea of realizing research presented
in this paper. This small local government
administration – from now on referred to as SLGA
is a kind of “miniature” replica of almost all
government functions from national to regional level
and – thanks to having so many functions
concentrated in a few persons – was chosen to be a
test pilot for the e-government project, later to be
extended to all government entities of the main
island. This project has three main aspects: (1) the
implementation of a work flow system to simplify
and automate many operational processes currently
paper based and/or – although using Word/Excel
documents – lacking in structure and coherence; (2)
the development of an online portal to automate as
much as possible the interactions and services
currently provided at a local physical Citizen Service
Desk (CSD), so that the citizens can initiate such
interactions in the comfort of their homes; and (3)
the development of an IT integration layer with
other regional and national government entities that
end up executing most of the processes. In this
context, our research team was assigned with the
responsibility of applying DEMO to model the
processes, interactions and information flows
occurring in the SLGA, to be used as a base for the
production of a strategic roadmap of organizational
changes that will have to occur for several
alternative scenarios of e-government
implementations, according to the possible levels of
integration and change in current government
entities and/or their IT systems. Our team comprised
4 DEMO experts, 2 working in the project full-time
138
Pinto D. and Aveiro D..
An e-Government Project Case Study - Interview based DEMO Axioms’ Benefits Validation.
DOI: 10.5220/0005171201380149
In Proceedings of the International Conference on Knowledge Engineering and Ontology Development (KEOD-2014), pages 138-149
ISBN: 978-989-758-049-9
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
and 2 part-time – one 50% and the other 25% –
totaling 55 man-days in a month of project
execution. Many interviews were made to officials
head of each of the SLGA's departments and also to
most of the officials responsible for each unit of
each department. Interviews were made both for
information collection and model validation. A final
global workshop with the presence of all
interviewees was made for final validation where
most models were deemed adequately correct and
complete after some small corrections and additions.
In the end we specified: 216 transactions – and their
associated result types; and 232 fact types – these
include classes/categories and fact types and exclude
properties. We additionally specified 250
ontological transaction kinds that followed a certain
repetitive pattern in certain departments and,
because of that, were abstracted into a small subset
of generic transactions of the above mentioned 216
transactions set. So, in fact, we specified almost 500
transaction kinds in this project.
Ten months after our main project activities
summarized above, we decided to conduct another
round of interviews having, as the main purpose, a
qualitative evaluation of DEMO's qualities of
conciseness and comprehensiveness brought about
by the transaction and distinction axioms and also
the stability of its ontological models. We took the
opportunity to re-validate all previously collected
data, and update existing models in case of
organizational changes. Very few changes and/or
corrections were needed demonstrating the stability
of the DEMO models which are, by nature, highly
abstracted from the human and technological means
implementing and operating an organization. The
qualities of conciseness and comprehensiveness
were also validated by the vast majority of the
interviewees. Regarding the interviews and their
analysis, we used a qualitative research method,
collecting the data with a previously conceived set
of questions specific for this case, most open ended
but with short answers. The outcomes in most
questions were mostly as expected but there were,
however, some peculiar answers.
In the remainder of this paper, section 2 presents
our Motivation, problem and research method. In
section 3, we present a brief introduction of DEMO -
Operation, Transaction and Distinction Axioms.
Section 4 has our Case details and Example
including some models of this case study. Section 5
explores the Interview questions and results based
on our experience and states the intentions behind
each question. Section 6 wraps it up with a Results
analysis and evaluation, and finally, in section 7, we
present our Conclusions.
2 MOTIVATION, PROBLEM AND
RESEARCH METHOD
Figure 1: Design Science Research Cycles.
We frame our motivation and research method in the
Design Science Research paradigm (Hevner et al.
2004)(Hevner 2007) which claims that all design
science research should take in account the three
cycles presented in Figure 1.
Regarding the relevance cycle, the motivation of
this study is the following problem: it is claimed in
(Dietz 2006) that DEMO possesses several qualities
but no formal proofs or studies are provided that
validate such claims. So our purpose was to validate
DEMO's qualities of conciseness,
comprehensiveness and stability of the ontological
models as to bring more weight and value in practice
to this method and associated theories. As for clear
definitions of these qualities, we adopt the ones from
(Dietz 2006). Namely by conciseness we mean that
no superfluous matters are contained in it, that the
whole is compact and succinct (Dietz 2006). That is,
models should provide a view containing the essence
that is a global picture of an organization out of
which all details can be properly specified.
Comprehensiveness implies that all relevant issues
are covered, that the whole is complete (Dietz 2006).
That is, all relevant perspectives like the dynamic
and static aspects of operation, human
responsibilities, operation flow and inter-
dependencies should be clearly understandable and
covered by the models. Stability of the ontological
models is supposedly guaranteed by the
implementation independence of DEMO models.
And by implementation it is understood the
assignment of human and/or computer resources to
operationalize an organization (Dietz 2006).
Looking at the rigor cycle we ground our study
on the sound formal theories behind DEMO and aim
to provide expertise to the Knowledge base while
contributing with a validation case study.
