DEC2H 2020

8th International Workshop on DEClarative, DECision and Hybrid approaches to processes (DEC2H 2020) In conjunction with BPM 2020

Workshop: , Seville, Spain

About DEC2H

Workshop Theme

Processes and business process models involve rules and decisions describing the premises and possible outcomes of specific situations. However, rules and decisions are often implicit in process flows, process activities or in the head of employees (tacit knowledge). To make them explicit, they must be discovered using state-of-art techniques. For knowledge-intensive processes, it is common that rules and decisions, as opposed to the process-flow, define the allowed behaviour of a process. For example, the major purpose of an insurance claim process is to ensure that the rules governing the claim are being followed and to arrive at a final decision.

In particular, Knowledge-intensive Processes (KiPs) rely on such rules and decisions to guide the workflows of all process stakeholders. Examples include checking creditworthiness in a financial process, claim acceptance in an insurance process, eligibility decisions in social security, etc. These are highly variable situations, and the processes supporting them are very flexible by nature. This variability and flexibility calls for explicit statement of the underlying rules and decisions when describing such processes.

While traditional imperative notations such as BPMN excel at describing “happy paths”, they are inadequate for modelling rules and decisions. This follows from the fact that imperative notations describe possible behaviour as alternative, restricted flows. Encompassing all possible variations makes imperative models cluttered and thus impractical in highly flexible scenarios. Against this background, a new declarative modelling paradigm has been proposed that aims to directly capture the business rules or constraints underlying the process. The approach has gained momentum, and in recent years several declarative notations have emerged, such as Declare, DCR Graphs, DMN, CMMN, GSM, eCRG, and DPIL.

In this workshop, we are interested in the application and challenges of decision- and rule-based modelling in all phases of the BPM lifecycle: identification, discovery, analysis, redesign, implementation and monitoring. Contributions may include completed work (research, case studies and tools), but also work-in-progress and position papers.

Download the Call for Papers

Purpose of the workshop

The purpose of the workshop is:

  • To examine the relationship between rules, decisions and processes, including models; not only to model the process, but also to model the rules and decisions.
  • To enhance rule and 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 rule- and decision-based modelling and flow-based modelling.
  • To examine standards (DMN, CMMN, BPMN) and their integration.
  • 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 study the integration between declarative models with traditional imperative models.
  • To show best practices in separating process, rule and decision concerns.

Topics

Topics of interest include, but are not limited to:

Declarative and hybrid (process modelling) approaches

  • Declarative notations (Declare, DCR Graphs, CMMN, GSM, eCRG, DPIL, …)
  • Decision & goal notations (DMN, PDM, …)
  • Declarative and hybrid modelling methodologies
  • Process metrics
  • Process maintenance and flexibility
  • Human-centered and flexible processes
  • Decision rules and processes
  • Decision models and structures
  • Formal analysis (e.g., expressiveness proofs) of declarative and hybrid notations
  • Formal verification (e.g. model-checking and static analysis) of declarative and hybrid models
  • Run-time adaptation of declarative and hybrid process models

Decision mining and declarative/hybrid process mining

  • Decision mining
  • Declarative process mining
  • Hybrid process mining
  • Data mining for decision and declarative/hybrid process analysis
  • Rule mining for decision and declarative/hybrid process analysis
  • Modeling challenges to combine static information of business

Applications of decision- and rule-modelling in BPM

  • Goal-driven processes
  • Knowledge-intensive processes
  • Business process compliance
  • Knowledge workflow management
  • Usability and understandability studies
  • Case studies
  • Tools

Virtual event

COVID-19 Due to the consequences of the COVID-19 pandemic, it has been decided that BPM 2020 will be a fully virtual DEC2H. The same will apply to DEC2H. Further details will follow as we get nearer to the workshop date.

Paper submission

Submission guidelines

Prospective authors are invited to submit papers on any of the topics of the workshop. We are interested in research, work-in-progress, position, case-study and tool papers, either in long (not exceeding 12 pages) or short (not exceeding 6 pages) format. Only papers in English will be considered. Submitted papers must present original research contributions not concurrently submitted elsewhere.

Authors are requested to prepare submissions according to the Lecture Notes in Business Information Processing (LNBIP) format specified by Springer (see the instructions). The title page must contain a short abstract and a list of keywords, preferably using the list of topics given above. Papers must be submitted electronically via EasyChair: enter the main conference installation (BPM 2020) and select “(Workshop on DEClarative, DECision and Hybrid approaches to processes)” as the submission track.

At least one author of each accepted manuscript must register for the workshop and present the paper. Registration is subject to the terms, conditions and procedure of the main BPM 2020 conference to be found on its website.

Each manuscript 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. The pre-prints will be made available before the workshop starts. The post-proceedings will be published by Springer in the Lecture Notes in Business Information Processing (LNBIP) series, in a single volume dedicated to the proceedings of all BPM 2020 workshops. During a time window after the conference the workshop participants will be granted the free download of the papers.

