Prediction of Safety Objective of an Enterprise Using Fuzzy Neural Network

Fuzzy neural network (FNN) method is presented where a functional model assists in a goal oriented prediction of safety objective and the FNN model based on time series is developed for prediction of the accident frequency rate which is one of important indices for safety objective management in an enterprise. The model is built in accordance with a connection link between fuzzy logic and neural computing. The fuzzy cluster identification algorithm is adopted to confirm the number of the best hidden layer nodes of the network. The FNN has capability to develop complex, nonlinear functional relationships between input-output patterns based on limited and sometimes inconsistent data, and is therefore suitable to predict the accident frequency rate on the basis of data obtained from historical data of the enterprise. Comparing with prediction results of the experiment of BP neural network model, the results of FNN model show that the prediction accuracy in function is reliable and practical. (C) 2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the Capital University of Economics and Business, China Academy of Safety Science and Technology.