DAPPA

21.11.2025 12:04
Paylaş
Paylaş
A+
A-

DAPPA

Overview

DAPPA is a bilateral Türkiye-Romania research project focused on developing a new generation of PLC-IoT industrial networks augmented with digital twins and artificial intelligence algorithms for predictive maintenance in smart factories.

Project Information

Project title: "Digital Twins of a New Generation Industrial PLC-IoT Network for AI-based Predictive Factory Maintenance" (acronym DAPPA).

Partners:

- Turkish partner: Izmir Katip Çelebi University (IKCU), TR Project Coordinator: Dr. Kamil Çetin; To be added with the corresponding role: EGE University - Bilge

- Romanian partner: National University of Science and Technology POLITEHNICA Bucharest (UNSTPB), RO Project Coordinator: Dr. Cornelia Ionela Bădoi.

Funding:

- TÜBİTAK (Scientific and Technological Research Council of Turkey), grant no. 124N78;

- UEFISCDI (Romanian Executive Agency for Higher Education, Research, Development and Innovation Funding), grant no. 6BMTR/2025.

Duration:

- Turkish partner: May 1, 2025 - May 1, 2027;

- Romanian partner: April 1, 2025 - December 31, 2026.

Project Objectives

DAPPA aims to revolutionize predictive maintenance in mass production factories by leveraging cutting-edge technologies. The project proposes a novel approach that integrates next-generation Industrial Embedded Personal Computers (IEPCs) with Industrial Internet of Things (IIoT) and creates a digital twin of the entire production environment. By harnessing the capabilities of AI and ML, this project seeks to develop a robust system for predicting potential equipment problems before they occur, ultimately minimizing downtime and maximizing production efficiency.

The main objectives are:

Implementing IIoT network: Determine and implement the optimal configuration for the IEPC-IIoT network at various production volumes, considering a range of wired and wireless network topologies (mesh, tree, hybrid etc) and communication technologies (Wi-Fi, Bluetooth, Zigbee etc). These considerations factor in essential elements like latency, data rate, and reliability, ensuring compatibility with existing systems.
Upgrading the Laboratory Setup: Establish a prototype with nine individual local automation setups in the control and industrial automation laboratory at IKCU. Each setup consists of PLC, encoder, photo-electric, capacitive, inductive sensors, AC and servo motors. Existing PLCs will be upgraded to IEPCs, and new sensors (current-voltage, temperature, vibration, audio) will be added to each setup's IIoT network.
Digital Twin Development: Create a digital twin of the physical setup in the laboratory, encompassing IEPCs, motors, and sensors, using Microsoft Azure's platform due to its functionalities like rapid algorithm development, factory automation simulations, and remote monitoring.
Predictive Maintenance with AI/ML: Sample factory production scenarios will be simulated across the nine setups. Real-time sensor data will be collected and analyzed using AI, particularly ML. This focuses on predictive maintenance for sample conveyor belt motors by analyzing multi-sensor data and automatically adjusting position and speeds based on potential errors identified by AI.

Architecture and Key Technologies

The DAPPA architecture proposes a hierarchical framework that connects IIoT networks, Azure IoT Hub, and digital twins, ensuring bidirectional data-control flow between physical equipment and their virtual models. The IEPC-IIoT infrastructure becomes the backbone of industrial communications, using IP and MQTT protocols, extended with wireless technologies (i.e., ZigBee) for redundancy and scalability.

Key features:

Next-generation IEPCs with enhanced processing power, flexible programming options, increased memory capacity, and seamless connectivity to the IIoT infrastructure;
Multi-sensor upgrades (vibration, temperature, sound, encoders) for rich data acquisition;
IEPC-IIoT network as the backbone of industrial communications, using IP and MQTT protocols;
IEPC-IIoT network extended with wireless technologies for redundancy and scalability;
Azure IoT Hub as the central cloud gateway;
Azure Digital Twins integrated with CoppeliaSim for high-fidelity simulation;
AI models for predictive maintenance and anomaly detection using unsupervised ML, including AI Anomaly Detector, Azure Machine Learning with PCA, and Azure Service Bus.

An experimental case study on a multi-conveyor robotic cell demonstrates the architecture's capability to distinguish operating regimes and highlight patterns of mechanical and thermal stress relevant for predictive maintenance.

Scientific Dissemination

The DAPPA project has resulted in several scientific contributions:

Özbek, M.E.; Çetin, K.; Kartal Çetin, B.; Karataş, Ç.; Şahin, S.; Bădoi, C.I. PLC Otomasyon Sistemlerinin Önleyici Bakım için Dijital İkizinin Oluşturulması (Digital Twin Development for Preventive Maintenance of PLC Automation Systems). Elektrik Tesisleri Ulusal Kongre ve Sergisi (ETUK 2025), 22-24 October 2025, Izmir, Turkey (to be indexed).
Bădoi, C.I.; Kartal Çetin, B.; Çetin, K.; Karataş, Ç.; Özbek, M.E.; Şahin, S. Architecting the Future Factory: Industrial PLC-IoT, Digital Twins and Predictive AI-based Maintenance. 28th International Symposium on Wireless Personal Multimedia Communications (WPMC 2025), Sofia, Bulgaria, 9-12 November 2025 (to be indexed).
Bădoi, C.I.; Kartal Çetin, B.; Çetin, K.; Karataş, Ç.; Özbek, M.E.; Şahin, S. A Hierarchical Framework Leveraging IIoT Networks, IoT Hub, and Device Twins for Intelligent Industrial Automation. Appl. Sci. 2026, 16, 645. https://doi.org/10.3390/app16020645.

Expected Impact

The potential benefits of this project are very important for smart factories that want to have sustainable production technologies. Such factories can expect to achieve substantial cost savings through increased efficiency and minimized downtime. Furthermore, the project offers valuable opportunities for young researchers, fosters future collaborative projects, and contributes to the advancement of the field through academic publications and postgraduate theses. Ultimately, the long-term vision is for the developed solutions to be widely adopted by the manufacturing industry, leading to a paradigm shift towards smarter and more efficient production practices.

Contact Information

Turkish Partner:

Izmir Katip Çelebi University (IKCU)

Department of Electrical and Electronics Engineering

Smart Factory Systems Application and Research Center (AFSUAM)

Project Leader: Dr. Kamil Çetin

Email: kamil.cetin@ikcu.edu.tr

Romanian Partner:

National University of Science and Technology POLITEHNICA Bucharest (UNSTPB)

Department of Telecommunications

Project Director: Dr. Cornelia Ionela Bădoi

Email: cornelia.badoi@upb.ro