You are here: Home / Departments and Research Centers / LIRIS / Research / DeMiMoP’13 / Home | 1st Int. Workshop on Decision Mining & Modeling for Business Processes (DeMiMoP’13)

Home | 1st Int. Workshop on Decision Mining & Modeling for Business Processes (DeMiMoP’13)

1st International Workshop on Decision Mining & Modeling for Business Processes (DeMiMoP’13)

In conjunction with BPM 2013 Beijing, China, August 26th - 30th 2013
 

1st International Workshop on Decision Mining & Modeling for Business Processes (DeMiMoP’13)

Held in conjunction with BPM 2013 Beijing China, August 26th - 30th 2013

Submissions are now closed.

The DeMiMoP’13 is the 1st workshop aiming at discovering and utilizing the decision process. The goals of this workshop include:
 - (i) to examine the relationship between decisions and processes;
 - (ii) to enhance decision mining based on process data;
 - (iii) to examine decision goals, structures, and their connection with business processes, in order to find a good integration between decision modeling and process modeling;
 - (iv) to study how different process models can be designed to fit a decision process, according to various optimization criteria, such as throughput time, use of resources, etc.;
 - (v) to show best practices in separating process and decision concerns.


Important Dates

Deadline for workshop paper submission: 25 May 2013 8 June 2013
Notification of acceptance: 25 June 2013
Camera-ready final submission: 23 July 2013
Workshop day: 26 August 2013

Workshop Program

The DeMiMoP’13 workshop will be held on 26 August 2013, from 9:00h until 17:00h. The workshop will start with a keynote talk followed by presentations of the accepted papers.

09:00 - 09:10 Opening
09:10 - 10:00 Introduction by the Chairs: "Business decisions and business processes"
10:00 - 10:30 Coffee break
10:30 - 12:00

Paper session 1

  • Automatic Generation of Business Process Models based on Attribute Relationship Diagrams
    Krzysztof Kluza and Grzegorz J. Nalepa
  • Enriching Business Process Models with Decision Rules
    Semra Catalkaya, David Knuplesch, Carolina Chiao and Manfred Reichert
  • Validating and Enhancing Declarative Business Process Models Based on Allowed and Non-Occurring Past Behavior
    Seppe Vanden Broucke, Filip Caron, Jan Vanthienen and Bart Baesens
12:00 - 13:00 Lunch break
13:00 - 14:00

Paper session 2

  • Constructing Decision Trees from Process Logs for Performer Recommendation
    Aekyung Kim, Josue Obregon and Jae-Yoon Jung
  • An Exploratory Approach for Understanding Customer Behavior Processes Based on Clustering and Sequence Mining
    Alex Seret, Seppe Vanden Broucke, Bart Baesens and Jan Vanthienen
14:00 - 15:00 Keynote: "Decision mining: are we looking under the streetlight?"
Pnina Soffer, University of Haifa, Israel (keynote details)
15:00 - 15:30 Coffee break
15:30 - 16:30 Closing roundtable: "Discussion, standards and future plans"
18:00 - 20:00 Workshop dinner

Description

Contemporary socio-economic factors, such as globalization and mergers & acquisitions, have resulted in a need for streamlining business operations and induced a plethora of business process optimization and reengineering projects. As such, researchers and practitioners focus heavily on the optimization of end-to-end business processes. Business Process Management (BPM) and its life cycle activities – design, modeling, execution, monitoring and optimization of business processes – have indeed become a crucial part in business management.

Most processes and business process models incorporate decisions of some kind. Decisions are typically based upon a number of business (decision) rules that describe the premises and possible outcomes of a specific situation. Since these decisions guide the activities and workflows of all process stakeholders (participants, owners), they should be regarded as first-class citizens in Business Process Management. Sometimes, the entire decision can be included as a decision activity or as a service (a decision service). Typical decisions are: creditworthiness of the customer in a financial process, claim acceptance in an insurance process, eligibility decision in social security, etc. The process then handles a number of steps, shows the appropriate decision points and represents the path to follow for each of the alternatives.