Ane-GovernmentProjectCaseStudy-InterviewbasedDEMOAxioms'BenefitsValidation
139
In respect to the design cycle, the research reported
in this paper aims to apply the DEMO artifact itself
and evaluate its claimed qualities by means of
interviews with key collaborators on the
organization target of study. Regarding such
evaluation, qualitative methods can facilitate the
study of issues in both depth and detail. They do not
have the constraints of predetermined categories of
analysis therefore allowing for a bigger depth,
openness and detail in the inquiry. On the other end
we find that quantitative methods require
standardized measures so that the varying
perspectives and experiences can fit in a limited
number of predetermined response categories to
which numbers are assigned (Patton & Patton 2002).
The main advantage of a quantitative approach is the
possibility to measure reactions of a large amount of
people to a limited set of questions, therefore
making comparisons and statistical aggregations of
data easier, and allowing for it to be presented in a
succinct way (Patton & Patton 2002). The
qualitative approach produces far more detailed
information about a much smaller sample of
individuals and cases. The qualitative approach
therefore increases the depth of the understanding of
the study but reduces the chances of it being
generalized (Patton & Patton 2002). The validity of
a quantitative research depends on careful
instrument construction that assures that what is
measured is really what is supposed to be measured.
This instrument must be appropriate and
standardized according to prescribed procedures.
The focus is on the measuring instrument, i.e. the
testing of the items, such as survey questions or
other measurement tools. In the qualitative inquiry,
the researchers are the instruments. The credibility
of these methods hinge in a great extent on the skill,
competence and rigor of the person doing the
fieldwork as well as what's going on in that person's
personal life that might prove to be a distraction
(Patton & Patton 2002). There is a third approach
that consists on mixing both of these methods, by
mixing both approaches, in some cases a researcher
can provide a better understanding of the problem
not using either the quantitative or qualitative
methods alone (Creswell & Plano Clark 2007). A
research using this third approach is usually named a
mixed methods research and can be defined as the
“class of research where the researcher mixes or
combines quantitative and qualitative research
techniques, methods, approaches, concepts or
language into a single study” (Johnson &
Onwuegbuzie 2004). By using a mixed methods
research, the researchers can provide more
comprehensive evidence than either quantitative or
qualitative research alone. Thus the researchers are
given permission to use all tools of data collection
available rather then being restricted to the types of
data collection associated with either of the methods
alone (Creswell & Plano Clark 2007). Given the
dimension of the SGLA target of our project and
analysis we have chosen a qualitative method
approach as we were limited to a small amount of
subjects having the knowledge about the modeled
processes.
As previously mentioned to achieve this
evaluation we opted to interview the key
collaborators involved, using a standardized open-
ended format that although lacking flexibility still
allowed us the use of open ended questions while
facilitating their analyzes furthermore the
generalization of the results (Patton & Patton 2002)
The interview method used can also be framed in the
seven stages of an interview investigation proposed
in (Kvale 1996).
1. Thematizing: formulation of a purpose of the
investigation and description of the topic being
investigated before starting the interviews – in this
case the purpose was the validation of the DEMO's
axioms in terms of the qualities of conciseness and
comprehensiveness and also the stability of its
ontological models. We wanted also to evaluate the
interview method itself. To achieve this we specified
several key points that the interviews should cover,
namely: (1) the duration of the interviews, (2) ability
by the collaborators to answer the questions in the
initial stage of the project both in the terms used by
the interviewers and their knowledge of what was
being asked for, (3) their opinion on the interview
methodology, (4) their current view on the processes
and eventual changes, (5) their perception of the
modeled workflow and ability to relate to the real
workflow in operation, (6) the names used in the
models, either in the organizational functions or the
transactions, (7) their self knowledge of the
organization, (8) the models and their
correspondence to current reality after almost a year
passed and (9) questions regarding the application of
the DEMO methodology and benefits obtained
thanks to its axioms.
2. Designing: planing the study taking in
consideration all the stages before the interviews
take place – to achieve this we devised a set of 43
questions that met the criteria set in the thematizing
as to approach all those 9 subjects, being most of
them open ended, but with the expectancy of rather
short answers considering the extent of the subjects
being inquired.
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
140
3. Interviewing: conducting the interviews based
on a guide and with a reflective approach
considering the desired knowledge – the round of
interviews was conducted with eleven SLGA
collaborators, ten that had been previously
interviewed, all either head of a department or chief
of a division, and one that, although not previously
interviewed, was now the current head of the human
resources department. Interviews took place
individually and were composed by the previously
mentioned set of 43 questions, placed after a re-
validation of the models that had been created for
the interviewee's department. Meetings were
previously scheduled and normally had a duration of
approximately one hour for all heads of department,
and two hours for the two chiefs of division so that
they could give their input on all the several
departments that they are responsible for.
4. Transcribing: preparing the interview results
for analysis; commonly translating oral speech into
written text – all answers were written down and
later organized into a spreadsheet containing all
participants together with the list of questions.
