Digital Value Dependency Framework for Digital Transformation
Suthamathy Sathananthan
1
, Dennis Gamrad
1
and Johanna Myrzik
2
1
Digital Solutions & Operations, Evonik Technology & Infrastructure GmbH, Essen, Germany
2
Institute for Automation Technology, University Bremen, Bremen, Germany
Keywords: Digital Transformation, Business Value, Dependency, Digitalization, Benefits, Business Model, Framework,
Value Network.
Abstract: Knowing there is considerable value in digitalization, enterprises have started to transform their operations
utilizing digital technologies. However, current methods used in estimating benefits are methods typically
used in capital budgeting projects which do not consider the value interdependencies or uncertainties from
digitalization into account. Therefore, a standardized yet comprehensive framework and a mathematical
model have been developed to estimate and measure potential from digitalization. The framework and model
together were applied in an industrial digital project where results show overall value of the project based on
economic and qualitatively measured impacts, and value contribution of transformational elements such as
technologies and organizational changes. The results have been used to form value networks which
demonstrate shared values between multiple digital projects with respect to digital capabilities. These results
bring transparency across projects for informed decision making and support in data-driven business model
innovation.
1 INTRODUCTION
According to World Economic Forum’s 2016 report
on digital transformation across industries, upward of
$100 Trillion potential value was estimated for
industries and society combined for the next ten years
(World Economic Forum, 2016). Such large
potentials are believed on a very abstract level to be
conceivable, but many industries have not tapped into
the full potential yet. Due to the uncertainties that
come with fast changing technologies, industries,
especially non-digital industries, which earn majority
of their profit through selling non-digital products or
services, are very slow in actively adapting to these
changes.
Although many companies understand that they
must move quickly to stay in the market and be
competitive, the volatile nature of digital technology
landscape makes it hard for them to decide if certain
technologies can bring the potential they promise to
deliver. For example, Industrial-Internet-of-Things
(IIoT) platforms play an important role in
automatization of production plants and in enabling
flexible digitalization across businesses. In order to
adapt to such platforms, brown-field plants must
change their system landscape completely which
requires a lot of commitment and trust financially.
However, the value potential from such platforms are
not completely transparent so are the costs and risks
associated with it. Therefore, only small-scale
projects are being done to test these technologies so
that the financial commitment is also low.
In order to estimate value from digital projects,
many companies rely on market research reports
which can be very different from industry to industry
therefore susceptible to over or under estimation. The
lack of common and concrete methodology to
estimate potentials makes it even harder for
companies to be agile.
The value from digital products or solutions are
attained through a complex network of value
interdependencies, and these networks must be
described in a formal way to be able to understand
potentials from different projects. Therefore, value
estimation in the digital era should look beyond single
project and give the possibility to see value networks
among different projects such that benefits and costs
sharing can be planned and managed. There exists no
standard methodology to assist business developers
and project managers to formalize particularly
potentials from digitalization in a comprehensive
manner.
Sathananthan, S., Gamrad, D. and Myrzik, J.
Digital Value Dependency Framework for Digital Transformation.
DOI: 10.5220/0009316106430655
In Proceedings of the 22nd International Conference on Enterprise Information Systems (ICEIS 2020) - Volume 2, pages 643-655
ISBN: 978-989-758-423-7
Copyright
c
2020 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
643
As described by Okhrimenko, 2019, digitalization
is a complex process where scientific approaches in
terms generalization and systemization are needed to
simplify and understand the process in many
industries and public sectors. As part of this study a
standardized digital value dependency framework has
been developed based on proven methodologies such
as benefit dependency mapping and empirical
findings of digitalization projects where digital
capabilities and organizational changes were
generalized. The framework and the model presented
allow the estimation of benefits from digital projects
in a systematic way and the network results that are
digitally stored therefore giving the possibility to
compare and understand value networks between
different digital projects.
In the following sections, background on existing
methods has been elaborated. Then the redefined
BDN for digitalization called, Digital Value
Dependency framework is introduced where each
column in the framework is described in detail, then
the procedure for using the framework is explained
which is followed by the introduction to mathematical
model. Finally, practical usage of the framework in
industrial setting has been demonstrated with an
example.
2 BACKGROUND
Benefit management has a strategic importance for
companies where the ability to create above-normal
value can lead organizations to sustainable
competitive advantage (Gomes &
Romão, 2014).
Systematically planning and managing benefits
especially from new technologies can be a very
demanding task. In order to estimate benefits in
general, organizations use different methods starting
from referring to market research reports to using
tools or techniques. One such tool is Benefit
Dependency Network (BDN) that was first
introduced in 1996 by Ward et al. in order to bring
business objectives, benefits and Information System
(IS)/Information Technology (IT) changes to realize
those benefits together (Ward et al., 1996). Although
there has many adaptation since, the most used BDN
has five steps: IS/IT enablers, enabling changes,
business changes, business benefits and investment
objectives (Ward & Daniel, 2006).
