The problem of planning of large cities water supply system modernization based on long-term water consumption forecast is reviewed. A review of existing models of long-term forecasting on the basis of time series is made. Advantages and disadvantages of forecasting models based on autoregressive models ARIMA, neural networks and exponential smoothing are formulated. Vivid example of a long-term forecasting of water consumption of large city using these methods is presented to identify the most efficient and adequate model.
Ключові слова: water supply, water consumption, forecast, forecasting, time series, model, statistical analysis, exponential smoothing, neural network
Planning of city water supply system modernization based on water consumption forecast,
Information Processing Systems,