From Raw Data to Field Decisions – Part 2
- 4 days ago
- 2 min read
Executive Summary
The discussion focused on how utilities are presenting, scrubbing, automating, and using reliability data, especially OMS outage data, reliability dashboards, cause coding, and QA/QC practices. Participants shared how they are building dashboards to track reliability performance, support leadership reporting, and help operating teams better understand what is driving outages.
A clear theme was that organizations are at different stages of maturity, but they are moving in the same general direction: more automated dashboards, cleaner outage data, better root-cause coding, and clearer views for both leadership and field-facing teams. Some teams are using near-daily refreshed dashboards connected to backend databases, while others are working from scrubbed monthly data that is usually a little behind. The group also spent time discussing how to reduce “unknown” outage causes, add guardrails around self-service data, and make reliability reporting easier to use without losing accuracy.
Key Takeaways
Reliability dashboards are becoming the main way teams share performance data.
Participants are moving toward interactive dashboards that show SAIFI, SAIDI, CMI, CI, major event days, outages, targets, trends, and drill-downs by feeder, circuit, district, device, and cause. These dashboards help teams quickly answer basic questions like how reliability is performing year to date, how results compare to targets, and which areas are driving the biggest impacts.
Raw OMS data is useful, but it needs scrubbing before it can be trusted.
OMS data is the main starting point for many reports and dashboards, but participants were clear that raw outage data often needs cleanup. Teams are reviewing start and stop times, outage durations, customer counts, device information, comments, and cause codes before using the data for formal reporting or decision-making.
Cause coding is a major challenge and a major opportunity.
Initial cause codes entered during restoration are not always correct. Several examples showed that an outage first marked as equipment failure, vegetation, or unknown may later be corrected after more review. Participants discussed the need to reduce “unknowns” and get to decision-grade cause codes that can actually support reliability planning.
Automation is helping reduce reporting effort and improve consistency.
One team reduced a monthly leadership report from about 40 pages to about 10 pages, with much of the wording, graphs, and report content generated automatically through an in-house script. The team still reviews the output, but automation has made the process faster, more consistent, and easier to maintain.
Some teams are adding outside data sources to improve outage analysis.
Examples included merging weather data into reliability metrics and using lightning data to help investigate unknown outages. These additional data sources can help teams better understand whether events were weather-related, lightning-related, or tied to another cause.
Teams are improving cause-code accuracy by having the owning department validate the final cause.
Rather than leaving cause coding only with the initial field entry or reporting team, some teams are sending events back to the department that owns the issue. For example, vegetation-related events may be reviewed by vegetation management, while reliability teams may help investigate unknowns or questionable outage causes. This helps turn initial outage notes into more reliable, decision-grade data.
Part 3 Scheduled for Thursday, June 25 | 11:05 AM – 11:55 AM EST.
Rob Earle
First Quartile Consulting
315.944.7610
Comments