A trend on digitization, digitalization, and digital transformation associated with several challenges and complexity can be observed. In this context, the usage of conceptual modelling, particularly business process modelling, is a widely established approach to structure complex activities. However, innovative approaches are required to tackle its liaison with digital technologies. In this paper, after an analysis of the state of the art for business process management approaches and related digital technology aspects, an interactive process modelbased approach - the so called “Scenario Scanner” - is proposed and applied in the context of the EU project Change2Twin. Finally, insights on future research consider conceptual modelling in an era of digital transformation.
Digital transformation encompasses a broad spectrum of initiatives ranging from IT
modernization considering legacy systems, over streamlining processes through digital
optimization to adopting innovative business models. To develop a business strategy for
digital transformation, aligning information and operational technology can be supported
by using various methods to discover, model, analyze, measure, improve, and optimize
business processes – known as business process management (BPM). Based on Gartner’s
definitions [
Starting with, a targeted set of exact search terms was elected to provide an indication
on research trends in a BPM context. The results (see Table 1) deem from a selection of
international renowned databases and illustrate that a plethora of literature on BPM exists
(till October 2023), while the numbers significantly decrease regarding digital technologies,
innovative approaches, or digital transformation. Although digital twins (DTs) are already
widely researched, DTs of processes are barely established and cannot found before 2018.
Nevertheless, the well-established discipline BPM can be considered as a driver of
organizational efficiency. However, contemporary ecosystems are marked by agility and
uncertainty resulting in disruption and business model innovation. Hence, traditional BPM
requires transformation towards more flexibility, modularity and context-sensitivity [
With respect to digital technologies, the European H2020 project Change2Twin [20] aims at supporting small and medium-sized manufacturing enterprises in their digital twinning efforts. The project includes the provisioning of appropriate DT solutions, which is far from being trivial. This paper attempts to bridge conceptual modelling – particularly in form of (business) process models – and the establishment of DTs. Specifically, scenario building is used to elaborate on the connection between the digital models and the real world. The paint production pilot case of Change2Twin is used to demonstrate DT challenges related to the production process.
Concluding the introduction, the overlying question for this work is how conceptual modelling (starting with BPM) must transform to a more agile, interactive, and innovative approach, allowing to gain benefits from digital technologies, while keeping the human in the loop. The following chapters present a selection of related work that is followed by the proposed model-based approach referred to as Scenario Scanner. The next chapter illustrates the experimental prototyping for the Change2Twin pilot case. The work is concluded with brief insights on further research.
A deeper look into selected publications on BPM-based approaches revealed diverse
application domains – ranging from adaptive processes for learning purposes to the
management and monitoring of IoT devices – underpinned with related industrial tools.
Contemporary, BPM is considered as a key success factor for digitization initiatives
harmonizing business and IT perspectives [10] building upon process optimization
considering context and dependencies of process-related models influencing process
architectures and their evaluation. Going beyond digitization, the implementation of a
multi-sensor approach [11] for monitoring hand-operated process parts revealed that a
combination of sensors, context and process data is needed to digitally monitor manual
business processes. However, there is still a gap in more integrative and cyber-physical BPM
approaches. The plethora of existing modelling and formalization approaches can serve to
tackle this gap, where an essential aspect is the readability of the models influencing the
validation by human experts [12]. Here, business process modelling can support process
analysis for instance by applying process automation and mining techniques facilitating the
translation of sensor measurements and logs into human actions. Especially in hybrid and
heterogeneous environments across industries, model-driven workflow automation – e.g.
using the BPMN 2.0 standard – facilitates human system interaction for both automated and
manual subprocesses [
With regard to technologies, a strive for the integration of conceptual models with digital technologies and related services makes sense in the long run to foster transparency and understandability in complex heterogeneous and agile environments. Starting from a business perspective, the OMiLAB Community of Practice [21] offers such an innovation environment that looks into business ideas and drills them down to the level of feasibility experiments. Here, conceptual modelling serves as an intermediary between business and cyber-physical experimentation. Digital twinning is an underlying concept of the OMiLAB environment, establishing conceptual representations bridging business and technological perspectives. In [22] the integration of DTs into organizational structures and business models is proposed. Technology-wise there has been a lot of research on DT development, whereas the connection to business aspects is deficient. Hence, the authors suggest extending the DT design towards capability driven development such as developing DT management dashboards based on capability design and context models including KPIs and historical data about components’ performances. [23] points out that targeted DTs of organizations come across challenges such as interdependencies requiring innovative ways, like more comprehensive vocabulary, compared to current business process models.
The Scenario Scanner builds upon the OMiLAB Innovation Environment [16] aiming at
facilitating the tackling of digitalization challenges. OMiLAB’s architecture is structured in
three main layers, which are business, conceptual modelling, and proof-of-concept layer.
