Managing water company supply interruptions using Network Analysis

The Challenge

Supply interruptions are one of the Operational Delivery Incentives (ODIs) a water company is measured against, with the Ofwat definition of a supply interruption being when a customer goes without water for over 3 hours (taken as being operationally equivalent to =<3m pressure at the main).

The average supply interruption duration for each customer is calculated as follows: the duration of each supply interruption event and the number of properties impacted results in the property minutes for that event, with all the minutes accrued across the full financial year is then divided by the total number of properties supplied by the water company. The event official start time is when it is confirmed that customers are without water supply, either by customer contact, flow or pressure monitor indications, or verified modelled data.

During an incident, the control room asks questions relating to the supply interruption start time, cause, location, impacted customers, time to restore supply and required resources. The efficiency at answering these types of questions and understanding all the factors and options relating to the event play a significant role in either preventing a supply interruption event taking place or minimizing the impact and/or duration of any events which progress beyond 3 hours. The importance of a fast reaction to an event while ensuring the decisions made are correct cannot be overstated, and are crucial to saving property minutes. 

It is important to answer as many of these questions as possible in the first “golden” hour to ensure the team managing the incident has as much information available within a time frame that ensures they can manage the incident as efficiently as possible.

The Solution

Network Analysis contributes to answering some of these questions quickly using batched modelled data and network analytics so that front end users are not having to spend time processing data/models in the initial stages of an incident.

This includes:

• Burst Location Analysis: allowing users to input network pressures at certain locations to understand the most likely location(s) a burst could be occurring within the network.
• Burst Impact Analysis: allowing users to input the location of a burst to understand the impact to the wider system and highlight and quantify those customers which are impacted while the burst is running.
• Location Analysis / estimated repair time: Every asset in the network is geospatially assessed for the factors which could cause engineering difficulty if a repair was required. These factors in addition to isolation configuration analytics are used to provide an estimated repair time for every asset in the network.
• Isolation Impact Analysis: allowing users to input the location of a required isolation to understand the impact to the wider system and highlight and quantify those customers which are impacted while the isolation is in place.
• Rezone Option Analysis: The data within network analysis allows users to identify all the potential rezone valves that provide resupply options for each given isolation. These valves and their associated attributes can either be exported in tabular form or mapped geospatially so that users can appreciate where they are in relation to the impacted area.
• Asset Report Tool: allowing the user to input a pipe reference of interest, creating a bespoke report containing the following asset information: location, physical characteristics, isolation configuration details, impact of failure on wider network, rezone considerations, and resource management estimation.