The framework is completed either from right to
left, starting from investment objectives to IS/IT
enablers or from left to right (
Chaves & Pedron, 2015).
The most important aspect of this tool is that it brings
enablers and changes together which guide in
estimating the benefits from those changes. For
example, the organization will operate the same way
when there are no changes and no additional benefits
could be realized, but when changes are introduced or
identified then estimating benefits from those
changes are unpretentious. While utilizing this
mapping tool, project owners brainstorm enablers and
changes specific to their project and put them as the
elements of the steps.
Although this activity supports the planning of
individual projects, it lacks in two ways to make use
of it in digital transformational projects. One is that
every user defines their own enablers, changes and
benefits within the tool to understand their potential
outcomes, therefore results of the mappings from
many projects cannot be compared for strategic
planning of digital transformation because each
mapping’s entries will be completely different from
the other. Second is that although the steps help
logically identify benefits, the network outcome itself
are not stored or measured for better utilization of the
results. In digital transformation however, digital
projects show value interdependencies where benefits
and costs are shared among digital projects and/or
users of the end solution. Also, certain technologies,
capabilities or organizational changes are more
important than others in realizing the value, but
currently there are no measurement system integrated
into the BDN to understand and assess value
contribution of these elements of a resulting network.
Being able to measure how much value each
technology, capability or organizational change
contribute to the projects can help in prioritizing and
managing them effectively. In order to easily
compare, the dependency framework used in digital
projects should have some consistency in its entries,
and the resulting networks should have a
measurement system and model so that the network
elements can be measured and analysed.
Therefore, in this work these two points are
addressed by redefining benefit dependency network
suitable for digital transformation projects through
generalizing elements in the framework that are
specific for digitalization, and a web-tool and a
mathematical model have been developed to store the
estimated dependencies and then the value
contributions of elements are analysed through the
model. This framework not only helps individual
projects to estimate benefits logically, but it also helps
in strategic planning of digital transformation where
resulting value networks from multiple projects can
be compared and analysed.
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
644
3 DIGITAL VALUE
DEPENDENCY FRAMEWORK
The Digital Value Dependency (DigVD) framework
follows traditional BDN and has five steps, starting
from digital technologies, digital capabilities,
organizational changes, specific positive and negative
impacts to high-level business goals/values. The
entries or rows under each of these steps are called
elements. As depicted from the screenshot of the web-
tool in Figure 1, the elements of digital capabilities,
organizational changes and business values are
generalized and fixed, and this as described earlier the
main difference from traditional BDN to DigVD
framework. Elements of digital technologies and
specific impacts are defined as per the project during
the value estimation process.
The DigVD framework is built as a digital web-
tool and every element is given a unique ID. In order
to standardize these elements, digitalization-based
taxonomy findings, and empirical findings from
multiple digital projects in house at the chemical
company were considered. These were then evaluated
with case-studies from other industries. The elements
considered within each step has been described in
detail in the following subsections.
3.1 Digital Technologies
Digital technologies are defined as technologies that
enable digitalization for the intended group of users.
As defined in many papers and reports of recent
times, smart sensors and edge devices, protocols,
network, human-machine interface technologies such
as AR/VR, cloud and platform-based technologies,
big data solutions, machine learning and artificial
intelligence are all considered important technologies
for digitalization (Lee et al., 2015,
Probst, 2018).
Those technologies that are purchased, established
and/or considered necessary to successfully enable
the digital project are the elements in this first step.
In addition to digital technologies in the market,
any internal digital project which becomes a required
or enabling technology for another digital project can
be entered. Furthermore, protocols or connections to
existing IT systems and specific Application
Programming Interfaces (API) are also considered
within this column as enabling technologies along with
existing data sources required to enable the project. For
example, specific API’s to Enterprise Resource
Management systems (ERP) or Process
Instrumentation Measurement Systems (PIMS) which
already contain the necessary data for additional digital
projects/products are considered. This column is kept
open for the framework user, therefore the user can
either enter new technologies or chose suggested
technologies from a drop-down list which shows
technologies entered by previous users of the
framework. Each technology is assigned a unique ID
with DT (Digital Technology) as its prefix.
3.2 Digital Capabilities
The abilities of a digital product user or of the
company that are improved or developed using digital
technologies are defined as digital capabilities.
Elements under this second step are fixed on the
framework. Many studies have come up with
Figure 1: Screenshot of the web-tool for Digital Value Dependency Framework.
Digital Value Dependency Framework for Digital Transformation
645
taxonomies or classifications of digital capabilities
based on current and commonly considered digital
technologies (
Freitas Junior et al., 2016; Lenka et al.,
2017; Cenamor et al., 2017; Rizk et al., 2018). These
classifications from various findings were
consolidated, and taxonomy-based research process
was applied to classify them into appropriate
categories and verified with their fit to existing
projects.
