Publicaciones

The ability to track live monitoring data and use AI-based software to detect anomalous events – breaks, leaks, meter malfunction, abnormal consumption, etc – while fully integrating AMI/AMR with SCADA monitoring streams enables faster response, finds more leaks, finds them earlier, and keeps automatically accrued all the key components of real losses and consumption. This paper presents a Spanish public utility, Consorcio de Aguas Bilbao-Bizcaia (CABB) implementing a DMA program as part of their water & energy losses management effort. The system is fully covered with DMAs equipped to track down and address breaks and leaks, quantify consumption and manage NRW and the system in a more systematic and efficient way.

Having gained traction mostly as a consultant’s tool, hydraulic modelling software exists often in isolation from a data standpoint, even when it facilitates imports from GIS or SCADA. Infrastructure data and consumption data evolve continuously. Significant effort is needed when creating or updating a model. This leads to relatively high maintenance costs if the model is to be permanently accurate and relevant. The result is that few utilities can actually rely on permanently updated models, ready to run immediately, at any given moment in time. This paper discusses how new cloud-based, ‘always-on’ connected software paradigms1 enables massified continuous generation and update of hydraulic models for everyday use in utility environments – in software where GIS, SCADA, and billing or AMI data are integrated and continuously updated. The paper explores real case applications in water utilities from Spain, Portugal and the USA with varying degrees of data maturity.

Embracing digital transformation has become an imperative for water utilities, who are witnessing a significant enhancement in their toolset, driven by improved sensor technology, a proliferation of data sources, and an ever-expanding volume of data. However, the true value of digital transformation can only be realized when it leads to tangible results that align with the core objectives of the utility, which have not changed: delivering high-quality, efficient, and sustainable services. This paper details the successful implementation of a water losses management system in Andratx, a municipality in the Balearic Island of Majorca, Spain. This management system reconciles live data from the SCADA systems (253 sensors) in a 115 km network, with the utility’s billing system, as well as with the CMMS/work orders system. A decrease of 20% has been achieved in non-revenue water in the first year of implementation, attributable to the utility being able to better identify and focus efforts on sectors with higher water losses, as well as to better track the progress of their actions.

Monitoring water demand is extremely helpful in the early detection of issues and malfunctions in water distribution networks. Therefore, distinguishing abnormal water meter readings is an important problem in the management process. Although much focus has been given to this problem in the last few years, many of the approaches still lack efficiency and scalability when adapted to (near) real-time settings. To tackle these issues, we propose the Online Curvature Matcher, a real-time algorithm for the detection of pipe burst events in flow readings of District Metering Areas (DMA), which requires only a few weeks of data clean of burst events. In the proposed methodology, libraries of segments of the time series are constructed and then used to find abnormal patterns in consumption. We also propose the incorporation of additional constraints to address the complexities of real-world data (such as the existence of irrigation events, seasonal changes, and discrepancies in behavior due to weekday). We perform widespread testing on data from 32 DMA. The results indicate an increase in detected bursts compared to the state-of-the-art, while a decrease in the number of generated false positive events was achieved. Additionally, it was found that six weeks of history is the minimum amount of data necessary to perform online burst detection, while Dynamic Time Warping (DTW) is a suitable distance metric to measure the similarity between different flow reading segments.

Monitoring water demand is extremely helpful in the early detection of issues and malfunctions in water distribution networks. Baseform develops forward-thinking software for networked water infrastructures that assists water utilities in monitoring the entire network in its daily management. Several methodologies are continuously applied by Baseform to analyze and accurately predict water demand patterns and to establish whether the consumption recorded during a specific period is normal or may reveal abnormal events. These include pipe bursts or leaks, a Legitimate change in consumer behavior, and Irrigation systems. However, it is often found that the appropriate post-processing that the identified events require is laborious, and it is, therefore, essential to develop mechanisms that automate many of the decisions in events' processing and reclassification. Our approach starts with defining a set of new variables that try to capture the behavior of several types of events and then developing and comparing several statistical and machine learning methods to continuously classify those events. Through the analysis carried out, it was possible to extract and better understand what features are relevant in the discrimination between different event categories. The first developed classifier still does not possess reliable enough performance to distinguish rare events, such as those related to pipe bursts. In fact, it will be necessary to conduct a more thorough examination of the extracted variables to make it clearer which further pre-processing steps, like event pruning, might be required to lessen the amount of noise in the data and promote a better classifier performance.

Baseform develops software for networked water infrastructures that assists utilities in monitoring the entire network in its daily management. Already, Baseform possesses methodologies to detect instances where a significant deviation from the expected consumption levels was observed. These are called events, and they include pipe bursts, legitimate changes in consumer behavior, and abnormal consumption levels, among others. However, it is often found that the appropriate post-processing that the identified events require is laborious, and it is, therefore, essential to develop mechanisms that automate the events’ classification. Two approaches are tested with this goal in mind. The first starts with defining a set of new variables that try to capture the behavior of several types of events and then developing a classifier trained in a supervised manner on a history of the available events. Through the analysis carried out, it was possible to extract and better understand what features are relevant in the discrimination between different event categories. However, it was found that this offline methodology leads to lackluster performances. In an attempt to improve the classification process, and create a mechanism capable of handling changes in consumer behavior over time, a real-time approach was developed, particularly, in the detection of bursts. The algorithm chosen is called the Curvature Matcher and consists of the calculation of a dissimilarity score between the current window readings and the previously registered readings. The Curvature Matcher proved effective in discriminating between Bursts and other events, correctly identifying 73% of the Bursts detected by Baseform’s software.

In a California public utility serving about 20 200 pop. through 250mi of network, a program has been implemented in the last 3 years to objectively quantify the impact of CIP projects and support the prioritization of resources, maximizing the number of avoided leaks and other failures.

Water, a vital resource for life, is being used more rapidly than Nature can replenish. With the increase of population density, the water supply distribution system is more pressured, resulting in many challenges, including the leakage and wastage of water during transportation. To detect such events and allow for a rapid and more precise response, water flow modelling and prediction can be used, with the intent of comparing the expected and registered water flow. This project was developed in a business context, with the aim to better characterise, model and forecast the water flow demand measured at the entry of Spanish District Metered Areas (DMA). Several distribution models, alongside their parametrization (range of input historical data and types of the grouping of the days of the week), are studied to model the water consumption data of the yearly, monthly and some weekly periods of 2018 and 2019. The models are evaluated with the Weighted Absolute Percentage Error (WAPE). The results show that the proposed distribution models outperform the company’s current best model for the majority of the Winter months (October to March). The results also evidence that smaller periods of input data lead to better predictions, for most of the DMAs under study. Additionally, a new implementation for the holidays’ prediction is suggested.

Proactive management for water pipelines is a crucial aspect in the field of water infrastructure assessment. One of the main aspects of this approach is to rehabilitate or replace pipes prior to failure, to avoid the deterioration of the network. This work presents two gradient boosting algorithms: XGBoost and LightGBM, for pipe failure prediction. The two models, together with the current approach used by Baseform, were compared when applied to two water supply systems, one in Spain and another in the USA. We considered three different problems: a binary classification problem that aims to predict the probability of failure for a given pipe in a given year; a counting problem through a hurdle model that aims to predict how many failures a pipe will experience in a given year; and a regression problem through a Tweedie distribution with the purpose of prediction the cost of repairing failed pipes. Including the Baseform models, all the developed models had satisfactory results, with a particular emphasis on the LightGBM when applied to the counting problem, which strikes a good balance between performance and running time. The proposed models predict a conservative fail rate when in comparison with the Baseform models. The regression models under-predicted the real values of the repair costs. Nevertheless, it opened a new approach to pipe failure study. Gradient Boosting algorithms revealed themselves to be an upgrade and an improvement to failure analysis in water distribution systems.

The Gwinnett County Department of Water Resources has established an advanced, multi-pronged approach to short- and long-term asset management for its 3,000-mile gravity sanitary sewer system, with the goals of predicting 1) areas with high probability of operational failure, 2) areas with high probability of structural failure, and 3) deterioration curves for the various pipe cohorts, informing the long-term renewal process. An application of software-based analytics is described that represents a step forward in decision-making used to drive asset inspection, maintenance and capital project prioritization. The components above are jointly addressed in a mutually informative, permanently available software, providing clear, defensible, quantified overall prioritization of capital and operational expenses. Advanced analytics are brought together in a multi-criteria, optimized framework designed to continuously update itself and provide both operational and structural support on demand. The paper presents the analytics methods used, a detail of key results and their discussion, and provides an overview of the multi-criteria framework deployed to support the prioritization of assets, sub-basins and projects.

La predicción de fallos y la priorización de los esfuerzos en la rehabilitación desempeñan un papel muy importante en el marco de la gestión patrimonial de infraestructuras. En los últimos años han surgido diversos enfoques científicos dirigidos a determinar tiempos de vida y previsión de fallos, con el objetivo de identificar las tuberías más propensas a fallos y ayudar a priorizar rehabilitaciones del sistema. El análisis de fallos de Baseform está diseñado para utilizar los datos reales con sus problemas habituales, combinando herramientas estadísticas de vanguardia con la adquisición continua de datos. Por lo tanto, está preparado para hacer frente a las lagunas de datos y la información inconsistente, y aprovecha al máximo los escenarios de mayor cantidad de datos, siempre y cuando estén disponibles. Este artículo ilustra el análisis de fallos a través de su aplicación real en el Consorcio de Aguas Bilbao Bizkaia (CABB).

Principal Component Analysis is a popular statistical method for dimensionality reduction and has been extended to interval-valued data by several authors. Four of this most common symbolic principal components estimation methods have a unified formulation, namely the Centers and Vertices Methods, Complete Information Principal Component Analysis, and Symbolic Covariance Principal Component Analysis. A meticulous simulation study is conducted to compare several robust estimators of symbolic Principal Components for interval data, which use robust symbolic covariance matrices. The robust versions of the four estimation methods in study are compared with their classical counterparts. The simulation study is accompanied by visualisations of the various types of contamination used, which helps to understand what are considered to be outliers in interval data. In addition, an outlier detection method based on robust interval principal components was devel- oped. An initial comparison study was conducted, to assess the performance of the method proposed, in particular to compare with another existing proposal for outlier detection in interval data. The potentialities of the developed methodologies are illustrated with their application to a real- life dataset of drinking water consumption of more than 90 000 clients, served by a large urban water supply system in a Portuguese utility. Using the geographic identification of the clients provided by the utility, the results of the symbolic principal components are also represented in maps through the use of the Baseform software. The outlier detection procedure is used to identify and characterise clients with anomalous consumption behaviours.

Drinking water is provided to the majority of population through a complex water distribution system, which is constantly affected by economic constraints, climate changes, infrastructure deterioration, and increasing urbanisation and customer demand. Hence, water utilities are challenged with a demanding task and, they are forced to venture in long-term strategies which direct their efforts in planning and prioritisation of rehabilitation actions. Despite the common reactive approach in which water utilities base their efforts, they are recently moving towards a more proactive approach and using failure prediction strategies in infrastructure asset management. This project’s goal is to apply machine learning techniques to build prediction models and evaluate their performance when compared to existing approaches such as the one implemented by Baseform. Failure data is characterised by some important specificities, such as recurrent events, left-truncation and right-censoring, which distinguish it from other types of data. Tree-based models emerge as ap- pealing approaches to model time until the next failure, which combine the strengths of survival analysis with the most recent developments in machine learning. To prevent the potential instability caused by individual trees, ensemble techniques arise as an alternative methodology. In pair with the Baseform model, all methods are capable of obtaining equally satisfactory results (sometimes surpassing the existing approaches) and ranking pipes with higher risks of failing. From the work developed so far, it is noticeable that survival analysis combined with machine learning techniques is a broad field of possibilities, which can bring many benefits to failure analysis in water distribution systems.

