Forecasting supply and demand for water in a growth area

Managing a water network is an exercise in balancing supply and demand when both variables are constantly changing. So how do you plan for the future in such a volatile environment?

 

One of our clients is a large metropolitan water utility that services one of the fastest-growing residential areas in Australia, with a population that is forecast to double over the next 30 years.

The planning team at this water provider uses forecast.id - our independent forecasts of population and housing which are underpinned by our Residential Development Forecasts.

These forecasts are a key input to the team's own demand forecast modelling, which underpins infrastructure investments, and network operations and supports their corporate strategy.

 

What do water utilities need to plan for?

Planning teams at water utilities are responsible for modelling water supply and demand in their catchment. This is called Mass Balance Modelling, but simply put, it's a calculation of where water comes from, and where it goes.

Of course, in practice, this is a complex modelling exercise, accounting for rivers, rainfall and weather events, water from neighbouring networks, evaporation and climate change.

And that's just the supply side!

On the demand side, there are two main factors that impact future demand for water.

1. How people use water

These behaviours and technologies determine how much water each household uses.

2. How many people use water

This is where .id's forecasts play a key role in the modelling. 

This is an important number to get right. Population growth is a critical factor to consider because it is the multiplier in the demand forecasting equation:

Future demand = Average rates of consumption per household x The number of future households.

 

A foundation for planning across the organisation

This population forecast is a critical assumption that underpins many of the organisation's strategic planning decisions. This is the first piece of information that goes into everything else down the line.

These are significant investment decisions with a long lead time. What size pump to specify? What size pipe to specify?

Water planners go to extraordinary lengths to test how their water infrastructure network would perform under a range of scenarios, modelling for system performance on an average day, peak days, during maintenance, shutdown or system failures.  In all these scenarios, understanding total system demand is critical. 

Building it right the first time

When planning capital works, water providers need to ensure they build infrastructure that is fit for purpose over the long term (they don't want to have to dig up a road twice) but they also need to avoid 'gold plating', or over-investing in a network. A water treatment plant is not a quick build.

Planning for both building and maintenance

Many large metropolitan utilities service both established and growth areas. In this case, it's vital to have a single forecast that shows how new development is forecast to play out, but also how the population of established areas is likely to change.

For this reason, there is value in the fact that the residential development forecasts that underpin forecast.id provides both the timing (each year of the forecast), sequence (the order of different developments) and the total capacity of growth expected in a given area in the forecast period.

Beyond new infrastructure, this demand forecasting also underpins planning for network maintenance and upgrades and is used to calculate pricing based on revenue forecasts, to balance forecast revenues with the costs of maintaining this essential service. 

A thorough and defendable view of future demand

As these forecasts underpin many significant decisions, they are subject to scrutiny and oversight.

Planning managers at water utilities need to justify their forecast modelling to internal stakeholders and are subject to auditing from bodies such as State Essential Services Commissions.

In this process, being able to present an open and transparent model with clear and explicit assumptions means they can have a more productive discussion with these other stakeholders.

Building on official forecasts

As essential service providers, utilities are given access to information from State Government Planning Departments. However, there are instances where these organisations will present an evidence-based argument for planning with different forecasts of population growth to those in the official projections. 

Using the most up-to-date information available

Many utilities update their corporate plans each year, and complete pricing submissions every five years. This doesn't always coincide with the release of official State Government projections. 

In these cases, forecast.id provides a more up-to-date assessment of how changes in macro-demographic factors such as births, deaths and migration, and more localised changes in land use policy and planned development are forecast to impact future population growth.

A more detailed forecast

When trying to understand how growth is forecast to occur within a catchment area, an SA2 is simply too big. We've heard from our clients that one of the major advantages of using forecast.id is it provides a finer resolution. 

This is a challenge for many organisations who plan in growth areas, as the geography created to present current census data isn't granular enough to show a useful view of these growth areas as they will be in the future. In these growth areas, what is sparsely population farmland today will be a housing estate with thousands of residents in the years to come.

This is why forecast.id is published on our own geography that matches standard ABS geography in established areas but provides a more detailed picture of how the population is forecast to be distributed in growth areas.  

Explict assumptions about future development

Research about the timing and location of future housing development is an important assumption that underpins forecast.id. Most utilities also collect information about current and proposed future development from their own conversations with developers, however this is typically limited to the next few years of proposed development activity.

.id's Residential Development Forecasts are based on a comprehensive research process and are a critical evidence base that underpins forecast.id. They provide forecasts of housing development for the full forecast period, organised by the location, timing, capacity and sequence of that development. 

By making our development assumptions available, utilities can understand and interrogate this important assumption underpinning the forecasts, better integrate forecast.id with their own modelling, and have more productive and transparent conversations with developers and other stakeholders. 

Watch this short video to learn more about our land use research →

Evidence-based negotiations with developers 

These applications often lead to negotiations between utilities and developers about future infrastructure planning and the contributions they make to those projects.  

forecast.id is developed with a top-down model that provides localised information in a regional and national context. This gives our clients an independent assessment of the location and timing of future development - a balanced view that accounts for both forecast demand from population growth but also forecast supply from other competing sources of housing development. 

Local Government relationships

Local Governments are the gatekeepers to all development. For over 20 years, forecast.id has been developed as a product for, and in close consultation with, local government.

This work gives us unique knowledge of the nuances of local planning policies and a deep knowledge of the functioning of the demographic, economic and housing conditions in these areas. 

We can help you plan with confidence

The planning and modelling done by our essential service providers is quite remarkable. Our clients have modelled how their networks would perform on each day (and, in some cases, for each hour in that day) for the next 50 years.

This combination of big-picture perspective and granular detail mirrors our approach to population forecasting.

forecast.id gives our clients a detailed forecast for a place in its wider regional and national context, to help them invest in the right place, at the right time. In this way, we help our clients plan with confidence, knowing the many factors that influence population change on a national and localised scale have been considered and distilled to a single-scenario forecast of population and housing.

Read more about how forecast.id can support decision-making in your organisation, or book a time using the scheduler below to discuss your project with one of our consultants.