Automated Extraction of Key Performance Indicators Based on PNML and Decision-makers input

  • Type:Master Thesis
  • Supervisor:

    Ohad Daniel

    Prof. Sanja Lazarova-Molnar
     

Description

Problem: 

Current extraction methodologies of KPIs for simulation experiments of Digital Twins are often made manually or extracted automatically in a simplified manner (based on one extracted transition or place). This gap hinders the decision-support capabilities of Digital Twins.

Goal: 

The thesis aims to formalize the decision-maker input required for the systematic extraction of KPIs from PNML files. This goal of this formalization is to enable the automatic extraction of KPIs from PNML files and decision-maker input. The thesis will include natural language processing methodologies to accurately extract the most relevant KPIs based on the extracted Petri nets and the decision-maker input.  

Required Skills and Knowledge: 

  • Understanding of Digital Twin technology and its applications.

  • Discrete Event Simulation.

  • Knowledge of Python programming language.

  • Experience with NLP is a plus.