In the last decade, use of Internet of Things (IoT) in industrial applications is enhanced in most industries to maximize the production and development. However, as its popularity continues, it has become an increasingly attractive target for industrial network damage and various cyber attacks. These increasing number of cyber attacks in Industrial IoT (IIoT) systems, can lead to significant business failures and damages, sometimes even posing life-threatening situations Network attacks and their timely detection has became crucial among research community. Numerous attempts to counteract these threats through intelligent detection techniques have been developed. However, ever-evolving cultivation of network attacks and evaluation on latest datasets have rendered many approaches obsolete. Confronting these challenges, this paper introduces a modern framework for Bot Detection in IIoT that harnesses the power of machine learning algorithms through ensemble modeling. The proposed ensemble model integrates the upsides of three machine learning models, i.e., K-nearest neighbor (kNN), Gradient Boosting Decision Trees (GBDT), Random Forest (RF), and one deep learning model, Multilayer Perceptron (MLP). Validation results demonstrate the robustness of our proposed framework against detection of most recent attacks. On the ToN-IoT dataset having 127 features and 35,975 samples, our approach achieves an impressive accuracy rate with the ensemble model. The dataset includes IoT and IIoT sensor generated seven Bot attacks, providing comprehensive dataset for the proposed model’s effectiveness. Compared to state-of-the-art, ensemble model outperformed and achieved 98.20% accuracy and 1.7% Mean Squared Error (MSE). Further, we analyzed the proposed approach and other approaches to find utilized time, and memory using Scalene, a Git project.