A Tool for Developing and Evaluating Methods (TDEM)
Murad Huseynli, Michael Chima Ogbuachi and Udo Bub
Faculty of Informatics, E
¨
otv
¨
os Lor
´
and University (ELTE), P
´
azmany P
´
eter s
´
et
´
any 1/C, H-1117 Budapest, Hungary
Keywords:
Information Systems, Digital Innovation, Design Science Research, Method Engineering, Method Design,
Method Evaluation, Process Evaluation, Microservices.
Abstract:
Digital Innovation (DI) is the use of digital technologies during the process of innovation or as the result of in-
novation. While we see a huge amount of interest in DI research from IS scholars, there is still scarcity in terms
of using Design Science Research (DSR) for the engineering of DI, specifically from a method perspective and
a possible integration of both fields. In this paper, we propose a tool for developing and evaluating methods to
build method design theories, not just in the DI field, but also generally applicable to method engineering in
other fields as well. Thus, we contribute systematically to the body of knowledge for information systems with
a tool that shall be an effective assistant in method engineering. The tool builds on the eight components of a
design theory first introduced by (Gregor and Jones, 2007) and conceptualized by (Offermann et al., 2010a)
for the artifact type “method”. Finally, we analyze the role and utility of the tool in detail, concentrating on a
method for the engineering of DI, in a DI project related to the microservices architecture, and lastly, we show
generalizability of the tool in terms of design process evaluation, not specifically following a DSR paradigm.
1 INTRODUCTION
Design science research is generally applied to classes
of artifacts, including algorithms, computer or hu-
man interfaces, design methods and techniques, and
process models. It is primarily applied in engineer-
ing and computer science, but is not limited to these
fields and can be found in many other disciplines and
fields. Since the research paradigm of design science
focuses on the creation of an innovative artifact, it
contributes to the field of digital innovation from both
a research and practical perspective. Digital innova-
tion describes the use of digital technologies as an aid
during the innovation process, or as an outcome of
innovation. Digital innovation is increasingly becom-
ing a dominant research focus in the field of informa-
tion systems. Although the pursuit of innovation is
crucial for forward-looking companies (Bub, 2018),
little attention is paid to the systematic development
of digital innovation in related fields such as IS and
computer science, which play a key role in digital
transformation. Innovation plays an important role in
economic progress and consequently, in human well-
being (Baumol, 2002). Information technologies are
developing at a rapid pace, so digital innovation based
on digital technologies has become the focus of at-
tention. Basically, digital innovation combines phys-
ical and digital entities and creates new products on
this basis (Hevner and Gregor, 2020). As digitized
products and services are among the key drivers of
innovation, digital innovation is now practiced by a
growing number of companies (Berntsson-Svensson
and Taghavianfar, 2015). At its core, digital innova-
tion is about novel digital technologies, entrepreneur-
ship and innovation management, and digitization of
information and services.
At its essence, DSR in Information Systems (IS)
is about DI (Hevner and Gregor, 2020). While this
much synergy exists between both fields, not many
studies have been conducted regarding method engi-
neering for DI, using DSR. This gap was the moti-
vation for this research. As we looked into existing
literature we found that we could not identify a scien-
tific tool that allowed researchers to perform proper
method engineering, not only in DI, but also in any
other IS fields where DSR methodologies can make a
huge impact (considering research processes and out-
comes). DSR is a generally accepted paradigm of In-
formation Systems (IS) research (Hevner et al., 2004;
Peffers et al., 2008). Among the artifact types that
are subject to design in IS, “method” is of high rel-
evance and subject to intensive research work in the
IS community (Hevner et al., 2004; Bucher and Win-
ter, 2008; Offermann et al., 2010a). Reviews of ex-
Huseynli, M., Ogbuachi, M. and Bub, U.
A Tool for Developing and Evaluating Methods (TDEM).
DOI: 10.5220/0011077700003179
In Proceedings of the 24th International Conference on Enterprise Information Systems (ICEIS 2022) - Volume 2, pages 543-552
ISBN: 978-989-758-569-2; ISSN: 2184-4992
Copyright
c
2022 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
543
isting literature (Huseynli et al., 2021) showed that
there is still scarcity regarding methods for the engi-
neering of DI using DSR. (Bub, 2018) takes the ini-
tiative for the development of such a method and puts
the basis for it. This effort, though, needs further val-
idation and improvement. Designing such a method
requires a thorough, comprehensive knowledge base
with both scientific rigor and practical relevance. Re-
searchers face challenges in identifying academic in-
sight and progress in comparison to the state of the
art, due to the complex nature and diverse ways of
describing method artifacts. While it is unavoidable
to put much effort into identifying how to design and
evaluate methods, to support these activities, we in-
troduce a tool for use in method development and
evaluation. Our proposal is scientifically based on
IS Design Theories (Gregor, 2006) and preliminary
practical considerations (Bub, 2020). In their seminal
work, (Gregor and Jones, 2007) have identified eight
components for an IS Design Theory: “Purpose and
Scope”, “Constructs”, “Principles of Form and Func-
tion”, Artifact Mutability”, “Testable Propositions”,
“Justificatory Knowledge”, “Principles of Implemen-
tation”, and “Expository Instantiation”. Their work
targets components of a Design Theory of Design and
Action, meant to be independent of any artifact types
and that require further detailing for concrete imple-
mentation (i.e. to define method artifacts). (Offer-
mann et al., 2010a) have built on their work and have
proposed detailed advice on how to describe methods
in a scientific manner, by leveraging the aforemen-
tioned components to build method Design Theories.
