Der Lehrstuhl auf der ECIS 2021
Unsere Forschung auf der ECIS 2021
Neues Jahr, neue Erfolge!
Wir sind stolz darauf, dass im Jahr 2021 drei Publikationen unseres Lehrstuhls auf einer der wichtigsten internationalen Konferenzen der Wirtschaftsinformatik – der European Conference on Information Systems (ECIS 2021) – vorgestellt wurden.
Herzlichen Glückwunsch an alle Autoren!
Sie können die Beiträge über die Buttons aufrufen.
Smart Service Systems in Mexico: Implications of the Socio-Economic Context on the Willingness to Co-Create
The increasing availability of data generated by IoT opens the door for new opportunities to create and deliver value.
At the same time, IoT technology gets more complex and start to involve a growing number of stakeholders who can create value in a joint process.
Following the trend that will integrate IoT and their stockholders into smart service systems, it is important to ensure willingness to co-create from both – customers and companies sides.
Therefore, Emanuel Marx, Willi Tang, and Martin Matzner shed light on how the socio-economic context influences IoT adoption by affecting a company’s willingness to co-create and provide an overview of potential fields of IoT applications for which there is an urgent need.
Predictive Business Process Deviation Monitoring
Organisations strive to align as-is and to-be processes to avoid undesired process flows.
Conformance checking, being one of the solutions, can only identify deviations after it has occurred.
Therefore other methods, such as predictive business process monitoring (PBPM), that can identify deviations before they occur could enable handling deviations proactively.
While existing PBPM methods do not rely on labels generated with conformance checking, Sven Weinzierl, Sebastian Dunzer, Johannes Tenschert, Sandra Zilker, and Martin Matzner developed a new method, Predictive Business Process Deviation Monitoring (PDM), using conformance checking and deep learning. The method predicts process deviations with high predictive quality.
Bringing Light Into the Darkness – A Systematic Literature Review on Explainable Predictive Business Process Monitoring Techniques
More recent PBPM techniques rely on machine learning (ML) algorithms for more accurate models, lacking explainability for humans.
This makes ML-based PBPM systems a “black box” for stakeholders and induces their scepticism in this regard.
Hence, the research field of Explainable Artificial Intelligence (XAI) is a promising avenue for the broader adoption of ML-based systems in practice.
By providing an overview and classification of existing Explainable Predictive Business Process Monitoring (XPBPM) approaches, Matthias Stierle, Sven Weinzierl, Sandra Zilker, and Martin Matzner highlight the importance to design and develop XPBPM techniques that advance the field.