Asset maintenance

Optimising asset management and uptimes through predictive maintenance

Infrastructure, Urban Infrastructure

A common challenge across industries is finding methods to harness the huge volumes of data generated by their operations.

Download PDF

Goal

At one urban infrastructure company, senior leadership and on-the-ground engineers found it extremely difficult to make sense of the volume of data produced from more than 250,000 sensors. The company needed to consolidate and clarify its processes for gathering and analysing data in an effort to improve risk management on a £15 billion infrastructure project.

Insight and Action

QuantumBlack first aggregated data from the range of sources and extended the traditional Gaussian models with spatio-temporal correlation derived from machine-learning techniques. Next, our team automated the basic analysis, allowing data engineers to go beyond issue detection to focus on value-added interpretation. This ability to hunt for patterns between sensors is changing the industry by enabling real-time detection of anomalies, event forecasting, and optimisation of the monitoring regime. The infrastructure company successfully deployed this process at two stations as a pilot to gauge its effectiveness in advance of a more extensive rollout.

Results

  • 20Percentreduction in monitoring costs
  • 7Daysof forecasting capability added to historical processes