1st International Workshop on Business Process Innovation with Artificial Intelligence (BPAI 2017)

Transcription

1 1st International Workshop on Business Process Innovation with Artificial Intelligence (BPAI 2017)

2 Introduction to the 1st International Workshop on Business Process Innovation with Artificial Intelligence (BPAI 2017) Riccardo De Masellis 1, Chiara Di Francescomarino 1, Jana Koehler 2, Fabrizio Maria Maggi 3, Marco Montali 4, Arik Senderovich 5, Biplav Srivastava 6, and Heiner Stuckenschmidt 7 1 Fondazione Bruno Kessler, Trento, Italy {r.demasellis,dfmchiara}@fbk.eu 2 Lucerne University of Applied Sciences and Arts, Lucerne, Switzerland jana.koehler@hslu.ch 3 University of Tartu, Tartu, Estonia f.m.maggi@ut.ee 4 Free University of Bolzano, Bolzano, Italy montali@inf.unibz.it 5 Technion - Israel Institute of Technology, Haifa, Israel sariks@technion.ac.il 6 IBM Research, Yorktown Heights, USA biplavs@us.ibm.com 7 University of Mannheim, Mannheim, Germany heiner@informatik.uni-mannheim.de Abstract. Artificial Intelligence (AI) is receiving high interest from academics, business professionals, and media. It is considered as the next disruptive technology that will significantly impact the workplace and change, innovate, and automate a manifold of business activities. The goal of the workshop was to foster the exchange between AI and Business Process Management (BPM) by taking a closer look at how BPM inspires novel application domains for AI, as well as at how BPM and related fields can benefit from AI solutions. Six full and four short papers were accepted for presentation at the workshop. They stimulated an interesting discussion on potential future synergies between the two disciplines. Keywords: Artificial intelligence Machine learning Business process modelling Business process mining 1 Aims and Scope AI is receiving high interest from academics, business professionals, and media, being considered as the next disruptive technology that will impact millions of jobs and change, innovate, and automate a manifold of business activities. AI examples are:

3 Introduction to the 1st International Workshop on Business Process Innovation 3 Machine Learning and decision-theoretic models for data analysis, e.g., predictive monitoring and customer segmentation; Constraint reasoning and related algorithms as the key technology underlying business rule engines; Search algorithms for process optimization; Intelligent assistants and companions. These are only a few examples of what AI technology can do to improve and reengineer business processes. Recently, major IT companies have developed cognitive services to make AI technology ready to use for developing applications. Many large companies started projects to plug these services into their processes or to develop their own AI solutions based on these services. At the same time, AI researchers are discussing safety issues and identifying important sources of risks in AI solutions. The workshop identified many potential sources for synergies between AI and BPM. On the one hand, several AI solutions can be used in the context of BPM, e.g., planning for adapting or composing business processes, machine learning for process mining and analysis, constraint reasoning for process transformation, verification, and compliance checking. On the other hand, AI will influence the role of humans within a business process and solutions should be developed to address questions such as novel requirements on employee qualification, shared responsibilities between AI and humans, control and impact of automated decision making. Six full and four short papers were presented at the workshop. The focus of the papers ranged from leveraging Machine Learning approaches for addressing BPM problems to applying planning approaches in BPM scenarios on the one hand, and from analyzing event logs using AI techniques to finding optimal paths in business processes on the other hand. In particular, Hinkka and colleagues addressed the problem of classifying business process instances based on structural features derived from event logs. Back and colleagues investigated how different measures for entropy could be used to give insights on the complexity of an event log and could act as an indicator of which paradigm (imperative or declarative) should be used for process mining. Comuzzi presented a framework for calculating optimal execution paths in business processes by relying on workflow hypergraph abstraction and using an ant-colony optimisation customised for the hypergraph traversal. Koehler and colleagues showed how AI can support service processes in a variety of ways by proposing three intelligent assistants that support service employees in their complex tasks. Baldoni and colleagues proposed to enrich the definition of business artifact with a normative layer by relying on a multiagent systems approach. De Masellis and colleagues focused on automatically repairing traces with missing information by notably considering not only activities but also data manipulated by them. Wiśniewski and Kluza presented a method of business process composition based on constraint programming. Rietzke and colleagues presented a semantically oriented business process visualization approach developed using a knowledge-based system. Chesani and colleagues leveraged on the abductive declarative language SCIFF for the realization of an event log generator. Finally, Sulis and Di

4 4 R. De Masellis et al. Leva applied simulation approaches to a real case study of an hospital emergency department. The importance of the synergy between the AI and BPM fields also came out in the keynote by Andrea Marrella. He showed how automated planning techniques can be leveraged to enable new levels of automation and support for BPM. He discussed several concrete examples of successful application of AI techniques to the different stages of the BPM lifecycle. The workshop ended with an interesting panel session about the relevance of the synergy between AI and BPM for both fields. Moreover, the discussion covered the future of the BPAI workshop. Everybody agreed to continue this experience at the next year s BPM conference and to take the chance to further foster the interaction between AI and BPM by targeting with the call for paper also other research areas at the intersection between the two fields. To sum up, the workshop received a significant number of submissions and a good attendance. It clearly showed the strong interest of the BPM community in how the AI and BPM fields can potentially fertilize each other. Of course, many of the questions raised in the call for papers remained unaddressed by this year s submissions, e.g., assessing the business risks of AI technologies, understanding the interplay of humans and robots in business processes, using AI technologies such as AI planning to create dynamic business processes. However, we believe that this only shows how much more can be done in the future in this research direction. 2 Workshop Co-organizers Riccardo De Masellis Chiara Di Francescomarino Jana Koehler Fabrizio Maria Maggi Marco Montali Arik Senderovich Biplav Srivastava Heiner Stuckenschmidt Lucerne University of Applied Sciences and Arts, Switzerland Free University of Bolzano, Italy Technion - Israel Institute of Technology, Israel IBM Research, USA University of Mannheim, Germany 3 Program Committee Rama Akkiraju Ralph Bergmann Andrea Burattin Federico Chesani Raffaele Conforti Giuseppe De Giacomo Claudio Di Ciccio IBM Almaden Research, USA University of Trier, Germany University of Innsbruck, Austria University of Bologna, Italy Queensland University, Australia Vienna University of Economics and Business, Austria

5 Introduction to the 1st International Workshop on Business Process Innovation 5 Peter Fettke Chiara Ghidini Anna Leontjeva Henrik Leopold Andrea Marrella Paola Mello Fabio Patrizi Giulio Petrucci Andreas Rogge-Solti Ute Schmid Irene Teinemaa Stefano Teso Ingo Weber German Research Center for Artificial Intelligence (DKFI), Germany Vrije Universiteit Amsterdam, Netherlands University of Bologna, Italy Vienna University of Economics and Business, Austria University of Bamberg, Germany University of Trento, Italy University of New South Wales, Australia