Business decisions are important, but are often hidden in process flows, process activities or in the head of employees (tacit knowledge), so that they need to be discovered using state-of-art intelligent techniques. Decisions can be straightforward, based on a number of simple rules, or can be the result of complex analytics (decision mining). Moreover, in a large number of cases, a particular business process does not just contain decisions, but the entire process is about making a decision. The major purpose of a loan process e.g., or an insurance claim process, etc., is to prepare and make a final decision. The process shows different steps, models the communication between parties, records the decision and returns the result.

It is not considered good practice to model the detailed decision paths in the business process model. Separating rules and decisions from the process simplifies the process model (separation of concerns). Using this workshop we try to extend the reach of the BPM 2013 audience towards the decisions and rules community and increase the integration between different modeling perspectives.

Purpose of the workshop

The decision process is not the same thing as the decision structure (including requirements, dependencies, goals, data sources, etc.) because a specific process is only one possible way to model and implement a decision. There may be more possible process models and implementations for a specific decision. And the same decision could be used in multiple processes.

The purpose of the workshop, therefore, is:

  • To examine the relationship between decisions and processes, including models not only to model the process, but also to model the decisions.
  • To enhance decision mining based on process data (e.g. event logs).
  • To examine decision goals, structures, and their connection with business processes, in order to find a good integration between decision modeling and process modeling.
  • To study how different process models can be designed to fit a decision process, according to various optimization criteria, such as throughput time, use of resources, etc.
  • To show best practices in separating process and decision concerns.

 

Topics of interest

Topics of interest include, but are not limited to:

  • Decisions, rules and processes
  • Decision mining
  • Decision models and structures
  • Data mining, rule mining, process mining
  • Goal driven processes
  • Process metrics
  • Process maintenance and flexibility
  • Human-centered and flexible processes
  • Case studies

Keynote

The workshop features a keynote by Pnina Soffer, University of Haifa, Israel, entitled: "Decision mining: are we looking under the streetlight?"

Abstract
Decision mining in business processes aims at capturing the rationale and guiding decisions that are made in business processes. A basic premise is that decisions are made in a specific context, intending to achieve defined goals.  Hence, to understand decisions we should consider three elements: decisions, context, and goals, and the relationships among them. The talk highlights challenges related to each. 
  • Decisions: a major challenge is to gain a complete view of the decisions entailed in a business process. We argue that relying on activity-based process models leads to overlooking decisions that are not explicitly represented in these models. Imperative process models emphasize path selection decisions, represented by choice splits. Declarative models relate to decisions of selecting an activity to perform from the ones available at a given situation. However, many of the decisions made in the course of a business process are embedded within activities. Such decisions become apparent when a state-based view of a process is adopted, so the detailed state that follows an activity reflects decisions that have been made. A state-based view, however, is difficult to manage and visualize, and requires research efforts to produce methods that are computationally efficient and usable for human reasoning.
  • Context: context plays a major role in decision making. The context of a specific decision would be the detailed state known at decision time. It can reflect previous actions taken through the business process, some initial state and case properties, and events that occurred during the process. The challenge here is twofold. First, given the large amount of data available in business situations, we need to identify the parameters that should be considered as the relevant context of a certain decision. Assuming that different contextual conditions should drive different decisions when striving to achieve specific goals, context identification can rely on mining the decisions and the outcomes achieved by them. Second, even with the large amount of available data, context identification should always take a partial information assumption, since we do not have a complete deterministic knowledge of all the variables that might affect the outcomes of a decision, nor can we assume they are included in the available data.
  • Goals: decisions are made with the intention of achieving desirable goals. Since a goal can be specified as conditions on the state, a state-based view supports goal specification as well. Three main challenges can be related to goal specification. First, detailed goals might depend on the context, as what would be considered an excellent achievement given some initial state might be considered below expectations with a different starting point. Second, goals can be "hard", namely, their achievement is possible to assess on True/False scale, or "soft", namely goals whose achievement can only be assessed relatively, by comparison of several cases. Decision mining should be able to relate to both kinds of goals. Third, while most business processes have a clear ending point where goal achievement can be assessed, some processes are long-lasting ones without such clear ending. For these processes, means of goal assessment and assessment points need to be defined.
In conclusion, many challenges are still faced by decision mining research. A main issue raised in the talk is the need to anchor decision mining in a state-based process view, since the common activity-based view can only provide a partial understanding. Developing models as well as methods that build upon this view is a main challenge still to be addressed.
SPEAKER BIO

