Search

In Spain, the water supply service is a municipal responsibility and in general is a sector without competitors. For this reason, an efficiency analysis attains greater significance. This study uses Data Envelopment Analysis (DEA) to estimate the efficiency of different urban water utilities. An extensive data search, where several variables such as the capital expenditure, the cost of material, or the labor have been recorded, has allowed evaluating the relative efficiency of the most important Spanish water distribution networks in using their resources. Furthermore, their sustainable efficiency has also been evaluated by including a variable representing the percentage of water losses. Results reveal the weaknesses of inefficient utilities and help to detect potential aspects that these companies should improve. For instance, there is an evident incorrect management of the costs of material by many urban water utilities, which does not happen with the labor. Additionally, the most efficient water utilities regarding the sustainable efficiency help to discover target percentages of water losses for the inefficient ones.

The water supply networks of many countries are experiencing a drastic increase in the number of pipe failures. To reverse this tendency, it is essential to optimise the replacement plans of pipes. For this reason, companies demand pioneering techniques to predict which pipes are more prone to fail. In this study, an Artificial Neural Network (ANN) is designed to classify pipes according to their predisposition to fail based on physical and operational input variables. In addition, the usefulness and effectiveness of two sampling methods, under-sampling and over-sampling, are analysed. The implementation of the model is done using the open-source software Weka, which is specialised in machine-learning algorithms. The system is tested with a database from a real water network in Spain, obtaining high-accurate results. It is verified that the balance of the training set is imperative to increase the predictions’ accurateness. Furthermore, under-sampling prioritises true positive rates, whereas over-sampling makes the system learn to predict failures and non-failures with the same precision.