We further took the identified concepts from both re-
search works and built a tool for developing and eval-
uating methods based on them. We abbreviate this
tool as TDEM (A Tool for Developing and Evaluat-
ing Methods) and we illustrate the utility of TDEM
with a thorough analysis.
We structured the overall process as follows. The
concept of TDEM for Design Science in IS is intro-
duced and its components are thoroughly explained.
Subsequently, an analysis of the potential uses of
TDEM is provided and two aspects are investigated:
the analytical and ideational points of view. For the
first one, an example is shown by providing a brief
study (evaluation) on a scientific paper, while for
the latter, a design project of DI in microservices
was developed and is introduced, which discusses
how TDEM participates in the ideation process and
helps establish scientific knowledge in Design Sci-
ence Research. Finally, we show the generalizability
of TDEM on a non-DSR paper in terms of process
evaluation.
2 BACKGROUND
In this section, we closely look into DI, DSR, and
Method Engineering from a theoretical point of view.
2.1 Digital Innovation
Digital technologies transform organizations world-
wide to a great extent. DI can be seen as the appli-
cation of digital technologies to business problems.
(Hevner et al., 2019) explains digital innovation as
the appropriation of digital technologies during the
process of and as the outcome of innovation. DI en-
compasses several directives just like developing a
novel technology or technology strategy in a business
context, adopting, and implementing technology re-
sources, and digitizing work activities within an or-
ganization. Organizations that innovate or transform
come across with several benefits, such as streamlin-
ing business processes by automating and integrating
different technological artifacts and cutting costs as a
result of having digital solutions, having a better re-
turn on investment, and boosting revenue streams.
2.2 Design Science Research
Design Science Research (DSR) has seen a vast
amount of interest in Information Systems (IS), as
an outcome-based research methodology that seeks
to create novel technological artifacts. In its essence,
Design Science (DS) pursues exploring, comprehend-
ing, and rectifying potential components, constructs,
and methods to create and position original artifacts
which are capable of solving an organizational prob-
lem in a natural setting (Baskerville, 2008). DSR
aims to further improve the designed artifacts through
its iterative approach, focusing on their development
and evaluation. The output artifact resulting from
DSR can be software, an algorithm, a mathemati-
cal equation, frameworks, process models, and so on.
DSR outputs can be categorized in constructs, mod-
els, methods, implementations, architectures, design
principles, frameworks, instantiations, and theories as
described by researchers (Chatterjee, 2015; Vaishnavi
and Kuechler, 2004). The all-round learning and es-
tablishment of knowledge, where it is essential to de-
liver innovative artifacts, is at the heart of Design Sci-
ence Research. The actual creation process generally
also covers the improvement of design artifacts and
processes in their setting. As the iterative evaluation
is the main part of DSR, it provides a finer explica-
tion of the problem and provides feedback to improve
the quality and efficacy of the artifact and its design
process.
ICEIS 2022 - 24th International Conference on Enterprise Information Systems
544
2.3 Method Engineering
Initially, (Bergstra et al., 1985) came up with the term
“Method Engineering” (ME), then (van Slooten and
Brinkkemper, 1993) and (Brinkkemper, 1996) fur-
ther supported the idea of ME. ME in IS focuses
on designing, constructing, and evaluating methods,
procedures, tools, and techniques to enhance infor-
mation systems development (Brinkkemper, 1996).
While computer engineering deals with hardware and
firmware aspects, ME focuses on methods, proce-
dures, and techniques in all engineering activities,
since ME can relate to other research areas including
project management, software configuration manage-
ment, software engineering environments, and soft-
ware process modelling (Brinkkemper, 1996). In
addition, methods are the fundament of goal-driven
and tactical actions in various application fields. A
Method incorporates several parts that make the in-
formation systems development process more man-
ageable (Sunyaev et al., 2008). There are several
concepts regarding methods, such as Method Chains
and Alliances (Nilsson, 1999), which are necessary
when performing integration of methods and meth-
ods parts. Among the latter, we have: Method Com-
ponents (Karlsson and Wistrand, 2003), which are de-
fined as being part of a method, consisting of concept,
notation, and process; Method Fragments (Harmsen
et al., 1994), which are building blocks of methods;
Method Chunks (Ralyt
´
e and Rolland, 2001), which
are the combination of the existing process and prod-
uct fragments.
