Using machine learning to better understand consumer needs

At a Glance

  • $960K

    Project Cost (AUD)

  • 20

    Seconds to Forecast

  • 1.3

    Million Customers

Location
Adelaide, South Australia
Offices
Client
  • SA Water
Partners
  • Bentley Systems
  • Chamonix
Award
SA Water Awards 2018 - Infrastructure Project Innovation Award

Enhanced Water Demand Forecasting

Across the globe, utility companies aim to maximize efficiency in their water and wastewater networks. With increasing regulatory pressures, limited capital funding, and the reduction in operating budgets as well as global water availability (and quality), utilities are challenged to do more with less. When SA Water wanted to more accurately forecast their utility performance, we led the teams that implemented a range of improvements to their demand forecast tool (DFT).

These upgrades included a new process for developing and calibrating the DFT regression model and new data sources that would validate and improve accuracy. The tool’s reach was geographically expanded by adding additional water treatment plants and supply zones, and its precision was improved through the addition of Microsoft Azure Machine Learning, a cloud-based platform that automates forecast model updates.

Water utilities need to understand and forecast water demands to accurately plan operations, water supply requirements, and short-term water treatment production plans when conditions change. With these forecast improvements, SA Water can manage their costs and risks more easily.

At a Glance

  • $960K

    Project Cost (AUD)

  • 20

    Seconds to Forecast

  • 1.3

    Million Customers

Location
Adelaide, South Australia
Offices
Client
  • SA Water
Partners
  • Bentley Systems
  • Chamonix
Award
SA Water Awards 2018 - Infrastructure Project Innovation Award

Services

Markets