Applications of Neural Networks: Data Centers, Predictive Maintenance, Production, Sustainable Computing, and Healthcare

Artificial Neural Network teaches computers to process data like human brains. ANN uses interconnected and layered nodes or neurons that resemble a human brain. The following eight case studies are reported in this paper: (1) scheduling of tasks in cloud computing , (2) scheduling on heterogeneous platforms, (3) predictive maintenance, (4) production scheduling using ANN, (5) train rescheduling, (6) micro-grid scheduling, (7) detection of obesity, and (8) prediction of falls in older patients. In every case study, the model is described, and the operating method is explained in detail. The recursive ANN is shown to accommodate large data. Neural network employing inhibitor neurons use 50% of neurons used by regular ANN. Genetic algorithm (GA) shows good performance in train rescheduling when the data size is small and ANN performs well when the data size is large. Multi-layer perceptron (MLP) method is found to work with 97% accuracy in predicting falls in older patients. Applications of each case study are highlighted to help the readers apply the right model for a particular application.