Those layers are considered being in line with the building blocks of the Scenario Scanner.
A set of (modelling) tools and approaches accompany the laboratory ecosystem such as the
meta modelling platform ADOxx [
The scenario building starts from a business perspective drilled down to experimental
prototypes of DTs. An innovative design thinking approach is followed to create physical
scenes describing business processes in a co-creative way. This can be done either following
a pen and paper-based approach or in a technology supported way based on the
Scene2Model [
Identifying the right balance between top-down and bottom-up approach is not trivial. On the one hand, disruptive business models must be supported to deal with the contemporary agile ecosystems, potentially resulting in additional hardware (sensors or actuators) needs. On the other hand, existing hardware should be exploited to limit infrastructure costs and risks. The Scenario Scanner is intended to support in this balancing act by establishing experiments – where possible considering underlying processes and BPM efforts that are already in place in application scenarios – focusing on relevant aspects of digital transformation and managing the transition from the domain concepts to physical devices and vice versa via a model-based approach. Digitization patterns [15] are applied – on both, the experiment, and the real application case – in order to facilitate the selection of proper measures and technologies for implementing the scenario characteristics. However, the reader should bear in mind that the scenario building will evolve with further experiments towards a more holistic method (e.g. applying patterns or providing recommendations) supporting conceptual modelling and exploiting contemporary digital technologies.
Change2Twin perceives DTs as “a digital replica of an artefact, process or service that is so accurate that it can be the basis for decisions. The digital replica and physical world are often connected by streams of data”. This experiment illustrates the paint production pilot revealing the potential of DTs and related technologies in manufacturing. Originally, the paint producer strongly relied on paper-based documentation, where an error-proneness of the manual activities could be observed. Therefore, the company intended the installation of a real-time inventory, leveraging the paper-based documentation. Starting with the design building block of the Scenario Scanner, a series of process modelling workshops with representatives of the paint producer and digital technology providers was conducted. Based on the process map of the production plant a focus was set on the raw material flow, the production process itself, and the labelling process. Detailed digital process models were designed (involving the middle building block of the Scenario Scanner) in guided workshops to depict the activities of the production and the labelling as well as their execution sequence. Means of abstraction were applied onto the defined processes to reduce the complexity and focus on the digitization relevant aspects. The digitization (involving the bottom building block of the Scenario Scanner) was tackled building upon the digital models (considered as an early stage of a DT). Referring to the digitization building block of the Scenario Scanner, (process) modelling served as a foundation for the technology specification and documentation following an event-based approach. Ideas on digitizing a production use case were presented in [13], a focus on integration considering models, data and products was set in [18] and the usage of physical experiments to reduce the use case complexity were researched in [14].
Summarizing, three digitization challenges were extracted: the digitization of (a) the production process, (b) the raw material warehouse, and (c) the product information. Starting from completely analogous machines was particularly challenging for the development of a DT. Hence, the DT experiment in the OMiLAB mirroring the production process was used to transparently discuss the digitization challenges among the heterogeneous stakeholders such as technical or business experts. The experiment (see Figure 2) used a simplified, familiar association – the making of tea – covering the major pilot characteristics to identify appropriate infrastructure. RFID technology – one digitization option – was used to digitize production orders, material slots and production process steps. Going through the process, the captured information enables monitoring in real-time paving the way for further analysis such as process simulations and predictions. Haptic process simulations were used to collect event data in form of timestamps allowing conclusions on stock levels or maintenance plans. After the finalization and evaluation of the paint production experiment, the pilot company decided to equip the production stages and the warehouses with RFID technology (similar to the one used in the experiment) allowing for a continuous collection of data throughout the process highly leveraging the manual documentation towards a real time DT approach.
In line with traditional BPM approaches, the Scenario Scanner is intended to support the establishment of DTs of processes. So far, the automation of currently manual steps was researched by implementing promising digital technologies and services. Showcased in the Change2Twin project pilot, digital twinning was facilitated by scenario building in the manufacturing domain. The usage of physical experimentation is considered promising to leverage conceptual modelling towards more interactivity and innovation. However, in future a more comprehensive methodology on how to apply the Scenario Scanner by conducting application scenario experimentation in heterogeneous domains is expected to contribute to conceptual modelling in general and to targeted concepts for innovative BPM.
This work was supported by the European H2020 project Change2Twin (“Create and Harvest Offerings to support Manufacturing SMEs to become Digital Twin Champions”) with grant agreement ID 951956 that started on 1st of June 2020 and will end on 30th of November 2024. Details can be found on the project webpage: www.change2twin.eu.