Declarative Modeling - Semantic Scholar
Declarative Modeling An Academic Dream or the Future for BPM?
Hajo A. Reijers, Tijs Slaats, and Christian Stahl
Modeling approaches imperative
declarative
• overspecification underspecification • academic and industrial tools and languages
PAGE 1
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Research question
?
PAGE 2
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Outline • Workshop
• Quantitative Evaluation
• Qualitative Evaluation
PAGE 3
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
I. The method
PAGE 4
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Organization of the workshop Intro to declarative modeling
Assignment
Discussion + Questionnaire
Declare
DCR-graphs PAGE 5
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
II. Quantitative Evaluation
PAGE 6
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Participants • Ten professionals • Five consultants • Five developers
• Average experience in BPM: >11 years • Average #models read in last 12 months: 15
PAGE 7
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Insight 1: Technique does not matter • consider usefulness and ease of use • pen-and-paper evaluation Declare
PAGE 8
DCR-graph
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Insight 2: Perceived usefulness/ease of use
• The more experienced, the more optimistic regarding usefulness PAGE 9
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
III. Qualitative Evaluation
PAGE 10
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Assignment • Correct solutions for Declare and DCR-graphs Declarative modeling can be taught • Graphical notation • Too academic • Neither convincing nor intuitive
• Informal descriptions of constraints not always helpful
PAGE 11
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Opportunities for a declarative approach • “Always at least some structured subprocess”
• Hybrid approach
PAGE 12
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Requirements • Need for declarative specifications
• Readability
• Candidates: spaghetti models, model freedom
PAGE 13
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Use cases • Process evolution
• Communication model
PAGE 14
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Limitations • Get a declarative specification
• Component based systems
• Existing tools and their usability
PAGE 15
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Tool requirements
Recommendations
Constraint visualization
BPM System Usability Derive missing constraints
PAGE 16
Detect conflicting constraints
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Take-home picture
?
• Complete BPM life cycle for hybrid models
PAGE 17
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Hajo A. Reijers, Tijs Slaats, and Christian Stahl
Modeling approaches imperative
declarative
• overspecification underspecification • academic and industrial tools and languages
PAGE 1
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Research question
?
PAGE 2
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Outline • Workshop
• Quantitative Evaluation
• Qualitative Evaluation
PAGE 3
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
I. The method
PAGE 4
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Organization of the workshop Intro to declarative modeling
Assignment
Discussion + Questionnaire
Declare
DCR-graphs PAGE 5
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
II. Quantitative Evaluation
PAGE 6
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Participants • Ten professionals • Five consultants • Five developers
• Average experience in BPM: >11 years • Average #models read in last 12 months: 15
PAGE 7
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Insight 1: Technique does not matter • consider usefulness and ease of use • pen-and-paper evaluation Declare
PAGE 8
DCR-graph
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Insight 2: Perceived usefulness/ease of use
• The more experienced, the more optimistic regarding usefulness PAGE 9
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
III. Qualitative Evaluation
PAGE 10
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Assignment • Correct solutions for Declare and DCR-graphs Declarative modeling can be taught • Graphical notation • Too academic • Neither convincing nor intuitive
• Informal descriptions of constraints not always helpful
PAGE 11
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Opportunities for a declarative approach • “Always at least some structured subprocess”
• Hybrid approach
PAGE 12
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Requirements • Need for declarative specifications
• Readability
• Candidates: spaghetti models, model freedom
PAGE 13
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Use cases • Process evolution
• Communication model
PAGE 14
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Limitations • Get a declarative specification
• Component based systems
• Existing tools and their usability
PAGE 15
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Tool requirements
Recommendations
Constraint visualization
BPM System Usability Derive missing constraints
PAGE 16
Detect conflicting constraints
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
Take-home picture
?
• Complete BPM life cycle for hybrid models
PAGE 17
Reijers et al.: Declarative Modeling—An Academic Dream or the Future for BPM?
