DTFL‐DF: Digital twin architecture powered by federated learning decision forest to mitigate fire accidents in mining industry

Abstract Automation is the guiding principle of this new era, and despite the problems that humanity faces as a result of automation, technology has greatly benefitted people by streamlining challenging jobs across many industries. The mining business, where there are frequently unforeseen mishaps, is one such industry that requires complete automation. In this work, a new simulative processing environment termed DTFL‐DF—Digital twin federated learning decision forest a digital twin environment that is tailored to handle unforeseen fire incidents—is offered as a means of avoiding these unplanned catastrophes in the mining industry. Although the design presented here is intended for usage in the mining sector, it can also be applied to other sectors. The overall technological contribution of this study is to guarantee the processing of real‐time data in order to successfully handle mission‐critical operations without relying on past data. This is accomplished by adapting the digital twin's original design and distributing the processing environment within the edge‐fog layer. Results analysis in the form of robustness analysis, performance evaluation of the classification model, etc. provides strong support for the suggested methodology. For handling the decentralized training procedure, a brand‐new algorithm termed FL‐DF is put forth in order to speed up classification and prevent any sort of catastrophe.