How can the Internet of Things close the gap between processes in the physical world and their digital representation? A research team led by HSG Professor Barbara Weber at the Institute of Computer Science (ICS-HSG) is investigating this question.
The digitalisation of processes is well advanced in many areas of our daily lives. Simple examples of how manual processes are being replaced by automated ones are the control of automatic shading in modern buildings or the automatic notification of flight delays on smartphones. The aim is to reduce the workload, increase efficiency, minimise errors and check quality. "Digitalisation driven by IT systems offers great potential for the automation, analysis and optimisation of processes," emphasises Barbara Weber, director at the Institute of Computer Science and member of HSG's President's Board. In a purely digital world, processes can be easily tracked by IT systems. "Here, we know how to analyse and optimise processes."
In the physical world, however, there is usually no central IT system to take over the tracing of processes. The digital traces of a real-world process are often incomplete. "This is particularly true for human interactions, which limits the possibilities for automated process monitoring." Another difficulty is that in the physical world, process descriptions are often created in the form of guidelines or checklists that can be interpreted in different ways. This leads to ambiguities in the corresponding process models, which must be taken into account during the conformity check.
“Our work aims to advance the state of the art in computer science research and technology.”
With the ProAmbitIon project, Babara Weber and her team are proposing to use the Internet of Things (IoT) to close the gap between process flows in the real physical world and their digital representations. The IoT offers a wealth of intelligent objects that collect data about activities and their surroundings. However, this data is often not directly linked to processes, nor can it understand them in all their complexity. "The challenge is to select the relevant data and correlate it with the activities in the processes. By providing novel tools and methods to find these correlations, we can help to analyse processes in the physical world and check whether they are being executed correctly," says Marco Franceschetti, who is involved in the project as a post-doctoral researcher.
Based on highly automated processes in smart factories, in which production machines, robots and people collaborate with each other, the researchers are developing methods and techniques to monitor the error-free execution of process steps. To this end, data streams from different sensors and actuators are combined to enable the processes to be observed as accurately as possible.
The techniques developed are then tested in processes that mainly involve manual activities carried out by people. To this end, a laboratory environment was designed in collaboration with the Cantonal Hospital of St.Gallen, in which non-invasive sensors are used to observe a blood donation process. In training environments, it is possible to check whether medical and nursing students are carrying out the necessary sequence of steps in the treatment processes correctly and, for example, whether they are following the hand hygiene guidelines. "By observing the processes using the IoT, direct feedback can be provided to medical staff in real time and, for example, the risk of infection can be reduced or processes optimised," emphasises Ronny Seiger, assistant professor at the HSG, who is also researching in the project.
The team of researchers emphasises that neither the specific area of application of the blood donation process nor the application example of smart factories are the overarching goal of the project. Rather, it is about developing generally applicable solutions that can be used in different areas. The aim is not to monitor, patronise or expose people at work. "We want to research new approaches to IoT-based conformance testing of processes that can be useful in a wide variety of areas. This ranges from production and healthcare to mobility and even cooking."
“For example, using IoT to monitor processes can reduce the risk of infection.”
New approaches to IoT-based process conformance testing include a user-friendly approach to enriching process descriptions with IoT-related, pattern-based monitoring points provided by domain experts. New concepts for providing feedback on process conformance would be developed to make the results of the conformance check understandable for the end user and to resolve ambiguities interactively. "Our work aims to advance the state of the art in computer science with several contributions," concludes Barbara Weber. The project will be conducted according to the principles of design science research and will develop new software artefacts, including frameworks, models, methods and architectures.
Prof Dr Barbara Weber holds the Chair at the Institute of Computer Science (ICS-HSG) and is Vice-president of Studies & Teaching. In the project ‘ProAmbitIon: Online testing of process conformity in the case of ambiguities through the Internet of Things’, she is working with Prof. Dr Ronny Seiger, Assistant Professor of Computer Science with a focus on software engineering methods and techniques, and research assistant Marco Franceschetti, PhD.