For over a decade, the Department of Water Resources (DWR) of Gwinnett County (GA, USA) has developed a robust work order and inspection history within its Computerized Maintenance Management System. DWR has recently taken the next step in this asset management journey, leveraging over 43,000 pipeline inspection records and 30,000 work orders, in conjunction with comprehensive flow monitoring coverage, to 1) predict areas with high probability of operational failure, 2) predict areas with high probability of structural failure, and 3) develop deterioration curves for the various cohorts of pipe, informing the long-term renewal process. This prediction allows DWR to target inspections and proactive O&M where failure is most likely, as well as definitively plan for long-term renewal investment (Suttles et al., 2019). The three components of the approach concurrently evaluate infiltration and inflow (I&I), inspections and work orders, in order to address: (i) excessive I&I, a symptom of structural deterioration and a cause of increased pumping, treatment and reclamation costs, lower efficiencies and service levels, and public and environmental liabilities; (ii) excessive reactive O&M levels, driving higher costs and impacting levels of service (overflows, backups); and (iii) capital expense, crucially tied to controlling asset deterioration and determining how best to prioritize available funds to renew or rehabilitate for maximum sustainability.

Although reacting to fast events such as pipe breaks is customarily at the top of a utility’s O&M priorities, given the mission-critical nature of providing a lifeline service such as water supply, numbers show that lesser visible leaks that develop deceptively over long periods of time are often responsible for 50-70% of real losses. Conversely, they are also a more challenging NRW component to identify, quantify and address. This paper describes a procedure for identification of trends in minimum night flow over mid- term to long- term horizons (i.e., a few days to several months). This approach is particularly useful in the early detection of ongoing leaks that, due to their progressive nature, are less detectable to the human eye for a long time, during which significant NRW develops unchecked.

The prediction of pipeline and service line failures, and the prioritization of rehabilitation efforts, play an important role in infrastructure asset management. Traditional condition assessment methods for water pipes are commonly associated with time- and resource-consuming tasks, that are either expensive, inaccurate, or both. Often, these physical methods cannot be applied to the entire system, so the results cannot inform the decision-making process in a structured manner. Several scientific approaches focused on lifetime assessment and failure forecasting have emerged in recent years aiming to identify the most failure-prone pipes and help prioritize system rehabilitation actions in a non-invasive but highly informed and utility-specific way. Failure Analysis, like other Baseform apps, is designed for real data with common, everyday problems, combining breakthrough statistical tools with continuous utility data acquisition. It is therefore prepared to cope with data gaps and inconsistent information, and it takes full advantage of more data-rich scenarios if and when they become available. This paper presents some key technological aspects and capabilities available in this application, illustrating them through examples from a case study at a large utility where the software is in daily use.

When defining the variables to predict sewer failure and therefore optimise sewer systems maintenance, it is important to identify the ones that most significantly influence the quality of the predictions or to define the smallest number of variables that is sufficient to obtain accurate predictions. In this study, three different statistical variable selection algorithms are applied for the first time to identify the most important variables for sewer failure prediction: the mutual information indicator, the out-of-bag samples concept, based on the random forest algorithm, and the stepwise search approach. The methods were applied to a real data-set that consists of the categorisation of sewer condition and associated physical characteristics. The mutual information and the stepwise search methods provided good predictions while those obtained using out-of-bag samples based on random forest were somewhat different, justified by the lack of robustness to imbalanced class distributions.

Infrastructure asset management (IAM) of mature urban water infrastructures should address the overall process of balancing out capital improvement, rehab and operational & maintenance expenditures over the long term, faced with given strategic objectives, translated into service (performance), cost and risk targets. The ability to follow the entire network in its daily management is at the base of structured short- to long- term IAM procedures. When data collection for these purposes is embedded in everyday activity, continuously making available the key metrics used for diagnosis and prioritization of assets, zones and systems, it gains in validation and repeatability and avoids becoming a tedious chore to be undertaken once a year with piecemeal results. Baseform is a web-served SaaS (Software as a Service) environment designed from scratch to aggregate multiple data sources on a continuous basis, in order to make them available specifically to the analytics and procedures required to fully support IAM in the water utility through a process- focused applicational portfolio. This paper presents the main technology aspects and analyses available in the software, the advantages of serving it as SaaS, the benefits of a fast-growing scale, and illustrates the above with concrete real-life cases.

Failure prediction plays an important role in the management of urban water systems infrastructures. An accurate description of the deterioration of urban drainage systems is essential for optimal investment and rehabilitation planning. In the study presented in this paper, a new method to predict sewer pipe failure based on robust decision trees is proposed. Five other different stochastic failure prediction models – the non-homogeneous Poisson process, the zero-inflated non-homogeneous Poisson process, classical decision tress (CART and Random Forest algorithms), the Weibull accelerated lifetime model and the linear extended Yule process – are also implemented and explored in order to identify models that combine good failure prediction results with robustness. The six models were tested on the asset register and pipe failure register of a large US wastewater utility; only pipe blockage failures were considered in this study. The linear extended Yule process and the zero-inflated non-homogeneous...

Infrastructure asset management of mature urban water infrastructures should address the overall process of balancing out capital improvement, rehab and operational & maintenance expenditures over the long term, faced with given strategic objectives, translated into service (performance), cost and risk targets. As sensors and data sources multiply (SCADA, telemetry, AMR, billing, work orders, GIS, inspections, models), urban water networks generate very large amounts of information on a daily, weekly, monthly basis. Engineering, architecture, and urban planning are undergoing a fundamental revolution — instead of relying only on projected versions of reality (often made decades ago), we can now increasingly observe that reality in detail and continuously adapt the designed landscape, facilities, infrastructures, etc., to the changing needs and intensities. Baseform is a web-served SaaS (Software as a Service) environment where a portfolio of process-focused apps are made available to a managed, unlimited universe of users inside the organization. The data are acquired from their sources through automatable upload, avoiding data migration or the need to replace the existing IT systems. This paper presents the main technology aspects and analyses available in the software, the advantages of serving it as SaaS, the benefits of a fast-growing scale, and illustrates the above with concrete real-life cases.

Real-time monitoring of DMA can be a powerful tool for improving the efficiency of water supply systems and reduce losses. However, in large systems it is unrealistic to be permanently attentive to all zones without automatic monitoring tools, assisting the operations teams in thoroughly monitoring the entire network and systematically tracking down, collecting and validating all relevant information produced. Several sophisticated algorithms, based on robust statistical models and historical flow data, are continuously used by Baseform to accurately predict water demand patterns and to establish whether the consumption recorded, during a specific period, is normal or may reveal abnormal events. The strategic concern of the utility with prioritization of DMA and planning of O&M drive a search for reliable mechanisms to detect outliers, i.e., monitored values that lie outside of normality. This paper introduces a procedure for identification of trends in minimum night flow for mid- to long-term horizons, particularly useful in the early detection of ongoing leaks before they escalate into catastrophic bursts. The main advantages of this approach are illustrated through selected cases of detected events that empowered benefits to day-to-day operations, in a large urban water supply system in a Portuguese utility where the software is in use.

Methods to detect outliers in network flow measurements that may be due to pipe bursts or unusual consumptions are fundamental to improve water distribution system on-line operation and management, and to ensure reliable historical data for sustainable planning and design of these systems. To detect and classify anomalous events in flow data from district metering areas a four-step methodology was adopted, implemented and tested: i) data acquisition, ii) data validation and normalization, iii) anomalous observation detection, iv) anomalous event detection and characterization. This approach is based on the renewed concept of outlier regions and depends on a reduced number of configuration parameters: the number of past observations, the true positive rate and the false positive rate. Results indicate that this approach is flexible and applicable to the detection of different types of events (e.g., pipe burst, unusual consumption) and to different flow time series (e.g., instantaneous, minimum night flow).

A decision support system (DSS) tool for the assessment of intervention\nstrategies (Alternatives) in an urban water system (UWS) with an\nintegral simulation model called `WaterMet(2)' is presented. The DSS\npermits the user to identify one or more optimal Alternatives over a\nfixed long-term planning horizon using performance metrics mapped to the\nTRUST sustainability criteria. The DSS exposes lists of in-built\nintervention options and system performance metrics for the user to\ncompose new Alternatives. The quantitative metrics are calculated by the\nWaterMet(2) model, and further qualitative or user-defined metrics may\nbe specified by the user or by external tools feeding into the DSS. A\nmulti-criteria decision analysis approach is employed within the DSS to\ncompare the defined Alternatives and to rank them with respect to a\npre-specified weighting scheme for different Scenarios. Two rich,\ninteractive graphical user interfaces, one desktop and one web-based,\nare employed to assist with guiding the end user through the stages of\ndefining the problem, evaluating and ranking Alternatives. This\nmechanism provides a useful tool for decision makers to compare\ndifferent strategies for the planning of UWS with respect to multiple\nScenarios. The efficacy of the DSS is demonstrated on a northern\nEuropean case study inspired by a real-life UWS for a mixture of\nquantitative and qualitative criteria. The results demonstrate how the\nDSS, integrated with an UWS modelling approach, can be used to assist\nplanners in meeting their long-term, strategic-level sustainability\nobjectives.

iGPI, the National Initiative for Infrastructure Asset Management is a Portuguese collaborative project led by LNEC (National Civil Engineering Laboratory, Portugal) through which 19 water utilities develop their own infrastructure asset management (IAM) systems and plans in a joint training and capacitation programme. Technical assistance to the participating utilities is ensured by LNEC, IST (Technical University of Lisbon) and Addition, a software development company. The water utilities get collective as well as one-on-one support and specific training. They benefit from networking with the other utilities in a common and simultaneous process, with similar difficulties and challenges, leading to an effective sharing of solutions. The developed products, including training materials, templates and guidelines for developing strategic and tactical IAM plans, are available to the general public. This project has greatly contributed to the establishment of reference methodologies and standards for IAM planning, in a range of utilities of widely diverse size and context, effectively defining an accepted best practice. This paper discusses the project's format and its advantages, and goes on to describe the main outcomes, including selected cases and final products.

Several methodologies can be applied to predict the 24-hour demand pattern in district metering areas (DMA) of water distribution systems. However, in most cases, it is essential to know in advance the characteristics of each flow time-series, to choose the most suitable combination of parameters to use as the input of the method. In particular, the demand pattern derived is dynamic and may be influenced by the period of time considered. An adaptive predictive model of the pattern based on some selection criteria, which automatically choose the most appropriate parameterizations is described, in an attempt to overcome these drawbacks. This new approach is built on the concept of weighted percentiles, allowing the prediction to quickly adjust to changes in consumption habits, seasonality and other trends. The proposed procedure has undergone extensive testing and validated through the application to a large number of DMA, as preparation for use in a commercially deployed software.