Download the Call for Papers

Important dates

Abstract submission (optional)

AoE

Manuscript submission

AoE

Notification

Camera-ready version

Workshop


Read the programme of the workshop

Register

Programme

Format

The workshop will begin with a keynote, followed by presentations of accepted papers. For every paper, a slot of 20 minutes is dedicated: 15 minutes for the presentation plus 5 minutes for discussion and Q&A.

Keynote

Andrea Burattin, A robot may not injure a human being: improving (declarative) processes

Abstract

One of the benefits of managing business processes is the ability to improve them, for example, by applying the BPM lifecycle. An increasing effort is put in devising autonomous systems capable of providing such improvements, for example by analyzing historic data or by applying optimization heuristics. This presentation will focus on two key problems related to these technologies: observing the current status of the process and understanding the models used for training autonomous systems. The former suggests analyzing live data (as opposed to historic executions) to ensure that suggested improvements are timely and relevant. The latter relates to the algorithms underlying autonomous systems that require some type of training (that directly affects the way processes are improved): if we do not understand in all details the models use for training, we will end up with unexpected results. Examples of solutions for the two problems will be presented as well.

The keynote speaker

Andrea Burattin is Associate Professor at the Technical University of Denmark. Previously, he worked as Assistant Professor at the same university, and as postdoctoral researcher at the University Innsbruck (Austria) and at the University of Padua (Italy). In 2013 he obtained his Ph.D. degree from a joint Ph.D. School between the University of Bologna and Padua (Italy). The IEEE Task Force on Process Mining awarded to his Ph.D. thesis the Best Process Mining Dissertation Award for theses defended in 2012-2013. A reworked version of his thesis has then been published as Springer Monograph in the LNBIP series. He served as organizer of BPI workshop since 2015, and special sessions on process mining at IEEE CIDM since 2013. He is also in the program committee of several conferences such as BPM (senior member), CAiSE, and IJCAI.

Timetable

Time Event
Welcome and opening.
Keynote: Andrea Burattin, A robot may not injure a human being: improving (declarative) processes
On-line coffee break
Konstantinos Varvoutas and Anastasios Gounaris,
Evaluation of Heuristics for Product Data Models
Vedavyas Etikala, Ziboud Van Veldhoven and Jan Vanthienen,
Text2Dec: Extracting Decision Dependencies from Natural Language Text for Automated DMN Decision Modelling
On-line coffee break
Stephan Haarmann and Mathias Weske,
Data Object Cardinalities in Flexible Business Processes
Leandro Galrinho, João Costa Seco, Søren Debois, Thomas Hildebrandt, Håkon Normann and Tijs Slaats,
ReDa: Reactive Data-driven Processes as Graph Databases
Closing

People

Program Committee

  • Rafael Accorsi, PwC, Switzerland
  • Bart Baesens, KU Leuven, Belgium
  • Andrea Burattin, Technical University of Denmark, Denmark
  • Josep Carmona, Universitat Politècnica de Catalunya, Spain
  • João Costa Seco, Universidade Nova de Lisboa, Portugal
  • Massimiliano de Leoni, University of Padua, Italy
  • Johannes De Smedt, University of Edinburgh, United Kingdom
  • Jochen De Weerdt, KU Leuven, Belgium
  • Chiara Di Francescomarino, Fondazione Bruno Kessler, Italy
  • Rik Eshuis, Eindhoven University of Technology, The Netherlands
  • Robert Golan, DBmind Technologies Inc., United States
  • María Teresa Gómez-López, Universidad de Sevilla, Spain
  • Xunhua Guo, Tsinghua University, China
  • Thomas Hildebrandt, University of Copenhagen, Denmark
  • Amin Jalali, Stockholm University, Sweden
  • Krzysztof Kluza, AGH University of Science and Technology, Poland
  • Fabrizio M. Maggi, University of Tartu, Estonia
  • Andrea Marrella, Sapienza University of Rome, Italy
  • Marco Montali, Free University of Bozen-Bolzano, Italy
  • Artem Polyvyanyy, University of Melbourne, Australia
  • Hajo A. Reijers, Utrecht University, The Netherlands
  • Flavia M. Santoro, Universidade do Estado do Rio de Janeiro, Brazil
  • Stefan Schönig, Universität Regensburg, Germany
  • Lucinéia H. Thom, Universidade Federal do Rio Grande do Sul, Brazil
  • Han van der Aa, Humboldt University of Berlin, Germany
  • Wil M.P. van der Aalst, RWTH Aachen University, Germany
  • Barbara Weber, University of St. Gallen, Switzerland
  • Mathias Weske, Hasso Plattner Institute, University of Potsdam, Germany

Organisers

Contact the chairs at the following email address: dec2h [at] di [dot] uniroma1 [dot] it