Two major challenges for a water utility's finance department are to forecastwater consumption revenue with a reasonable degree of confidence and to explainvariations between the forecast and actual revenue. Business leaders are demanding morereliable forecast information and better explanations in an increasingly competitiveindustry. This paper demonstrates that Artificial Neural Network (ANN) modelingtechnology is a valuable tool to complete these tasks. The authors highlight the ANNmodeling process to transform the information from a utility billing system into a revenueforecasting model. The authors emphasize the user involvement in the modeling processto ensure the success of the models and to improve the user confidence in the model.Finally, the authors show the actual model performance in the real world scenario in theYear 2000 and how the models help the finance department to evaluate the risk inweather-based revenue hedging. Includes 8 references, tables, figures.
| Edition : | Vol. - No. |
| File Size : | 1
file
, 260 KB |
| Note : | This product is unavailable in Ukraine, Russia, Belarus |
| Number of Pages : | 18 |
| Published : | 06/01/2001 |
