Skip to main content

The Challenge

YPF's Vaca Muerta operations face fundamental limitations with traditional inspection methods that cannot scale to meet operational, safety, and regulatory demands.

Core Problem Statement

YPF cannot achieve continuous asset monitoring across 2.9 million acres using manual inspection methods, exposing the organization to safety incidents, environmental violations, and operational inefficiencies.

1. Coverage and Frequency Gap

Current Inspection Limitations

Traditional ground-based and piloted aerial inspections cannot provide the coverage frequency that modern operations require:

Asset TypeCurrent FrequencyRequired FrequencyGap
Wellhead inspectionWeeklyDaily7x
Pipeline patrolMonthlyWeekly4x
Facility thermal scanQuarterlyDaily90x
Methane monitoringAnnual surveyContinuousN/A
Perimeter securityPeriodic patrol24/7N/A

Scale Challenge

With 800+ horizontal wells spread across remote desert terrain:

  • Driving distances between wellheads make daily manual inspection impractical
  • Weather conditions (high winds, dust) frequently ground piloted aircraft
  • Night operations require specialized equipment and personnel

2. Safety and Compliance Exposure

Incident History

  • 2018 Bandurria Sur blowout: 45-85 hectares contaminated, 36-hour uncontrolled spill, license suspended
  • 2023 labor strike: 25,000 workers citing "business contempt for safety" after arm amputation incident

Regulatory Requirements

RegulationRequirementCurrent Capability
Resolution 143/199824-hour flaring/venting notificationManual monitoring
Neuquén Province500% gas price penalty for violationsInfrequent checks
YPF 2026 Target10% methane reductionNo baseline monitoring
YPF 2030 Target30% methane reductionNo tracking system

Super-Emitter Risk

Research shows super-emitters (≥100 kg/h) represent only 7% of emission events but cause 80% of total methane emissions. Without continuous monitoring, these events go undetected for hours or days.

3. Personnel Exposure

Field Safety Concerns

Manual inspections expose personnel to:

  • Remote locations with limited emergency response access
  • Hazardous atmospheres near potential leak sites
  • Extreme temperatures in desert environment
  • H2S exposure risk at certain well sites

Operational Impact

  • Field technician time spent on routine patrols rather than maintenance
  • Vehicle costs and fuel consumption for ground-based inspection
  • Limited skilled personnel available for remote site coverage

4. Traceability and Evidence Requirements

Current Documentation Gaps

RequirementCurrent StateImpact
Visual evidence of inspectionsPhotos manually uploadedInconsistent coverage
Timestamp and GPS correlationManual logsAudit vulnerability
Historical trend analysisScattered in multiple systemsNo predictive capability
Regulatory reportingManual compilationTime-consuming, error-prone

17 Mandatory AI Use Cases

The RFP specifies 17 AI-powered detection and monitoring capabilities that must be operational:

  1. Fluid detection on ground surfaces
  2. Flow detection in lines and wells
  3. Liquid and gas leak detection
  4. Methane (CH4) detection and quantification
  5. Medium voltage line inspection
  6. Construction progress measurement
  7. Unauthorized objects at operational sites
  8. Liquid levels in chemical skids
  9. Person and vehicle intrusion detection
  10. Volumetric calculation (sand, materials)
  11. Hot spots in electrical/mechanical equipment
  12. Road and access anomalies
  13. Electromechanical protection verification
  14. Perimeter fence integrity
  15. Mechanical integrity evaluation
  16. Valve and tank operational status
  17. Underground interference detection (GPR)

These use cases require integrated multi-sensor data analysis that manual inspection cannot provide.

Why Autonomous Aerial Monitoring

The solution requires:

NeedAutonomous Drone Capability
Coverage at scale24/7 operations from multiple autonomous bases
Consistent frequencyProgrammable flight schedules regardless of staffing
Night operationsThermal and spotlight capabilities
Hazard avoidanceUnmanned platforms reduce personnel exposure
Sensor integrationMulti-modal data collection in single mission
Immediate evidenceAutomatic timestamped, geolocated recording
AI-powered analysisReal-time detection exceeding human observation