ICSCASD-MPLC: Industrial Control System Cyber-Attack Simulation Dataset for Mitsubishi PLCs

The ICSCASD-MPLC dataset captures realistic network traffic from a physical ICS testbed built using Mitsubishi R04ENCPU PLCs, Siemens industrial switches, Node-RED for process control, and Wireshark integrated with Python scripts for automated network traffic collection. The dataset includes labeled network flows representing both normal operations and seven types of cyberattacks: Denial-of-Service (DoS), Man-in-the-Middle (MITM), ARP Spoofing, Data Injection, and Reconnaissance attacks.

2.6 million records were collected, comprising 57 features per entry. The data is available in CSV format and is suitable for training and evaluating Machine Learning and Deep Learning models for binary and multi-class intrusion detection tasks.

This dataset was created using real hardware under controlled conditions, making it more representative of operational ICS environments than synthetic or virtual datasets. It supports academic and industrial research in the fields of cybersecurity, anomaly detection, and AI-driven intrusion detection systems for ICS.

Format: CSV
Number of Records: 2.6 million
Number of Features: 57
Use Cases: Intrusion Detection, Machine Learning, Cyberattack Classification in ICS

Please cite the associated paper when using this dataset.

" Houkan, A., Sahoo, A.K., Gochhayat, S.P. et al. Artificial intelligence approach to intrusion detection in industrial control systems with real world dataset generation and model evaluation. Discov Artif Intell 5, 307 (2025). https://doi.org/10.1007/s44163-025-00507-2. "
and this work pl 
" A. Houkan, et al., Dynamic Tanh–enhanced transformer architecture for scalable and high accuracy cyber threat detection in IoT environments, Eng. Res. Express (in press) (2025).  https://doi.org/10.1088/2631-8695/adf527. "

🔗 Explore more:

📖 Research Article (Springer Nature): https://rdcu.be/eOj38

📥 Dataset: https://zenodo.org/records/16351397

🎥 YouTube Overview Video: https://youtu.be/RRQRDK4v6Tw