Abstract:The artificial neural network approaches have been extensively utilized in various engineering and science aspects because it can incorporate both nonlinear and linear systems without needing to make assumptions as a regulatory in many traditional statistical models. The specific objectives of this study first was to examine various forecasting methods in healthcare services demand using artificial neural network model, secondly to develop a model for data mining in order to facilitate forecasting of healthcare demand services. Thirdly to analyse the prediction of demand of outpatient healthcare services using the Artificial Neural Network and lastly to evaluate artificial neural networks model for forecast of healthcare services. Health care managers and planners therefore must make future decisions about healthcare services delivery without knowing what will happen in the future. Forecasts would enable the managers to anticipate the future demand and plan accordingly.  This study aimed at examining artificial neural network as an approach to health services demand forecast in Nairobi County in Kenya. The model was trained under the WEKA environment and applied to predict the demand for health services in various categories for private health care providers in Nairobi county Kenya. These results show that WEKA Forecasts using neural network algorithm gives a more accurate forecast results than the Moving Averages and Linear Regression and hence gives a more reliable results for the demand forecast. Testing of accuracy considered multiple variables – Since several factors affect the demand of the health services in Public hospitals as outlined above, this was also put into test. The MSE, RMSE, and MAPE can be used to measure the expected level of fit of a predictive model. If a model fits the training data set very well but does not fit the validation data, it is called overfitting. A good predictive model is supposed to generate consistent results in both training and validating data sets. This is confirmed by regression model and artificial neural network model.
Title:PREDICTING DEMAND FOR OUTPATIENT HEALTHCARE SERVICES USING ARTIFICIAL NEURAL NETWORKS