3 A TOOL FOR DEVELOPING
AND EVALUATING METHODS
(TDEM)
(Gregor and Jones, 2007) established a design the-
ory meant for information systems research, consti-
tuted by eight components. These eight components
were taken by (Offermann et al., 2010a) and refined,
characterizing an implementation that was specific for
methods. In fact, it is important to remember that the
original components were meant to define a process
or structure for building Design Theory, which means
they were meant for a much more generic scope. The
aforementioned refinements added important opera-
tional detail in the structure, mostly in two forms: at-
tributes of a component and evaluation criteria.
The attributes give added focus and structure and
help define the elements that are most important for
the definition of a method. A simple way of men-
tally visualizing these, which is particularly close to
people working in Information Systems, would be by
thinking of them like the attributes that define a class
in Object Oriented Programming (OOP), which are
the elements that constitute the internal properties of
a type, thus characterize it (where a class is one of the
eight components by (Gregor and Jones, 2007)). And
again, as with the OOP class attributes, these can also
be either singletons or collections of elements (some
attributes may be specified by a plurality of elements,
depending on the type of method that is being de-
signed). See Figure 1 for the structure of TDEM.
The evaluation criteria constitute an inbound ana-
lytical tool for the process and are expressed in the
form of questions. Each component has its set of
questions (criteria), which allow a segmented study
of the method that is being created, specific for each
component. The questions are meant to be able to ad-
mit both Boolean (of the yes/no type) and complex
answers. This level of detail, considering segmenta-
tion and structure of the questions, is very important,
as it makes it possible to understand if the process
for building a method is being utilized “correctly”, or
at least in a coherent manner. These questions were
selected through a hermeneutical analysis of the com-
ponents and revisited several times (Offermann et al.,
2010a). They are posed in a way that would push the
user to deeply assess the strategy that is being used to
structure the method.
So, to briefly recapitulate, the process to structure
a method should include the use of these eight compo-
nents of Design Theory, tailored to be used within the
area of Design Science Research. Each of these com-
ponents is structured in such a way that they have at-
tributes, to define them, and evaluation criteria, which
help make an in-depth analysis of the method that is
being built.
We gather all this important preparatory work as
input for creating a new tool, TDEM. This can be fun-
damentally explained as a one-pager tool that gives
space to all the augmented versions of the eight com-
ponents of Design Theory, as explained in the previ-
ous paragraphs. This characterization as a tool makes
it easier to understand the usage of the components
and provides an easy way to arrange, also at a graph-
ical level, such an abstract analytical process. Specif-
ically, this can be seen as a tool to perform an in-
depth study of a method and the artefacts that are pro-
duced as a result of the Design Science Research pro-
cess. Further in this section, the eight components are
shown in detail.
Purpose and Scope - This component describes
what a method is meant to achieve and what area of a
research field in Design Science it should cover. Pur-
pose statements should provide information about the
A Tool for Developing and Evaluating Methods (TDEM)
545
A Tool for Developing and Evaluating Methods (TDEM): <Name of the Method>
Purpose and Scope
Project type
Project context
Lifecycle coverage
Activity coverage
Evaluation criteria: Given a problem, is the purpose described in a way to tell if the method solves the problem? Is the product of the method clear? Given a problem context, is the scope described in a way to tell if
the context falls within the scope? Are all relevant dimensions specified within the scope? Are the covered lifecycle phases, roles and activities specified?
Constructs
Method specific constructs, meta-model
Output-specific constructs
Concepts in the application context
Evaluation criteria: Are terms and concepts presented to describe the method? ... describe the resulting
structure? ... describe the application context? ... describe the purpose and scope? ... create the testable
propositions? Are the concepts introduced comprehensively in one location and derived systematically?
Testable Propositions
Utility statement about theory
Optional: truth statements about match between method design and requirements of
method output
Evaluation criteria: Are testable propositions concerning the method utility given? Are the
propositions sufficiently well operationalised to allow testing by other scientists?
Principles of Form and Function
Description of the method (according to the meta-model selected above)
Evaluation criteria: Is the method described extensively enough to be useful and transferable? For each
possible role, is it clear which activities have to be done in what order?
Justificatory Knowledge
Design theories that this method is based on
Theories about the application context
Other aspects of interest
Evaluation Criteria: Are theories presented for key method features not covered by testable
propositions? ... to support the testable propositions? ... regarding the method product?
... regarding the method setting?
Artifact Mutability
Foreseeable changes (which part of the method, method base or method instance, kind of change)
Optional: support for change offered by method design paradigm (i.e. situational, method rationale)
Evaluation criteria: Is the method described extensively enough to understand the functioning of each
component? Are conditions for changes described? Does it become clear, which parts change and what they
change into? Are method tailoring mechanisms defined?
Principles of Implementation
Tailoring/assembly advice
Advice regarding introduction into real-life setting
Evaluation criteria: Is the tailoring/assembly advice comprehensive with regard to the combinatorial
possibility? Does the advice for introducing the method cover
different settings within the scope, or is at least clear for which it applies?