Strategic asset management (AM) of urban water infrastructures faces the challenge of dealing with expensive and long-lasting assets of a very diverse nature and wide-ranging useful lives and costs. Typically, utility managers inherit an infrastructure with assets in assorted conditions and stages in their lifecycle. They are expected to manage their value in order to ensure adequate service, and make sure that what they pass on to their successors is capable of continuing to do so. Long-term vision is needed, and sound transition paths must be sought in order to ensure that urban water services are sustainable, without jeopardizing the quality of the service provided during the transition. Focusing on the transition to more sustainable services, the TRUST project revisited the concept of sustainability, explored pressures and drivers, and developed and tested a roadmapping methodology to cope with the implementation of disruptive solutions. Leading-edge analysis and assessment methods and tools were developed, multiple technologies were tested, scientific and guidance publications were produced. All of this can be explored and absorbed into a coherent management process. This paper describes the TRUST approach for infrastructure AM, addressing strategic, tactical and operational planning, supported by professional-grade software. © 2016 IWA Publishing.

The requirement to provide continuously the urban water services while infrastructures are ageing, impose the need for increasingly sustainable infrastructure asset management (IAM). To achieve and maintain the adequate levels of service, the AWARE-P IAM methodology has been applied in collaborative projects launched by the National Civil Engineering Laboratory, in partnership with IST (Technical University of Lisbon), Addition (software company) and several water utilities. The objective of these projects is to support urban water utilities in the development, implementation and maintenance of IAM plans. To guarantee the success of IAM planning, following the AWARE-P IAM methodology, utilities are required to: consider that the infrastructure has system behaviour and lifespan is indefinite and guarantee the full-alignment of IAM planning with organisation objectives. By analysing the strategic and tactical plans of participating utilities, the proposed methodology principles are discussed and supported. The main innovation results from the implementation of IAM planning are also presented and discussed, including the challenges of setting up an IAM process, together with the major benefits and drawbacks that come up when developing IAM plans. The results were demonstrated by the effective implementation of 16 strategic and 14 tactical IAM plans by the participating utilities.

Managing urban water infrastructures faces the challenge of jointly dealing with assets of diverse types, useful life, cost, ages and condition. Service quality and sustainability require sound long-term planning, well aligned with tactical and operational planning and management. In summary, the objective of an integrated approach to infrastructure asset management is to assist utilities answer the following questions: The AWARE-P approach ([www.aware-p.org][1]) offers a coherent methodological framework and a valuable portfolio of software tools. It is designed to assist water supply and wastewater utility decision-makers in their analyses and planning processes. It is based on a Plan-Do-Check-Act process and is in accordance with the key principles of the International Standards Organization (ISO) 55000 standards on asset management. It is compatible with, and complementary to WERF's SIMPLE framework. The software assists in strategic, tactical, and operational planning, through a non-intrusive, web-based, collaborative environment where objectives and metrics drive IAM planning. It is aimed at industry professionals and managers, as well as at the consultants and technical experts that support them. It is easy to use and maximizes the value of information from multiple existing data sources, both in data-rich and data-stressed environments. This project aimed at testing, validating, and tailoring the AWARE-P approach and software in the U.S. wastewater services context; developing new open-source software tools.

The requirement to provide continuously the urban water services while infrastructures are ageing, impose the need for increasingly sustainable infrastructure asset management (IAM). To achieve and maintain the adequate levels of service, the AWARE-P IAM methodology has been applied in collaborative projects launched by the National Civil Engineering Laboratory, in partnership with IST (Technical University of Lisbon), Addition (software company) and several water utilities. The objective of these projects is to support urban water utilities in the development, implementation and maintenance of IAM plans. To guarantee the success of IAM planning, following the AWARE-P IAM methodology, utilities are required to: consider that the infrastructure has system behaviour and lifespan is indefinite and guarantee the full-alignment of IAM planning with organisation objectives. By analysing the strategic and tactical plans of participating utilities, the proposed methodology principles are discussed and supported. The main innovation results from the implementation of IAM planning are also presented and discussed, including the challenges of setting up an IAM process, together with the major benefits and drawbacks that come up when developing IAM plans. The results were demonstrated by the effective implementation of 16 strategic and 14 tactical IAM plans by the participating utilities.

The report presents a detailed description of a Decision Support System (DSS) methodology and software tool for use as a decision support tool to assist in the management of Urban Water Systems (UWS). The report is divided into the following three principal sections: An initial description of the DSS methodology and modelling concept is followed by a definition of the DSS problem and the elements of the DSS decision matrix and, finally, the ranking of alternatives in the DSS. Specifically, this section describes the DSS structure which encapsulates a framework for the assessment of intervention strategies in an UWS. The internal structure of the DSS engine comprises three principle modules including Environment, Performance and Multi-Criteria Decision Analysis (MCDA). The ‘Environment’ module manages the specifications of the analysis including timing, intervention strategies, PIs, scenarios and customised model input. The ‘Performance’ module is responsible for evaluating the two categories of metrics: (1) quantitative performance metrics calculated by WaterMet2 and the Risk Assessment module; (2) qualitative metrics defined within the DSS and quantified by external tools or third-parties outside the immediate scope of the DSS. The MCDA module applies a user-configured ranking approach to the specified intervention strategies for the purposes of scoring and ranking them for each scenario and user preference combination. The principal stages of the DSS map to four steps including 1) problem definition, 2) population of decision matrix and calculation of metrics (or impact assessment), 3) ranking of alternatives and viewing detailed results and 4) viewing result modification and re-evaluation of intervention strategies. The second part describes the DSS software tool itself. Two complementary interface instances are presented for the DSS, representing Desktop and web-based tools. The overviews of both tools consist of an introduction to how input data are prepared, how to run a simulation and, finally, how to interpret results in different formats. The final part of this report illustrates the use of DSS applied to the case study problem. This describes the configuration of the DSS for the case study problem for a city which faces water scarcity problems over a 30-year planning horizon, starting from year 2015. Seven intervention strategies for ameliorating this issue are examined through both implementations of the DSS. Six performance metrics are considered including five quantitative measures and a single qualitative criterion. The analysis accommodates two rates of future population growth (i.e. low and high) can be envisaged as two individual scenarios for a 30 year planning period starting from 2010. Comparison of the intervention strategies with respect to these performance metrics is also conducted based on three weighting schemes representing differing stakeholder perspectives. These weighting schemes include equal weights, and the perspectives of the Water Company and Consumers. The DSS is able to rank and prioritise the proposed intervention strategies under different individual specified scenarios and weighting schemes and to ultimately combine them to produce a single ranking for each intervention strategy.

A key worldwide challenge in most sectors is to boost the effective adoption of innovation, as underpinned by the new European Union research programme Horizon 2020, which focuses on increasing innovation in Europe from 2014 to 2020. This is particularly relevant in the water sector, often perceived as conservative and averse to change. This paper discusses the role that collaborative knowledge-transfer projects can play in effectively rolling out R&D in the water industry. LNEC (Laboratório Nacional de Engenharia Civil) has designed a structured model based on a phased programme and a network of utilities and researchers. The paper presents the core principles, the rationale, the model and methods used, and the theoretical background, as well as the project's impact, outcomes and products. The discussion highlights the lessons learnt and provides a formal analysis of the advantages of focusing on middle management as an effective entry point, even if innovation is needed across the organization. Making training materials, guidelines, use cases, data and software publicly available after the project's end has proven to have a decisive multiplying effect. The paper also argues in favour of the collaborative model as a basis for R&D sustainability, and details on-going and planned developments. © IWA Publishing 2015.

A Decision Support System (DSS) tool for the assessment of intervention strategies (Alternatives) in an Urban Water System (UWS) with an integral simulation model called “WaterMet2” is presented. The DSS permits the user to identify one or more optimal Alternatives over a fixed long-term planning horizon using performance metrics mapped to the TRUST sustainability criteria (Alegre et al., 2012). The DSS exposes lists of in-built intervention options and system performance metrics for the user to compose new Alternatives. The quantitative metrics are calculated by the WaterMet2 model and further qualitative or user-defined metrics may be specified by the user or by external tools feeding into the DSS. A Multi-Criteria Decision Analysis (MCDA) approach is employed within the DSS to compare the defined Alternatives and to rank them with respect to a pre-specified weighting scheme for different Scenarios. Two rich, interactive Graphical User Interfaces, one desktop and one web- based, are employed to assist with guiding the end user through the stages of defining the problem, evaluating and ranking Alternatives. This mechanism provides a useful tool for decision makers to compare different strategies for the planning of UWS with respect to multiple Scenarios. The efficacy of the DSS is demonstrated on a northern European case study inspired by a real-life urban water system for a mixture of quantitative and qualitative criteria. The results demonstrate how the DSS, integrated with an UWS modelling approach, can be used to assist planners in meeting their long-term, strategic level sustainability objectives.

The proper management of wastewater pipes is an important issue in today’s society, with serious financial and health implications. The ability to adequately prioritise maintenance inspections on these pipes may consequently significantly increase the quality of life of the affected populations. In this thesis, techniques to achieve this are studied. Using Data Mining procedures the goal is to be able to predict which pipes are more likely to be close to failure. The main prediction method used is the random forest, although other algorithms are also studied, in particular, conditional forests, logistic regression and Naive Bayes. To have a more global view of the problem, as well as to be able to obtain better predictions, variable selection techniques are also studied and applied, namely, mutual informa- tion and step-backwards search. Since it was found that (fortunately) only very rarely are pipes found to be in poor state, the examination of the problem extends to class balancing methodologies. Results show that there is no clear winning algorithm, although both ensembling and class balancing techniques manage to boost the performance of the tested algorithms. The results obtained are com- plemented by confidence measures to try to placate variability. Parallel to this investigation, a restriction of the problem to the pipes having a specific type of damage is conducted, and its results compared with the more general model.

The aim of this paper is to present the collaborative project iPERDAS, on water losses and energy efficiency management, and provide examples of the main results achieved by the participants during 2014. iPERDAS, led by LNEC (National Civil Engineering Laboratory, Portugal), was a project through which 17 water utilities developed their own water-energy losses programmes, following a joint training and capacitation approach (www.iPERDAS.org). The group of participating water utilities was distributed across the country and supplies water to nearly 16% of the Portuguese population along 11633 km of pipes. The group was quite heterogeneous in terms of dimension, corporate structure and level of maturity in water losses management which enabled rich discussions and sharing of best practices. Utilities have received collective as well as one-on-one support, specific training, and benefited from networking with each other in a common and simultaneous process, with similar difficulties and challenges, leading to an effective sharing of solutions. The most significant results of this project include the utilities’ Water and Energy Losses Management Plans, the establishment of solid procedures to carry out water and energy audits, the improvement of inter- departmental communication and the development of an integrated and sound organisational process for water-energy management.

The control of water losses is a major concern in the sustainability of urban water utilities and in promoting the efficient use of this natural resource. Recent advances in telemetry technology provide high-resolution consumption data at the consumer level, allowing for a remarkable knowledge improvement on the different water balance components. However, few studies focus on systematic approaches for improving system operation and maintenance by processing and analysing large amounts of consumption data. This paper presents a new methodology to calculate real losses and apparent losses in distribution networks using data collected from telemetry systems. The methodology proposes a set of algorithms that are simple to implement. These algorithmswere tested on different districtmetered areas (DMA) to improve understanding about water loss components and have already been included in commercial software. The results showed that these algorithms are robust and allow for accurately estimating the background leakage level (and unreported leaks and bursts), detecting earlier the occurrence of bursts and providing important insights into the type of illegal water uses. The use of these approaches reduced non-revenue water by more than 10% in the majority of the DMA tested. These findings are promising and demonstrate the strong potential of telemetry systems to reduce water losses and to improve the understanding of water uses. Key words | apparent losses, real losses, smart metering, water consumption, water losses

The present paper aims at providing reference values of consumption variables and daily consumption patterns for a variety of scenarios based on rigorous analysis work over 37 different urban and peri-urban DMA (District Metering Areas) in the northern and southern regions of Portugal. A comprehensive methodology that involves collecting, processing, characterizing and profiling demand, infrastructure, billing and socio-demographic data was adopted. This paper focuses on the results obtained on consumption data characterization step. Main results have shown that the analysed regions have distinct socio-demographic, infrastructure and billing make-ups that reveal differences in terms of water consumption trends. In the north region, families are bigger, consume less water per capita comparatively to the south and have a more regular consumption throughout the day. In the south region, population is more active and consume water essentially in the morning and night periods.

