ARC Data Analytics Handbook

Version 0.0.2

All things data analytics at ARC Resources.

Frac Stage Classification System Overview

Frac Stage Classification System Overview

The Frac Stage Classification System is an intelligent automation tool that helps our completions engineers understand what happens during frac operations. Instead of engineers spending hours manually reviewing complex sensor data from each frac stage, this system automatically identifies and categorizes important events, giving our team immediate insights to make better decisions about well completions.

This is a living document and will be continually updated. Please refer back for the latest enhancements.

What This System Does for ARC

During a frac operation, sensors continuously monitor dozens of measurements every second - pressure readings, sand concentrations, fluid rates, and many other critical parameters. Previously, engineers had to manually examine this data to identify important events like when sand first enters the formation, when pressure spikes occur, or when other significant operational events happen. This manual process was time-consuming, subjective, and difficult to scale across all our operations.

The Frac Stage Classification System automates this process. It continuously analyzes the sensor data and automatically identifies and labels these critical events in real-time. This means engineers can immediately see what’s happening during a frac operation without having to dig through thousands of data points manually.

Key Benefits for Operations

Faster Decision Making: Engineers now receive instant, consistent insights about frac stage performance, allowing them to quickly identify successful techniques and problematic patterns across our operations.

Improved Consistency: The system applies the same analytical standards to every frac stage, eliminating the variability that comes from different engineers interpreting the same data differently.

Better Learning from Experience: By automatically categorizing events across all our frac operations, the system helps us build a comprehensive database of what works and what doesn’t, enabling continuous improvement in our completion designs.

Reduced Manual Work: Engineers can focus their time on analysis and decision-making rather than spending hours reviewing raw sensor data to find important events.

Enhanced Pattern Recognition: The system can identify subtle patterns and relationships in the data that might be missed during manual review, helping us discover new optimization opportunities.

What the System Identifies

The system automatically detects and categorizes various types of events that occur during frac operations:

CategoryDescription
Category 0Default (Represents baseline/normal operations or segments not otherwise classified by specific rules/signatures).
Category 1Initiation.
Category 2Early Unexplained.
Category 3Sand Entry.
Category 4Hydrostatic.
Category 5Shutdown.
Category 6Equipment Issues.
Category 7Sustained Pressure Spike - Cluster Loss.
Category 8Pressure Drop - Not From Sand.

How Engineers Use This Information

Post-Operation Analysis: After a frac is complete, engineers have a detailed, automatically-generated summary of all events that occurred, including when they happened, how long they lasted, and their characteristics.

This information helps engineers answer critical questions: Was the frac treatment executed as planned? Did any unexpected events occur? How did this frac compare to similar operations? What can we learn to improve future operations?

Data Quality and Reliability

The system combines advanced analytical techniques with established engineering knowledge to ensure reliable event detection. It uses both automated pattern recognition and engineering rules developed by our completions experts, providing a balanced approach that captures both subtle data patterns and well-understood operational indicators.

The system continuously monitors its own performance and automatically alerts our technical team if any issues arise, ensuring consistent, high-quality analysis of all frac operations.

For detailed technical information, see the How It Works section.

Contacts

See the contacts page for project leads and support.