In the oil and gas IIoT environment, fire detection systems heavily depend on fire sensor data, which can be prone to inaccuracies due to faulty or unreliable sensors. These sensor issues, such as noise, missing values, outliers, sensor drift, and faulty readings, can lead to delayed or missed fire predictions, posing significant safety and operational risks in the oil and gas industrial IoT environment. This paper presents an approach for handling faulty sensors in edge servers within an IIoT environment to enhance the reliability and accuracy of fire prediction through multi-sensor fusion preprocessing, machine learning (ML)-driven probabilistic model adjustment, and uncertainty handling. First, a real-time anomaly detection and statistical assessment mechanism is employed to preprocess sensor data, filtering out faulty readings and normalizing data from multiple sensor types using dynamic thresholding, which adapts to sensor behavior in real-time. The proposed approach also deploys machine learning algorithms to dynamically adjust probabilistic models based on real-time sensor reliability, thereby improving prediction accuracy even in the presence of unreliable sensor data. A belief mass assignment mechanism is introduced, giving more weight to reliable sensors to ensure they have a stronger influence on fire detection. Simultaneously, a dynamic belief update strategy continuously adjusts sensor trust levels, reducing the impact of faulty readings over time. Additionally, uncertainty measurements using Hellinger and Deng entropy, along with Dempster–Shafer Theory, enable the integration of conflicting sensor inputs and enhance decision-making in fire detection. This approach improves decision-making by managing sensor discrepancies and provides a reliable solution for real-time fire predictions, even in the presence of faulty sensor readings, thereby mitigating the fire risks in IIoT environments.

In high-risk IoT-enabled Industrial Environments such as oil and gas industries, early and accurate fire detection is crucial to prevent catastrophic damage. Traditional fire detection systems that rely on single IoT sensors, like heat or smoke detectors, are prone to inaccuracies and delayed responses, particularly in noisy or complex industrial environments. This research proposes an advanced fire prediction approach aiming to enhance decision-making accuracy with uncertain or incomplete fire sensor data in an edge computing IoT complex industrial environment that integrates multiple supervised machine learning algorithms for each sensor types and Dempster-Shafer theory (DST) with multi-sensor fusion. By combining data from multiple different types of sensors (RGB, thermal, gas, smoke, and flame), the proposed architecture enhances the reliability of fire prediction and detection as each sensor detects different parameters of a fire and this ensures every parameter is considered for fire detection ensuring early detection and it reduces false positives. By preprocessing sensor data and by combining machine learning algorithms (MobileNet, EfficientNet, Random Forest, SVM, and Decision Trees) where each ML processes different sensor data types, this supports providing more accurate data to DST. The D-S theory helps in handling uncertainty by fusing conflicting multi-sensor data to generate a more reliable decision. The results demonstrate that this multiintegrated methodology proposed significantly outperforms traditional methods in predicting and detecting fires early and accurately, even under challenging and complex environmental conditions of Oil and Gas Industries. The goal of this research contribute to the development of advanced fire prediction protocols offering a robust solution for managing and mitigating fire risks in dynamic, data-rich IoT environments. Predicting fire and reducing Fire in a critical oil and gas IoT industrial environment remains a critical challenge. The proposed method aims to enhance fire prediction and detection accuracy through algorithms developed using sensor preprocessing, machine learning, and Dempster Shafer fusion handling uncertain and varying multi-sensor data and demonstrates real-time, data-driven decision-making in critical industrial environments. The results show an accuracy of 98.2%, a precision of 98.5%, and a recall of 98.3% significantly confirming the prediction over other models.