Dr. Pnina Soffer is a senior lecturer in the Information Systems Department at the University of Haifa. She received her BSc (1991) and MSc (1993) in Industrial Engineering from the Technion, Ph.D in Information Systems Engineering from the Technion (2002). Her research deals with business process modelling and management, requirements engineering, and conceptual modelling, addressing issues such as goal orientation, flexibility, learning techniques for improving decision making, and context-aware adaptation. Her research has appeared in over 90 conference and journal papers, in journals such as Journal of the AIS, European J of IS, Requirements Engineering, Information Systems, and others. She has served as a guest editor of a number of journal special issues related to various business process topics, and is a member of the editorial board of several journals, including Journal of the AIS. Pnina has served in program committees of numerous conferences, including CAiSE, BPM, ER, and many more. She is a member of the CAiSE steering committee and an organizer of the BPMDS working conference since 2004. She has served in different roles in conference organizing committees, such as Forum Chair and Workshop Chair in CAiSE and in BPM, and leads the organizing team of BPM 2014 in Haifa.

Submission

Papers should be submitted in advance and will be reviewed by at least three members of the program committee. Only papers in English will be accepted and must present original research contributions not concurrently submitted elsewhere. The length of a paper must not exceed 12 pages. Authors are requested to prepare submissions according to the LNCS/LNBIP format specified by Springer. The title page must contain a short abstract and a list of keywords, preferably using the list of topics given above.

Each paper will be reviewed by at least three program committee members guaranteeing that only papers presenting high quality work and innovative research in areas relevant to the workshop theme will be accepted. All accepted papers will appear in the workshop proceedings published by Springer in the Lecture Notes in Business Information Processing (LNBIP) series. There will be a single LNBIP volume dedicated to the proceedings of all BPM workshops.

Accepted papers imply that at least one of the authors will register for the BPM 2013 conference and present the paper at the DeMiMoP’13 workshop.

Papers are submitted electronically through EasyChair:

 

Organizers

Prof. dr. Bart Baesens
Department of Management Informatics
KU Leuven
Naamsestraat 69 - bus 3555, 3000 Leuven, Belgium
Tel: +32 16 326884
E-mail: bart.baesens@kuleuven.be
URL: http://www.bartbaesens.com
Prof. dr. Guoqing Chen
Department of Management Science and Engineering
School of Economics and Management
Tsinghua University
258 Weilun Building, Tsinghua University, Haidian, 100084 Beijing, China
Tel: +86 (10) 62772940
E-mail: chengq@sem.tsinghua.edu.cn
URL: http://www.sem.tsinghua.edu.cn/en/chengq
Prof. dr. Jan Vanthienen (corresponding organizer)
Department of Management Informatics
KU Leuven
Naamsestraat 69 - bus 3555, 3000 Leuven, Belgium
Tel: +32 16 326878
E-mail: jan.vanthienen@kuleuven.be
URL: http://www.econ.kuleuven.ac.be/.../vthienen/default.htm
Prof. dr. Qiang Wei
Department of Management Science and Engineering
School of Economics and Management
Tsinghua University
443 Weilun Building, Tsinghua University, Haidian, 100084 Beijing, China
Tel: +86 (10) 62789824
E-mail: weiq@sem.tsinghua.edu.cn
URL: http://www.sem.tsinghua.edu.cn/en/weiq

Program Committee

  • Dimitris Karagiannis, Universität Wien, Austria
  • Xunhua Guo, Tsinhua University, China
  • Hajo A. Reijers, Eindhoven University of Technology, The Netherlands
  • Robert Golan, DBmind technologies, United States
  • Markus Helfert, Dublin City University, Ireland
  • Leszek Maciaszek, Wroclaw University of Economics, Poland
  • Pericles Loucopoulos, Loughborough University, England
  • Josep Carmona, Universitat Politècnica de Catalunya, Spain
  • Jochen De Weerdt, Queensland University of Technology, Australia
  • Seppe vanden Broucke, KU Leuven, Belgium
  • Filip Caron, KU Leuven, Belgium