Water pipes: why ‘lifetime’ is not an adequate concept on which to base pipe renewal strategies

Abstract

Stakeholders involved in the renewal of drinking water pipes almost always use mean ‘lifetime’ values, which are set by an expert based on material type. A common approach to estimating replacement requirements is to assume that all pipes of the same class of material are renewed when they reach an age equal to their average ‘lifetime’. However, experience shows that for the same type of pipes, very different average ‘lifetime’ values are used, leading to inaccurate estimations of renewal requirements. Furthermore, the notion of ‘lifetime’ can lead to confusion between ‘financial requirements’ and ‘practical rehab requirements’. This study highlights the advantage of considering that pipes of the same class of material can have different lifetimes, the values of which can then be represented in a statistical ‘survival curve’. Replacement needs generated through this curve are then smoothed, allowing realistic annual renewal budgets to be calculated. Given the risk of incorrect evaluation that the term ‘lifetime’ may entail, the term should be avoided when discussing pipe renewal. It would appear preferable to use the term ‘service life’, which clearly expresses the choice to be made by decision makers between repair and renewal.

Reviewing the Long-Term Sustainability of Urban Water System Rehabilitation Strategies with an Alternative Approach

“…ISO 55000 describes an asset life as the period from its creation to the end of its life [3]. This is what is normally defined as the 'lifetime' of an asset, or 'average lifetime' of a group of assets [4]. ISO 14040 defines a life cycle as 'consecutive and interlinked stages of a product system, from raw material acquisition or generation from natural resources to final disposal' [5] (p. 2).…”

“…The end of an asset life can be caused by a number of different reasons, e.g., physical deterioration, area-based rehabilitation, coordinated rehabilitation with another infrastructure, need for hydraulic capacity increase, or phasing out of old pipe materials. The actual life of a pipe can, therefore, vary greatly between pipes within the same cohort and should, therefore, be referred to as 'service lives' [4]. Many of these decisions for replacing assets are not optimized and will cause a premature end of asset life with regards to their physical condition.…”

“…The 'service life' function represents the statistically distributed service lives of the assets within a cohort. These functions are normally referred to as 'survival functions' [4] (see Figure 1), which give the probability that a pipe within a cohort will survive beyond a given age. A survival function for urban water infrastructure was first defined by Herz [7], which is referred to as the 'Herz function'.…”

“…The problem arises, however, when defining the expected lifetime. By using an average expected lifetime of a cohort, it would lead to a misrepresentation of reality, as asserted in Renaud et al [4]. As stated in Section 1.1, and in Renaud et al [4], pipes are renewed for a range of different reasons.…”

“…By using an average expected lifetime of a cohort, it would lead to a misrepresentation of reality, as asserted in Renaud et al [4]. As stated in Section 1.1, and in Renaud et al [4], pipes are renewed for a range of different reasons. Survival functions presents this range of lifetimes, and the method proposed in this paper, therefore, represents a statistically reasonable distribution of actual lifetimes.…”