Research Project J2‑4470

Reliability and Efficiency Research of Edge Computing in Smart Factories Using 5G Technologies

(1 October 2022 – 30 September 2025)

1. Basic information about the project and the (co)funder

The J2‑4470 research project is a fundamental research project (co)funded by ARIS – the Slovenian Research and Innovation Agency. The project is carried out at the University of Ljubljana, Faculty of Mechanical Engineering, within the field of Industrial Production and Technology.

The objective of the project is to develop a methodology for integrating 5G technology and edge computing into smart manufacturing systems and to upgrade the reference architectural model LASFA. The project contributes to increasing the flexibility, responsiveness, and efficiency of production processes, as well as to the development of future digital factories, digital twins, and Industry 4.0/5.0 systems.

2. Project team

Project Leader

• prof. dr. Niko Herakovič – https://cris.cobiss.net/ecris/si/sl/researcher/7006

Člani projektne skupine

• dr. Marko Šimic – https://cris.cobiss.net/ecris/si/sl/researcher/34044
• dr. Miha Pipan – https://cris.cobiss.net/ecris/si/en/researcher/38205
• dr. Denis Jankovič – https://cris.cobiss.net/ecris/si/sl/researcher/49966
• dr. Mihael Debevec – https://cris.cobiss.net/ecris/si/sl/researcher/13146
• dr. Matevž Resman – https://cris.cobiss.net/ecris/si/sl/researcher/45530
• dr. Jure Murovec
• dr. Hugo Zupan
• Filip Jure Vuzem – https://cris.cobiss.net/ecris/si/sl/researcher/55566

The project team also closely collaborated with Telekom Slovenije d.d., AKOS, and several industrial partners.

3. Project description

Manufacturing processes are generating an increasing amount of data, which poses a challenge for traditional centralized processing systems. The purpose of the project was to:

• develop connectivity architectures using 5G technology,
• develop methods for distributed edge data processing,
• upgrade the UL FS Smart Factory experimental environment with 5G infrastructure,
• develop algorithms, digital twins, and simulation models,
• analyze the performance of various communication protocols (OPC UA, MQTT, AMQP, Modbus, CoAP, EtherCAT, etc.) in 5G SA and NSA networks,
• develop a methodology for systematic evaluation of communication network parameters (latency, jitter, packet loss, bandwidth).

The research was conducted under real-life conditions in the Smart Factory Demo Center, which was upgraded with a 5G base station and industrial 5G modems, enabling in‑depth experimental analysis.

4. Project phases and implemented activities

Phase 1 – LASFA architecture upgrade

• integration of edge devices and 5G communication level,
• adaptation of the architectural framework for modular and distributed production systems.

Result: new LASFA+EC+5G architecture for Industry 4.0/5.0.

Phase 2 – Analysis of communication technologies

• extensive analysis of industrial WiFi, LTE, 5G, LoRaWAN, Sigfox, NB‑IoT, ZigBee, BLE, Z‑Wave,
• identification of IoT technology limitations in industrial environments,
• confirmation of the advantages of 5G SA for IIoT applications.

Result: performance matrix and communication suitability criteria.

Phase 3 – Experimental environment upgrade

• implementation of 5G SA/NSA network in cooperation with Telekom Slovenije,
• integration of edge devices, modules, and local decision‑making algorithms,
• installation of new infrastructure in Smart Factory Demo Center 2.

Result: fully 5G‑supported research ecosystem.

Phase 4 – Algorithm and data method development

• network parameter measurement algorithms,
• data acquisition and edge processing algorithms,
• algorithms for local control of production processes,
• development of an expert system for a hydraulic press (98.7% improvement).

Result: data‑driven models for real‑time process optimization.

Phase 5 – Digital twins and simulation models

• automated generation of simulation models,
• integration of machine vision and AAS for data acquisition,
• connection of the digital twin with the 5G communication layer.

Result: first comprehensive digital twin methodology that incorporates 5G network parameters.

Phase 6 – Experimental validations

• comparative analyses of protocols in LAN/WiFi/5G,
• analyses of latency, throughput, and reliability,
• testing with industrial scenarios.

Result: experimentally validated models for industrial practice.

Phase 7 – Dissemination, collaborations, and impact

• collaboration with AKOS on regulatory guidelines,
• presentations at ASM’23 and ASM’24 conferences,
• integration of results into teaching and industrial workshops,
• participation in new EU projects (Widera Excellence HUBS – 5G, AI; H2020 RIA).

5. Bibliographic references (direct project outputs)

Scientific articles

• Vuzem F. J., Pipan M., Zupan H., Šimic M., Herakovič N.: Automated Generation of Simulation Models and a Digital Twin Framework for Modular Production (Systems, 2025).
• Pipan M., Šimic M., Herakovič N.: Performance Benchmarking of 5G SA and NSA Networks for Wireless Data Transfer (Journal of Sensor and Actuator Networks, 2026).
• Jankovič D., Pipan M., Šimic M., Herakovič N.: Polynomial Regression‑Based Predictive Expert System for Enhancing Hydraulic Press Performance over a 5G Network (Applied Sciences, 2024).
• Pipan M., Debevec M., Herakovič N.: Improved Static Model for Pneumatic Artificial Muscle (Actuators, 2025).
• Pipan M., Šimic M., Vončina L., Herakovič N.: Use of 5G Technology in Manufacturing Processes and Systems (Ventil, 2024).
• Pipan M., Herakovič N.: Povezovanje industrijske opreme v 5G omrežje (Ventil, 2025).

All publications are open access and include appropriate project acknowledgements (project number, funder, affiliations).