Production Challenges


Failure Troubleshooting (HITs, Rod Parts)

Ambyint continues to learn from its unparalleled data lake of 33 million dynocards with expert-marked failures. The platform evaluates the numerical signatures underlying dynocard data in real time and is able to diagnose and predict common downhole failures like hole in tubing, rod parts failures, and upstroke wear, among others.

Ambyint identifies HIT immediately

Simplifying & Automating Daily Actions

Morning meetings in the field often rely on a quick SCADA review of wells down and greasebook notes from the prior day to prioritize discussion. This leads to suboptimal planning of manpower and route prioritization. With Ambyint workflow tools, both operations and engineering are effectively aligned to optimize activities from the field and office each day.

Identifying well problems for review in a user interface
Ambyint’s Production Optimization Platform allows users to identify problem wells for review

Root Cause Failure Analysis

For most operators data relevant to RCFA is fragmented and stored in multiple places and formats in a best-case scenario, and of poor quality or non-existent in a worst-case scenario. Given the many demands on production staff, there is no time to properly gather this data and perform insightful analytics. Ambyint offers a comprehensive failure analysis dashboard and tools to maintain all key failure analysis data, including operational and wellbore data.

Equipment Selection / Technology Validation

Highly deviated, horizontal wells create so much friction that traditional optimization algorithms have trouble accurately diagnosing downhole conditions. Ambyint uses high resolution data combined with cognitive card recognition data science to evaluate the micro-level patterns in dynocard data. With this resolution of data, engineers can evaluate new downhole technologies and rod designs in real-time to assess if the new technologies are adding value to the well’s operation. This capability can save dozens of suboptimal designs from going into the hole, compared to the iterative approach of conventional lift optimization.

Deviated horizontal well creates high downhole friction

Finding Outliers

Current well optimization review methodology calls for manually reviewing each and every well, leading to much time spent reviewing wells running optimally and limiting the time available to review the true problem wells. Ambyint automatically flags and sorts wells in need of review, allowing reviews to focus on true value-add potential. Ambyint is like your best Tech at your well, watching every stroke, every day, and notifying you when an outlying event occurs.

Automating Well Review

Well reviews are heavily manual, time-consuming processes for engineers and operations staff. Compounding that is the inefficient process for assigning and tracking well-specific action items. Ambyint automatically sorts wells and highlights bad actors for easy well review prioritization. There is also an “assign task” feature to log and track actions in the platform. When a Tech drives on location, the tasks for that well will pop up on the mobile app automatically. Optimization actions like fluid shots or set point changes can be tagged and counted. This allows for teams to quantify the volume and value of optimization activities.

Easily track well performance and assign action items

Reactive Nature of Failures

Well remediation cycle times are lengthy due to the reactive nature of the troubleshooting process. Ambyint data science models pair physics-based features with cutting edge data science to move failure diagnosis and resolution from reactive to proactive. This creates more time for procedures to be written, rigs to be scheduled, and proper quantitative-based design work to be thoughtfully performed. Improved data inputs and logistics allocation allows operators to reduce their cycle time from well down to return to production (RTP).