5. Analyzing: deciding, considering the purpose
of the interview and the interview material, what
methods are appropriate for analysis – in order to
facilitate analysis, our answer data was grouped in
sets according to common-theme questions. We then
studied the outcomes of each of those sets taking in
account the devised goals. All answers were also
analyzed individually for particularities and properly
considered in the presented results.
6. Verifying: ascertain the generalizability,
reliability, and validity of the interview findings i.e.
the possibility to apply the results in other contexts,
the consistency of the results, and if the study meets
the intended purpose – the findings of our research
are presented in chapter 6 Results analysis and
evaluation as also are the considerations relating to
those findings.
7. Reporting: communicate the findings and the
methods applied in a form that lives up to scientific
criteria, while taking the ethical aspects of the
investigation into consideration, and that having the
results in a readable and usable product for its
audience – in our case, to communicate our findings
we are using this paper, presenting the background,
contextualization and outcomes as well as a
description of the process used.
3 DEMO - OPERATION,
TRANSACTION AND
DISTINCTION AXIOMS
In the Ψ-theory (Dietz 2009) – on which DEMO is
based – the operation axiom (Dietz 2006) states that,
in organizations, subjects perform two kinds of acts:
production acts that have an effect in the production
world or P-world and coordination acts that have an
effect on the coordination world or C-world.
Subjects are actors performing an actor role
responsible for the execution of these acts. At any
moment, these worlds are in a particular state
specified by the C-facts and P-facts respectively
occurred until that moment in time. When active,
actors take the current state of the P-world and the
C-world into account. C-facts serve as agenda for
actors, which they constantly try to deal with. In
other words, actors interact by means of creating and
dealing with C-facts. This interaction between the
actors and the worlds is illustrated in Figure 3. It
depicts the operational principle of organizations
where actors are committed to deal adequately with
their agenda. The production acts contribute towards
the organization's objectives by bringing about or
delivering products and/or services to the
organization's environment and coordination acts are
the way actors enter into and comply with
commitments towards achieving a certain production
fact (Dietz 2008b).
Figure 2: Basic Transaction Pattern.
Figure 3: Actors Interaction with Production and
Coordination Worlds.
According to the Ψ-theory's transaction axiom the
Ane-GovernmentProjectCaseStudy-InterviewbasedDEMOAxioms'BenefitsValidation
141
coordination acts follow a certain path along a
generic universal pattern called transaction (Dietz
2006). The transaction pattern has three phases: (1)
the order phase, were the initiating actor role of the
transaction expresses his wishes in the shape of a
request, and the executing actor role promises to
produce the desired result; (2) the execution phase
where the executing actor role produces in fact the
desired result; and (3) the result phase, where the
executing actor role states the produced result and
the initiating actor role accepts that result, thus
effectively concluding the transaction. This
sequence is known as the basic transaction pattern,
illustrated in Figure 2, and only considers the “happy
case” where everything happens according to the
expected outcomes. All these five mandatory steps
must happen so that a new production fact is
realized. In (Dietz 2008b) we find the universal
transaction pattern that also considers many other
coordination acts, including cancellations and
rejections that may happen at every step of the
“happy path”.
Even though all transactions go through the four
– social commitment – coordination acts of request,
promise, state and accept, these may be performed
tacitly, i.e. without any kind of explicit
communication happening. This may happen due to
the traditional “no news is good news” rule or pure
forgetfulness which can lead to severe business
breakdown. Thus the importance of always
considering the full transaction pattern and the
initiator and executor roles when designing
organizations (Dietz 2008b).
The distinction axiom from the Ψ-theory states
that three human abilities play a significant role in
an organization's operation: (1) the forma ability that
concerns datalogical actions; (2) the informa that
concerns infological actions; and (3) the performa
that concerns ontological actions (Dietz 2006).
Regarding coordination acts, the performa ability
may be considered the essential human ability for
doing any kind of business as it concerns being able
to engage into commitments either as a performer or
as an addressee of a coordination act (Dietz 2008b).
When it comes to production, the performa ability
concerns the business actors. Those are the actors
who perform production acts like deciding or
judging or producing new and original (non
derivable) things, thus realizing the organization's
production facts. The informa ability on the other
hand concerns the intellectual actors, the ones who
perform infological acts like deriving or computing
already existing facts. And finally the forma ability
concerns the datalogical actors, the ones who
perform datalogical acts like gathering, distributing
or storing documents and or data. The organization
theorem states that actors in each of these abilities
form three kinds of systems whereas the D-
organization supports the I-organization with
datalogical services and the I-organization supports
the B-organization (from Business=Ontological)
with informational services (Dietz & Albani 2005).
By applying these axioms, DEMO is claimed to be
able to produce concise, coherent and complete
models with a reduction of around 90% in
complexity, compared to traditional approaches like
flowcharts and BPMN (Dietz 2008a).
