Business Activity Monitoring
Business process management focuses on the working procedures of a company to foster the company’s competitive advantage on the market. One can decompose process management into several tasks with process monitoring among them. The goal of process monitoring is to trace the state of individual processes and deliver statistics on their performance. An example of a process performance monitoring task is the evaluation of the mean value of handling one customer request. One particular type of monitoring is monitoring of business process activities. Its goal is to provide information about the state of specific business operations. Thereby, activity monitoring allows business users to discover process performance bottlenecks rapidly, align processes, and make better informed decisions. Seminar “Business Activity Monitoring” provides the master students an opportunity to make their contribution to this emerging field of business process management. While the textual description for some topics can be found further, you can also consider the
opening presentation slides.
Important Dates
The submission deadline is 23:59:59 CET.
29 July 2011 - final paper submission (participants submit the final paper version)
Slides
Seminar Structure
The seminar will contain two introductory lectures:
- how to write a research paper
- how to present a scientific talk
Topic Proposals
- Ahmed Awad
- From Compliance Patterns to ECA Monitoring Rules
- Context: Compliance monitoring
- Task: Given a set of compliance patterns about inter/intra task timing constraints and resource allocation constraints, develop an approach that generates the corresponding Event-Condition-Actions rules that predict/detect the violation of such rules. To develop such mapping, the activity lifecycle supported by the process execution engine must be taken into consideration. All intermediate events relevant to certain compliance pattern must be as well defined based on the raw events emitted by the engine
- Literature:
- Christopher Giblin, Samuel Mueller, and Birgit Pfitzmann. From Regulatory Policies to Event Monitoring Rules: Towards Model-Driven Compliance Automation. Research Report RZ 3662, IBM Zurich Research Laboratory, 2006.
- Kioumars Namiri and Nenad Stojanovic. Pattern-Based Design and Validation of Business Process Compliance. In Robert Meersman and Zahir Tari, editors, OTM Conferences (1), volume 4803 of Lecture Notes in Computer Science, pages 59-76. Springer, 2007.
- Wil M. P. van der Aalst and Ana Karla A. de Medeiros. Process mining and security: Detecting anomalous process executions and checking process conformance. Electr. Notes Theor. Comput. Sci., 2005.
- Christian Wolter and Andreas Schaad. Modeling of Task-Based Authorization Constraints in BPMN. In Gustavo Alonso, Peter Dadam, and Michael Rosemann, editors, BPM, volume 4714 of Lecture Notes in Computer Science, pages 64-79. Springer, 2007.
- Rami Eid-Sabbagh
- Business Activity Monitoring in the Public Administration
- Context: The use of business activity monitoring is not explored in the public sector. Modernization efforts in the public sectors increase. However, many processes in the public sector are based on laws and regulations, often they are not real time critical. Traits which do not suggest using BAM in the public domain. However in some parts of administration business activity monitoring are used.
- Task: Your task will be to research general aspects of business activity monitoring and matching your findings to a real world scenario, the single point of contact in berlin.
- Research Business activity monitoring aspects from literature
- Identify the requirements for performing BAM
- Identify key areas for the use of BAM
- Match your findings to a real world scenario, Berlin single point of contact
- Evaluate and elaborate on your findings
- Monitoring Web-Application Interaction on the Example of the National Process Library
- Context: Monitoring Web-Application interaction activities gathers important insights on the user’s focus and his habits while visiting the web platform. In the course of the research project “National Process Library” a web platform that allows the exchange of process models and their information is developed. Finding out the interaction patterns and focus areas of its user is of utmost importance for the evaluation and improvement of its functions and structure.
- Task: Your task will be to do theoretical research on monitoring web-application interaction activities, create a general research setup and apply this scenario to the national process library. Evaluate and elaborate on the findings of your research and derive process models out of your results.
- Alexander Luebbe
- Business Activity Monitoring - A Technology Review
- Context: Business Activity Monitoring can be realized by system experts that dig deep to extract the KPIs. But what are the chances for users and savvy admins to monitor process instances and tailor their view to their specific needs?
- Task: Find BAM scenarios in the literature. Identify the main use cases for business activity monitoring. Derive a framework of requirements for BAM. Evaluate exisitng BPMS (maybe engines only) for their BAM capabilities.