Expository Instantiation
Example of an instantiated method
Report on a real-life case study in which an instantiated method has been enacted
Evaluation criteria: Does the example cover a case within the scope and is it covering the main concepts of the method? Is the example specific enough to be illustrative but not too idiosyncratic to be
incomprehensible?
Figure 1: TDEM based on (Offermann et al., 2010a; Gregor and Jones, 2007).
ICEIS 2022 - 24th International Conference on Enterprise Information Systems
546
type of output that the usage of a method should yield,
the properties of the output, and some statements re-
garding the method itself. On top of that, the scope of
the method is also described from an external and an
internal point of view.
Scientific Role: Purpose and scope give the basis
for the scientific significance of a theory or method, as
they provide a clear vision of the end goal for which
they are being developed, and describe what the re-
search area should “look like”. This strongly con-
tributes to providing the generalisability that should
be typical of a proper method or theory. When the
scope is well defined, it is easier to generalize the va-
lidity of the scientific process.
Constructs - Constructs are properties that can
be identified either from the method itself, from the
structure of its output, or the context of the enactment
of such method. One of the concepts that concretely
participate in the constructs is that of Method itself, as
methods can be generated from other method models
or meta-models. Other constructs might come from
product classes or justificatory knowledge as well.
Scientific Role: The constructs do not provide in-
formation that would directly help the evaluation of a
theory or method, but they do help in setting the fun-
dament for the description of all the other components
of the process.
Principles of Form and Function: This compo-
nent describes what the abstract architecture of the
artefact should look like, especially when the arte-
fact is a system. In case the artefact is a method,
then there is no specification for different architec-
tural models, but situational context is considered. A
general method is described, which can be seen as
a set of individual solutions. The principles of form
and function establish a strict relation between the el-
ements of the method and the meta-model that defined
the internal structures. The way such elements are de-
scribed must comply with the representation coming
from the metamodel and having justificatory knowl-
edge would be an added point.
Scientific Role: This component is strictly neces-
sary for a method to be useful. In fact, it provides
the information that allows the understanding of the
purpose of the method, which is necessary for its co-
herent use.
Artefact Mutability: Mutability during a re-
search process can happen both on the design stage of
the method, or on a particular instance of the method
itself. For methods, specifically, two types of mutabil-
ity are identifiable. The first kind identifies the muta-
bility, changes on the method itself, that the designer
can predict. The second kind identifies the mutability
that can be expected and is described in the original
method design, but happens during the instantiation
of the method.
Scientific Role: This is another component that
greatly helps increase the possibility of generalization
of a method, such as the principles of form and func-
tion. Understanding what changes (and in what man-
ner) a method can undergo also helps to understand
whether it is possible to transfer it across research do-
mains.
Testable Propositions: Here it is possible to
make an analysis and have a complete view of what
are the statements made about the method or arte-
fact of the process in general, that can be measured
and tested. For simplicity, an overview of what these
propositions are can be obtained using the following
three questions: What are the statements about? What
is asserted by the statement, utility or truth? How
complete should the collection of testable proposi-
tions be?
Scientific Role: Testable propositions have the
purpose of supporting the validity of the method, es-
pecially considering its utility statement. As in (Of-
fermann et al., 2010a), “. . . a method is valid if it is
useful in respect to its purpose.”. Of course, an ob-
jective and testable analysis can only be made if the
hypotheses are made in such a way that they can be
reproduced by someone else.
Justificatory Knowledge: This component is
meant to clearly state all the possible theories sup-
porting the method, the product, and the context in
which the method is being applied. As in (Gregor,
2009), there are different kinds of theory that can be
selected, based on specific situations.
Scientific role: Justificatory knowledge is meant
to support the possibility of transferring a method into
different research domains and to prove its validity. It
provides elements that can be used to repeat a proof
of validity for the method in different contexts.
Principles of Implementation: This component
helps explore what the necessary steps are during the
implementation phase, which usually imply adapting
a method to the specific scope (situation or organiza-
tion) it is being used in. This is also known as tailor-
ing process.
Scientific role: These principles help again sup-
port the statements about the utility of a method and
define how such a method is transferable into different
scopes.
Expository Instantiation: This component is
meant to show a real-life application of the method
that is being designed. This can be represented either
as a description of the method applied in a specific
context or as information about the execution of the
method.
A Tool for Developing and Evaluating Methods (TDEM)
547
Scientific role: These principles, as the previous
one, also support transferability and generalization,
but provide more concrete information about the re-
sults coming from an actual tentative instantiation and
use of the method.
4 USAGE DUALITY OF TDEM
In this section, we explore the possibilities that
TDEM offers in Design Science. Specifically, two
distinct modes are taken into consideration, which we
named the analytical (evaluation) and the ideational
(development) processes. Both processes are ex-
plored and illustrated by giving a specific example for
each.
