How ElectraNet transformed asset management operations through advanced image recognition in KNIME
História de sucesso do cliente

How ElectraNet transformed asset management operations through advanced image recognition in KNIME

How ElectraNet transformed asset management with KNIME and Prophix Australia

Disponível em inglês

ElectraNet manages South Australia’s energy transmission network, spanning more than 200,000 square kilometres and over 100,000 assets worth $2.5 billion AUD. Working with Prophix Australia (formerly Forest Grove) and KNIME, the team built an image analytics solution that identifies insulator sub-classes across 15,000 transmission towers and corrects discrepancies in SAP, freeing subject matter experts from manual audits.

Key results include:

  • Reduced time spent on manual auditing by 80% across 72,000 insulator entries in SAP.
  • Achieved 99.5% model accuracy, improving SAP data quality.
  • Identified insulator sub-class across 15,000 transmission towers.
  • Deployed as an automated workflow and a web app, so SMEs can upload images and run analysis themselves.
Share

How ElectraNet transformed asset management operations through advanced image recognition in KNIME

Baixe o PDF para visualizá-lo no seu dispositivo.

Baixar PDF

Recursos relacionados

AI Use in Construction Finance: Foundation First, Framing Second

AI Use in Construction Finance: Foundation First, Framing Second

You wouldn’t pour a foundation without checking the soil first. That’s exactly what’s happening with AI in construction finance Boards and leadership are pushing hard for AI adoption—the promise is real: faster close cycles, forecast accuracy and a finance team that can do more without adding headcount. But most teams are being asked to deliver on that promise faster than their data and processes can support and they’re not sure what to trust or where to start. Here’s the real issue: a DIY AI tool or bolted-on chatbot can answer a question once. But finance needs a job-cost figure or WIP schedule that will be 100% accurate every time, auditable and reliable even after the person who built the tool moves on. That’s an architecture problem, not a prompting problem—which is why AI still barely cracks contractors’ top concerns for 2026, despite all the pressure to adopt it. This webinar lays out a proven, three-stage path from shaky ground to AI tooling that features finance-native, governed workflows and trusted number. This is the delegation era, where agentic AI can handle the grunt work while finance oversees, validates and decides.

Ver o webinar sob demanda
Should CFOs Build, Buy or Blend? The True Cost of AI in Finance

Should CFOs Build, Buy or Blend? The True Cost of AI in Finance

AI pricing is moving away from the traditional licence fee towards consumption-based models, while the range of deployment options keeps growing. The result is a technology that is harder to evaluate and harder for CFOs to control with confidence.

Ver o webinar sob demanda
MCP Server Product Tour

MCP Server Product Tour

Using financial data in AI tools like Claude and ChatGPT is powerful, but risky. Data gets stale when it leaves your source system, calculations can’t be trusted, and general AI lacks financial context. Prophix One MCP grounds every AI prompt in live, governed finance data. Connect directly to your Prophix models, overlay Reporting Agent intelligence, and prompt with confidence. Prophix One access controls are enforced, and every calculation runs on Prophix One’s deterministic foundation.

Ver o tour do produto