- Literature:
- Business Activity Monitoring (thesis, J Kolár)
- Design and Implementation of a Personalized Business Activity Monitoring System (HCI 2007:581-590)
- Richard Welk, Gabriel Cavalheiro, Ajantha Dahanayake: Improving IT-Enabled Sense and Respond Capabilities: An Application of Business Activity Monitoring at Southern International Airlines. J. Cases on Inf. Techn. (JCIT-IGI) 9(4):40-56 (2007)
- Torsten Greiner, Willy Düster, Francis Pouatcha, Rainer von Ammon, Hans-Martin Brandl, David Guschakowski: Business activity monitoring of norisbank taking the example of the application easyCredit and the future adoption of Complex Event Processing (CEP). PPPJ 2006:237-242
- Some easily available engines: ActiveVOS?, Activity/jBPM,apache ODE, YAWL
- Matthias Kunze
- Introducing KPIs to BPMN
- Context: BPMN 2.0 has, for the first time, a metamodel formally defined by means of the Meta Object Facility (MOF). This opens the field to extend the language through well defined methods.
- Task: The student shall introduce the definition of KPIs to BPMN 2.0. For this, they need to (1) understand how KPIs are correlated with process instances, which KPIs are useful in the context of BPM, and how they can be captured. They will also (2) need to study the BPMN specification and how it can be formally extended. Finally, and as the main contribution of this topic (3), the student shall extend BPMN 2.0 with means to capture KPIs in process models. This includes to extend the BPMN metamodel, including a specification of notation (concepts and graphical representation), syntax (properties of valid models), and semantics (esp. transformation semantics towards XSD and operational semantics of KPIs).
- Literature:
- BPMN 2.0 specification: http://www.omg.org/spec/BPMN/2.0/
- Adela del-Río-Ortega, Manuel Resinas und Antonio Ruiz-Cortés. Defining Process Performance Indicators: An Ontological Approach. OTM 2010, Vol. 6426 of LNCS. Springer 2010.
- Design of a Process Data Warehouse
- Context: Business Activity Monitoring (BAM) provides a wide range of opportunities, including analysis, interpretation, controlling, optimization. The basis for them is a warehouse that stores process data and process related data. Such a warehouse can answer short-term questions to to support operational decisions as well as long-term questions to improve process quality or support strategic decisions, if it has been well designed with regard to fast query processing. Similar work has long been done for general business data: Data Warehouses (OLAP).
- Task: The student shall (1) review literature towards data warehousing, architectures, best practices, etc to get an understanding of the basics. Additionally, (2) questions that are relevant in terms of BAM and Business Process Intelligence shall be explored and discussed---these questions will guide the design of a data ware house. Finally (3), the student shall design a process data warehouse (formally) that stores history of processes (logs) and related data. An example shall support the theoretic concerns.
- Literature:
- J. Eder, G. E. Olivotto, and W. Gruber. A Data Warehouse for Workflow Logs. In Proceedings of the First International Conference on Engineering and Deployment of Cooperative Information Systems, EDCIS ’02, pages 1–15, London, UK, UK, 2002. Springer-Verlag.
- D. Grigori, F. Casati, M. Castellanos, U. Dayal, M. Sayal, and M.-C. Shan. Business process intelligence. Computers in Industry, 53(3):321–343, April 2004.
- M. Z. Muehlen. Process-driven Management Information Systems - Combining Data Warehouses and Workflow Technology. In Proceedings of the 4th International Conference on Electronic Commerce Research (ICECR-4), Dallas (TX, 2001.
- B. Mutschler and M. Reichert. Aktuelles schlagwort: Business process intelligence. EMISA Forum, 26(1):27–31, January 2006.
- K. Shahzad and P. Johannesson. An evaluation of process warehousing approaches for business process analysis. In Proceedings of the International Workshop on Enterprises & Organizational Modeling and Simulation, EOMAS ’09, pages 10:1–10:14, New York, NY, USA, 2009. ACM.
- Activity warehouse: Data management for business activity monitoring. ICEIS 2007:137-144
- Gabriel Cavalheiro, Ajantha Dahanayake, Richard J. Welke: Combining Business Activity Monitoring with the Data Warehouse for Event-Context Correlation - Examining the Practical Applicability of this BAM Approach. ICEIS 2006:263-268
- What do you want to know about your processes (running)? A user study
- Context: Business Activity Monitoring (BAM) and Business Process Intelligence (BPI) address insights about processes that are currently running or have been run in the past. Many interesting questions already exist from a scientific point of view, however, we lack understanding of the desires of the prospected users of these methods, i.e., process designers, process owners, management, controlling, etc.