4.1 TDEM as an Analytical Tool
From what concerns the analytical point of view,
TDEM proved to be a tool that is effective in the
evaluation of artifacts of Design Science, specifically
methods. In an earlier work, we explored a poten-
tial usage of TDEM as a framework component for
process evaluation (Ogbuachi et al., 2021). In this
subsection, the evaluation process is shown through
the analysis of a method for the engineering of DI by
(Bub, 2018), which describes design activities, design
outcomes, and the necessary roles for organizations
that deliver innovation. The essential construct is the
process model, which shows the sequence of design
activities, essentially concentrating on method engi-
neering.
The analysis conducted on the approach using
TDEM, revealed several interesting matters, which
are illustrated here for each of the eight components
of Design Science Research.
Purpose and Scope: The research was carried out
within the scope of method engineering for digital in-
novation. The purpose of the paper is to develop a
method for the engineering of DI and integrating a
DSR process into the process model.
Constructs: Fundamental constructs can be eas-
ily identified from the method, such as an enhanced
meta-model for situational method engineering, a
stage-gate-oriented idea-to-launch process model, a
stage-gate-oriented process model integrated with a
DSR process, and a combined process model.
Testable Propositions: The utility of the pro-
posed method is shown in the paper, through a de-
scription of how the method is applied in a university-
industry co-innovation lab. The research process and
implementation are also described.
Principles of Form and Function: The process
model is clearly described in the paper. Design activ-
ities are in sequential order and they have to be per-
formed “from left to right”, completing each stage one
after the other.
Justificatory Knowledge: Method design theo-
ries are presented, such as theories about the applica-
tion context (including innovation management) and
theories about DSR.
Artifact Mutability: After application in the
original method setting, situational tailoring for
project types (and similar adaptations) is encouraged
for other settings, since the number of gates for inno-
vation processes can be adjusted to specific needs of
the respective enterprise.
Principles of Implementation: The method re-
quires an already implemented idea-to-launch pro-
cess, or willingness to implement it at the enterprise.
Expository Instantiation: The paper is illustrat-
ing a case study in which an instantiated method has
been enacted.
4.2 TDEM as an Ideation Tool
TDEM worked well also as a creational tool for De-
sign Science, thus representing a sort of duality if
compared to the analytical approach that was de-
scribed in the previous section. This new point of
view is illustrated here, by a study that was conducted
in a DI internal project, from the structuring of its fun-
damental parts to the verification of its completion.
The project aimed at representing the subdivision of
a simple monolithic application into a structure that
tried to replicate the microservices architecture.
The project is based on the Stencil Parallel Pat-
tern. This pattern suits specifically areas like image
processing. In Stencil, every step of computation is
done starting from a memory address (or an array in-
dex) and operating on offsets from that address. The
result is then saved to an output array, specifically to
an index that corresponds to the starting memory lo-
cation for the input. The image processing program
consists of a set of algorithms that run consecutively
on a given image dataset, using this pattern.
The pipeline is constituted by execution units
(functions) that are executed in sequence. Given
the structure of the application, the plan consisted in
analysing what the functional units were, that could
be defined independently from all the others, and
transforming those into actual applications that would
run as small services.
As mentioned earlier, in this specific case the
functional units are the functions that implement the
image processing algorithms. Those functions are
ICEIS 2022 - 24th International Conference on Enterprise Information Systems
548
therefore encapsulated into fine-grained services that
are able to retrieve an image as an input, process it
by applying one specific filter to the image, and give
back an output as a result, having only a few mi-
nor requirements on the input image. Each service
is able to operate independently from who sends the
input, thus fulfilling one of the core characteristics of
a microservices-based architecture.
It is important to remember, though, that in this
specific example the functionality of the full mono-
lithic application should also be preserved, after the
transition to a microservices-based architecture. De-
spite the structure being very different now from an
architectural point of view, the original functionality
could still be preserved by keeping the temporal point
of view unchanged. By this, we mean that the orig-
inal order of operations was sequential in the origi-
nal monolithic application, as it is typical of the gen-
eral imperative programming approach, but that same
order could be replicated in the new architecture as
well, by enforcing the sequentiality of the operations
through a structured series of service calls and data
transfers. The intended sequence then looks as shown
in Figure 2.
Image Upload
Edge
ASCII
Resize
Output
Service
Figure 2: Sequential order of operations.
Therefore, the initial image is now uploaded to an
endpoint of the service, which acts as the only com-
ponent the user is allowed to directly interact with.
This endpoint should then send the image that was
uploaded by the user to the first step of the image fil-
tering process, which is the Grayscale function. After
the execution is complete, the output gray image is
then sent to the next component, Resize, and so on,
until the last one, ASCII text-based rendering, com-
pletes its elaboration and returns the image as a file
to the user, as a response from the same aforemen-
tioned service endpoint. In our set-up, each func-
tion provides an input to the next one, completing
the pipeline. By restructuring the application in a
microservice architecture, we explored the possibility
of enforcing synchronous behavior (originating from
the monolithic application) in an environment (in this
case, the network) that is inherently asynchronous.
Having a guideline to define the development process,
helped us identify the potential issues (such as com-
munication issues, segregation of data, scalability of
workload, etc.) that could arise when building a ser-
vice that strongly relies on asynchronous interactions.
