A failure registration system for use by water companies, providing an exceptional database for statistical analysis is described in an article published in Water Asset Management International. Insight into the deterioration process of individual water mains and pipe groups can help to predict when replacement is due.
Ageing drinking water distribution networks are subject to increasing failure rates and leaks. Underground networks cannot be widely inspected and therefore little is known about their condition. Deterioration of the water mains causes supply interruptions, pressure loss and decreasing water quality, which in turn leads to increasing costs over time.
Since monitoring of the mains' conditions is relatively expensive, these techniques are not appropriate for widespread application. Alternatively, maintenance and failure data can be useful sources of information to estimate remaining lifetime or expected failure frequencies. Statistical analysis of registered failure data can reveal patterns in the occurrence of failures. The registered failure data from a number of Dutch water companies were analysed in 2007 showing that, despite the incomplete registration and varying data quality, some trends were evident (e.g. relations between age and the occurrence of failures in asbestos cement pipes).
However, the knowledge which can be derived from the failure occurrences of individual water companies is limited. One method of overcoming this problem is by gathering failure information from a larger network, i.e. sharing failure information between water utilities. The authors, of KWR Watercycle Research Institute, describe the development of U-STORE, a failure registration system designed for use by water companies. By encouraging water companies to log failure data in a uniform and consistent manner, a more accurate statistical analysis of failures can be performed.
The authors describe what commitment to participation means for the water companies: employees must be trained to record data in a uniform and accurate manner. Values of agreed parameters must be recorded using standardised categories. A research study revealed that parameters such as weather, environment (e.g. soil type and presence of trees and/or traffic), network (e.g. pipe material, age and dimensions) and failure (e.g. cause) are all important. For example, a study on the relationship between asbestos cement pipes and a number of aspects of weather has shown that the weather has an effect on the occurrence of failures in asbestos cement pipes older than 40 years of age.
The U-STORE system is more than just a database, however. By encouraging water companies to record data in a uniform way, opportunities for analysis will not be lost. By exchanging data, KWR can analyse a larger database than would be available from a single water company. KWR can then provide water companies with information helping to influence policy and management. The results of the analysis can be a rich source of information for the repair or replacement dilemma if the right parameters are registered. The information will also be used to evaluate and improve the U-STORE system itself.
Using results achieved since seven companies registered with the scheme in January 2009, the authors demonstrate the wealth of information which the companies now have access to. For example, a company may receive information on steel pipe failure despite not having registered any steel pipe failures themselves. The authors conclude by stating that postponement of failure registration simply translates as postponement of access to useful knowledge.
Source: Irene Vloerbergh and Mirjam Blokker, 2010. Sharing failure data to gain insight into network deterioration. Water Asset Management International, 6(2), 9-14.