4 CASE DETAILS AND
EXAMPLE
The SLGA currently has two divisions (had three at
the time of the first round of interviews in the
beginning of our project) which include ten main
departments. The first of those two is the Division of
Natural Resources Management (DNRM) that
includes the Veterinary, Fish, Parks and Agriculture
departments, all with a collaborator in charge, being
that the chief of the DNRM division is also in charge
of the Veterinary department. The second division is
the Division of Administration Finances,
Maintenance and Infrastructure Management
(DAFMIM) and comprises the departments of
Human Resources, Supply, Finance, Fleet,
Maintenance and the Citizen Service Desk, each also
with a different collaborator in charge. Each of these
departments deals with specific different aspects of
the SLGA. For example, the Veterinary department
has the only available veterinarian on the island and
deals mostly with farm animals health, safety and
well-being, and food safety issues regarding animal
based food and animal food itself. The Fish
department makes the bridge between the local
fisherman and the local commerce but also deals
with matters related with fishing boats diesel oil, the
selling of ice to local businesses and cold storage
units rental. On the DAFMIM division we find the
department of Human Resources that deals with the
allocation of the workers to the different
departments, their vacations, their evaluations, and
training programs in their unit as well as their day to
day task management realized by the head of each
department. The CSD, although included in the
DAFMIM is barely connected to the other
departments as it works as a local proxy for services
offered by multiple regional divisions located at the
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
142
Figure 4: CSD - Actor Transaction Diagram.
main island, such as employment related issues,
housing, driving related issues and so on. In Figure 4
we can see, an excerpt of the Actor Transaction
Diagram, produced for one of the SLGA
departments, the CSD, and the description can be
found in the following paragraphs.
In Figure 4 we can notice four clear clusters in
the ATD diagram, the first being the transactions
initiated by the citizen; it starts with a citizen service
that may or may not lead to a process realization –
e.g. the case that what the citizen needs is not
provided at this desk but in another specific
government office. If there is a process realization
then there will be a creation of process. The process
realization may have an associated cost
communicated in the process payment transaction.
But there are many processes with no costs
associated that may be target of an emission of proof
of receipt of the request for the realization of the
process. Hence why a step that usually would simply
be the state act of a payment transaction deserves to
be a transaction on its own. The second cluster is the
funds deposit cluster, this one is isolated from the
rest due to its nature. It is a daily transaction that can
only be done by one CSD coordinator at the end of
the day.
The third cluster is the process management
cluster, here are the transactions related with the
process done in the back-office when the CSD
collaborators are free from attending citizens. Two
of the datalogical transactions, the scanning of
documents of the process and archiving of these
documents of the process take place whenever the
CSD collaborators have free time, while the
forwarding process documents may vary depending
on the process forwarding method. If it's sent by fax
it takes place at the time of scanning. If it is sent by
paper through the ferry boat it takes place every
afternoon sometime before the ferry trip. The last
transaction in this cluster also takes place when the
CSD collaborators have free time, but it does not
happen every day as many of the CSD processes
have no returning documents, and in most that do,
those documents are sent directly to the citizen by
postal mail instead of returning to the CSD building.
Finally in the fourth cluster we have the process kind
management and its related transactions that, as we
previously stated, are meant to deal with the constant
change in processes and their related documents and
conditions.
The diagram described previously is a typical
example of contents presented to the interviewees
and already give an idea of the conciseness quality
of DEMO, something to which, as we will see just
next, almost all interviewees agreed to.
5 INTERVIEW QUESTIONS AND
RESULTS
The research pool for this interviews was rather
Ane-GovernmentProjectCaseStudy-InterviewbasedDEMOAxioms'BenefitsValidation
143
small with only eleven individuals but, considering
their positions within the organization and the
objectives of these interviews, this was a very
significant and useful sample. As previously
mentioned, the interview's questions were mostly
based on open ended questions, although most with
short sentence responses. When a simple yes or no
question was made, the usual follow up question
consisted in asking the reasons for such answer.
Out of 43 questions made in the interviews 20
are listed below in Table 1 with a summary of the
multiple outcomes. The other 23 questions were not
relevant for the focus of this paper, as they mostly
focused on: our interview method, approach and
language used in the first round of interviews in the
beginning of the project; as well as on comparisons
with results of two other modeling initiatives that
had occurred prior to ours. Unfortunately we were
unable to obtain enough information to compare our
approach with the others, as the key collaborators
involved in these other initiatives were no longer
working at the SGLA. Therefore, the other 23
questions are not present here, not only due to the
mentioned reasons but also due to space constraints.
The results obtained from these other questions will
be target of another paper. As many of the placed
questions were open ended, we have opted to
summarize and group the related answers in order to
present them in a more compact and intuitive
manner. For each question presented in the
following table, we present the number of
interviewees that answered with each of the
alternative or generally given responses, as well as
the total number of collaborators that in fact
answered to each specific question. After the table
we explain the goals and/or reasoning for each
question asked while already providing some
comments on the results. In the next section we do a
more thorough analysis on the outcome.
Table 1: Interview Answers.
7. Have you ever felt
difficulty with the framing
of the questions that were
made to you? (Regarding
the terms used)
14. How do you evaluate
the workflow in the models
when compared with the
real operational flow of
your work?
Had difficulty 1/10 Corresponds to the
real Work flow
11/11
Had no difficulty
(or quickly
answered)
9/10
15. Looking at the names
assigned to the transactions
would you change any?
16. Which one(s) would
you change?