According to this analysis and as illustrated on
Figure 1, Sensing & capturing, Connectivity of data
& information, Data contextualization & analytics,
Information sharing & collaboration, Visualization &
decisiveness, and Monitoring & control are
capabilities distinctively defined as abilities
developed or improved with current digital
technologies. ‘Sensing & capturing’ is the ability to
sense and capture the digital information where
technologies like sensors and edge devices are used
in realizing this capability. Secondly, ‘connectivity
is defined as the ability to connect various data and
information where the source of the data, interfaces,
platforms, and cloud technologies are all considered
in enabling this capability. ‘Data contextualization &
analytics’ is defined as the capability to make sense
of the connected data and information by structuring
and analysing them, with the use of digital
technologies such as data engineering tools, data
modelling, databases, basic analytics and big data
analytics.
The ‘Information sharing & collaboration’ is the
ability to share information enabling the ability to
work together using technologies such as mail clients,
notifications, social media or other network related
technologies specifically established for the product
in consideration. Next is ‘Visualization & decision-
making support’ and this capability allows the digital
product user to visualize the data, information and
knowledge with the use of dashboard technologies
and events-based suggestions and alerts are integrated
where decision making ability of the user has been
made easy. Finally, the ‘Monitoring & control’
capability is defined as the ability to monitor and
directly control the outcome through the product itself
and alter the outcome, where the control function can
be either fully automatically or semi-automatically
embedded into product. These predefined elements
on the digital capability step are given unique ID’s
with DC as their prefixes.
3.3 Organizational Changes
Organizational changes are listed on the third step,
where elements are fixed and categorized based on
empirical findings from in house digital projects,
which were categorized based on taxonomy research
methods and verified with findings in other studies
(
Mikusz, 2014; Cenamor et al., 2017; Gerber et al.,
2017;
Westermann & Dumitrescu, 2018). What changes
will be influenced by the capabilities of the product
to the organization on the frontend and on the
backend during and after the implementation are
defined as the elements as shown in
Figure 1.
From these findings, it was evident that processes
or workflows in an organization are changed
primarily by digitalization in two ways. Firstly,
certain processes are changed due to user’s direct
interaction with data on the physical level using
digital devices. Within this interaction-based changes
there are three elements. 1. Changes in terms of
consuming data and information, which means the
user can now consume data from different sources
easily therefore the current frontend process has to be
changed. 2. Changes with regards to feeding in raw
data and information. This means, user is now able to
input data directly, for example instead of writing on
a paper the user inputs the data into a digitally
readable format which changes the current workflow.
3. Changes related to completing a workflow that
does not have to do with consuming or feeding in
data, but conducting the overall process with the use
of digital devices. This means, with a digital
tool/application extended work processes are
changed and/or simplified.
Secondly, within the process related changes, the
new knowledge and insights that are derived from
raw data change the existing processes, where those
insights from data help the user on an intellectual
level. The elements within this category are separated
by the type of insights possible with data. As
mentioned in number of taxonomy-based studies,
there are three types of insights possible and they are
descriptive, predictive and prescriptive (
Rizk et al.,
2018)
. Descriptive insights are related to historical
knowledge and reveals to the user what has happened
in the past, changing the workflow to accommodate
this new knowledge. Predictive insights are
knowledge related to what will happen in the future,
which can also change the existing process to
accommodate the new predictive knowledge. Lastly,
the prescriptive insights are knowledge about what
should be done about something where it can also
change the ways the current workflow is completed.
When these types of insights are given to the user the
current front-end processes must account for some
changes and therefore considered within the frontend
process related changes.
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
646
The second part of organizational changes are the
backend architectural changes within its current
infrastructure and system level which act as the
backbone for frontend processes to take place. Under
this category, three levels are identified. Network
based changes, hardware-based changes and finally
software-based changes. Within network-based
changes, any changes in firewalls, proxy, and WLAN
are considered together as one element. Under the
hardware-based changes category, it has been split
into three parts based on data gathering, data handling
and data storing hardware. Changes in data gathering
hardware components would be sensors and
actuators, changes in data handling devices would be
mobile devices, and changes in data storing devices
would be server related hardware components.
Within the software-based changes category, there
are three elements, which are grouped based on the
level of coding/programming would be required.
First, changes with respect to software-as-a-service
(SAAS) components or applications, platforms and
operating system are grouped into one element as
they all belong to system related software. Second
element is software interfaces or protocols which are
required to connect other data sources or systems.
Third element is changes relevant to software within
data storing servers. These fixed elements of the
organizational changes step are given unique ID’s
starting with OC as their prefixes.
3.4 Specific Impacts
The fourth step is to identify specific impacts or
effects from the project due to the organizational
changes identified in previous step. These impacts are
estimated for the company through the product user
perspective that is being considered. Impacts in this
case are defined as any influence on a business’s
values. These impacts could be positively or
negatively influencing the business values.