Condition prediction and deterioration forecasts are crucial for long-term planning of water and wastewater systems. Although sufficiently beneficial to inspire full coverage strategies, particularly in sewers, inspections are costly and it may be valuable to prioritize them based on correlations between previous results and potentially explanatory factors (e.g., age, material). The paper reports on a detailed application of a Random Forest classification algorithm in the context of a utility's thorough ongoing CCTV inspection program and AM planning, where it has been used both for estimating the importance of potential co-variates in predicting sewer condition and in directing investment in inspections.

Nowadays a variety of methodologies (e.g. black-box predictors, absolute minimums of night flows) are applied to extract relevant information for leakage control, network operation, and system planning and rehabilitation. However, the value of profiling the 24-hour demand cycle as a central element for statistical analysis, as well as a definition of normality and behavior, is often overlooked. A reliable set of methods and algorithms including data acquisition, validation and normalization procedures to obtain reliable 24-hour demand profiling is described, aided by an efficient software implementation, validated through the application in a large number of DMA from over 40 different utilities.

Smartmetering is already being used inmany utilities tomanage water loss and customer consumption, but there is also the potential to improve network planning and operation and the overall system’s water-energy efficiency. PAULA VIEIRA, DÁLIA LOUREIRO, JOSÉ BARATEIRO, RITA RIBEIRO,MARGARIDA REBELO and SÉRGIO TEIXEIRA COELHO discuss the development of a new ICT tool, iWIDGET, which brings together a set of applications for smartmetering data.

Smart metering (SM) has the potential to improve distribution systems planning and operation and the overall system’s water-energy efficiency. Although numerous water utilities already explore SM systems, the main focus has been water loss management and control of large consumers’ consumption and most of the existing SM technologies are often used in isolation. Therefore, there is a need for a more integrated and complete approach that covers non-current applications of SM data on water and energy consumption. The work presented in this paper identified and validated a set of relevant applications derived from SM real-time data, that are valuable for the water utilities. Moreover, it is presented a method to test the ICT tool development that implements the applications.

The National Initiative for Infrastructure Asset Management (iGPI) led by LNEC, IST and Addition, and its twin initiative PGPI, led by AGS, were collaborative projects through which a total of 30 Portuguese water utilities developed IAM systems and plans through a collective training, capacitation and R&D rollout programme (Apr 2012 – Oct 2013). The utilities received specialized training, collective and individual support, and networked with the other utilities in a common process with similar difficulties and challenges, leading to an effective sharing of solutions. The products, including software, training materials, guidelines and templates for developing strategic and tactical IAM plans, are in the public domain. The projects have significantly contributed to the rollout of reference methodologies and standards for IAM planning, demonstrated in a range of utilities of varied size and context, successfully pushing a consensual best practice. The methodology and outcomes are discussed and illustrated with selected cases. Keywords:

Strategic asset management of urban water infrastructures jointly deals with assets of diverse nature, useful life, cost, age and condition. Service sustainability requires a sound long-term planning, which needs assessing, among other aspects: the value of the infrastructure over time; the need for reinvestments; the impact of long-term re-investment policies. The infrastructure value index (IVI) was proven to be a powerful modelling tool for combined long-term planning of linear and vertical assets. An open-source software enables IVI assessment for both asset-by-asset detailed inventory and for simplified cohort-based infrastructure description. This paper presents the formulation, discusses the underlying assumptions and applicability, and illustrates its use for strategic planning.

Water utilities have exponentially increased the usage of software and data solutions in recent years, from daily operation to long-term planning and corporate management. The cumulative refinement of generic software approaches has led to major advances, but marginal improvements are shrinking, as the limiting factors imposed by legacy software paradigms take over. This poster describes a collaborative and pluggable software platform and project, hosting an open range of apps focused on specific reengineering of core water utility engineering processes. Software development is combined with a long-term R&D roll-out strategy, connecting with the industry and a network of public research organizations. With references to practical workflows in water utilities, the text illustrates the increased efficiency of this new software paradigm, resulting in measurable alignment of operational, tactical and strategic objectives, processes and users.

The failure prediction process plays an important role in the management of urban water systems infrastructures. An accurate description of the deterioration of urban drainage systems is necessary for optimal investment and rehabilitation planning. Predicting urban drainage system and sewer pipe failures is an important challenge since available failure data often have a short failure history and incomplete records. In the study presented in this thesis, five different failure prediction models, the non- homogeneous Poisson process, the zero-inflated non-homogeneous Poisson process, decision trees, the Weibull accelerated lifetime model and the linear extended Yule process (LEYP), were implemented and explored in order to identify models that combine good failure predic- tion results with robustness, when applied to different pipe samples. The five models were applied to a failure data set provided by The LEYP and the zero-inflated non-homogeneous Poisson process presented the overall best results throughout the models comparison, presenting accurate predictions and a good ability to detect pipes with high likelihood of failure. Decision trees only detected pipes with high likelihood of failure when considering a short- term predicting window and the accuracy of the predictions was one of the best when using robust decision trees. TheWeibull accelerated lifetime model was one of the best when considering a medium-term prediction window but failed to predict when considering a shorter one and the expected number of failures could only be obtained using a Monte Carlo simulation, which can be a time-consuming procedure.

The modelling of chlorine residual in water supply systems is of great importance in managing disinfectant concentrations throughout the network. First order decay kinetics are currently often used to describe both bulk and wall chlorine consumption. However, more complex approaches have been proposed, namely a parallel, two-reactant second order decay model (2R model) that has been reported as yielding better accuracy for simulating chlorine bulk decay in laboratory tests. The recent EPANET Multi-Species Extension (EPANET MSX) brought enhanced capabilities for the simulation of chlorine residuals in water supply systems, including the use of the 2R model or any other formulation. In the current paper, the performance of the 2R model as well as of first and nthorder decay kinetics was assessed for full scale modelling of chlorine in a transmission system. Results have shown that a similar level of accuracy can be achieved with the three tested kinetic models, provided that a good calibration of the wall decay coefficient is accomplished. Although with improved modelling capabilities, the use of the standalone EPANET MSX was less user-friendly than normal EPANET application by the lack of a graphical interface allowing for the visualization of chlorine concentration profiles along the system. The use of the 3D-enabled Epanet Java web application circumvented such limitations. This tool together with a better characterization and estimation of the bulk and wall decay components allow for a more practical and accurate modelling of chlorine in water supply systems, while taking advantage of EPANET MSX's enhanced capabilities. © 2013 The Authors.

The AWARE-P IAM planning software offers a non-intrusive, web-based, collaborative integration environment for a wide variety of data and processes that may be relevant to the IAM decision-making process, including maps, GIS shapefiles and geodatabases; inventory records; work orders, maintenance, inspections/CCTV records; network models, performance indicators, asset valuation records, among others. The software provides an organized framework for evaluating and comparing planning alternatives or competing IAM solutions, through selected performance, risk and cost metrics. It comprises a portfolio of system metrics and network analysis tools that may also be used individually for diagnosis and sensitivity gain. The public beta release in early 2012 garnered significant numbers of users worldwide, and subsequent versions and a growing number of utility deployments and pilots have been steadily confirming the potential of its system-based approach. It is based on the collaborative, web-based and highly modular Baseform platform (www.baseform.org), which runs wherever Java is supported, and materializes as an integrated and expandable suite of plug-in tools, taking advantage of the platform's user management, common data integration services and next-generation 2D/3D visualization capabilities with Google Earth® integration among other features. The paper describes the software’s design and main features, and illustrates its main use cases.

Water utilities have addressed the operation of sewer systems with a reactive approach, and lack of data on the condition of sewers hinders the development of predictive models. In this study, customer complaint/ failure databases were used to identify the various problems that are occurring within the sewer system. Two state of the art software tools were used for comparing different statistical models that use complaints/failure databases: Fail model and SIMA model. We found that in general, the NHPP model tends to overestimate failure probability when compared with the Poisson and LEYP models; and that such overestimation is higher for smaller grid cell (or sub-catchment) areas.

This paper presents the integrated methodology for infrastructure asset management (IAM) developed in AWARE-P, an R&D project aimed at producing adequate and effective support tools for assisting urban water utilities in decision making and rehabilitation planning (www.aware-p.org). The proposed methodology addresses all three planning and decisional levels (strategic, tactical and operational) and follows a PDCA-inspired cycle. The methodology assesses the water systems and any planning solutions along the cost, risk and performance dimensions, with a variety of standardized assessment methods and models proposed for each. It differs from existing approaches in the incorporation, into a single organized framework, of the entire IAM process with an integrated, long-term vision of the urban water network – looking at the infrastructure as a system, and not as a mere sum of individual assets. The paper describes the approach and illustrates its application with reference to the several business cases already undertaken.

Reliable water forecasting provides the basis for making operational, tactical and strategic decisions for water utilities. The aim of this paper is to present the results of a spatial and temporal forecasting of water consumption patterns that has been carried out at the District Metered Area (DMA) level. Flow, infrastructure and billing data were collected from water utilities and combined with recently published census data. Results include empiric relations for estimating design and operational parameters for different seasonal and weekday scenarios, as well as average demand patterns for different consumer profiles taking into consideration key variables.

iGPI, the National Initiative for Infrastructure Asset Management (April 2012 - October 2013), is a Portuguese collaborative project led by LNEC (National Civil Engineering Laboratory, Portugal) through which 19 water utilities develop their own IAM systems and plans through a joint training and capacitation programme. Technical assistance to the participating utilities is ensured by LNEC, by a team from IST (Technical University of Lisbon) and by Addition, a software development company. The water utilities get collective as well as one-on-one support, specific training, and benefit from networking with the other utilities in a common and simultaneous process, with similar difficulties and challenges, leading to an effective sharing of solutions. The developed products, including training materials, templates and guidelines for developing strategic and tactical IAM plans, are available to the general public. This project has greatly contributed to the establishment of reference methodologies and standards for IAM planning, demonstrated in a range of utilities of widely diverse size and context, effectively defining an accepted best practice. This paper discusses the project’s format and its advantages, and goes on to describe the main outcomes, including selected cases and final products.

The prediction of pipe failures in urban water systems is a complex process because the available failure records, originating in work orders, are often short and incomplete. To identify a robust and simple model with good failure prediction results using short data history, three existing models were compared in this study: the single-variate Poisson process, the Weibull accelerated lifetime model, and the linear-extended Yule process. This work also presents modifications to these models that enable them to produce more accurate predictions and overcome computational issues for practical software implementation. The three models, together with the improvements where applicable, were applied to water supply system data provided by a Portuguese water utility, and the results were comparatively analysed to assess the accuracy of each model. The Weibull accelerated lifetime model yielded the best results among the three models, accurately predicting failures and detecting pipes with high failure likelihood; however, it is based on Monte Carlo simulations, which can be time-consuming. The linear extended Yule process could also effectively detect pipes with higher failure likelihood; however, it presented a clear tendency to overestimate the number of future failures. The single-variate Poisson process is the simplest of the three models and produced failure prediction results of lower quality.

