Declarative Behavioural Rules with Declare

This project demonstrates the use of declarative process modeling using the Declare framework in Python. It focuses on defining and analyzing business process constraints through a flexible, rule-based approach.

🧩 Project Overview

The repository provides an example of how to model and evaluate business processes through behavioral rules. Unlike traditional imperative models, Declare allows the specification of process constraints without prescribing exact execution paths. This flexibility is ideal for scenarios where processes are loosely structured or evolve frequently.

Main Goals

• Explore declarative process modeling using the Declare framework
• Apply constraints to event logs for process analysis
• Test and validate business logic against real or synthetic data

---

📂 Structure

📁 Declarative-Behavioural-Rules
├── declare.ipynb # Main notebook implementing declarative rules
├── unidade_teste_1.csv # Sample dataset representing a process log
└── README.md # Project documentation

Key Components

• declare.ipynb: Contains Python code and explanations for defining Declare constraints, importing and analyzing event logs, and interpreting results.
• unidade_teste_1.csv: A sample CSV file representing an event log, used for testing the behavioral constraints.
• Declare Constraints:
  • Response(A, B): If activity A occurs, B must eventually follow
  • Precedence(B, A): B can only occur if A has occurred before
  • Not Co-Existence(A, B): A and B cannot both appear in the same case (and others)
  • Any other constraints can be checked on the references, page 111.

---

🔧 Technologies Used

• Python 3.10.12
• Jupyter Notebook
• Pandas
• Custom implementation of Declare constraints

---

📈 Skills Demonstrated

• Process mining fundamentals
• Declarative modeling (Declare)
• Data manipulation and validation using event logs
• Rule-based logic implementation
• Use of Jupyter for reproducible data science workflows

---

📌 Use Case

This type of modeling is useful in:

• Compliance checking
• Audit and legal process validation
• Discovering hidden process constraints from log data
• Business process flexibility modeling

---

🚀 Getting Started

To run the notebook:

1. Clone the repository:

   git clone https://github.com/joaopmarinho/Declarative-Behavioural-Rules
   cd Declarative-Behavioural-Rules

2. Install the required Python packages (typically pandas and jupyter)

3. Run declare.ipynb in Jupyter Notebook or Jupyter Lab

📫 About This repository was created as a practical exploration of declarative process logic. It is ideal for showcasing skills in process mining, business rules modeling, and logic validation with real-world applications in BPM (Business Process Management).

---

Let me know if you’d like to tailor it for a specific company or role (e.g., data analyst, BPM consultant, compliance tech), or add a personal note at the end like "Developed as part of a process mining portfolio."

📚 References

• VAN DER AALST, Wil M. P.; CARMONA, Josep (Ed.). Process Mining Handbook. Cham: Springer, 2022. (Lecture Notes in Business Information Processing, v. 448). Disponível em: https://doi.org/10.1007/978-3-031-08848-3. Acesso em: 30 maio 2025.