Using AI and business intelligence to forecast energy asset health
One of our energy clients — an organization that addresses pressing energy challenges with transformative solutions — uses drone photography to inspect utility infrastructure. These drones make site inspections safer, faster and more cost-effective by eliminating the need for field workers to visit each site and manually collect images.
However, manually analyzing thousands of images became too difficult and time-consuming for the client. So, the client sought our help to learn how to use AI and business intelligence together to automate image analysis and forecast asset maintenance needs.
With the support of Microsoft’s energy-focused team, we built an end-to-end functional prototype that not only provides asset health insights at a glance, but also sets the groundwork for future development phases and deployment at scale. Using cutting-edge data science techniques and tools like Microsoft Power BI, the prototype demonstrates how the client can reduce OpEx costs, maximize equipment lifespans, and forecast anomalies with a deployed solution.
- Automating image analysis to reduce manual hours of image analysis and data entry
- Building a low-risk prototype to justify an AI and analytics solution before investing in it
- Developing an automated, end-to-end prototype that not only extracts data from drone photos, but also produces rich, interactive dashboards and predictive insights for proactive decision-making
Using state-of-the-art image recognition technologies and tools like Microsoft Power BI and Azure cloud services, we developed a demonstrable prototype that allows the client to:
- Ingest data – the client can consume images and transport image data into a centralized data warehouse, where it is easily accessed, used and analyzed.
- Process images – using object detection and recognition, the client can identify and classify patterns, create metadata, and optimize data for reporting.
- Visualize insights – by translating the processed data into an extensible data model along with interactive Microsoft Power BI dashboards, the client can easily visualize data, filter and search for specific anomalies, and view anomalies on a map.
Alongside Microsoft, we demonstrated the prototype to the client, illustrating how the solution can help the organization forecast anomalies and make more informed predictive maintenance decisions. Now, with a working prototype, the client has the framework to build a fully deployed solution — and it’s positioned to improve service uptime and lower OpEx costs by shifting from a reactive to a proactive maintenance approach.
Why Use Drones, AI & Analytics Together?
Adopting advanced analytics to power predictive maintenance offers a new avenue to improve performance while reducing asset management costs by as much as 10 to 20 percent.1 Here are other benefits of an end-to-end image analytics solution: