bp Cuts Maintenance Response Time to 30 Seconds with Genie
With the launch of Genie, bp transforms maintenance operations, enabling rapid responses to equipment inquiries in just 30 seconds. This advancement not only boosts efficiency but also empowers engineers with immediate access to vital data.
Key Facts
- bp reduced maintenance query response time from weeks to 30 seconds, enhancing operational efficiency.
- AI-ready data model increased Genie Agent setup speed to 10 minutes, improving user accessibility.
- Properly modeled data improved Genie’s response speed by 400%, showcasing competitive data management.
- 75% of data modeling is automated, reducing costs and time for maintenance analysis significantly.
- Expansion from 400 to 9,000 users indicates strong demand for AI solutions, signaling market readiness.
Summary
Summary
bp, a global energy company, faced significant delays in maintenance analysis due to fragmented data across 60 systems. By implementing an AI-driven maintenance foundation using Databricks and Genie Agents, bp reduced response times for equipment failure inquiries from weeks to approximately 30 seconds.
Background
bp operates in the energy sector, managing offshore production, refining, terminals, and pipelines. The company relies on safe and available equipment, but prior to the deployment, maintenance data was scattered across 60 different systems, including 47 SAP instances and Maximo. This disorganization made it difficult for maintenance engineers to quickly access the information they needed.
Challenge
The primary challenge was the lengthy process required to gather maintenance data. Analysts became bottlenecks, as it could take weeks to collect requirements and provide answers regarding potential equipment failures. The inconsistency in data terminology further complicated matters, with multiple ways to refer to the same equipment.
Solution
bp unified its maintenance data on Databricks, utilizing Unity Catalog for governance. They created a shared data model that standardized terms and established clear relationships among data points. An AI data-readiness scoring model was implemented to assess data quality and usability. Setting up a Genie Agent became a streamlined process, taking only about 10 minutes. This agent was designed to answer maintenance-related questions, significantly improving response times.
Results
With the deployment of Genie Agents, bp transformed its maintenance inquiry process. Questions that previously took weeks to answer can now be resolved in about 30 seconds. The initial rollout to 400 users received positive feedback, and bp is preparing to expand access to 9,000 users. In tests, Genie provided answers four times faster when using well-modeled data compared to poorly structured data.
Key Insights
Standardizing data terminology and implementing a robust governance framework can drastically reduce response times and improve efficiency. The use of AI-ready data models enables quicker access to critical information, which is essential for operational success in complex industries.
Customer Testimonial
No direct quote is available in the provided source material.
Entities Mentioned
Companies
Products
Technologies
Key Concepts
Definitions
- Genie Agent
- An AI-driven tool developed by bp to provide quick answers to maintenance-related questions using AI-ready data.
- Unity Catalog
- A governance solution that manages metadata and data readiness for AI applications.
- data readiness scoring model
- A model that assesses data quality and usability to determine if it is suitable for AI applications.
- mean time between failure
- A standard metric used to predict the average time between equipment failures.
- MVP
- Minimum Viable Product; a version of a product with just enough features to satisfy early adopters.
Use Cases
- →Predicting equipment failure within 30 days
- →Scaling Genie Agents to thousands of users
- →Creating digital twins from maintenance data
- →Improving production management processes
- →Enhancing risk management strategies
Frequently Asked Questions
What is the purpose of the Genie Agent?
The Genie Agent is designed to provide rapid answers to maintenance questions using AI-ready data. It significantly reduces the time required to obtain insights from weeks to about 30 seconds.
How does bp ensure data readiness for AI?
bp uses a data readiness scoring model that evaluates data quality, standardization, and governance. Data must achieve a score of four or five to be considered ready for AI applications.
What systems does bp integrate for maintenance data?
bp integrates data from approximately 60 source systems, including 47 SAP instances and Maximo, to create a unified maintenance data foundation.
How quickly can a Genie Agent be set up?
Setting up a Genie Agent takes about 10 minutes once the AI-ready data is in place. This allows for rapid deployment across the organization.
What benefits does metadata governance provide?
Metadata governance ensures that data is well-defined and easily accessible, which enhances the speed and accuracy of data queries. It also supports the certification of gold data products for sharing across various applications.