AI Assistant for Centralized Work Protocol Retrieval

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Large engineering and manufacturing firms face challenges in accessing and utilizing dispersed technical documentation, requiring manual aggregation and extensive tacit knowledge, leading to inefficiencies and errors in work protocols.

Innovation Solution

A system that aggregates disparate documentation and metadata into centralized Large Language Models (LLMs) using an artificially intelligent assistant, enabling quick and efficient retrieval of relevant information through a digital work environment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If manual processes are used to access and review technical documentation, then engineers can access information stored in multiple locations, but the process requires significant time and effort to aggregate disparate documentation and metadata

Engineering Contradiction:
Improveaccess to technical documentationVSAvoidtime to aggregate documentation
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The patent combines disparate documentation and metadata from multiple programs and sources into a centralized LLM. The system merges technical documentation, procedures, policies, and CAD metadata into a unified knowledge base that can be queried simultaneously, eliminating the need to manually aggregate information from separate locations.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The AI assistant acts as an intermediary between engineers and the centralized LLM. It receives natural language queries from users, retrieves relevant information from the LLM, and presents contextualized responses, thereby mediating the complex process of information retrieval and making it simple and efficient.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If engineers manually access documentation from multiple programs and sources, then they can gather required information, but extensive tacit knowledge is required to navigate each system

Engineering Contradiction:
Improveretrieval of technical informationVSAvoidease of accessing documentation
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The centralized LLM provides a universal access point that consolidates multiple documentation systems into a single interface. The AI assistant supports various query types and delivers information across different contexts, making the system versatile and eliminating the need for engineers to learn multiple systems.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The AI assistant serves as an intermediary that handles the complexity of navigating multiple systems. It translates user intent into appropriate queries across the centralized LLM and presents results in a user-friendly format, shielding users from the underlying system complexity.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If very few employees hold all institutional knowledge needed to generate blanket coverage, then expertise is concentrated, but most employees cannot access comprehensive information independently

Engineering Contradiction:
Improvequality of work protocolsVSAvoidability of employees to access information
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system creates a digital copy of institutional knowledge stored in the centralized LLM. This copy encapsulates the expertise of subject matter experts and makes it universally accessible to all employees, allowing anyone to query the knowledge base without needing direct access to the original experts.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The AI assistant enables employees to independently retrieve the information they need through natural language queries. The system empowers users to serve themselves by providing on-demand access to comprehensive technical documentation and procedures without requiring manual intervention from experts.

Inventive Principle:
Principle #25Self-service

4Productivity

If centralized LLMs are used to aggregate documentation, then information retrieval becomes efficient, but the system requires significant computational resources and infrastructure

Engineering Contradiction:
Improveinformation retrieval efficiencyVSAvoidsystem infrastructure
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The AI assistant acts as an intermediary layer between users and the complex LLM infrastructure. It handles query processing, result synthesis, and presentation, thereby shielding users from the underlying system complexity while maintaining high retrieval efficiency.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system uses a centralized LLM that creates efficient digital copies and representations of documentation. This allows rapid retrieval and processing of information without requiring physical access to multiple documentation systems, improving productivity while managing infrastructure through virtualization.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP4597356A1Artificially intelligent assistant for work protocols
Publication Date: 2025.08.06 THE BOEING CO
  • EP4597356A1 patent drawingFigure 1
  • EP4597356A1 patent drawingFigure 2
  • EP4597356A1 patent drawingFigure 3

AI summary

A computer-implemented method (400) comprises storing (210) technical documents (104) in a storage device (102). The technical documents (104) include at least work protocols (112) and technical data files tagged with metadata (124). One or more large language models (134) are trained (230) on at least the work protocols (112), and text (122) and metadata (124) extracted from the technical documents (104). An artificially intelligent assistant (142) is commissioned (240) for the large language models (134), the artificially intelligent assistant (142) configured to at least retrieve text (122) and metadata (124) in response to receiving a prompt (502). A digital work environment (140) including an interface for the artificially intelligent assistant (142) is provided (250). At the artificially intelligent assistant (142), a prompt (502) related to a first work protocol (308) is received (410). Text (122) and metadata (124) related to the first work protocol (308) are retrieved (420) via the large language model (134) based on the received prompt (502). A contextual response (504) is provided (430) via the digital work environment (140) based on the retrieved text (122) and metadata (124).