Therefore, this step is separated into positive impacts
and negative impacts. Within positive impacts all
financial or economically measurable benefits as well
as soft or qualitative benefits and opportunities which
are not measurable in economic terms are considered.
Within the negative impacts, all financial or
economically measurable costs and soft or qualitative
costs and risks that are not economically defined are
considered. This step is kept open therefore the
framework user only enters specific benefits and costs
of the digital product being evaluated. Every element
in the impact step is also stored by unique IDs, with
prefixes SI1 and SI2 for specific benefits and costs
respectively.
3.5 Business Goals/Values
Business Goals/Values are defined as the high-level
value categories that are important for the business.
Based on the various benefits and costs from digital
projects, five categories of business values are
derived for this framework. These goals could be
uniquely defined for each company’s own set of
goals, but the goals here are more generically defined
to be able to classify highly relevant impacts of
digitalization. Mostly considered business value
categories in BDN networks such as profitability,
productivity and customer value were examined but
certain impacts which could not be directly linked to
these three major categories but have been rated as
equally important by businesses, were grouped under
two additional categories.
As part of the framework, profitability, producti-
vity and customer value are listed as Financial Value,
Internal Process Value and External Stakeholder
Value respectively. Any specific impact that affects
the return on investment, profit, revenue or costs and
can be measurable in economic terms are linked to the
Financial Value element. All other impacts that are
not measurable in economic terms are connected to
one of the other four business value categories and a
qualitative impact level is chosen. Specific impacts
relating to internal processes such as operations and
administration processes or productivity
improvement and efficiency gains or loss in the
processes that are not measurable in economic terms
are connected to Internal Process Value element.
External stakeholder’s value which are not yet
measurable in economic terms such as customer
satisfaction is connected to External Stakeholder
Value element.
Although every impact in a business
hypothetically linked to Financial, Internal Process or
External Stakeholder there are certain benefits from
digitalization that can be classified into other
categories due to uncertainty in when they will start
impacting economically and some impacts’
qualitative nature in general. For example, safety is
paramount in many industries. Digitalization plays an
important role in enhancing safety value by allowing
real-time sensing/monitoring of unsafe conditions
and allowing businesses to act fast to avoid
catastrophic incidents from occurring altogether.
Such values are indirectly related to avoidance of
losing productivity therefore profit, from which the
safety impact can be economically measured.
However, there is uncertainty to account for in this
case, where ‘if and when’ the risk event is avoided
should be considered in the benefit calculation and
Digital Value Dependency Framework for Digital Transformation
647
can be hard to estimate. Another way to look at this
would be, it improves the safe operation of plants,
therefore provides safe working conditions for
employees. Therefore, such impact with a qualitative
nature should be estimated as well. Similarly,
employee satisfaction is one of the main benefits of
digitalization where workflows are improved through
digital products and such benefits are significantly
important for businesses’ sustainability.
Although these impacts can indirectly influence
the productivity of the business, it would not always
be possible to translate it into an economic value due
to its inherent qualitative nature. Such specific
impacts have significant importance to the business
and directly related to internal stakeholders, but when
they cannot be economically defined just yet they
should be connected to a category that relates to the
internal stakeholders. Therefore, Internal Stakeholder
Value has been introduced.
Furthermore, digitalization offers opportunities
for growth and be innovative, which in the long run
improves financial value. It is however not possible
at early stages to predict with accuracy how much
benefit such innovation related impacts will bring.
For example, initiatives to deliver new digital tools,
new ways of working or providing training using
digital technologies such as AR, increase the learning
and put the organization in the growth path.
Such futuristic impacts can be qualitatively
described but they would not fit into the currently
defined categories of business goals. Therefore, in
order to classify innovation related impacts, Learning
and Growth Value has been introduced. Each
element in this step has a unique ID, indicated by FV,
IPV, ESV, ISV and LGV for Financial Value, Internal
Process Value, External Stakeholder Value, Internal
Stakeholder Value, and Learning and Growth Value
respectively as in
Figure 1.
3.6 Procedure
The owner of a digital product/project and its
stakeholders are the intended users of the DigVD
framework which helps them systematically
determine the impacts of their digital product before
development starts. A facilitator who is trained on the
framework can conduct the estimation process in a
one-on-one interview basis, when the product owner
can estimate benefits and costs from all stakeholders’
perspectives. Otherwise, it can also be conducted as a
facilitated workshop where product owner and
stakeholders of the digital product/project are
involved. The framework has been developed as a
digital web-tool and this tool has been directly used
in this estimation process. The results are saved in the
database of the tool, such that the dependencies and
elements can be queried and analyzed later.
At the beginning, the product owner identifies
unique user groups in order to capture impacts from
all user perspectives. User-groups here are the type of
users of the digital product being developed. First,
one user-perspective is chosen, then the framework
elements are connected from left to right according to
Figure 1, and the steps described in Figure 2 are
followed.