As before, an analysis based on the eight compo-
nents of TDEM is provided, with the different focus
presented in this section.
Purpose and Scope: The purpose of the project
can be explained easily as the creation of a method to
define microservices from an existing monolithic ap-
plication. Trying to identify the context, the best op-
tion is software engineering, specifically image pro-
cessing and networking. The definition of a lifecycle
coverage process was not entirely defined, as it was
partly out of the scope of the process (only the design
phase is considered). While structuring the method,
we described the purpose in a way that can tell if the
method solves the proposed problem, and it does, es-
pecially the problem of splitting a pre-existing appli-
cation into smaller parts. The approach we adopted is
general enough to be adaptable to the specific prob-
lem.
Constructs: The construct that comes as an out-
put of this project is a modelling approach to the def-
inition of a microservice from a pre-existing applica-
tion. Thus, the main construct is a purely theoretical
high-level approach. The application context would
be Autonomous processing services and Microservice
candidates. We described the basic concepts needed
to present the method, the resulting structure, and the
application context.
Testable Propositions: As a utility statement, we
agreed that this approach should allow an easier def-
inition for the subdivision of components when cre-
ating a new microservice system by splitting existing
systems and, in this specific case, the method tries to
fulfil completely the problem it was used for. The
work that has been done should be enough informa-
tion to allow at least the unit testing of components,
in most cases, testable propositions about the method
A Tool for Developing and Evaluating Methods (TDEM)
549
utility are also given.
Principles of Form and Function: The method
described in this paper is meant to be a starting point
for the creation of microservices based on the seg-
mentation of an application. The method we pro-
duced was extensively described, and the amount of
information provided is enough to understand that it
is adaptable. For each role in the project, the neces-
sary activities and their order were clearly specified.
In fact, the method is described with step-by-step clar-
ity (keeping the descriptions very high-level).
Justificatory Knowledge: The proposed method
is based on Domain Driven Design and Microser-
vices. It is still general, but the instantiation is im-
age processing as a domain. The main theory is that
monolithic, computationally expensive applications
can be split into smaller parts, to which we added at-
tention to the fact that the knowledge base of the ap-
plications can be singular or plural, both before and
after the splitting process, and both approaches could
actually be beneficial (indirectly), especially while
considering efficiency.
Artifact Mutability: Considering the foreseeable
changes, we estimated that, because of the generic
nature of this method, which is also based on very
simple patterns, such changes are minor and would
not extensively change the defined structure. More
specifically, the foreseeable changes are mostly based
on small adaptations to the application scopes. The
only method tailoring mechanism that was defined
was for the given example, trying to put it in a more
generic context.
Principles of Implementation: About the tailor-
ing or assembly advice we provided, most of it is
about the definition of “autonomous entities”, and
mostly about the use in the suggested scope, with-
out excluding the consideration of several other areas.
Particular attention has been put in making it clear for
the proposed scope.
Expository Instantiation: The method is instan-
tiated only within the provided example. There is no
report of the sort of real-life case study in the paper.
Mostly the use we made was for laboratory experi-
ments in a controlled environment. The provided ex-
ample defined the operational scope and it is specific
enough to be illustrative about the applied method.
5 GENERALIZABILITY OF
TDEM
TDEM can also be used as a tool for evaluating pro-
cesses, which are not being necessarily developed us-
ing DSR paradigm. We show this through the analysis
of a design approach structured by O’Connor, Elger,
and Clarke in (O’Connor et al., 2016), which is meant
to help enhance development (especially in IT-related
fields) by taking into consideration the so-called situ-
ational context of a project.
As described in the paper, situational context can
be of utmost importance when designing a process.
It can be defined as the set of elements that may
have a direct influence on a project and its outcome,
from the Enterprise level to the actual development
phases. In particular, (O’Connor et al., 2016) show
that the elements identified as components of this con-
text are the personnel, requirements, application (the
software project itself), technology, organization, op-
eration, management, and business. The first four el-
ements are the ones having the highest impact on the
design stages. The personnel is identified as the non-
managerial characters participating first-hand in the
development of the software application (the software
engineers), while the requirements are all the imple-
mentation needs and rules as defined in the software
engineering standard ISO 12207 (Singh, 1996), which
can be characterized by the commissioner or the de-
velopment team. The technology that is available in
the company affects directly the capabilities of the
aforementioned personnel and the required develop-
ment time.
The analysis conducted on this approach using
TDEM, revealed several interesting matters, which
are illustrated here for each of the eight components.
Purpose and Scope: The type, context, life-
cycle, and activity coverage of the project could
be explained fairly easily. The project was car-
ried out within the scope of software engineering for
microservices, and the lifecycle management corre-
sponds to what is often seen in the paradigm known
as “agile development”. The purpose of the paper was
not to show a method, specifically, but rather showing
through a concrete example how situational context
affects project management.