High levels of deferred maintenance and rehabilitation and overwhelming investment needs in urban water services infrastructure demand wise spending and innovative, efficient planning. Rather than component-centric AM approaches, such as like-for-like prioritization and replacement, the complexity of the problem must be addressed by system-centric methods that ensure the best possible compromise between performance (level-of-service), risk and financial effort. An open-source software system based on a set of tools and models that assist in the analyses and decision support involved in the planning process has been developed in order to host the methods mentioned above. This paper presents the methodology and the software's objectives and features, describes the context and vision that led to its inception and details the main design requirements and technology options.

The most widely used method for assessing real water losses in a district metering area (DMA) involves measuring minimum night flow and estimating what fraction of it is not ascribed to real losses. It may be further broken down into authorized household consumption, authorized non-household consumption and non-authorised uses. Household consumption is in general the most relevant and harder to estimate component. It is very much context-dependent and very few studies are published outside of the UK. This paper summarises the results achieved in a recent study carried out in two Portuguese municipalities, and describes a set of simplified rules developed to predict DMA night household consumption. These rules depend on the number of households and on their social-demographic and billing features. Once household night consumption is estimated, it can then be deducted from to minimum nigh flow monitored in a DMA or part of a DMA, contributing to a more accurate estimate of real water losses. The results obtained show significant differences from case to case, but in most cases lower values than the published UK references. This is believed to show that residential night consumption is rather context-dependent and that real losses may in fact be higher than estimated so far in this region.

This paper presents the application of a Multiple Criteria Decision Analysis (MCDA) framework to help selecting the best rehabilitation solution for a water service. MCDA is performed using different criteria from three dimension of analysis (performance, cost and risk) and is applied to a case study in Portugal. Two external scenarios, four rehabilitation alternatives and six evaluation criteria were considered to test the decision context. The ELECTRE III technique was used to rank the rehabilitation alternatives. The results obtained and the main conclusions are presented in this paper.

The aim of this paper is to compare sorting and ranking methods for prioritization of rehabilitation interventions of sewers, taking into account risk, performance and cost. For that purpose multiple criteria decision-aid (MCDA) methods such as ELECTRE TRI for sorting and ELECTRE III for ranking are applied in a real case-study and the results obtained are compared. The case study is a small sanitary sewer system from a Portuguese utility located in the metropolitan area of Lisbon. The problem to investigate is the prioritization of the sewer candidates for rehabilitation. The decision maker (a panel group of specialists) has chosen five assessment measures: water level and maximum flow velocity (hydraulic performance indices), sewer importance and failure repair cost (collapse-related consequences of failure) and the risk of collapse. The results show that the outcomes from ELECTRE III are easier to understand than those from ELECTRE TRI method. Two different sets of weights were used, and the sorting and ranking results from both methods were found to be sensitive to them. ELECTRE TRI method is not straightforward as it involves technical parameters that are difficult to define, such as reference profiles and cut levels.

This paper presents the integrated methodology for infrastructure asset management (IAM) developed in AWARE-P, an R&D project aimed at producing adequate and effective support tools for assisting urban water utilities in decision making and rehabilitation planning (www.aware-p.org). The proposed methodology addresses all three planning and decisional levels (strategic, tactical and operational) and follows a PDCA-inspired cycle. The methodology assesses the water systems and any planning solutions along the cost, risk and performance dimensions, with a variety of standardised assessment methods and models proposed for each. It differs from existing approaches in the incorporation, into a single organised framework, of the entire IAM process with an integrated, long-term vision of the urban water network – looking at the infrastructure as a system, and not as a mere sum of individual assets. The paper describes the approach and illustrates its application with reference to the several business cases already undertaken.

Water services are a strategic sector of large social and economic relevance. It is therefore essential that they are managed rationally and efficiently. Advanced water supply and wastewater infrastructure asset management (lAM) is key in achieving adequate levels of service in the future, particularly with regard to reliable and high quality drinking water supply, prevention of urban flooding, efficient use of natural resources and prevention of pollution. This paper presents a methodology for supporting the development of urban water lAM, developed during the AWARE-P project as well as an appraisal of its implementation in four water utilities. Both water supply and wastewater systems were considered. Due to the different contexts and features of the utilities, the main concerns vary from case to case; some problems essentially are related to performance, others to risk. Cost is a common deciding factor. The paper describes the procedure applied, focusing on the diversity of drivers, constraints, benefits and outcomes. It also points out the main challenges and the results obtained through the implementation of a structured procedure for supporting urban water lAM.

A paradigm shift is required towards managing urban water services in a more integrated and efficient way, which includes the use of advanced asset management in order to ensure high quality and adequate water supplies in the future. Here, Helena Alegre, Dídia Covas, Sérgio T Coelho, Maria do Céu Almeida and Maria Adriana Cardoso present the AWARE-P R&D project, which aims to produce adequate and effective support tools for assisting urban water utilities in decision making and rehabilitation planning.

High levels of deferred maintenance and rehabilitation and overwhelming investment needs in urban water services infrastructure demand wise spending and innovative, efficient planning. Rather than component-centric AM approaches, such as like-for-like prioritization and replacement, the complexity of the problem must be addressed by system-centric methods that ensure the best possible compromise between performance (level-of-service), risk and financial effort. An open-source software system based on a set of tools and models that assist in the analyses and decision support involved in the planning process has been developed in order to host the methods mentioned above. This paper presents the methodology and the software’s objectives and features, describes the context and vision that led to its inception and details the main design requirements and technology options.

Urban water systems are essential tomodern societies, butmanaging themstrategically is a challenge as they require continuous and costly capitalmaintenance to ensure service sustainability.Wise rehabilitation andmaintenance planning are therefore essential. JMarques, AP Saramago,MH Silva, C Paiva, S Coelho, A Pina, SC Oliveira, JP Teixeira, PC Camacho, JP Leitão and ST Coelho present an application of an innovative urban water infrastructure assetmanagement (IAM)methodology, developed by project AWARE-P, to the case of the Oeiras & Amadora water utility (SMAS O&A), where the application of the AWARE-P approach by SMAS O&A contributed to a shift of paradigmwithin the organisation with regard to rehabilitation andmaintenance planning.

The aim of this paper is to compare sorting and ranking methods for prioritization of rehabilitation interventions of sewers, taking into account risk, performance and cost. For that purpose multiple criteria decision-aid (MCDA) methods such as ELECTRE TRI for sorting and ELECTRE III for ranking are applied in a real case-study and the results obtained are compared. The case study is a small sanitary sewer system from a Portuguese utility located in the metropolitan area of Lisbon. The problem to investigate is the prioritization of the sewer candidates for rehabilitation. The decision maker (a panel group of specialists) has chosen five assessment measures: water level and maximum flow velocity (hydraulic performance indices), sewer importance and failure repair cost (collapse-related consequences of failure) and the risk of collapse. The results show that the outcomes from ELECTRE III are easier to understand than those from ELECTRE TRI method. Two different sets of weights were used, and the sorting and ranking results from both methods were found to be sensitive to them. ELECTRE TRI method is not straightforward as it involves technical parameters that are difficult to define, such as reference profiles and cut levels.

The aim of this paper is to compare sorting and ranking methods for prioritization of rehabilitation interventions of sewers, taking into account risk, performance and cost. For that purpose multiple criteria decision-aid (MCDA) methods such as ELECTRE TRI for sorting and ELECTRE III for ranking are applied in a real case-study and the results obtained are compared. The case study is a small sanitary sewer system from a Portuguese utility located in the metropolitan area of Lisbon. The problem to investigate is the prioritization of the sewer candidates for rehabilitation. The decision maker (a panel group of specialists) has chosen five assessment measures: water level and maximum flow velocity (hydraulic performance indices), sewer importance and failure repair cost (collapse-related consequences of failure) and the risk of collapse. The results show that the outcomes from ELECTRE III are easier to understand than those from ELECTRE TRI method. Two different sets of weights were used, and the sorting and ranking results from both methods were found to be sensitive to them. ELECTRE TRI method is not straightforward as it involves technical parameters that are difficult to define, such as reference profiles and cut levels.

The failure prediction process plays an important role in infrastructure asset management of urban water systems. This process aims at assessing the future behaviour of a urban water network. However, failure prediction in urban water systems is a complex process, since the available failure data often present a short failure history and incomplete records. In the study presented in this thesis, three di erent failure prediction models, the single- variate Poisson process, theWeibull accelerated lifetime model and the linear extended Yule process, were implemented and explored in order to identify robust and simple models that combine good failure prediction results using short data history. The three models were applied to a Portuguese urban water supply system. TheWeibull accelerated lifetime model presented the best results throughout the comparison of the three models, presenting accurate predictions and a good ability to detect pipes with high likelihood of failure. Nevertheless, the expected number of failures can only be obtained using a Monte Carlo simulation, which can be a time-consuming procedure. The linear extended Yule process could also e ectively detect pipes more prone to fail. However, it presented a clear tendency to overestimate the number of future failures. The single-variate Poisson process is a very simple stochastic process, that is easy to un- derstand and to apply. Due to its simplicity, this prediction process lacks the ability of di erentiating e ectively the water network pipes, which leads to lower quality individual failure predictions. It is noteworthy that no signi cant di erence between the three models results was found when predicting failures in pipes with no failure history.

The AWARE-P project aims to promote the application of integrated and risk-based approaches to the rehabilitation of urban water supply and wastewater drainage systems. Central to the project is the development of a software platform based on a set of computational components, which assist in the analyses and decision support involved in the planning process for sustainable infrastructural asset management. The AWARE-P software system brings together onto a common platform the infrastructural geo- referenced data that describe the physical water supply or sewerage systems, as well as a variety of data originating in the several information systems that exist in the water utility corporate environment, such as maintenance and work orders data, billing and costs databases, or network analysis models. Building on those, it makes available an organized set of user- configurable assessment algorithms related to performance, cost and risk, which are used to evaluate alternative system configurations or planning solutions, against current and projected or hypothetical scenarios.

The current paper briefly presents the infrastructure asset management (IAM) methodology developed in the framework of the AWARE-P project for the comparison of alternative rehabilitation solutions and applies it to a Portuguese water supply system taking into consideration different criteria. These criteria incorporate three assessment dimensions – performance, cost and risk – expressed by several assessment measures (e.g., investment cost, risks of pipe leakage, risk of pipe burst and infrastructure index value). Criteria are aggregated by a Multiple Criteria Decision-Aid (MCDA) technique – the ELECTRE III method. Alternative rehabilitation solutions were established and assessed using this approach. Results are presented and discussed.

Water services are a strategic sector of large social and economic relevance. Thus, it is essential that they are managed rationally and efficiently. Advanced water supply and wastewater infrastructure asset management is key in achieving adequate levels of service in the future, particularly with regard to reliable and high quality drinking water supply, prevention of urban flooding, efficient use of natural resources and prevention of pollution. This paper presents an appraisal of the implementation in four Portuguese water utilities of a methodology for supporting the development of urban water infrastructure asset management (IAM), developed during the AWARE-P project. Both water supply systems and wastewater systems were considered. Due to the different characteristics of the operators, the main concerns vary from case to case; some problems are related essentially to performance, others to risk. Cost is a deciding factor that is common to all utilities. The paper describes the applied procedure focusing on the existing differences, drivers, constraints, major benefits and outcomes. It also points out the main challenges and the results obtained by the implementation of a structured procedure for supporting urban water IAM.