- Task: The student shall (1) get in contact with several business process experts in organizations and discuss with them the topic of BAM/BPI. These interviews shall give a direction for an online survey (2) that shall be conducted by the student. The result (3) shall be analyzed and aggregated to get an overview of topics currently relevant to companies.
- Literature:
- J. Eder, G. E. Olivotto, and W. Gruber. A Data Warehouse for Workflow Logs. In Proceedings of the First International Conference on Engineering and Deployment of Cooperative Information Systems, EDCIS ’02, pages 1–15, London, UK, UK, 2002. Springer-Verlag.
- B. Mutschler and M. Reichert. Aktuelles schlagwort: Business process intelligence. EMISA Forum, 26(1):27–31, January 2006.
- Andreas Rogge-Solti
- Techniques for Mining Medical Processes from Unstructured Data
- Context: Hospitals typically use an information system (HIS) to track their patient's data and treatment findings. However, since these treatments are not coordinated by an execution engine, as it is the case in typical business process management scenarios, the data in these HIS are not structured and readily interpretable. Typically, the track of a patient's treatment has to be reconstructed based on the produced documents on the way along the treatment procedure, e.g., the time stamps on the files indicate roughly the duration of the treatment steps.
- Task: Analyze existing use cases and give an overview of different techniques proposed by researchers to tackle particular healthcare specific problems. Find techniques for monitoring live process instances under the assumption, that process models and links to the produced documents exist.
- Literature:
- [1] Mans, R.S, Schonenberg, M. H., Song, M., Aalst, W.M.P. and Bakker, P.J.M. Application of process Mining in health care-a case study in a Dutch hospital. Biomedical engineering systems and technologies, International Joint conference, BIOSTEC 2008 Funchal, Madeira, Portugal, January 28-31, 2008.
- [2] Poelmans, J., Dedene, G., Verheyden, G., Van der Mussele, H., Viaene, S. and Peters, E. Combining business process and data discovery techniques for analyzing and improving integrated care pathways. Lecture Notes in Computer Science, Advances in Data Mining. Applications and Theoretical Aspects, 10th Industrial Conference (ICDM), Leipzig, Germany, July 12-14, 2010.
- Business Process Management Systems vs. Clinical Decision Support Systems
- Context: Business Process Management Systems and Clinical Decision Support Systems have been developed independently from one another, yet there are some similarities. There has been research to apply business process management techniques in healthcare to deal with specific issues like flexibility [1], however, this research is mostly of academic nature.
- Task: Develop a taxonomy to compare the two different systems and point out the biggest differences and similarities. Is it possible to derive an abstract interface based on this taxonomy? If so, develop the conceptual architecture to integrate these two systems into one hybrid system. Otherwise, pick a BPMS of your choice and sketch extension points for decision support.
- Literature:
- [1] R. Lenz and M. Reichert, IT Support for Healthcare Processes-Premises, Challenges, Perspectives., Data and Knowledge Engineering, Volume 1, 2007.
- [2] Terenziani, P. A hybrid multi-layered approach to the integration of workflow and clinical guideline approaches. ProHealth? 2009, Business Process Management Workshops, 2010.
- Sergey Smirnov
- State Consistency of Process Models Belonging to Different Abstraction Levels
- Context: One business process can be described through several process models. Among these multiple models we study those that describe the process with different precision. To simplify the task we can study model pairs, an abstract model and a more detailed model. We assume that each activity of abstract model is refined with several activities of a detailed model. If the detialed model is used for process execution, the abstract model can be used for monitoring of the business process. In this case the process state is decribed by the activity states. We assume that there is a state propagation mechanism that defines the states of activities in the abstract model given the activity states in the detailed model. However, activity refinement coupled with state propagation leads to behavioral inconsistencies within the abstract model.
- Task: The researcher has to investigate the reasons for such inconsistencies, provide their classification, and possibly suggest methods to resolve them.
- Literature:
- Sergey Smirnov, Armin Zamani Farahani, and Mathias Weske. State Propagation in Abstracted Business Processes. Technical Report of Hasso Plattner Institute, 47 (2011).
- Matthias Weidlich, Remco Dijkman, and Mathias Weske. Deciding Behaviour Compatibility of Complex Correspondences between Process Models. Proceedings of the 8th International Conference on Business Process Management (BPM'10). Hoboken, NJ, US, September 14-15, 2010. LNCS 6336, pp. 78-94, Springer-Verlag.