Constructs: The “products” of the project can be
plainly pointed out. Specifically, the output-specific
construct here is a framework for situational analy-
sis which is fundamental to understand the necessary
adaptation steps required for general methods, when
applying them to any concrete project. One notable
remark is that the use of terms and concepts in the pa-
per has been made in such a way that allows the entire
framework to be easily understandable, to the point
that an entire section was dedicated to the definition
of concepts and their utility.
Testable Propositions: The utility of the pro-
posed theory is illustrated across the paper and can
be summarized with one proposition: the suggested
ICEIS 2022 - 24th International Conference on Enterprise Information Systems
550
analysis should provide a better view of how to anal-
yse the effects of context on development. Although
there are no specifically defined measures, the infor-
mation provided by the authors includes the results of
actual tests conducted in an enterprise.
Principles of Form and Function: This paper
provides a framework to analyze the environment in
which a microservices-based software system is be-
ing developed, to easily adapt generic methods to the
needs of specific cases. The suggested approach is de-
scribed very extensively and the amount of informa-
tion provided is enough to understand how the adapta-
tion process works. The framework is described with
precise detail and great clarity (as mentioned earlier,
a use case example is also shown).
Justificatory Knowledge: This component helps
identify some important properties regarding the
background knowledge needed to justify the theories
that are discussed in the paper. Starting from the de-
sign theories, which are based on the agile and lean
development principles and the microservices archi-
tecture, it is understandable that the majorly observed
scope is that of web applications and full-stack opera-
tions. Other aspects of interest are the precision of the
framework and the possibility of extending it. Theo-
ries supporting the testable propositions and regarding
the product of the research are also presented.
Artifact Mutability: Foreseeable changes are to
be taken into consideration. In fact, because of the
generic nature of this framework, which is also based
on very simple discernment strategies, it is correct
to say that the foreseeable changes are minor and
wouldn’t extensively change the defined structure of
the framework itself. It can also be added that the
foreseeable changes are mostly situational, based on
adaptations to the application scopes, and they are
supposed to be like that. The framework resulting
from this work is described extensively to allow its
use as a method component and, since change is con-
sidered to be a fundamental factor through the whole
process, tailoring mechanisms have been defined for
the given example and for more generic contexts.
Principles of Implementation: It would not be
wrong to affirm that the paper is meant to empha-
size and prove the importance of tailoring as a prin-
ciple. Some advice regarding the introduction of the
approach into real-life settings is also included, to-
gether with an extensive example conducted in a real
company. Generally speaking, the advice is mostly
clear for the setting in which it is applied, but also
helps a general understanding for an adaptation into
different scopes.
Expository Instantiation: An expository instan-
tiation is not missing, as the approach has been instan-
tiated in the provided example, which is practically an
enactment within a real-life case study. The example
covers the main concepts illustrated by the proposed
approach and tackles the issues of a case that fully
belongs to the predefined scope.
6 CONCLUSION AND FUTURE
WORK
There exists a huge amount of interest in DI re-
search from IS scholars. As of our previous research
(Huseynli et al., 2021), we found out that there is
still scarcity in terms of using Design Science Re-
search (DSR) for the engineering of DI, especially
from a method perspective, and a possible integration
of both fields. While getting motivation for develop-
ing a method for the engineering of DI, using and in-
tegrating DSR, we came across early research works
- (Gregor and Jones, 2007) and (Offermann et al.,
2010a) - and based on the concepts illustrated in both
works, we proposed a tool for developing and evaluat-
ing methods which we called TDEM. We showed the
utility of TDEM as an evaluation tool by analysing
a paper from the DI-DSR research area and also il-
lustrated its generalizability capabilities in non-DSR
works, on a paper based on continuous software engi-
neering from a microservices perspective. TDEM is
also useful as a development tool for methods, and we
described this through a DI internal project. We be-
lieve that TDEM is a useful contribution to the body
of knowledge for information systems and shall be
utilized while developing and evaluating methods.
Our work focuses on the very important artifact
type “method”. (March and Smith, 1995) identify
four artifact types from design science: constructs,
models, methods, and instantiations. Moreover, (Of-
fermann et al., 2010b) identified eight different arti-
fact types from an extensive literature review of de-
sign science publications: Systems Design, Method,
Language/Notation, Algorithm, Guideline, Require-
ments, Pattern, Metric. In future work, we would like
to develop similar tools also for artifact types other
than methods, possibly starting from system design.
REFERENCES
Baskerville, R. (2008). What design science is not. Euro-
pean Journal of Information Systems, 17:441–443.
Baumol, W. J. (2002). The free-market innovation machine
- analyzing the growth miracle of capitalism. Journal
of Economics, 82:93–97.
Bergstra, J., Jonkers, H., and Obbink, J. (1985). A software
A Tool for Developing and Evaluating Methods (TDEM)
551
development model for method engineering. Esprit,
84:84–94.
Berntsson-Svensson, R. and Taghavianfar, M. (2015). Se-
lecting creativity techniques for creative require-
ments: An evaluation of four techniques using cre-
ativity workshops. 2015 IEEE 23rd International Re-
quirements Engineering Conference (RE), pages 66–
75.