In this process, mutually exclusive positive and
negative impacts are considered and described, and
these impacts are then connected to most fitting
business goals. When this connection is established,
the estimation of the impact must be entered. When
an impact can be economically measurable, it will be
connected to the Financial Value category and a
calculated monetary value is entered together with an
estimated probability percentage. When it is not
possible to estimate an economic value the product
owner estimates the impact qualitatively from a 5
level ratings from ‘Insignificant’ to ‘Extremely
significant’, and the selected qualitative impact level
is then converted to a score as described in section
3.7.1.
These steps are then repeated for every user
perspective defined earlier for the digital project in
question, and the results are saved and analyzed.
When every element on the left-side step is connected
to an element on the right-side step, user has
the
option to enter the influence weightage (defined
Figure 2: Dependency mapping procedure.
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
648
in Section 3.7.3) on the edge, otherwise it will be
assumed to be equally influencing the adjacent
element during analysis.
3.7 Mathematical Model
In order to measure and estimate the value
contribution of every element in a network, a
mathematical model has been developed and the
terms, variables and equations used in the
measurement system are explained in this section.
Although the specific impact is estimated, this value
gets passed onto and distributed to all elements that
are connected on its left. Figure 6 summarizes the
variables and equations used when the value of a
successive element on the right side gets passed onto
the predecessor element.
3.7.1 Value Weight of Specific Impact
Value Weight (VW) is defined as the measurable
quantity of the impact. There are two types of value
weights based on whether the impact’s economic
value is measurable or can only be defined
qualitatively. Economically measurable impacts are
estimated for its corresponding economic value in
euros and a probability (%) of occurrence is also
entered. The qualitative impacts are based on ordinal
levels where the significance of impact is selected
along with a probability of occurrence. Impact levels
are insignificant, minor, moderate, major and
significant which range from 1 to 5 for positive
impacts and -1 to -5 for negative impacts. Probability
of occurrence starts from rare to very likely with 10%
to 90% rating. Then using impact-probability matrix
an impact score is given for every impact.
Therefore, an economic value or qualitative score,
represented by € and Q respectively is stored for each
impact. The positive and negative impacts are stored
separately as well. Therefore, there are four types of
value weights are possible, where positive impacts
are stored as €+ and Q+, and negative impacts are
stored as €- and Q-. Value Weights are specified by
impact IDs on the subscript (i and j indicating
elements within categories) and value weight types on
superscript as below:
𝑉𝑊

€
> 0 , 𝑉𝑊

€
< 0 Economic
Impacts
𝑉𝑊


> 0 , 𝑉𝑊


< 0
Qualitative Impacts
3.7.2 Value Weight of Business Value
Since only the economically measured impacts are
connected to Financial Value (FV) category, all
economic benefits (€+) and costs (€-) are summed up
and positive and negative impacts are saved
separately. On the other hand, those impacts that are
not measured economically, gets a qualitative score
as explained in previous section and these qualitative
impact scores are connected to one of the qualitative
business value categories (IPV, LGV, ISV, ESV)
which are then averaged to estimate the overall
qualitative impact of every business value category,
as shown for IPV in (2).
Economic:
𝑉𝑊

€
=
(𝑉𝑊

→
€
)
(1)
Qualitative Impact:
(2)
3.7.3 Influence Weightage
Influence weightage (µ) is defined as how much a
predecessor element influences the successive
element’s value. This weightage is either defined by
the user by estimating the percentage of influence
directly when edges are drawn, or calculated
afterwards based on how many edges connect to a
successive element, where every undefined edge that
connects to the successor element is assumed equally
influential as follows:
(3)
EW
pi->sj
= VW
si
* µ
pi->sj
(4)
The value weights of predecessor-elements
(VW
pi
) are calculated by summing up every edge
weight going from the predecessor element as follow.
VW
pi
=
𝐸𝑊


(j: successive elements
dependent on pi) (5)
3.7.4 Edge Weight
Edge Weight (EW) is defined as the value weight of
the edge from a predecessor element to a successive
element. Using µ and VW of successive element, the
Edge Weight is calculated as below. Since the VW of
impacts are the starting point, VW
sj
=VW
B.
Digital Value Dependency Framework for Digital Transformation
649
3.7.5 Value Weights of Step 3, 2, 1’s
Elements
Value Weights of the specific impact elements are
distributed backwards to other elements in each
predecessor step according to edge weights that
originated from the predecessor element. This is
based on the logic that if predecessor element is
influencing a successive element, the value weight of
the successive element is distributed to the influencer.
The basic operations have been shown in Figure
3. The equation is applied to each value type
separately. For qualitative value types, when scores
are summed up to more than 4.5, then the maximum
score of 4.5 is assigned to the element, since the score
has a limit and that predecessor element has the
maximum possible score to influence the successive
element.