Specification of
intervention
1/3
Table 1: Interview Answers. (cont.)
Yes 3/11 Expense process 1/3
No 8/11 Records for
statistical purposes
1/3
17. Looking at the names
assigned to the
organizational functions
would you change any?
18. Which one(s) would
you change?
DGMI responsible 1/3
Requester of
vehicle
2/3
Yes 3/11 Names of the
regional offices
1/3
No 8/11 Warehouse
responsible
1/3
19. Can you identify
anything produced in your
organizational area that you
cannot find described in the
models?
20. (If yes) 16. What?
HACCP Control
(fishery)
2/5
Management of
lighting and air.
(multi-purpose
pavilion)
1/5
Technical opinion
(supply and
finances)
1/5
Yes 5/11 Registration of
commitment
1/5
No 6/11 Decision of the
selection of
budgets (supply
and finances)
2/5
21. In your personal
opinion, do you feel that
these models can give you a
concise and unambiguous
notion of what goes on in
the organizational area
where you perform your
work?
28. Do you agree with all
transactions in the areas
under your responsibility?
Yes 11/11 Yes 11/11
No 0/11 No 0/11
29. (If no) Which are the
ones you do not agree with?
30. Can you find any of
your transactions, or one in
an area under your
responsibility that you had
a different perception of
the actors involved before
this modeling?
Yes 0/11
Answers 0/0 No 11/11
31. Which one (s)? 32. Do you consider that
the models that were
produced still describe the
reality of performed
transactions and involved
actors?
Yes 11/11
Answers 0/0 No 0/11
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
144
Table 1: Interview Answers. (cont.)
33. (If no) What has
changed?
35. Can you find any
reason for these models be
considered an important
resource in the knowledge
of the organization for their
own employees?
Yes 8/11
Answers 0/0 No 3/11
37. Suppose you had a new
employee under your
supervision and you had to
explain his roles within the
organization. Would you
consider using any of these
models as an aid in this
explanation?
38. Do you think these
models give you a
comprehensive and
summarized vision of the
organization's operation?
Yes 9/11 Yes 11/11
No 2/11 No 0/11
39. Why? 40. Do you believe that is
useful that these models
give a view abstracted of
implementation (regarding
people, technology,
technical, implementation
channels)?
All is discriminated
and summarized.
8/11
Reflects the steps
and processes and
who performs the
tasks.
1/11
Properly
diagrammed.
2/11 Yes 10/11
Generalized view. 1/11 No 1/11
41. Why? 42.Do you think that the
fact that these models
differentiate the initiator
and executor actor roles
and include the acts of
request, promise, execute,
accept and state a
transaction help understand
and clarify the
responsibilities of each
member of the
organization?
Because there is
constant change in
those items.
9/11
Because it is more
practical.
1/11 Yes 9/11
It makes it difficult
to give credit to the
proper person.
1/11 No 2/11
43. Why?
Because it clarifies
responsibilities
8/11
Needs to be well
explained so that
information does
not get lost
1/11
Can be used for
assigning blame
1/11
There is no need
because the process
is treated as a
whole
1/11
Question 7 tried to determine if the terms used
by us in the interviews were of difficult
understanding for the participants. By terms used,
we refer to more technical words such as actor, role,
and transaction, widely used in the DEMO
methodology. Nine of the participants stated that
either had no difficulty, or, if in fact some doubt
arose, it was promptly clarified by the interviewer's
explanation of the terms. One of the collaborators
however stated that she had in fact difficulty during
the questions as she had been “caught off guard”.
Question 14 had the intention to validate the
workflow modeled in the process step diagrams in
comparison to the real workflow in order to find any
flaws or changes in it. All the eleven collaborators
who answered this question agreed that the
workflow in the models was in fact similar to the
reality of their processes agreeing to what was
modeled in every step of each transaction.
Questions 15 and 16 were related to the names
specified for the transactions. Although there had
been already some discussion and validation on this
point around one year ago, we decided to re-evaluate
the appropriateness of these given names. Three of
collaborators found, each, just one name that they
would change in their department's models.
Questions 17 and 18 were on their turn related to
the names assigned to the organizational functions.
Three collaborators said they would change one or
more names. In fact it was somehow surprising that
one of the collaborators told us that some names
were not “generic” enough. As the positions within
the organization are in constant change the
DAFMIM chief of division did not agree that that
position was used as an organizational function, but
instead suggested that we used chief of said
department. In the same way, it was also suggested
to change the names of the regional authorities, as
they also suffer changes when another government is
elected. In this case it was suggested that we
changed to “regional direction with tutelage of said
service”. The other two suggestions by the other
collaborators were related to names that are more
commonly used, instead of the originally proposed
ones.
Questions 19 and 20 intended to identify possible
missing items that failed to be modeled originally.
Five of the collaborators were able to find items that
were not modeled. In the fish department the Hazard
analysis and critical control points (HACCP)
Control was not modeled initially as the head of the
department did not find it important at the time, but
as the paper print left in the process was significant
both him and the chief of the division in charge now
Ane-GovernmentProjectCaseStudy-InterviewbasedDEMOAxioms'BenefitsValidation
145
qualified it as significant. Finally on the supply and
finances department there were just three new
transactions proposed to be integrated in the current
steps of the product acquisition process. All-in-all,
the number of new items identified was quite small
compared to the vast amount of transactions that
kept stable during this whole year.
