Theoretical and Conceptual Framework - Pattern Recognition for Reliability Assessment of Water Distribution Networks - page 24

2.6 MAIN GAPS IN NETWORK RELIABILITY ANALYSES

The main gaps in water distribution system reliability analyses can be clustered per (A)
definitions, (B) methods and (C) tools, which are elaborated in the following paragraphs.

2.6.1 Definitions (A)

The major definitions are not seen as gaps in analysing the network reliability, as such.
However, they affect the uniformity of approaches to the reliability assessment. The key
questions to be considered with that respect are:

(A.1) What service level can be considered as acceptable?
Universal definition on this, which could be taken as a starting point for any reliability
assessment, does not exist. Consumers' satisfaction is the only real parameter for general
comparison of the service levels in various distribution systems. Nevertheless, that one rests
on a wide range of 'what is needed' and 'what is available and/or affordable'.

(A.2) When is a water distribution system sufficiently reliable?
Here as well, a universal conclusion is difficult to reach, although most would agree that in
case the issue of acceptable service level has been resolved, the system would be sufficiently
reliable if it can maintain that service level defined for regular and irregular demand
scenarios. Customised service levels based on various degrees of irregularity are acceptable
provided that the water shortages are reasonable in quantities and duration.

(A.3) What network parameters are the closets indicators of its reliability?
Nodal pressures and demands are the most convenient i.e. accessible parameters for network
reliability assessment. It is indeed that sufficiently high pressures will almost always mean
sufficient quantities of water in the system. Nevertheless, the pressure-demand dependency is
a 'chicken and egg story': the lower pressures affect the demand but the increased demand
will also affect the pressures! Ultimately, consumers are those who decide when to open the
tap, with or without sufficient pressure. Examples of trying to save excessive water quantities
once they have observed a drop in the pressure, anticipating forthcoming shortages, are well
known in water distribution practice.

Furthermore, more research could be done to assess the reliability by analysing economic
parameters. The most economic designs commonly start from some optimisation of the
investment and operational costs taking into consideration regular supply conditions. More
light should therefore be thrown on the costs of failures and shortages in relation to the
investment into the reliability improvement (larger diameter/parallel pipes, more pumps,
storage volume, etc.). Apart from guaranteeing the adopted service levels, the reliability is
also to be achieved at minimum costs for water companies and consumers. A robust tool that
can compare investments into the reliability improvement and clearly describe their
implications for the maintenance of service levels in irregular conditions would certainly be
helpful in decision making process.

Finally, the literature review showed little research done on the role of water quality
parameters in the reliability assessment; this aspect is missing even in the basic definitions of