3.7.6 Value Contribution Fraction
Value Contribution Fraction (VCF) is defined as an
element’s contribution percentage to the total value of
the project and calculated as follows:
VCF
pi
= VWpi/(Total VW of Project per value
type) (6)
Similar to value weights, there will be four value
contribution fractions to each element representing
each value types of €+, €-, Q+, and Q-.
3.7.7 Value Share
Value Share (δ) is defined as how much importance
the product owner estimates the economic impacts vs.
qualitative impacts towards the project, where δ
€+
+
δ
Q+
= 100 and δ
€-
+ δ
Q-
=100. If both economic and
qualitative value types are equally important then it
will be 50:50. Additionally, if total value weight of an
impact type is zero then the other value type’s share
becomes 100. (ie:
= 0), then, δ
Q-
= 0 and δ
€-
= 100.
3.7.8 Normalized Contribution
Normalized contribution (η) is the contribution of
both economic value and qualitative score for every
element per positive and negative impact. Since
economic value and qualitative scores are on two
different scales, they are both combined using the
Value Share and Value Contribution Fractions for
positive and negative impacts.

𝑉𝐶𝐹

€
€
𝑉𝐶𝐹



(7)


𝑉𝐶𝐹

€
€
𝑉𝐶𝐹



(8)
4 APPLICATION
4.1 Digital Project
In order to illustrate the applicability of the
framework, a typical industrial digital project has
been chosen. This is a test digital project within
energy management for industrial plants. Among
many user groups, a specific group of managers are
expected to benefit the most from such a solution,
therefore the results are only presented for this user
group’s perspective in this paper. The network
outcome from the framework mapping is shown in
Figure 4
as a Sankey diagram
.
Digital Capabilities, Org. Changes and Business
Values had fixed elements as described in section 3.2,
3.3 and 3.5 respectively. The Digital technologies and
Specific Impacts steps were left open and 13 digital
technologies, 9 benefits and 3 costs were identified by
following the process described in Section 3.6. The
thickness of the links indicates the influence
weightage % (µ) used in the calculation of edge
weights (EW) of each link and therefore the value
weights (VW) of elements.
Figure 3: Value Weight calculation of elements in steps OC, DC and DT.
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
650
Figure 4: DigVD network result for applied project.
Several digital technologies being considered for
this project but on the network only their IDs are
presented with respect to how it is saved in the tool’s
database. These technologies ranged from smart
devices, an application development environment,
connectors to existing IT systems, interface protocols,
an integration platform, databases, to data
engineering and analytics technologies. The nine
benefits are also specified on the network by their IDs
and ranged from reducing process times, saving costs
due to having quick access to data, better efficiency
management, to employees and external stakeholder
satisfactions due to the consolidation of data. Among
the nine identified benefits, 3 are economic benefits
(€+) and 6 are qualitative benefits (Q+). The user
group not only makes use of the product, it also pays
for the solution. Therefore, in terms of negative
impacts 3 costs elements were identified which were
all economic costs (€-) related to technology
purchases and maintenance of the solution.
Furthermore, other industrial test projects from
supply chain and site management are brought in to
illustrate user group specific results and multi projects
analysis possibilities of common elements.
This approach has helped in estimating actual
impacts in terms of economic and qualitative values
from the intended projects. However, on this paper,
for publication consideration instead of their true
economic values, the impacts and VW’s are
represented as low to very high levels so that results
from the framework can be illustrated.
4.2 Results and Discussion
From the estimated impacts, business values that are
influenced with positive and negative impacts are
identified according to their VWs
(Figure 5).
It
becomes evident that in terms of economic impacts
close to a very high-level benefits are estimated, and
less than 25% of the estimated benefits has been
estimated as costs. Furthermore, the estimation
process with the tool enabled identification and
recording of impacts which cannot be readily
measurable in economic terms. In terms of such
qualitative impacts, External Stakeholder Value,
Internal Process Value, and Learning & Growth
Value categories will be expected to have positive
impacts, where the averages for each value has been
calculated to be 1.9, 2.40, and 2.15 respectively out
of maximum score of 4.5 that is possible with highest
impact and highest probability. There were no
negative qualitative impacts identified. Although the
estimation is only for one year, when positive and
negative value weights of Financial Value are added,
the net is positive, indicating a positive turnover for
the project. In addition, overall positive qualitative
impact scores in the other three business values also
hint that the project has an overall benefit to the
business.
Now the importance of other elements such as
organizational changes and digital capabilities of the
dependency network with respect to the measured
impacts have been calculated using the formulas in
section 3.7. The economic value distribution across
all organizational changes has been presented in
Figure
6
, and each change element’s benefit and cost
Digital Value Dependency Framework for Digital Transformation
651
Figure 5: Economic and qualitative impacts of the project.
contributions are displayed. As seen on this figure,
OC20 relating to the change within the software
systems contribute to most of the cost. In terms of
economic benefits, OC20, OC10 and OC13 changes
contribute to a lot of benefits. These results can be
used in change management.