On question 21 we tried once again to receive
input by the members of the organization on how
DEMO was appropriate for the modeling, by asking
them if they found the models of their departments
concise and unambiguous. Every collaborator that
answered to this question confirmed this quality.
Questions 28 and 29 aimed at reinforcing the
comprehensiveness quality while confirming if all
collaborators agreed with all the listed transactions
in the models deeming them as needed or even as
essential.
With questions 30 and 31 we intended to find
any eventual discrepancy between the collaborators'
perception of reality compared to the modeled
transactions. No interviewed person had a different
perception of what was modeled, further reinforcing
the comprehensiveness quality.
Question 32 intended to capture the validity of
the work produced nearly one year before, and how
it was still applicable to the current reality. Even
though some collaborators and documents used
changed, all eleven interviewees agreed that all
models still correctly described the reality of the
organization.
Question 35 aimed to get a perception on the
relevance given to the DEMO models by the
collaborators. The answers here varied, and although
most collaborators said yes, three couldn’t find any
reason for the models to be relevant and another one
stated that their activities were already so mechanic
that the models were of little use.
Question 37 had the intention to obtain the
predisposition to use these models to explain
someone who was not familiarized with the
organization and their new tasks. Answers were
somehow similar to the previous question. Most
collaborators said yes, but one questioned the ability
of someone new to understand these models,
although another also mentioned that the actor
transaction diagram, would be a good model to
explain the procedures without complications. Still
another person that also said “no” mentioned also
that the models could be complicated, and it would
be more profitable time wise to show them the real
operations in practice.
With questions 38 and 39 we intended once more
to validate one of the claimed qualities of DEMO
models, and their perception by the organization's
members. To do so, we asked if the models gave a
comprehensive and summarized view of the
organization's operation. All interviewees answered
yes, and when asked why they thought like that,
there was little variation in their answers. Most
replies focused on how everything was
discriminated and properly summarized, others
stated that it was properly diagrammed and reflected
their department of the organization, one also
mentioned that it gave a generalized view of
everything, and finally it was also stated that it
clearly reflected step by step the processes of each
collaborator and their respective tasks.
The objective of question 40 was to validate the
level of abstraction used in DEMO and understand
to what extent this is assimilated by the collaborators
of an organization. Ten out of the eleven
interviewees answered that it was useful to use this
level of abstraction, while one said otherwise. When
asked why the responses showed the understanding
of the reasons as they were mostly based on the fact
that the organization is in constant change, new
employees join, old employees leave and the
documentation is also under constant updates,
therefore this level of abstraction allowed for the
models to remain correct after a long period of time,
and still reflect the reality of the organization, as
also demonstrated in the previous questions of the
interview. Although most answers were centered in
these aspects, one of the collaborators had a very
different opinion that may reflect some difficulty
understanding the method, as the reason used to
justify the “no” was that this level of abstraction
makes it difficult to give proper credit to a
collaborator when its due, because no person names
are used, but instead only the organizational
functions.
The last question of this interview focused on
determining if the collaborators found important the
fact that, in the models, there was a differentiation
between the initiator and executor roles as well as
the specification of each transaction's main steps of
request, promise, execute, state and accept. Nine of
the answers were positive, eight focused on how this
helped indeed to clarify the responsibilities, and how
important it is to know who is responsible for what
within each department. One answer had a different
justification: the organization needs to be well
explained so that information does not get lost, and
this way of modeling did exactly prevent that. There
were two collaborators on the “no” side, one stating
that there was no need for this differentiation in their
department because a single collaborator usually did
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
146
most of the transactions as being a single process,
and the other stated that clarifying the
responsibilities isn’t always good, as the goal of the
employees is to properly do their work, as such, they
normally don’t do mistakes on purpose. And, that
being the case, they should not look to assign blame
but instead work together as a group to fix what
went wrong.
6 RESULTS ANALYSIS AND
EVALUATION
With these interviews we were able to validate the
importance and relevance of the Ψ-theory's
operation, transaction and distinction axioms and the
qualities they bring about in the application of
DEMO. We will now see how these axioms affected
the modeling outcome and the perception the
interviewees have of their organization and of its
respective models, thus, validating the DEMO
qualities target of research in this paper.
In Table 2 we have a summary of our results
analysis where we present the three DEMO qualities
that we intended to validate in the research reported
in this paper, along with the questions (presented in
the first column) and the validating outcomes
(presented in the second column). The reasoning of
how the outcomes validate each quality can be found
after this table.
Table 2: Demo qualities analysis.
DEMO Quality - Concise
21. In your personal opinion, do you
feel that these models can give you a
concise and unambiguous notion of
what goes on in the organizational area
where you perform your work?
100% Yes
38, 39. Do you think these models give
you a comprehensive and summarized
vision of the organization Operation?