This paper presents the integrated methodology for infrastructure asset management (IAM) developed in AWARE-P, an R&D project aimed at producing adequate and effective support tools for assisting urban water utilities in decision making and rehabilitation planning (www.aware-p.org). The proposed methodology is organized into three planning and decisional levels (strategic, tactical and operational); at each level, a cycle based on PDCA (plan-do-check-act) principles runs through (i) definition of objectives and targets; (ii) diagnosis; (iii) plan development; (iv) plan implementation; and (v) plan monitoring and review. The methodology assesses the system and any planning solutions under consideration along three dimensions – cost, risk and performance – with a variety of standardised assessment methods and models proposed for each. Despite being based on IAM principles generally recommended and adopted by leading-edge research, consultant and utility organizations, the proposed methodology differs from existing approaches in the incorporation, in a single organised framework, of the entire IAM process with an integrated vision of the water networks – i.e., looking at the infrastructure as a system, and not as a mere sum of individual assets – in the long-term. Specific methods and software tools have been developed to assist water utilities in implementing the AWARE-P approach in the industry environment, and several business cases are available.

Urban water systems are essential to modern societies; managing them strategically is a challenge, as they require continuous and costly capital maintenance to ensure service sustainability. Wise rehabilitation and maintenance planning are therefore essential. This paper presents an application of an innovative urban water infrastructure asset management (IAM) methodology, developed by project AWARE-P, to the case of the Oeiras & Amadora water utility (SMAS O&A). This IAM approach considers three levels of planning: strategic, tactical and operational, corresponding to long-, medium- and short-term horizons. One of the main advances of the approach is efficiently integrating assessment dimensions other than cost, namely performance and risk. The entire utility was considered at the strategic level and at the first stage of the tactical level. A district metering area of the water supply system was used as a pilot for testing the procedure at the more detailed level of tactical planning. Five IAM alternatives were devised, including a conventional like-for-like pipe replacement alternative and a number of optimal design alternatives that take into account energy efficiency and system resilience. After assessment and comparison, the alternative that represented the best trade-off was selected using a simple multi-criteria decision method. The application of the AWARE-P approach by SMAS O&A contributed to a shift of paradigm within the organisation with regard to rehabilitation and maintenance planning.

This paper describes a study of residential consumption characterization at the district metering area (DMA) level, taking into consideration billing, socio-demographic and infrastructure-related factors. The article describes the methodology used and presents the results of its application to a set of 20 network districts in the Lisbon (Portugal) metropolitan region, over a period of 12 months of continuous monitoring, including the analysis of consumption scenarios; the estimation of daily consumption patterns; and the study of the most relevant factors that influence consumption. The modelling of consumption patterns involved the analysis of large consumers, the identification of consumption scenarios and the estimation of daily consumption patterns. Multivariate statistical analyses were carried out to obtain factors and identify the correlations with consumption statistics. The main correlations with daily, weekly and seasonal statistics are shown.

As the water and wastewater network in Europe is declining, the recommendation is to move towards Asset Management, as this is a proactive approach to system rehabilitation. This thesis has been written at LNEC, Portugal, in cooperation with the AWARE-P team, who is developing a software to deal with future planning of rehabilitation of the network. The work has been to collect the need of data in their modules, looking at data collected from other rehabilitation softwares and comparing it to what is available in Gemini VA to value the use of Gemini VA for rehabilitation planning support. Planning tools for rehabilitation of water and wastewater networks is still in the developing phase. The programs are often data hungry, and collecting data is time demanding. Gemini VA represent a good software for collecting base data used in rehabilitation planning, and is the most used water and wastewater maintenance software in Norway. It's important to identify the limits and possibilities of the software for future development, and the goal of this thesis has been to look into improvements of the software to give better data at a lower time cost. In the Asset Management approach the condition of the network must first be decided. This analysis can be done with basic data such as material, construction year, dimension and condition factor from surveys done on the network (CCTV etc). When the condition is determined, assessments within cost, performance and risk can be completed at the strategic, tactical and operational level. Gemini VA can provide a lot of base data, but even though the data is available, the preparations is time demanding, and the only data validation feature available, is a material-dimension-construction year- validation. Therefore historical data, and the rest of the structural data, must be validated manually. Gemini VA should try to move towards an Asset Managment approach by including more validation reports, pre made statistical reports, and visualisation effects of the statistics. The software should also include cost, more information on stormwater assets, water quality tests and more pumpingstation and treatment plant information. Gemini VA can be used to collect data for a national statistical database for utilities with low or poor record of historical data to retrieve statistical numbers for planning (failure rate for different materials, etc).

Estimation of household consumption is one of the key elements in the analysis of water losses in distribution networks. Household night consumption is very specific of each network and socio-demographic context, and can vary according to multiple factors. This paper describes an on-going study aimed at assisting a Portuguese water utility in the development and implementation of a methodology to quantify household night consumption based on expedite rules, dependent on socio-demographic and billing factors. A stochastic model was developed to simulate household night consumption and was tested in 8 study areas. The paper describes the selection of case studies, the collection and processing of field data, the establishment of significant night consumption metering periods based on cluster analysis, the identification and characterization of rectangular pulse events, and the process of identification of and correlation with possible explanatory factors.

A large number of distribution network pipes in Portugal is over-sized for human demand flows, due to regulatory fire-fighting minimum diameters, which has a direct bearing on water quality, given the potentially lowvelocities it may originate. In order to assess the magnitude of the problem, a large scale studywas undertaken, based on twenty water distribution system sectors, from networks across the country, representative of a wide range of conditions. Calibrated hydraulic models developed and updated by the utilities were provided. The extended period models were used for exhaustive runs of typical operational conditions, and the results - flow velocity, pressure, head loss and Reynolds number - were recorded for the sets of time periods concerned. The resultswere analyzed by using a computer application specifically developed for this purpose.Acomplete set of statistics for the above parameters was produced. Performance assessment of water distribution systems was also performed based on the results obtained. The main conclusions were that (i) the networks are over-designed with very low velocities in most of the time (due to being designed for the peak flow conditions in 40-year time and for fire-fighting conditions) and (ii) the systems are operated with pressures higher than necessary with evident potential of water losses and energy saving.

Urban water infrastructures provide essential services to modern societies and represent a major portion of the value of municipal physical assets. Managing these assets rationally is therefore fundamental for the sustainability of the services and to the economy of societies. “Asset Management” (AM) is a modern term for an old practice—assets have always been managed. In recent years, significant evolution occurred in terms of the AM formal approaches, of the monitoring and decision support tools and of the implementation success cases. However, most tools developed are too sophisticated and data seek for small utilities. The European R&D network COST Action C18 ( www.costc18.org) identified key research problems related to the management of urban water infrastructures, currently not covered by on-going projects of the European Framework Program. The top 1 topic is “Efficient management of small community”. This paper addresses challenges and opportunities for small and medium utilities with regard to infrastructure AM (IAM). To put this into context, the first sections discuss the need for IAM, highlight key recent developments, and present IAM drivers, as well as research and development gaps, priorities and products needed.

The Techneau WP5.5 Water Quality Model (Coelho, 2008) is a water quality modeling platform based on an event-driven, hydrodynamic water supply network model. It provides a basis for the integration of the results of Techneau research being carried out into specific physical-chemical and microbiological processes taking place in the water and its environment, the network (Wricke et al, 2007).

Water and Wastewater companies operating all around the world have faced rising asset management and replacement costs, often to levels that are financially unsustainable.Management of investment needs, while meeting regulatory and other goals, has required: A better understanding of what customers demand from the services they pay for, and the extent to which they are willing to pay for improvements or be compensated for a reduction in performanceDevelopment of models to predict asset failure and to identify and concentrate investment on critical assetsImproved management systemsImproved accounting for costs and benefits and their incorporation within an appropriate cost-benefit frameworkIncorporation of risk management techniquesUtilisation of advanced maintenance techniques including new rehabilitation failure detection technologiesEnhancements in pipeline materials, technologies and laying techniques.These papers developed from LESAM 2007 for inclusion in Strategic Asset Management of Water Supply and Wastewater Infrastructures are focused on the techniques, technologies and management approaches aiming at optimising the investment in infrastructure while achieving demanded customer service standards, and they provide an opportunity to gain access to the latest discussion and developments at the leading-edge in this field. This book will be essential reading for utility operators and managers, regulators and consultants.ISBN: 9781843391869 (Print)ISBN: 9781780401720 (eBook)

The current paper aims at the establishment of criteria and performance measures for the assessment of the effectiveness of energy management processes in water supply systems. The main objectives that generally motivate the implementation of energy management strategies are identified, assessment criteria for each objective and the corresponding performance indicators to be used are recommended. Three new performance indices for evaluating energy efficiency of systems are proposed and tested, and information provided and advantages and shortcomings of each indicator are discussed.

Urban water infrastructures provide essential services to modern societies and represent a major portion of the value of municipal physical assets. Managing these assets rationally is therefore fundamental for the sustainability of the services and for the economy of societies. “Asset Management” (AM) is a modern term for an old practice – assets have always been managed. In recent years, significant evolution occurred in terms of the AM formal approaches, of the monitoring and decision support tools and of the implementation success cases. However, a long way is still to go. This paper presents the highlights of key recent developments in IAM main asset management, with an emphasis on the on-going Portuguese project AWARE-P. The paper also provides an overview, inevitably subjective, of the current drivers, challenges and trends of infrastructure asset management, at an international level. It finishes with some basic guidance on how to implement an infrastructure asset management approach and with some concluding remarks.

Urban sewer systems constitute a significant patrimony in Europe and worldwide. The structural quality and functional efficiency of sewer systems are fundamental to guarantee their good performance. These systems are nowadays facing problems of ageing, malfunctioning and overloading caused by deterioration due to long life times, improper maintenance, population growth or catchment increase. In order to effectively plan the operation, maintenance and rehabilitation of sewer systems it is important to provide methods and tools for wastewater utilities in order to support the technical management of their systems. The importance of performance evaluation has been growing in most services and economic activities in recent years. Decisions must be economically and environmentally sustainable, the customer/consumer must be satisfied by the product or service provided, and legislation becomes ever more stringent and far-reaching. It is increasingly vital for the managers of an organization to know where, how and when changes must be introduced in order to reach a better performance. These decisions must be taken based on all the available relevant information. The objective of the methodology proposed in this paper is to establish, at the engineering level, the use of a performance assessment methodology suitable for the diagnosis, control and selection of solutions for the improvement of technical processes and systems performance. It can be applied to monitoring data or modelling results, in this case both to extended period operational scenarios and to a range of load factors, displaying appropriate dispersion bands (e.g. percentiles). The methodology is objective-oriented, quantitative and constitutes a standardised approach to technical performance assessment of sewerage systems as a useful contribution to the quest for a common language that would facilitate comparisons, prioritization and the setting of service standards. It is based on relevant performance variables or indicators, meaning the quantity which translates the system behaviour or properties from the point of view taken into consideration (e.g. hydraulic or environmental). It can be applied at catchment or system element level, depending on type of available data; performance functions scores the values of the performance variables or indicators against a scale of performance; for the element level assessment a generalising function is defined for extending the element-level calculation across the system, producing zonal or system-wide diagnoses. Through this methodology an overall view of system performance can be achieved facilitating performance comparisons. A case study is presented to illustrate the application.