Additionally, using a Value Share of 50:50,
meaning economic and qualitative impacts are
equally valued for this project, among the positive
impacts it is evident that OC_13 has been rated with
most normalized contribution, considering its large
contribution to both positive qualitative impact and
economic impact. This means that this particular
change relating to having descriptive knowledge
access, contributes to benefits the most. Being able to
access the knowledge about historic data brings
highest value to the project compared to all other
changes. In this example, front-end process related
changes are contributing to high benefits which mean
that these changes should be prioritized during
transformation. Furthermore, identification of which
changes influence the most benefits can help in
measuring the benefit after the product has been
implemented.
It is often the case that product owners find it
difficult to decide economic impacts from new
technology-based projects and therefore qualitative
impacts have been given as an option when they
estimate the impacts. Now with the Value Contribution
Fraction and the Value Share, the Normalized
Contribution is calculated which brings two different
value types together so that total positive impact and
total negative impact of an element can be calculated.
On the overall, applying such a measurement model
helps in understanding the dependencies better, and
product owners are able to prioritize changes,
technologies and capabilities more effectively.
Similarly, within Digital Capabilities and Digital
Technologies the most value contributing elements
are identified. As such Connectivity (DC_4) and
Visualization
(DC_3)
are
contributing
to
most
value,
Figure 6: Organizational changes’ economic impact
distribution.
and DT_65 which is an application service and a
technology influencing most of the capabilities
contribute to highest benefit when compared to all
other technologies.
Change management in transformational
undertakings like digitalization can be difficult
especially when potentials from a new digital project
is not transparent to all involved stakeholders. Since
the framework is done for every user perspective of a
project as presented for a site management project in
Figure 7, it allows users to have transparency on how
much value is being contributed by different user
groups within the same project. In this example, the
financial value of the project completely comes from
User 2’s perspective. However, qualitative values
such as external stakeholder value, internal
stakeholder value, internal process value and learning
& growth value are being influenced through impacts
that are estimated from User 1, User 2 and User 3
perspectives. Such results make the benefits shared
by all user groups more transparent and enable the
measurement and realization of benefits more viable.
Most talked about topic when it comes to change
management is people and culture. If certain group of
users are resistant to change, it is important to be open
about which benefits are expected and communicate
them
effectively
and
as
early
as
possible. From the
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
652
Figure 7: Impacts from different user perspectives.
results of DigVD such transparency is achievable.
Although any benefit-cost analysis must consider
user perspective in general, what is special about
applying DigVD approach is that it brings more
transparency onto the type of changes, capabilities
and technologies utilized by the different user groups
so that decisions can be made to maximize benefits
based on the dependency results. When the kind of
changes required for highest value realization is
identified the right training for the specific user group
can be decided to realize those dependent benefits in
timely manner. Likewise, if benefits are not high from
one user group’s perspective compared to other user
groups, it can be explained why this is the case from
the dependency mapping. Untapped organizational
changes that can influence high benefit contribution
can be initiated so that additional positive impacts for
that user group become possible.
These results show the kind of analysis possible
for single projects so that product owners can build
business cases and manage project and changes
effectively. Additionally, when benefit dependency
framework’s results from multiple projects are put
together, it brings additional transparency to the
overall digital transformation of the business, where
the results help in decision making. In this regard,
digital capabilities being developed within the
business can be brought together like in Figure 8,
where benefit and cost from four projects that are
influenced by the six digital capabilities’ of a business
are displayed. From this, Visualization capability is
contributing to high economic value and to high
positive qualitative score as well. Following this,
connectivity is the next digital capability contributing
to high positive impact overall. On the other hand,
when these four projects are considered together
information sharing capability accounts for the
highest cost followed by connectivity capability.
From this result, it is also evident that monitoring and
control capability has not been associated with values
of the four projects yet, where this capability’s
contribution to benefits and costs are very low.
Although this representation shows only four projects
and not the right representation of the current capabi-
lities mapping of the business, such results can tell
which capabilities they can potentially invest in future
to have a balanced digitalization effort in all areas.
Therefore, the DigVD framework and the model
behind it take benefit dependency network to where
the estimated outcome can be measured and
compared for further understanding of value
contribution and the development of business cases
driven by data. These results demonstrate project-
based analysis is possible where the results can help
prioritize transformational changes. Being able to
relate the projects values to other elements through
value weights, value contribution fraction and
normalized contribution open better understanding of
how other elements are influencing the project. The
framework applicability has been tested with four
process industry projects with outcomes showing
more impact estimations when it was done with the
DigVD than it was done before for similar projects.
