This thesis proposes a deep learning-based technique for detecting flooding attacks, a type of denial-of-service (DoS) attack, on Industrial Internet of Things (IIoT) systems. IIoT systems use interconnected devices, sensors, and machines to improve efficiency, reduce costs, and enhance safety. However, flooding attacks can cause disruptions and damage to these systems. The proposed technique uses Bidirectional Long Short-Term Memory (LSTM) networks to detect flooding attacks. The LSTM networks analyze the behavior of IIoT systems and detect anomalies that may indicate a flooding attack. The technique is evaluated using the EdgeIIoTset dataset, which is the latest and most comprehensive dataset available for IIoT systems. The EdgeIIoTset dataset includes multiple sets of attacks, including multiple DoS flooding attacks on IIoT systems. The dataset is designed to represent real-world scenarios, making it an ideal testing ground for the proposed technique. The evaluation results demonstrate the effectiveness and high accuracy of the proposed technique in detecting flooding attacks in IIoT systems. This provides a proactive and reliable solution for enhancing the security and reliability of IIoT systems, making them more resilient against flooding attacks.