Water distribution network analysis models are the single most influential tool in understanding existing or projected networks, and a precious aid in the assessment of evolution scenarios. As such, they are at the core of the decision-making process in any short-, medium-, or long-term asset management strategy. However, they are still less than well-established in many parts of the world, particularly in the operational and technical management environments. Experience shows that it is only possible to fully realise the potential of network models in a water utility when there is an internal structure explicitly devoted to the tasks of building, managing and updating them in a sustainable and efficient manner. Externally developed models, where often not enough attention is paid to capability building and organisational issues, have a much greater chance of being abandoned by the utility, and after relatively short life spans. Conversely, the assimilation of model-building knowledge by the utility and the establishment of specific, tried-and-tested internal procedures have shown to be a successful formula that spurs a number of parallel innovative advancements in related areas within the utility. The internal development of simulation models is a powerful strategic driver for the process of information integration across the various information systems (IS) in the utility – GIS, customer/billing system, SCADA system, maintenance IS, laboratory IS (LIMS), etc. Water utilities are massive data generators; such data is, in turn, a precious asset to the organization. Water network models are a prime example of an engineering-oriented application that is very demanding on the quantity, scope and quality of the data it deploys, particularly if correctly and sustainably developed. It so happens that such information is at the very core of the engineering decision-making process, and beyond. The information requirements involved in the well-supported establishment of a network model place demanding questions to the other systems – particularly the GIS, SCADA and billing IS – that inevitably prompt an increase in quality and depth. It makes sense, therefore, that the development of network modelling within a water utility is approached in a structured and systematic way, in order to ensure the highest possible benefit in terms of the effort and resources invested – both in the generation of the initial solutions and in their maintenance and expansion throughout their lifespan (Figure 2 shows a recommended phased strategy for model development). The National Initiative for the Development of Simulation Models (INSSAA), a program ran in Portugal between 2003 and 2006 with the objective of promoting the development and use of simulation models by the utilities, explicitly in the operational and technical management tasks, is thought to be a valuable example of the above principles. Based on phased development, training and knowledge-building strategies, common to the set of 9 participating utilities, the following main components were covered: -Human resources – modelling team set-up, mission definition and organisational fit; training in network analysis model development, use and maintenance. -Planning, development, and commissioning by the utility teams of fully functional, calibrated models of network sectors, with a strategy for full territorial coverage. -Set-up of data and organisational mechanisms and procedures for the integration of the developed tool, its efficient management and its integration with the other IS across the utility. This paper presents the INSSAA program and its stages of implementation, and focuses on the results obtained, including the core gains; the main data, human and organisational difficulties and hurdles; and the ways found to overcome them. A discussion is undertaken on the global issues involved in the efficient creation and maintenance of engineering-oriented data-intensive tools, for the operational and technical management environments of water supply utilities.

Water distribution pipe failure records are a primary source of information regarding the state of conservation of the buried infrastructure and the way it is operated. Failure can be defined as an event that requires an intervention to recover pipeline and system function. Such events are normally caused by break or collapse of pipes, and through leaks in unions. The functional effect of the physical damage is reflected in water supply interruptions that can last from a few hours to days, as well as pressure fluctuations, and flow availability and water quality changes at the consumers’ tap. Knowledge on the weakness of the network – the areas, materials or pipe types with the greatest numbers of occurrences, and the probable causes behind them, is a critical type of information for the network manager, who can act in anticipation to try and eliminate or reduce the failures, the associated costs and the drop in the levels of service at the consumers’. The purpose of this paper is to contribute to a systematic process of information collection on failures in distribution systems, and to the development of tools for the prediction of failure events, based on the corresponding database of historical failures. Two examples of the latter are made available by the proprietary Care-W procedure and software (http://care-w.unife.it/), particularly through a PhM and a Poisson-based tools used to predict failure rates for classes of explanatory variables chosen by the user in a water distribution system. The methods proposed here were developed without the need for sophisticated or expensive software, and are made available based on simple MS® Excel® spreadsheets. Pipe failure records are grouped into explanatory classes of variables, and data is tested for fit to the statistical distribution used (Poisson); statistical indicators are used to evaluate if the classes have been well selected. With the failure rate for each class, and assuming a given Poisson distribution for the failures, it is possible to calculate the failure probability of each individual pipe. A GIS system (ESRI® ArcGIS 9.1®) is used in order to geographically illustrate failure rates and probabilities (Figure 1). Given those, a tool was developed for assessing the functional consequence of failure of each pipe in the network, in terms of pressure and affected demand, using a network model to evaluate each scenario. This tool was developed using ESRI® ArcObjects® and the Epanet simulator library of functions, and is fully integrated in the GIS system. Running the two mechanisms in conjunction, a quantitative measure of the risk of failure for each pipe (risk=probability of failure x consequence) can be mapped out (Figure 2). This methodology has been tested and used on the distribution system of Lisbon region municipality, with full integration of the data and results on the GIS system. It is found that the insight into the functional effect of the failures, together with the careful exploration of failure rates and probabilities, affords better supported decision-making in both operational and investment terms.

Understanding the factors that influence daily and weekly water consumption patterns in distribution systems is fundamental for both long term tasks, such as planning, design, expansion or rehabilitation of the systems, and short term tasks, such as routine system operation or emergency management. This paper describes the development of a Portuguese nationwide program for characterizing residential water consumption, through the use of a specific consumption data analysis protocol and the build-up of a database of demand patterns; and for associating it with a range of easily obtained, potential relatable factors that include the technical features of the network (e.g., service pressure, network state and maintenance level, use of household tanks), billing and customer statistics and a range of social-demographic variables, such as age, social-economic level and mobility of consumers, age and type of buildings, and economic activities. The program is currently being test-run on a set of 20 metering districts from 9 water utilities, comprising a variety of network types and demand characteristics across representative urban areas of Portugal. The metering districts range in size from 2,000 to 12,000 connected properties, and have mostly been monitored since 2005 (in some cases, 2004). Total consumption is continuously monitored during the main seasonal scenarios (typically winter and summer) in the areas selected. Standardized daily demand patterns and consumption statistics are produced for each day of the week, as well as for each season, in case there is seasonality in the area. Social-demographic information is based on statistics obtained from the 2001 National Census for the basic statistical units, which are around 300 dwellings in size. After the initial run, the program is to be disseminated and made accessible to all water utilities willing to participate, on a voluntary basis and with data anonymity. This will be done mostly through a dedicated website containing the consumption database, together with query and analysis tools, as well as a supervised mechanism to submit new cases in standard format. It is believed that the publication of the result database and its continued growth will provide the water industry designers, consultants and managers with much more reliable and updated data on water consumption than previously available. It will also represent a means to estimate demand through correlation and profiling, in cases when there are no records to work from. The paper discusses the demand analysis performed, as well as the development of the set of potential explanatory factors (technical features of the network and social-demographic variables), and illustrates with the results obtained so far.

The structural quality and functional efficiency of sewer systems are key parameters to guarantee the transfer of domestic, commercial and industrial wastewater to treatment plants without infiltration nor exfiltration. Infiltration of groundwater is particularly detrimental to treatment plant efficiency, while exfiltration of wastewater can lead to groundwater contamination. The APUSS project associating universities, SMEs and municipalities in 7 European countries, developed new methods and techniques to assess and quantify infiltration and exfiltration (I/E) in sewer systems. This paper describes the establishment of a set of performance indicators (PI) developed to assess the impact of I/E on sewer systems and their application to three project case studies, in Italy and France, focusing on sewer systems characteristics, I/E measurements campaigns and PI application results. The methodology for PI definition consists in the selection and development of a sewer network property or state variable, which is expressive of aspects being scrutinized (I/E); the PI values are then calculated; finally, a classification of the PI values is made in relation to good or bad performance. The use of PI allows a standardized and objective comparison of the performance of sewer systems and constitutes a means to technically support the establishment of priorities for rehabilitation and/or construction investments taking into account I/E impacts.

A model to support decision systems regarding the quantification, location and opening adjustment of control valves in a network system, with the main objective to minimise pressures and consequently leakage levels is developed. This research work aims at a solution that allows simultaneously optimising the number of valves and its location, as well as valves opening adjustments for simulation in an extended period, dependently of the system characteristics. EPANET model is used for hydraulic network analysis and two operational models are developed based on the Genetic Algorithm optimisation method for pressure control, and consequently leakage reduction, since a leak is a pressure dependent function. In these two modules, this method has guaranteed an adequate technique performance, which demands a global evaluation of the system for different scenarios. A case study is presented to show the efficiency of the system by pressure control through valves management.

Performance assessment in urban water infrastructures is an increasingly important field of knowledge. Performance has traditionally been expressed in a variety of ways relating mostly to local design practice, with hardly any consensus on how it should be measured or compared. The efficient technical management of these systems deserves a specific approach, suited to the methodologies regularly employed while planning, designing, constructing, operating, and maintaining the systems. At the engineering level, decisions are based on operational, physical, and resources data and on analyses deploying simulation models, geographic information systems, or other information systems. However, such tools tend to produce vast amounts of insufficiently aggregated or performance-oriented information. This paper presents a performance assessment system that is based on the decisional concept of utility functions and designed as a technical analysis tool with the purpose of shifting the focus of technical management of urban drainage systems to a performance-oriented view.

The structural quality and functional efficiency of sewer systems are key parameters to guarantee the transfer of domestic, commercial and industrial wastewater to treatment plants without infiltration nor exfiltration. Infiltration of groundwater is particularly detrimental to treatment plant efficiency, while exfiltration of wastewater can lead to groundwater contamination. The APUSS project associating universities, SMEs and municipalities in 7 European countries, developed new methods and techniques to assess and quantify infiltration and exfiltration (I/E) in sewer systems. This paper describes the establishment of a set of performance indicators (PI) developed to assess the impact of I/E on sewer systems and their application to three project case studies, in Italy and France, focusing on sewer systems characteristics, I/E measurements campaigns and PI application results. The methodology for PI definition consists in the selection and development of a sewer network property or state variable, which is expressive of aspects being scrutinized (I/E); the PI values are then calculated; finally, a classification of the PI values is made in relation to good or bad performance. The use of PI allows a standardized and objective comparison of the performance of sewer systems and constitutes a means to technically support the establishment of priorities for rehabilitation and/or construction investments taking into account I/E impacts.

This paper addresses the issue of performance assessment in water and wastewater systems by reviewing two of the tools that are available to the companies operating in this field: systems of performance indicators and technical performance assessment through simulation. The paper focuses on a technical performance evaluation framework and its application to wastewater systems. The assessment is carried out through the application of utility functions to the network’s elements, and the production of performance- oriented graphs that yield concise and informative views about the system’s behaviour. Two case studies involving a combined sewer system and a separate domestic system are presented in order to illustrate the method.

Degradation of water quality in supply and distribution systems is one of the foremost problems that water companies are faced with. It is everyday practice in many countries to maintain a residual quantity of a disinfectant such as chlorine throughout the system, in order to ensure the safety of distributed water. However, the chlorine added at the water treatment plant disappears as it travels to the consumer tap, due to reactions in the bulk of the flow and at the pipe walls. This paper analyses the influence on free chlorine bulk decay of several water quality parameters, such as temperature, initial chlorine dosage, organic matter and iron content. The performance of five alternative kinetic models for describing the disinfectant behaviour is also assessed, considering both static and steady-state flow conditions. For the conditions tested, a parallel pseudo-first order model provided the best results.Although, in practice, parallel first order kinetic models fit experimental data better than the classic first order model, most water quality modelling software still relies on the latter as a base choice. The paper discusses practical conditions that may influence model selection for network simulation purposes, and the relevance of the associated errors.

