User-Specific Engineering Workflows From Multi-User Action Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems in multi-user engineering environments fail to generate personalized user action recommendations due to the lack of role-based classification and scope of operations identification, leading to irrelevant recommendations and high user rejection rates.
Innovation Solution
A method and system that generate user-specific engineering programs by creating user profiles based on analysis of user actions using unsupervised learning algorithms, and then training an artificial intelligence model to generate customized workflows for each user profile.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional systems use machine learning models to generate user action recommendations without role-based classification, then the system can provide automated recommendations, but the recommendations are not personalized and have low user acceptance rates
Solution Approach 1:
The patent segments users into distinct user profiles based on their roles and scopes of operations within the engineering system. By classifying users into different profiles (e.g., developer, tester, operator) with specific roles and permissions, the system delivers personalized recommendations tailored to each profile's needs, thereby improving personalization without overwhelming complexity through systematic categorization
Solution Approach 2:
The patent introduces user profiles as an intermediary layer between the machine learning model and individual users. These profiles aggregate common roles, scopes of operations, and action patterns, allowing the system to personalize recommendations efficiently by matching users to pre-defined profiles rather than creating entirely custom models for each user
2Reliability
If the system gathers and processes a gigantic number of user actions for training, then the model can learn comprehensive patterns, but the task becomes tedious and time consuming
Solution Approach 1:
The patent segments the gigantic set of user actions into manageable subsets corresponding to different user profiles. Each profile receives training data filtered by relevant roles, scopes of operations, and action types, reducing the effective training dataset size for each model while maintaining comprehensive coverage of relevant patterns through distributed training across multiple profile-specific models
Solution Approach 2:
The patent performs preliminary classification and organization of user actions into user profiles before training the machine learning model. By pre-aggregating actions according to user roles and scopes of operations, the system prepares structured training data that reduces processing time during model training while ensuring comprehensive pattern learning within each profile category
3Ease of operation
If the machine learning model generates generic user action recommendations without identifying scope of operations, then the model can operate with simpler logic, but the recommendations are irrelevant to specific user contexts
Solution Approach 1:
The patent applies local quality by tailoring recommendations to specific user profiles with distinct roles and scopes of operations. Each user profile receives customized recommendations based on their specific context (e.g., a developer profile receives coding-related suggestions while an operator profile receives operational suggestions), ensuring relevance without requiring complex generic logic for all users
4Measurement precision
If conventional systems store confidential user information for personalization, then the system can identify users accurately, but cybersecurity risks increase
Solution Approach 1:
The patent extracts and removes confidential user information (such as personal identifiers, passwords, and sensitive data) from the user profile system. Instead, it uses anonymized or pseudonymized identifiers that maintain user identification accuracy while eliminating security vulnerabilities associated with storing sensitive personal information
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
The present invention provides a method and system for generating engineering programs in a multi-user engineering environment (100). The method comprises generating a plurality of user profiles (108A-N) for the plurality of users, based on an analysis of a plurality of user actions performed by the plurality of users. The method further comprises training an artificial intelligence model to generate a workflow which is specific to each user profile of the plurality of user profiles (108A-N), based on the analysis of the plurality of user actions of the plurality of users. The method further comprises generating a workflow which is specific to the user based on the determined user profile and an application of the generated artificial intelligence model on the detected at least one user action performed by the user in the multi- user engineering environment.