As future work, further tests on consistency of the
resulting networks between similar projects are to be
evaluated. Additionally, influence weightage has
been assumed to be equal based on the number of
lines connecting to an element so that the
methodology can be studied. Since the methodology
has been studied with positive outcomes, the actual
influence weightages could be included but deciding
on these influence weightages is a separate process on
itself and must be evaluated together with product
owners more extensively. One of the main advantages
of this tools is that size of the project is not a
limitation. Project that is expected to have a small
return to large size projects with expected return in
millions of euros can be evaluated. However, the way
the tool is currently defined only one-year estimation
of impacts is possible. However, how the project
performs in the following years are important for
product owners and therefore by extending this work
and using the results from the current tool, calculation
of values over number of years using real option
valuation are being investigated.
Digital Value Dependency Framework for Digital Transformation
653
Figure 8: Values of Digital Capabilities of a business according to the four sample projects assessed.
5 CONCLUSION
Transparency on potential benefits from digital
projects is strategically important for companies. In
this paper, a redefined BDN for digital projects has
been presented where dependency results can be
saved, measured and compared. The elements within
the framework are generalized and a mathematical
model has been developed such that results from
multiple projects can be consolidated which then help
in implementing and better manage digital
transformation in a company.
The methodology and model presented here
are foundational for understanding value networks
within digitalization. Filling out the dependency
network for every project enables not only the
transparency of value potential in one project, but it
also aids in digital transformational activities, starting
from prioritizing changes to understanding
dependencies and value shared between multiple
projects and user groups. This allows for the creation
of value networks where relationships between digital
technologies, capabilities and organizational changes
and their contribution to value are made transparent.
The results from the applied projects show specific
identification of impacts, contribution of value from
different user perspectives, as well as reduce the
complexity in understanding the role of digital
technologies, capabilities and organizational changes
in brining value to the business. When multiple
projects use the same framework to assess their
projects’ benefits and costs, investment decision
making for digital transformation in a company
becomes more standardized, and changes or
capabilities the organization must enable for realizing
highest value are identified with this approach.
REFERENCES
Cenamor, J., Sjödin, D. R., & Parida, V., 2017. Adopting a
platform approach in servitization: Leveraging the
value of digitalization. International Journal of
Production Economics, 192, 54-65.
Chaves, M., & Pedron, C., 2015. Web 2.0 Technologies in
Project Management: A Proposal of a Benefits
Dependency Network. AMCIS 2015 Proceedings.
Digital transformation of industries, Demystifying Digital
and Securing $100 Trillion for Society and Industry by
2025., 2016. Industry Agenda, World Economic Forum.
Freitas Junior, J. C., Maçada, A. C., Brinkhues, R., &
Montesdioca, G. (2016). Digital capabilities as driver to
digital business performance. In Proceedings of AMCIS
2016.
Gerber, A., Baskerville, R., & Van der Merwe, A., 2017. A
taxonomy of classification approaches in IS research. N
Proceedings of Twenty-third Americas Conference on
Information Systems.
Gomes, J., & Romão, M., 2014. Benefits management
enhances the organizational competitive advantage. In
New Perspectives in Information Systems and
Technologies, Volume 1 (pp. 137-150).
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
654
Lee, I., & Lee, K., 2015. The Internet of Things (IoT):
Applications, investments, and challenges for
enterprises. Business Horizons, 58(4), 431-440.
Lenka, S., Parida, V., & Wincent, J., 2017. Digitalization
capabilities as enablers of value co creation in
servitizing firms. Psychology & marketing, 34(1), 92-
100.
Okhrimenko, I., Sovik, I., Pyankova, S., & Lukyanova, A.,
2019. Digital transformation of the socio-economic
system: prospects for digitalization in society. Revista
ESPACIOS, 40(38).
Probst, L., Lefebvre. V., Martinez-Diaz. C., 2018. Digital
Transformation Scoreboard 2018. EU businesses go
digital: Opportunities, outcomes and uptake.
Publications Office of the European Union,
Luxembourg, DOI, 10(691861), 10-2826.
Mikusz, M., 2014. Towards an understanding of cyber-
physical systems as industrial software-product-service
systems. Procedia Cirp, 16(1), 385-389.
Rizk, A., Bergvall-Kåreborn, B., & Elragal, A., 2018.
Towards a Taxonomy for Data-Driven Digital Services.
In Proceedings of the 51st Hawaii International
Conference on System Sciences.
Ward, J., Taylor, P., & Bond, P.,1996. Evaluation and
realisation of IS/IT benefits: an empirical study of
current practice. European Journal of Information
Systems, 4(4), 214-225. doi:10.1057/ejis.1996.3
Ward. J, and Daniel. E., 2006. Benefits Management:
Delivering Value from IS and IT Investments. John
Wiley & Sons.
Westermann, T., & Dumitrescu, R., 2018. Maturity model-
based planning of cyber-physical systems in the
machinery and plant engineering industry. In DS 92:
Proceedings of the DESIGN 2018 15th International
Design Conference (pp. 3041-3052).
Digital Value Dependency Framework for Digital Transformation
655