- Using Value Chain Models to Monitor Process Execution Progress
- Context: Recently several methods for BPMA has been developed. All of them discuss how an abstract process model can be obtained from the given process model in the same notation. For instance, an abstract EPC is generated from an EPC, or a high-level BPMN model is obtained from a BPD. However, in practice it is often the case that an abstrac model is created in another notation, e.g., value chain. While detailed process models are used for execution, high-level models can be used for monitoring of process execution progress. If the detailed model is annotated with activity execution information, it is possible to evaluate the process execution progress quantitevly. For instance, if average activity execution time is given, it is possible to predict the remaining process/activity execution time.
- Task: Within this topic the researcher has to get acquainted with existing abstraction methods and choose one of them. This method has to be extended towards value chain model generation. Further, the researcher has to suggest: 1) a method for evaluation of current value chain activity duration and 2) a method for evaluation of remaining value chain activity duration.
- Literature:
- Sergey Smirnov, Matthias Weidlich, and Jan Mendling. Business Process Model Abstraction based on Behavioral Profiles. Proceedings of the 8th International Conference on Service Oriented Computing, San Francisco, CA, USA, December 2010. Springer Verlag.
- Basten, T., Aalst, W.: Inheritance of Behavior. JLAP 47(2) (2001) 47–145
- Marlon Dumas, Luciano García-Bañuelos, Artem Polyvyanyy, Yong Yang, and Liang Zhang. Aggregate Quality of Service Computation for Composite Services. Proceedings of the 8th International Conference on Service Oriented Computing (ICSOC). San Francisco, CA, US, December 2010
- Matthias Weidlich
- Monitoring of Activity Cardinality Constraints
- Context: Business operations are often supported by process-aware information systems. Still, these systems typically do not directly execute the process model, but provide the flexibility to deviate from the normative model. Once processes are automated in a complex event processing environment, one may generate event queries to check whether the actual processing is in line with the process model.
- Task: There are results available on how order and exclusiveness constraints translate into complex event queries. Within the seminar, these results should be extended towards cardinality constraints (e.g., for each A there are two Bs). As a formal grounding boolean combinations of integer linear constraints may be used, as they have been applied in communication fingerprints.
- Literature:
- Olivia Oanea, Jan Sürmeli, Karsten Wolf: Service Discovery Using Communication Fingerprints. ICSOC 2010:612-618
- M. Weidlich, H. Ziekow, J. Mendling, O. Guenther, M. Weske, N. Desai: Event-based Monitoring of Process Execution Violations. Technical Report. 2011.
- Andreas Meyer
- Business Goals Monitoring
- Context: In Requirements engineering, goal modeling became much attention during the last decade. Generally, these goals are reflected in business processes and drive process control there.
- Task: 1) Define a mapping of the goals between the goal model and the process model (at least for EPC and BPMN 2.0, a generic solution is fine as well)! 2) Deal with the identification of (sub)goal completion in a process! 3) Provide means to alert/ notify if certain goals are not reachable due to current process execution or whether specific goals cannot be reached any more in the observed instance. 4) Provide means for process control based on a) priority of goals b) pure number of goals and c) further identified aspects to fulfill certain requirements, e.g. reaching the most important goals, reach as many goals as possible, ect.
- Literature:
- Lapouchnian - Goal-Oriented Requirements Engineering: An Overview of the Current Research
- Lamsweerde and Letier - From Object Orientation to Goal Orientation: A Paradigm Shift for Requirements Engineering
- Yu - Towards Modeling and Reasoning Support for Early-Phase Requirements Engineering
- Data Object State Related KPI Measuring
- Context: KPIs can be observed by BAM. They may also relate to data objects which are affected by state changes over time. Following, KPIs may be measured by state changes, e.g. the duration of a delivery is maximum three days. Therefore, the time allowed to pass between the states "order confirmed" and "product delivered" is maximum three days.
- Task: 1) Identify KPIs which may be linked to data objects! 2) Provide means to link the identified KPIs to data objects and their states! 3) Elaborate on an automatic linkage! 4) Visualize the results and statistics (you consider useful) in a dashboard. For these tasks, think about single and multiple process instances, multiple data objects, as well as single and aggregated KPIs. For the latter elaborate on means to aggregate KPIs!
- Literature:
- Ortega et al. – Defining Process Performance Indicators: An Ontological Approach
- Steiner - Applying an operational Dashboard to a Customer Support Process (Master Seminar on Process Intelligence, Summer Semester 2010)
Additional Information
- All submissions have to be send via e-mail to sergey..smirnov (at) hpi..uni-potsdam..de
