Balvant Rajani

Comprehensive review of structural deterioration of water mains: statistical models

Abstract This paper provides a comprehensive (although not exhaustive) overview of a large body of work carried out in the last 20 years to quantify the structural deterioration of water mains by analysing historical performance data. The physical mechanisms that lead to pipe failure often require data that are not readily available and are costly to obtain. Thus, physical models may currently be justified only for major transmission water mains, where the cost of failure is significant, whereas statistical models, which can be applied with various levels of input data, are useful for distribution water mains. The statistical methods are classified into two classes, deterministic and probabilistic models. Sub classes are probabilistic multi-variate and probabilistic single-variate group processing models. The review provides descriptions of the various models including their governing equations, as well as critiques, comparisons and identification of the types of data that are required for implementation. In some cases, a brief description of the methodology is provided where a decision support system was developed based on a specific statistical model. A companion paper “Comprehensive review of structural deterioration of water mains: physical models” helps to complete the picture of the work that has been done on the subject of water main deterioration and failure.

Comprehensive review of structural deterioration of water mains: physically based models

Abstract This paper provides a comprehensive (although not exhaustive) overview of the physical/mechanical models that have been developed to improve the understanding of the structural performance of water mains. Several components have to be considered in modelling this structural behaviour. The residual structural capacity of water mains is affected by material deterioration due to environmental and operational conditions as well as quality of manufacturing and installation. This residual structural capacity is subjected to external and internal loads exerted by the soil pressure, traffic loading, frost loads, operational pressure and third party interference. Some models address only one or a few of the numerous components of the physical process that lead to breakage, while others attempt to take a more comprehensive approach. Initial efforts were aimed mainly towards development of deterministic models, while more recent models use a probabilistic approach to deal with uncertainties in defining the deterioration and failure processes. The physical/mechanical models were classified into two classes: deterministic and probabilistic models. The effect of temperature on pipe breakage is discussed from three angles; the first deals with temperature effects on pipe-soil interaction, the second deals with frost load effects and the third provides a brief review of various attempts to statistically quantify influence of temperature on water main failure. This paper complements the companion paper “Comprehensive review of structural deterioration of water mains: statistical models”, which reviews statistical methods that explain, quantify and predict pipe breakage or structural failures of water mains.

Probabilistic risk analysis of corrosion associated failures in cast iron water mains

Abstract This paper proposes a method using probabilistic risk analysis for application to corrosion associated failures in grey cast iron water mains. External corrosion reduces the capacity of the pipeline to resist stresses. When external stresses exceed the residual ultimate strength, pipe breakage becomes imminent, and the overall reliability of a water distribution network is reduced. Modelling stresses and external corrosion acting on a pipe involves uncertainties inherent in the mechanistic/statistical models and their input parameters. Monte Carlo (MC) simulations were used to perform the probabilistic analysis. The reduction in the factor of safety (FOS) of water mains over time was computed, with a failure defined as a situation in which FOS becomes smaller than 1. The MC simulations yielded an empirical probability density function of time to failure, to which a lognormal distribution was fitted leading to the derivation of a failure hazard function. A sensitivity analysis revealed that the contribution of corrosion parameters to the variability of time to failure was more significant than the combined contributions of all other parameters. Areas where more research is needed are identified.

Rehabilitation of water networks ☆: Survey of research needs and on-going efforts

Abstract Several water utilities experience and expect a great future challenge due to shortage of water, economical constraints and ageing water supply networks. The current and expected amount of pipe bursts and leaks are important indicators on network condition. Documentation of network properties, failures and water leaks, therefore, are of crucial importance for an efficient management. The paper refers different ways of handling this scope in European countries and in North America. The use of statistical methods for estimating existing and future rehabilitation needs and the use of software tools for prioritising actions are discussed. Current development on technologies for detecting leaks and for measuring pipe wall thinning is commented. It is argued that there is still a knowledge gap, and that joint international research could be a way to improve the knowledge, create new technologies and improve the water network management. A possible frame for an international programme has been presented.

Modeling failure risk in buried pipes using fuzzy Markov deterioration process

Numerous models have been proposed in the last two decades for the deterioration of buried pipes. The most prominen t approach has been the Markovian deterioration processes (MDP), which requires that the condition of the deteriorating system be encoded as an ordinal condition state. This encoding is based on numerous distress indicators obtained possibly from direct an d indirect observations, as well as from non-destructive tests. To date, few buried pipes have been inspected and their condition assessed. In addition, the encoding of distress indicators into condition states is inherently imprecise and involves subjecti ve judgment. Furthermore, the consequences of failure for buried pipes are often difficult to quantify precisely due to lack of data. In this paper, a new approach is presented to model the deterioration of buried pipes using a fuzzy rule-based, non-homogeneous Markov process. This deterioration model yields possibility of failure at every point along the life of the pipe. The possibility of failure, expressed as a fuzzy number, is coupled with the failure consequence (also expressed as a fuzzy number) to o btain the failure risk as a function of the pipe age. The use of fuzzy sets and fuzzy techniques help to incorporate the inherent imprecision and subjectivity of the data, as well as to propagate these attributes throughout the model, yielding more realist ic results. At the time of submission, adequate and sufficient data to validate the model were not available.

