The increased need to maximize production from mature assets has resulted in the transformation of the oilfield surveillance workflow. Hitherto, wells were reviewed sequentially throughout the field on a calendar basis. This was a time-consuming process with the potential of considerable time lag between problem occurrence and diagnosis. The new data-driven workflow requires a tool to support a "review by exception" process through the automated identification and prioritization of exception wells.

The primary benefit of incorporating the surveillance tool in an integrated workflow is to shorten decision time and improve the quality of the decision through an automated process. Other benefits include timelier proactive problem identification, better use of the practitioner's time (focus on analysis rather than identification), elimination of repetitive data gathering and reformatting tasks, consistency and repeatability of evaluation, and better knowledge management.

Developed in San Joaquin Valley under the i-field initiative which espouses business transformation to the digital oilfield of the future, the tool's main role is to classify and diagnose well performance. It accomplishes this by encoding customer-derived business rules in a multi-channel framework. Using a sequential, frame-based expert system as the classification engine provided a superior diagnostic capability while limiting the number of unique diagnoses to maintain to an acceptable level. It also provided complete transparency in the analysis unlike other AI methods such as an artificial neural network (ANN). A web-based UI provided the user extensive filtering, sorting, and drill down capability by which to quickly identify and diagnose problem wells.

Deployment of the tool with the new integrated workflow has led to a 30% increase in the number of wells and patterns reviewed. In addition several specific well work opportunities were identified and led to increased production, effectively bridging the gap from data bytes to barrels.

You can access this article if you purchase or spend a download.