100% Yes
40, 41. Do you believe that is useful
that these models give a view abstracted
of implementation (regarding people,
technology, technical, implementation
channels)?
91% Yes
9% No
DEMO Quality - Comprehensive
7. Have you ever felt difficulty with the
framing of the questions that were made
to you? (Regarding the terms used)
10% Yes
90% No
14. How do you evaluate the workflow
in the models when compared with the
real operational flow of your work?
100%
Corresponds
fully
Table 2: Demo qualities analysis. (cont.)
19, 20. Can you identify anything
produced in your organizational area
that you cannot find described in the
models? (Note: in this question,
although almost half of the interviewees
found missing items, the percentage of
missing items in their area of
responsibility varied only from 1% to
6% and the other half reported 0% of
missing items)
45% Yes
55% No
28, 29. Do you agree with all
transactions in the areas under your
responsibility?
100% Yes
30, 31. Can you find any of your
transactions, or one in an area under
your responsibility that you had a
different perception of the actors
involved before this modeling?
100% No
35. Can you find any reason for these
models be considered an important
resource in the knowledge of the
organization for their own employees?
73% Yes
27% No
37. Suppose you had a new employee
under your supervision and you had to
explain his roles within the
organization. Would you consider using
any of these models as an aid in this
explanation?
82% Yes
18% No
38, 39. Do you think these models give
you a comprehensive and summarized
vision of the organization's operation?
100% Yes
42. Do you think that the fact that these
models differentiate the initiator and
executor actor roles and include the acts
of request, promise, execute, accept and
state of a transaction help understand
and clarify the responsibilities of each
member of the organization?
82% Yes
18% No
DEMO Quality - Stable
14. How do you evaluate the workflow
in the models when compared with the
real operational flow of your work?
100%
Corresponds
OK
15, 16. Looking at the names assigned
to the transactions would you change
any?
27% Yes
73% No
17, 18. Looking at the names assigned
to the organizational functions would
you change any?
27% Yes
73% No
32, 33. Do you consider that the models
that were produced still describe the
reality of performed transactions and
involved actors?
100% Yes
Question 7 was relevant to make sure that the
participants were at ease with the main concepts of
the DEMO approach – like actor and transaction
leading to a correct comprehension of the models.
Ane-GovernmentProjectCaseStudy-InterviewbasedDEMOAxioms'BenefitsValidation
147
The strong positive result in this question supports
the comprehensiveness quality of DEMO.
Question 14 was a very important question in
order to demonstrate 2 points. By having a
unanimous answer on how the process step diagram
models reflected the proper workflow of the
organization's departments, we realize the great
importance of the transaction axiom. Thanks to the
structuring of the many essential and common
process steps in the transaction pattern, we managed
to uncover some “hidden” (in the minds of the
persons) transactions and, on the other hand, because
the collaborators become aware that a single
transaction “automatically” includes the many kinds
of social interactions that can happen regarding
some production, they end up evaluating the
modeled process fully corresponds to their daily
work. Thus, this outcome also validates DEMO's
comprehensiveness. We were also able to verify that
the models remained current even after multiple
changes in the organization in terms of persons and
documentation, demonstrating DEMO's quality of
model stability, brought about thanks to the
distinction axiom and its separation of the human
abilities, where we normally abstract from
information processing, communication and
document aspects.
Questions 15 to 18 allowed us to demonstrate
how the naming’s determined for both
organizational functions and transactions were quite
adequate for the collaborators that realize the
respective transactions, something that sometimes
proves difficult when gathering this kind of
information. There were three suggestions of
transaction name changes and four organizational
function changes, but taking in account the huge
number of almost 500 transactions that had been
modeled and the number of twice as much actor
roles involved (even though many repeat themselves
multiple times), the amount of change suggestions is
of very small significance. This outcome strongly
validates the stability of DEMO models.
Questions 19 and 20 demonstrated that, although
the DEMO approach seems to be a very good
approach compared to other methods, it is not
infallible and, as such, these questions allowed us to
detect some transactions that were not modeled on
the first round of interviews around one year ago.
But to give the due relevance to the amount of new
transactions found in a more precise fashion, we
now analyze the answers of each of the five
collaborators that answered this question
individually. One of the collaborators, the chief of
the division DAFMIM, identified three lacking
transactions while 317 transactions were modeled as
being under his responsibility, that is, not much
more than 1% of missing items. The other chief of
division (DNRM) identified only one missing
transaction in the set of 162 transactions modeled for
the areas under her responsibility. Again a
percentage close to 1%. The other three
collaborators – all department heads – identified as
missing transactions in their area of responsibility,
respectively: 1 out of 18, 1 out of 26 and 1 out of 31,
that is, percentages from 6% to 4%. The other 5
department heads could not find any missing item,
leading to 0% of missing items. These figures also
highly contribute to validate DEMO's quality of
comprehensiveness.
With question 21 we validated the conciseness
quality thanks to a unanimous response on how the
models really gave a concise and unambiguous
notion of each of the organizational areas to their
respective heads
Questions 28 and 29 demonstrated the
importance of the operation axiom on how it
allowed that interviewees' interactions within the
organization were correctly modeled, by having a
once again unanimous positive outcome when asked
if they agreed with all the transactions under their
areas of responsibility, thus also contributing to
comprehensiveness.
