The IFS.ai Copilot for FMECA streamlines workflows by extracting critical information from uploaded documents, such as manuals and guidelines, as well as system data, thereby reducing manual effort. It dynamically responds to user queries and page contexts, including the Prepare FMECA and Perform FMECA pages, by identifying and executing the most appropriate functions.
These functions include retrieving fault types, critical combinations of Item and Process Classes, and suggesting maintenance strategies. By grounding its insights in authoritative sources and ensuring consistency across analyses, the assistant promotes reliable decision-making. Seamlessly integrated into FMECA workflows, it reduces workloads while enhancing accuracy and efficiency.
To learn how to upload documents to the data lake, refer to our documentation on Uploading Documents to Data Lake.
IFS.ai Copilot for FMECA is automatically activated when you open the IFS Cloud Copilot on the following pages:
Note: While the results displayed on individual pages are restricted to the sites the user has access to, please note that the answers provided by IFS.ai Copilot are not limited to a specific site. Instead, Copilot aggregates data from all sites within the installation for certain queries, which may cause some users to notice discrepancies between page-restricted data and Copilot responses.
The FMECA-related pages include predefined prompts to guide users in creating relevant queries. These prompts are not part of the IFS.ai Copilot itself but are integrated with the FMECA pages mentioned above. They serve as reminders of the available queries and use cases that the Copilot can assist with.
IFS.ai Copilot for FMECA supports the following queries and use cases:
This query retrieves and lists the most frequently occurring fault types using data from previous work tasks. The description guiding the LLM on which function to call and when could be “Retrieve the most common fault types based on the Item Class ID and Process Class ID”.
This function identifies fault types recorded on work tasks associated with a specified asset or equipment. It uses the Item Class ID and Process Class ID of the asset or equipment and defaults to returning the top five fault types, including their descriptions and counts. The default limit of five can be adjusted by instructing the Copilot to “limit the result to X records,” where X is the desired number of results. This functionality aids users in prioritizing equipment for FMECA preparation by highlighting recurring fault patterns.
Note: Copilot aggregates fault type data across all sites for this query, meaning results reflect faults from the entire installation, not just the user's connected sites.This is done on purpose.
Using criticality ratings (found under Maintenance > Equipment > Equipment Basic Data > Object Criticality), this query identifies highly critical equipment and lists their item and process classes along with the count for each combination. The data is pulled across all sites, providing a holistic view of critical equipment and their related classifications.
The prompt description for this functionality is "Suggest equipment for FMECA," which expands to a longer prompt: "List the top combinations of the Process Class and Item Class of the most critical equipment. Include the count for each combination. Answer like this: Process Class: A, Item Class: B, Count: C. Include the chosen criticality in the answer."
This functionality utilizes AI-driven algorithms to enhance FMECA analysis capabilities by accessing data stored in IFS Cloud data tables. The system leverages this structured information to suggest failure modes, causes, symptoms, and corresponding severity, probability, and detectability ratings, enabling more data-informed maintenance strategies.
The Prepare FMECA page now features AI-powered prompts integrated into the IFS.ai Co-Pilot Assistant, enabling FMECA facilitators, technicians, and reliability engineers to streamline their analyses with greater accuracy and efficiency.
The system uses item class data from the Failure Analysis Set Up Navigator and Work Task pages to suggest the most relevant fault types. Prompts also enable users to retrieve associated causes or symptoms, highlight the most commonly or recently used ones, and organize them alphabetically or by usage count. Additional enhancements include prompts for accessing Fault Reports, PM Action intervals, and Work Task Template intervals assisting in the assignment of Probability Ratings. Severity Rating suggestions are derived from criticality values linked to item and process classes or frequently used work task priorities, while Detectability Ratings provide context-sensitive recommendations for failure modes.
Designed with a focus on maintainability, performance, data integrity, and security, these enhancements significantly improve the FMECA process. They support informed decision-making and efficient maintenance strategies by incorporating Criticality Analysis on objects to retrieve Severity and Probability ratings, ensuring a more accurate and data-driven approach to risk assessment.