Brinkkemper, S. (1996). Method engineering: engineer-
ing of information systems development methods and
tools. Inf. Softw. Technol., 38:275–280.
Bub, U. (2018). Towards an integrated method for the
engineering of digital innovation and design science
research. In European Conference on Advances in
Databases and Information Systems, pages 327–338.
Springer.
Bub, U. (2020). Methodenvergleich und methodenentwick-
lung mit einem canvas. Wirtschaftsinformatik & Man-
agement, 12:146–157.
Bucher, T. and Winter, R. (2008). Dissemination and im-
portance of the ”method” artifact in the context of
design research for information systems. In Vaish-
navi, V. and Baskerville, R., editors, Proceedings of
the Third International Conference on Design Sci-
ence Research in Information Systems and Technology
(DESRIST 2008), pages 39–59, Atlanta, GA. Georgia
State University.
Chatterjee, S. (2015). Writing my next design science re-
search master-piece: But how do i make a theoretical
contribution to dsr? In ECIS 2015 Completed Re-
search Papers.
Gregor, S. (2009). Building theory in the sciences of the
artificial. In Proceedings of the 4th International Con-
ference on Design Science Research in Information
Systems and Technology, DESRIST ’09, New York,
NY, USA. Association for Computing Machinery.
Gregor, S. D. (2006). The nature of theory in information
systems. MIS Q., 30:611–642.
Gregor, S. D. and Jones, D. (2007). The anatomy of a design
theory. J. Assoc. Inf. Syst., 8:19.
Harmsen, F., Brinkkemper, S., and Oei, J. L. H. (1994). Sit-
uational method engineering for informational system
project approaches. In Methods and Associated Tools
for the Information Systems Life Cycle.
Hevner, A. R. and Gregor, S. D. (2020). Envisioning en-
trepreneurship and digital innovation through a design
science research lens: A matrix approach. Information
& Management, page 103350.
Hevner, A. R., March, S. T., Park, J., and Ram, S. (2004).
Design science in information systems research. MIS
Q., 28:75–105.
Hevner, A. R., vom Brocke, J., and Maedche, A. (2019).
Roles of digital innovation in design science research.
Business & Information Systems Engineering, 61:3–8.
Huseynli, M., Podder, I., Ogbuachi, M. C., and Bub, U.
(2021). Digital innovation and design science re-
search: The state-of-the-art review and integrated in-
novation strategies framework. In Proceedings of the
10th Jubilee Interdisciplinary Doctoral Conference
2021, pages 355 – 368.
Karlsson, F. and Wistrand, K. (2003). Method components
- extending the vision. In Proceedings of the 26th In-
formation Systems Research Seminar in Scandinavia
(IRIS 26), Haikko Manor, Finland.
March, S. T. and Smith, G. F. (1995). Design and natural
science research on information technology. Decis.
Support Syst., 15:251–266.
Nilsson, A. G. (1999). The business developer’s toolbox:
Chains and alliances between established methods. In
Perspectives on Business Modelling, pages 217–241.
Springer.
O’Connor, R. V., Elger, P., and Clarke, P. M. (2016). Ex-
ploring the impact of situational context a case
study of a software development process for a mi-
croservices architecture. 2016 IEEE/ACM Interna-
tional Conference on Software and System Processes
(ICSSP), pages 6–10.
Offermann, P., Blom, S., Levina, O., and Bub, U. (2010a).
Vorschlag f
¨
ur komponenten von methodendesignthe-
orien. WIRTSCHAFTSINFORMATIK, 52:287–297.
Offermann, P., Blom, S., Sch
¨
onherr, M., and Bub, U.
(2010b). Artifact types in information systems design
science - a literature review. In DESRIST.
Ogbuachi, M. C., Podder, I., Bub, U., and Huseynli, M.
(2021). A framework for quantifiable process im-
provement through method fragments in situational
method engineering. In 2021 3rd International Con-
ference on Advanced Information Science and System
(AISS 2021), AISS 2021, New York, NY, USA. Asso-
ciation for Computing Machinery.
Peffers, K., Tuunanen, T., Rothenberger, M. A., and Chat-
terjee, S. (2008). A design science research method-
ology for information systems research. Journal of
Management Information Systems, 24:45 – 77.
Ralyt
´
e, J. and Rolland, C. (2001). An approach for method
reengineering. In International Conference on Con-
ceptual Modeling, pages 471–484. Springer.
Singh, R. (1996). International standard iso/iec 12207 soft-
ware life cycle processes. Software Process: Improve-
ment and Practice, 2:35–50.
Sunyaev, A., Hansen, M., and Krcmar, H. (2008). Method
engineering: A formal description. In ISD.
Vaishnavi, V. K. and Kuechler, W. L. (2004). Design Sci-
ence Research in Information Systems. Ais, pages 1–
45.
van Slooten, K. and Brinkkemper, S. (1993). A method en-
gineering approach to information systems develop-
ment. In Information System Development Process.
ICEIS 2022 - 24th International Conference on Enterprise Information Systems
552