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.

We present a weighted sum model to calculate the risk index, which is equivalent to a priority index for the replacement of pipes in a network. This index is based on two factors related to pipe condition (probability of failure and supply pipe leakage flow), and three factors related to the consequences of pipe failure (non-served demand for supply pipes, maximum evacuation flow for sewer pipes, and pipe relevance). The procedure encompasses both the supply and sewer networks, to avoid the need for ex-ante budget allocations. We tested different configurations for the risk index function terms, and calculated the associated weights by means of a discrete choice experiment. We applied the procedure to a large water company in Spain, where the results were fully accepted as an equilibrium representation of the renovation opinions of its technical staff.

Asset management and renewal are usually planned at the annual and operational levels. However, at a strategic and tactical level, infrastructure companies are looking for management models that allow them to analyze their assets, their evolution, and the detection of those assets that must be renewed and incorporated before in medium- and long-term planning. Additionally, it is necessary to define the financial effort and its temporal application. This effort should allow the portfolio of assets that make up the water network to meet the organization's asset management objectives. The monitoring of objectives, including the sustainable maintenance of network infrastructure, can be developed on the basis of management indicators that facilitate their renewal. Indicators allow to analyze the state or strength of the organization, and also constitute an opportunity factor to improve efficiency in the management of assets when financial and non-financial strategies are combined.Following the ISO 55000 approach, asset management involves balancing the cost, opportunities, and risks that these assets play in the objectives of the organization. To achieve this balance, an alignment of different elements of influence in management is required.The decision-making process in asset management requires considering a risk management strategy. Risks can incorporate the interaction of several aspects that must be combined. In the process, multicriteria techniques could be incorporated into management to integrate both scientific knowledge and expert knowledge. These techniques contribute to defining the weight of the parameters that constitute the definition of risk in the choice of assets. The risk level combined with the financial factor allows you to define the replacement to support optimized strategic planning. Similarly, the model must consider the regulatory and financing framework that influences water companies. Finally, the model should be oriented toward a process of maximizing the renewal of assets based on a set of performance indicators.The model of the decision-making process presented is based on a) a phase of determination of investment and risk, b) a phase of construction of works based on operational criteria, c) a phase of analysis and planning proposal based on the current state of the network, and d) a phase of study of improvement scenarios based on the performance indicators of the organization, which should allow the generation of an action plan. The development of the methodological framework and its applications allows and improves decision making and provides a strategic vision from which managers benefit by obtaining a new perspective on the evolution of the network and the integration of the different departments or functional areas of the company when looking for a joint solution aligned with strategic objectives.