Documentation to Digital Model: A Python Based Simulation of a Drone Assembly Line

Abstract

Digital simulation models serve as the foundational component of Digital Twins for enabling real-time monitoring, predictive analysis and data driven decision making across production systems. As industries aim to evolve from Industry 4.0, the demand for flexible, transparent and cost efficient simulation tools continues to grow. Traditional DES platforms offer strong modelling capabilities, but are often limited by reduced customizability and challenges in integrating real-time data streams. To address these limitations, this thesis develops a Python based digital model of a semi-automated drone assembly line and evaluates its feasibility as a foundation for a future Digital Twin. The model was implemented using a modular architecture that replicates the physical system’s conveyor, docking-station logic, timing dependencies and PLC document workflows. Verification was performed against the factory’s PLC documentation and validation was conducted through scenario based experiments and cross comparison with a similar model in Siemens Plant Simulation. The python based digital model developed successfully reproduces both structural and behavioural characteristics of the real system, achieving an average accuracy of 87% similarity in production time performance and 84% average accuracy in station cycle time, though station cycle time performance varied between 60% to 98% in various scenarios. These findings demonstrate that Python provides a customizable and transparent platform for developing Digital Twin simulation models in discrete manufacturing environments.