Estimating risk of contaminant intrusion in water distribution networks using Dempster–Shafer theory of evidence

Intrusion of contaminants into water distribution networks requires the simultaneous presence of three elements: contamination source, pathway and driving force. The existence of each of these elements provides ‘partial’ evidence (typically incomplete and non-specific) to the occurrence of contaminant intrusion into distribution networks. Evidential reasoning, also called Dempster–Shafer theory, has proved useful to incorporate both aleatory and epistemic uncertainties in the inference mechanism. The application of evidential reasoning to assess risk of contaminant intrusion is demonstrated with the help of an example of a single pipe. The proposed approach can be extended to full-scale water distribution networks to establish risk-contours of contaminant intrusion. Risk-contours using GIS may help utilities to identify sensitive locations in the water distribution network and prioritize control and preventive strategies.

Fuzzy Expert System to Assess Corrosion of Cast/Ductile Iron Pipes from Backfill Properties

: Several factors may contribute to the structural failure of cast and ductile iron water mains, the most important of which is considered to be corrosion. The ANSI/AWWA C105/A21.5-99 10-point scoring (10-P) method is commonly used to predict the corrosivity potential of a given soil sample using certain soil properties. The 10-P and other scoring methods use binary logic to classify the soil as either corrosive or noncorrosive. Fuzzy logic extends binary logic in this context as it recognizes the real world phenomena using a certain degree of membership between 0 and 1. This article presents a fuzzy logic expert system capable of predicting the deterioration of cast and ductile iron water mains based on surrounding soil properties. The proposed model consists of two modules: a knowledge base and an inference mechanism. The knowledge base provides information for better decision making and is developed in a two-tier fuzzy modeling process. First in direct approach, the expert knowledge generates a subjective model to describe the characteristics of the system using fuzzy linguistic variables. Later in system identification, the field data are used to develop an objective model, which is eventually used in conjunction with the subjective model to provide a more reliable knowledge base for the expert system. The inference mechanism uses fuzzy approximate reasoning methods to process the encoded information of the knowledge base.

Exploring the Relationship between Soil Properties and Deterioration of Metallic Pipes Using Predictive Data Mining Methods

Soil corrosivity is considered to be a major factor for the deterioration of metallic water mains. Using a 10-point scoring method as suggested by the American Water Works Association, soil corrosivity potential can be estimated by five soil properties: (1) resistivity; (2) pH value; (3) redox potential; (4) sulfide; and (5) percentage of clay fines. However, the relationship between soil corrosivity and pipe deterioration is often ambiguous and not well-defined. In order to identify the direct relationship between soil properties and pipe deterioration, which is defined as the ratio of the maximum pit depth to pipe age, predictive data mining approaches are investigated in this study. Both single- and multipredictor based approaches are employed to model such relationship. The advantage of combining multiple predictors is also demonstrated. Among all approaches, rotation forest achieves the best result in terms of the prediction error to estimate pipe deterioration rate. Compared to the random forest method, which is next to the best, the normalized mean square error decreased 50%. With the proposed approaches, the assessment of pipe condition can be achieved by analyzing soil properties. This study also highlights the importance for collecting more reliable soil properties data.

Modeling Pipe Deterioration Using Soil Properties--An Application of Fuzzy Logic Expert System

Several factors may contribute to the structural failure of cast/ductile iron water mains, the most important of which is considered to be corrosion. The ANSI/AWWA C105/A21.5-99 10-point scoring (10-P) method is the most common method used to predict soil corrosivity potential, which is based on soil properties. For a given soil sample, each soil property is evaluated for its contribution towards the corrosivity of soil. The 10-P method uses binary logic to classify the soil, either as corrosive or non-corrosive. Fuzzy logic extends the binary logic in this context as it recognizes the real world phenomena in which each property has a certain degree of membership between 0 and 1. The main objective of the present research is to develop a fuzzy logic expert system capable of establishing a criterion (such as corrosion rate or breakage rate) for predicting the deterioration of cast/ductile iron water mains using soil properties. The proposed expert system includes a fuzzy model consisting of a series of IF-THEN rules to determine soil corrosivity potential (CoP) based on soil properties. The fuzzy model contains the data of linguistic variables (database) characterizing various soil properties, and a rule base that constructs relationships among those properties and CoP. Subsequently, the expert system uses a linear regression model to link CoP to the deterioration rate of metallic pipes. A case study on cast iron pipes is examined to illustrate the application of the proposed expert system.

IMPACT OF SOIL PROPERTIES ON PIPE CORROSION: RE-EXAMINATION OF TRADITIONAL CONVENTIONS

Soil corrosivity is not a directly measurable parameter and pipe corrosion is largely a random phenomenon. The literature is replete with methods and systems that attempts to predict soil corrosivity and resulting metallic pipe corrosion from soil properties (e.g., resistivity, pH, redox potential and others) surrounding the pipe. This paper describes research that endeavors to gain a thorough understanding of the geometry of external corrosion pits and the factors (e.g., soil properties, appurtenances, service connections, etc.) that influence this geometry. This understanding would lead to the ultimate objective of achieving a better ability to assess the remaining life of ductile iron pipes for a given set of circumstances. Varying lengths of ductile iron pipes were exhumed by several North American and Australian water utilities. The exhumed pipes were cut into sections, sandblasted and tagged. Soil samples extracted along the exhumed pipe were also provided. Pipe segments were scanned, using a specially developed laser scanner. Scanned data were processed using specially developed software. Statistical analyses were performed on three geometrical attributes, namely pit depth, pit area and pit volume. Various soil characteristics were investigated for their impact on the geometric properties of the corrosion pits. Preliminary findings indicate that the data not always support traditional conventions.