Prediction of Roof Fall Rate in Coal Mines Using Fuzzy Logic

Roof fall risk is a common problem in coal mines, and it is generally unpredictable due to variability in geological and mining parameters. In this study, a new fuzzy logic model was developed to predict roof fall rate in coal mines. Parameters such as coal mine roof rating (CMRR), primary roof support (PRSUP), intersection span, and depth of cover were considered as the model inputs. The model was developed based on experts? knowledge and also a database including about 109 datasets of roof performance from U.S. coal mines. Approximately 20% of these datasets were used to assess the performance of the fuzzy model. In each case, comparison between results from model and real roof fall rate showed that this model can predict roof fall rate to an acceptable extent.