The research described in this paper explores the suitability of a method for estimating the leakage-associated parameters and reaching a more plausible demand description than that afforded by traditional static estimates, when modelling the hydraulic behaviour of a water distribution system. Leakage is represented as a uniformly distributed, pressure-dependent demand and is deducted from the metered total system demand. The method relies on a genetic algorithm as the optimisation technique and uses de Epanet 2.0 programmable toolkit in order to create a general-purpose network analysis program.

The maintenance of a residual quantity of a disinfectant such as chlorine throughout the system, in order to ensure the safety of distributed water is current practice in many countries. Chlorine added at the water treatment plant decays as it travels to the consumer tap, due to reactions in the bulk of the flow and at pipe walls. The classic kinetic model used to describe chlorine bulk reactions in most water quality modelling software packages is a first order decay model. In practice, other models provide considerably better fits in laboratory decay tests. Therefore, when the first order model is preferred to other models for simulation purposes, it is important to evaluate the magnitude of the associated errors. This paper discusses practical conditions for the application of first order/parallel first order models, based on the study of the influence of several water quality parameters (temperature, initial chlorine dosage, organic matter and iron content) on chlorine decay. Laboratory decay tests were carried out using groundwater and the performance of five kinetic models for describing bulk decay has also been assessed.

In countries suffering from chronic water shortages, water distribution systems are often operated on an intermittent basis, leaving them unpressurised and allowing ingress of contaminated groundwater, particularly in areas lacking sewerage systems. The periods of stagnation can promote microbial regrowth, further compromising water quality. On the other hand, peaks in pressure and velocity in the pipe network lead to biofilm detachment events, thus releasing microbial cells that may escape the regular quality monitoring procedures and thus increase risk for the consumer. Finally, intermittent supply requires consumers to store water in household storage tanks, which often encourage bacterial regrowth. This paper reports on research undertaken in Jordan, Lebanon, Palestine, the UK and Portugal, which provided an improved understanding of the factors controlling bacterial concentrations in intermittent supplies, and presents an overview of the most important aspects of water quality control in intermittently operated networks.

Understanding chlorine residual decay kinetics and the factors that influence them are essential for such current tasks as siting chlorination facilities, dosage optimisation, choice of sampling locations and frequencies, and general design and operational control of drinking water networks, increasingly accomplished with the help of simulation models. Available constants for bulk chlorine decay are typically determined under static conditions. However, as for all fast reactions in water flows, chlorine consumption rates in drinking water pipes may be influenced by the existing mixing regimes, a function of flow turbulence, which is primarily controlled by flow velocity and pipe diameter. Flow velocities vary greatly in space and time in water transmission and distribution systems; pipe diameters are seldom uniform. Although both variables are readily available in the currently available network analysis simulators that implement chlorine models, such variations are not accounted for. Instead, a single preset decay rate constant is generally used for describing chlorine residual consumption throughout an entire system. In addition to highlighting the negligibleness of PVC pipe wall chlorine consumption, as such, this paper presents experimental evidence of a significant correlation between pipe flow velocity and bulk chlorine decay rates, and proposes a simple but effective approach to implement this dependency in current simulators.

The study and modelling of chlorine decay in distribution systems has been the subject of intense research in the last decade. This paper reports on a series of experiments carried out on a lifescale test rig in order to quantify any possible effect of hydraulic transients on residual chlorine concentrations in drinking water, comparatively to the decay in steady-state flow conditions. The results relate to two types of tests: steady-state tests were run in order to investigate the decay of residual chlorine during steady-state flow, for varying values of the Reynolds number; transient tests comprised an initial steady-state period, followed by successive transient events and a new steady-state period. The transient events were induced by rapid valve closing manoeuvres. The series of results obtained provide evidence towards the existence of a slowing down effect of hydraulic transients on chlorine decay rates, for the conditions represented by the experiment.

The authors developed a model with the capacity to support decisions regarding the quantification, localisation and opening adjustment of valves in a network system, with the objective to minimise pressures and consequently water losses. Two operational modules are establlished: (1) for the evaluation of an objective-function to optimise the localisation of valves in the network system; (2) for the opening valve adjustment in order to optimise pressures through the evaluation of an aptitude function. New values of these parameters are generated randomly by using the technique of genetic algorithms; in order the system reaches its optimum value in terms of pressure minimisation (i.e., consequently leaks are minimised), and EPA’s programmable EPANET routines as general-purpose for network.

Most European water supply stakeholders have seen a good number of their existing water mains reach their expected lifetime, and high capital investments are required to renovate them. Rehabilitation is nowadays an imperative topic in the water industry agenda. Well-devised methodologies for rehabilitation planning are needed, in order to define where, when and how to rehabilitate water transmission and distribution networks. In principle, asset condition should be directly taken into consideration. However, that is a challenging task because mains are buried infrastructures, not easily inspected. Indirect measures such as performance indicators (PI) must be considered. The use of PI to monitor the evolution of mains behaviour and assist in the decision-making process is seen as one of the tools with greatest potential in this framework. This paper focuses on the work developed with regard to the selection and use of rehabilitation PI, in the scope of research project CARE-W: Computed Aided REhabilitation of Water networks. Decision Support Tools for Sustainable Water Network Management.

The assessment of performance in public utilities infrastructure is an important, yet still embryonic field of knowledge, especially as regards urban sewer systems. Performance has traditionally been expressed in terms that relate to local design practices. This has resulted in a variety of ways of assessing performance in different countries without a consensus on how it should be expressed. While performance indicators are a suitable tool for addressing the utility s performance in all its sectors of activity, the efficient technical management of the systems deserves a specific approach, better suited to the type of methodologies regularly employed while planning, designing, constructing, operating, maintaining and rehabilitating the systems. At this engineering level, decisions are made based on operational, physical and natural resources data, and on the results of analyses deploying common engineering tools such as simulation models and information systems. However, such tools tend themselves to produce vast amounts of insufficiently aggregated information, which is scarcely performance-oriented. This paper presents the application of a performance assessment system, designed as a technical analysis tool, in order to shift the focus of technical management of urban drainage systems to a wider, more rigorous, performance-oriented view. The system draws from the experience gathered in performance assessment in water supply, and is based on the decisional concept of utility (performance) functions, in order to accommodate individual sensitivities and interpretations. The system is implemented as a post-processor to sewer modelling, showing significant potential in the performance evaluation of sewer systems, and provides a decision support basis for sewer system design, diagnosis, operation and rehabilitation. A comparative study involving several subsystems of the Estoril Coast interceptor system (Portugal) is presented in order to illustrate the method.

This paper presents the application of a performance assessment system, designed as a technical analysis tool, in order to shift the focus of technical management of urban drainage systems to a wider, more rigorous, performance-oriented view. The system draws from the experience gathered in performance assessment in water supply, and is based on the decisional concept of utility (performance) functions, in order to accommodate individual sensitivities and interpretations.

Understanding chlorine residual decay kinetics and the factors that influence them are essential for such current tasks as siting chlorination facilities, dosage optimisation, choice of sampling locations and frequencies, and general design and operational control of drinking water networks, increasingly accomplished with the help of simulation models. Available constants for bulk chlorine decay are typically determined under static conditions. However, as for all fast reactions in water flows, chlorine consumption rates in drinking water pipes may be influenced by the existing mixing regimes, a function of flow turbulence, which is primarily controlled by flow velocity and pipe diameter. Flow velocities vary greatly in space and time in water transmission and distribution systems; pipe diameters are seldom uniform. Although both variables are readily available in the currently available network analysis simulators that implement chlorine models, such variations are not accounted for. Instead, a single preset decay rate constant is generally used for describing chlorine residual consumption throughout an entire system. In addition to highlighting the negligibleness of PVC pipe wall chlorine consumption, as such, this paper presents experimental evidence of a significant correlation between pipe flow velocity and bulk chlorine decay rates, and proposes a simple but effective approach to implement this dependency in current simulators.

L'analyse de la performance est devenue tres importante pour la gestion des reseaux d'assainissement, pas uniquement comme le resultat d'une evolution naturelle des methodes de modelisation et de conception, mais egalement comme une reponse a la recherche permanente de l'amelioration du service fourni aux utilisateurs. Les indicateurs de performance constituent un outil puissant de comparaison de performances independantes ; ils peuvent etre exploites comme support pour etablir des strategies de rehabilitation et des options de gestion. Cette communication presente une methodologie systematique et standardisee d'evaluation de la performance des reseaux d'assainissement, basee sur des indicateurs de performance. Elle est illustree par une application au reseau de Laje, situe sur la zone cotiere du Portugal. L'objectif principal de ce travail est de developper un systeme d'evaluation applicable a tous les types de reseaux d'assainissement.

Computer programs have been available to support various tasks in water\nsupply and distribution, from network analysis and modelling\napplications to a variety of data processing software for billing,\naccounting, network data acquisition and geographical information\nsystems. These applications have mostly grown in a fragmented way as\nthey became necessary, and evolved on their own. For example, most\nhydraulic models in use today have typically evolved from their earlier\nmodel development through step-wise improvements. With rapid changes in\nthe computing technologies only introduced and made feasible in the\nrecent years, such incremental updates often suffer from various systems\nlimitations the original architectures were not designed for. This paper\nsummarises some important aspects of the authors' experience in adopting\nthe latest computing developments, together with the current\nadvancements in hydraulic modelling and engineering for the\nimplementation of a new generation of modelling tools.

Rigorous application of the method to different performance fields of water networks, illustrated by various case studies. Detailed analysis of key areas of water networks' behaviour, including advances in the analysis and modelling procedures that are currently available, e.g. in water quality modelling and reliability analysis. Applicable to a range of engineering problems in water distribution. Performance analysis is becoming a key issue in the engineering approach to the control of water supply and distribution systems, both as a natural process of evolution of the modelling and design methods available, and as a consequence of an ever increasing awareness of the quality of the service provided within the water industry today. This book presents a systematic approach to the analysis of technical performance of water networks by creating a framework in which a variety of concepts and criteria can be included. The system presented provides a shift in the way engineering problems are formulated in water supply, allowing greater control of analysis objectives and improved sensitivity.

Engineering information systems are characterized by both static infrastructure and dynamic time dependent data. Telemetered readings can often be relatively substantial over an extended operational period. The use of these data to support both engineering and operational decisions processes require the availability of an open and flexible data application. Design and implementation of such a system is presented in this paper.

Over the past twenty years a major financial effort has been made to improve the water supply in Portugal. A significant evolution can be observed, with 77% of the population currently connected to public systems and a target of 95% by the year 2000. Supplying a high percentage of the population is however not sufficient by itself. The quality of the service provided to the users is a fundamental issue, particularly where continuity of supply, water quality and reliability are concerned. Recent surveys based on simplified criteria indicate that 40% of the population supplied are thought to enjoy a good service, 54% an adequate service, and 6% a poor service.

Hydraulic simulation and demand analysis have been, for some years now, key research areas for the water supply team at the National Laboratory of Civil Engineering of Portugal. It is felt that the planning, design, operation and control of water supply and distribution system are greatly improved by the use of adequate analysis and processing tools.

The design, operations and management of water supply and distribution systems require considerable amounts of data and data processing and analysis. What is not available to the water industry in the present days is a comprehensive information system to offer the necessary facilities for both the data management and the desirable applications. This paper describes a joint effort in the development of such a system.