From reactive response to scalable operational intelligence

The crux of the challenge was the need to not only improve the speed and accuracy of methane leak detection, reduce safety and reputational risks, and strengthen operational efficiency – but also do so without creating a model that would be too costly or complex for the client to scale.

Several dimensions further underpinned this challenge:

• Coverage and speed: Existing manual inspections and point sensors left blind spots and were inherently reactive, with detection dependent on gas reaching a sensor or inspections occurring after the event
• Fragmentation: Detection systems, field data, and external inputs operated in silos, limiting integrated and timely decision-making
• Scalability and cost: Expanding dense sensor coverage across dispersed sites was financially and operationally difficult
• Measurement and adoption: High emissions uncertainty reduced decision usefulness; to drive real operational impact, the solution must be embedded into workflows

What was needed, therefore, was not just another detection tool, but a scalable operating model for methane visibility – one that would combine remote sensing, ground-based validation, analytics, and workflow integration.

Deloitte developed a methane detection platform POC to leverage and combine satellite data, ground-based sources, AI-enabled analytics, and geospatial visualisation. The objective was to integrate multiple data streams into a scalable platform to enable the identification, quantification, and prioritisation of flammable gas leaks.

This POC comprised three core components:

1. Satellite-powered methane detection: A range of hyperspectral satellite sources was assessed to build a broader detection window. Geospatial and analytical tools were used to process data, supporting methane plume detection, spatial analysis, and anomaly identification, while accounting for constraints such as revisit cycles and weather conditions.
   
2. Integrated detection and validation: Recognising the limits of satellite-only monitoring, the solution combined satellite observations with ground-based sources, existing fire and gas detection systems, and site inputs. This complementary model helped address coverage gaps and enabled more operationally relevant detection.
3. Insights orchestration for decision-making: The platform translated detection into action through visualisation, spatio-temporal analysis, and anomaly trending. Outputs were designed to integrate into operational tools, helping teams prioritise inspections, support maintenance, and manage risk more proactively.