Questions 30 and 31 helped to demonstrate the
importance of the transaction axiom because all
modeled transactions had all the proper responsible
participants identified, thus also contributing the
comprehensiveness quality.
Questions 32 and 33 demonstrated the
importance of the distinction axiom by proving that
all models were still up to date thanks to the
separation of the ontological, infological and
datalogical aspects, thus reinforcing the validation of
the stability quality.
With questions 35 and 37 we tried to understand
if the models could be considered useful both for
existing collaborators and to help train new ones.
The outcomes were not always the expected but the
answers confirmed that the models were considered
important for the collaborators to have, not only
knowledge of their individual tasks, but also of other
tasks all over the organization. Although some
interviewees pointed out that some of the diagrams
could be difficult to understand and thus answered
negatively, the outcome is strong enough to also
contribute to validate the comprehensiveness
quality.
Finally, the block of questions from 38 to 43
allowed us to demonstrate the understanding by the
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
148
organization's members of some key aspects of
DEMO such as the focus on abstraction from
implementation and the separation of the transaction
steps. Although not unanimously, the majority of
interviewed collaborators found these aspects
important and considered them a good quality of this
modeling process, as it can be seen from some of the
opinions we transcribed. So 38 and 39 clearly both
validate the qualities of comprehensiveness and also
conciseness. 40 and 41 distinctly validate the
conciseness quality, while 42 ends up also validating
comprehensiveness thanks to the clarification
provided by the clear identification of organizational
responsibilities.
7 CONCLUSIONS
Following the tenets of Design Science Research we
presented a relevant and needed contribution of an
interview based qualitative validation of some of
DEMO's axioms and claimed benefits – something
that, to our knowledge has never been done up to
now. Namely we looked at the qualities of
conciseness, comprehensiveness and stability of
DEMO's ontological models. This was done in the
context of a large scale practical DEMO project and,
to our knowledge, no publicly available study exists
that practically demonstrates such qualities. And
such studies like these – based in large scale projects
– are essential to contribute to a more widespread
and mainstream acceptance and adoption of DEMO
in enterprise change projects. We interviewed 11
key departments and division heads involved in a
large e-government project where around 500
ontological transactions were specified. Our research
was able to demonstrate that indeed DEMO's Ψ-
theory and its axioms contribute to provide a concise
and comprehensive view of the essential dynamic
and static aspects of an organization and that, even
after a year has passed, the majority of DEMO
models were still up to date and only needed to be
subjected to some minor changes. Our study has
limitations since the DEMO approach was evaluated
individually. In future studies we will apply also
other modeling approaches such as simple
flowcharts and/or BPMN based so that we can also
evaluate them in the same dimensions of analysis
target of this paper and we can compare the
performance of each approach according to the
perception of the organization's members.
Furthermore, the number of interviewees in the
research presented in this paper is not enough for a
pure quantitative validation which is something that
has to be done also to bring up even more solid
arguments supporting DEMO's claimed qualities.
We expect that, as the project advances from the
pilot stage in the small island to the full fledge stage
in the main island, then we will again apply DEMO
for modeling further processes to be implemented in
the e-government project and we will have a sample
of interviewees big enough for a pure quantitative
validation.
REFERENCES
Hevner, A. R., March, S. T., Park, J., Ram, S.: Design
science in information systems research. MIS
Quarterly. 28, 75–106 (2004).
Hevner, A.: A Three Cycle View of Design Science
Research. Scandinavian Journal of Information
Systems. 19, (2007).
Dietz, J. L. G.: Enterprise Ontology: Theory and
Methodology. Springer, Berlin Heidelberg (2006).
Patton, M.Q., Patton, M.Q.: Qualitative research and
evaluation methods. Sage, Thousand Oaks, Calif.
[etc.] (2002).
Creswell, J. W., Plano Clark, V. L.: Designing and
conducting mixed methods research. SAGE
Publications, Thousand Oaks, Calif. (2007).
Johnson, R. B., Onwuegbuzie, A. J.: Mixed Methods
Research: A Research Paradigm Whose Time Has
Come. EDUCATIONAL RESEARCHER. 33, 14–26
(2004).
Kvale, S.: InterViews: An Introduction to Qualitative
Research Interviewing. SAGE Publications (1996).
Dietz, J. L. G.: Is it PHI TAO PSI or Bullshit? Presented at
the Methodologies for Enterprise Engineering
symposium , Delft (2009).
Dietz, J. L. G.: On the Nature of Business Rules.
Advances in Enterprise Engineering I. 1–15 (2008).
Dietz, J. L. G., Albani, A.: Basic notions regarding
business processes and supporting information
systems. Requirements Eng. 10, 175–183 (2005).
Dietz, J. L. G.: Architecture - Building strategy into
design. Academic Service - Sdu Uitgevers bv (2008).
Ane-GovernmentProjectCaseStudy-InterviewbasedDEMOAxioms'BenefitsValidation
149