This development introduces AI-powered prompts in the Copilot Assistant to streamline the process of assigning PM Calendar Triggers and Event Triggers for PM Programs on the Perform FMECA page. By integrating these enhancements, technicians and reliability engineers can access data-driven suggestions tailored to specific Item Classes, Process Classes, and Work Task Templates, improving the efficiency and precision of maintenance strategies.
On the Perform FMECA page, prompts will provide users with insights into fault report patterns, such as the time gap between two fault reports and the average gap for a defined period (e.g., six months). These suggestions are generated by analyzing historical fault reports, offering valuable data to select the most suitable PM Calendar Triggers. For example, a technician can quickly identify recurring fault intervals to set optimal calendar-based PM actions.
Additionally, new prompts on the Prepare and Perform FMECA pages will assist users in determining the most suitable Event Triggers for PM Programs. These include suggestions for:
For example, when creating or updating a PM Program revision, prompts will guide users in selecting suitable Events for task templates, such as ‘WT-GEARBOX-SERVICE’ or ‘WTT06,’ where only one Event is allowed per template. These features enhance usability by analyzing Events and usage counts within PM Actions and Work Tasks, ensuring the most relevant triggers are selected. The technical design leverages dynamic data integration across fault reports, Work Task Templates, and historical maintenance records. It ensures accurate and secure recommendations, prioritizing maintainability, performance, and operational alignment. By automating complex data analyses, these enhancements streamline the FMECA process, support better decision-making, and enable the efficient update of PM Programs.
The technical design leverages dynamic data integration across fault reports, Work Task Templates, and historical maintenance records. It ensures accurate and secure recommendations, prioritizing maintainability, performance, and operational alignment. By automating complex data analyses, these enhancements streamline the FMECA process, support better decision-making, and enable the efficient update of PM Programs.
This development introduces advanced AI-driven prompts in the Copilot Assistant to enhance the process of managing and suggesting Maintenance Decision Trigger Parameters for PM Actions on the Prepare and Perform FMECA pages. By integrating these features, FMECA facilitators, technicians, and reliability engineers gain precise and actionable recommendations, ensuring that PM Actions are configured with optimal parameters for their intended equipment objects and maintenance scenarios.
On the Perform FMECA page, when creating a PM Action for a specific equipment object, users can prompt the Copilot to suggest the best matching PM Action. The suggestions are based on key parameters such as Action, WO Site, Maintenance Organization, Program ID, Work Task Template ID, and Revision. The system fetches all PM Actions related to the specified equipment object in a Preliminary or Active state and evaluates their suitability by comparing the input parameters. The PM Action with the highest number of matching parameters is recommended, with the latest action being prioritized in case of ties. This ensures the most relevant PM Action is selected, streamlining decision-making and saving time.
Additionally, on the Prepare FMECA page, new AI prompts are introduced to provide suggestions for condition trigger parameters.
These prompts utilize advanced data integration to analyze and return relevant condition parameters, enabling users to align Work Task Templates and PM Actions with operational requirements efficiently.
To support this functionality, a new permission set naming standard is introduced
for AI Assistants to ensure security and proper access management. Permission
sets will follow the format COPILOT_
These enhancements in trigger parameter management prioritize usability, operational efficiency, and secure decision-making, significantly advancing the functionality of FMECA analysis in maintenance planning.
This development introduces AI-driven functionality in the Copilot Assistant to enhance the process of updating maintenance strategies on the Perform FMECA page, specifically within the Object for Analysis tab. By leveraging this feature, FMECA facilitators gain valuable insights into existing maintenance data to refine strategies and optimize decision-making.
When a facilitator requests maintenance strategy suggestions for a specific object, the Copilot analyzes the item's Item Class and Process Class to provide a detailed overview. The suggestions include the count of action types associated with the object, offering a clear picture of the most frequently used maintenance actions. Additionally, the Copilot displays the total number of ongoing PM Actions and Fault Reports related to the object, enabling facilitators to assess current workloads and prioritize maintenance tasks effectively.
This integration ensures that maintenance strategies are data-driven and aligned with the object’s operational needs. By providing actionable insights, the Copilot streamlines strategy updates, enhances preventive maintenance planning, and supports decision-making with real-time, context-specific data. This functionality marks a significant step forward in making maintenance planning more efficient and intelligent.