Dependency Graph Effort Estimation for Process Plant Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating effort and implementing change requests in industrial automation systems are prone to human error and require extensive expert intervention, lacking accuracy and efficiency in determining the optimal schedule and cost for component modifications.
Innovation Solution
A system comprising a data acquisition unit, database unit, graph builder unit, and machine learning unit that builds a system dependency graph to compute impact and change parameter values, enabling accurate effort estimation and scheduling for change requests through machine learning techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional effort estimation techniques using Halstead's model and regression models are used, then expert input and feedback are incorporated, but human error and incorrect component recognition occur
Solution Approach 1:
The system enables self-service by automatically analyzing configuration files and system dependency graphs to identify impacted components without requiring expert intervention. The machine learning model autonomously performs effort estimation by traversing the dependency graph and computing impact parameters, eliminating human error while maintaining high accuracy.
Solution Approach 2:
The patent replaces the mechanical process of manual expert analysis with an automated computational system. The machine learning unit processes configuration files and dependency graphs algorithmically, substituting human cognitive processes with deterministic computational operations that eliminate human error in component identification and effort estimation.
2Reliability
If offline activity with human intervention is used for determining change request schedule, then expert knowledge is applied, but productivity and efficiency are reduced
Solution Approach 1:
The system performs self-service by automatically determining the schedule for implementing change requests without requiring offline human intervention. The machine learning model computes effort estimates and generates schedules by analyzing the system dependency graph, maintaining quality while dramatically improving productivity through automated, on-demand processing.
Solution Approach 2:
The system performs preliminary action by pre-computing system dependency graphs and storing them in the database. When a change request is received, the schedule determination is immediately available by querying the pre-built dependency structure, eliminating the need for time-consuming offline expert analysis while maintaining accurate scheduling.
3Measurement precision
If comprehensive assessment of all impacted components is performed, then accurate effort estimation is achieved, but device complexity and time consumption increase
Solution Approach 1:
The system applies segmentation by dividing the comprehensive system assessment into manageable components: configuration file parsing, dependency graph construction, impact parameter computation, and effort estimation. This modular approach maintains measurement precision by systematically analyzing each component while reducing overall system complexity through structured decomposition.
Solution Approach 2:
The system introduces an intermediary - the system dependency graph - that mediates between the change request and the effort estimation process. This graph structure organizes component relationships in a traversable format, enabling accurate identification of impacted components without requiring direct complex analysis of all system interactions, thus maintaining precision while reducing complexity.
4Measurement precision
If manual expert intervention is used for change request analysis, then accurate component recognition is achieved, but loss of time and efficiency occur
Solution Approach 1:
The system performs self-service by automatically recognizing impacted components through machine learning analysis of configuration files and dependency graphs. This eliminates the need for manual expert intervention while maintaining accurate component identification, thereby eliminating time loss without sacrificing recognition precision.
Solution Approach 2:
The system ensures continuity of useful action by maintaining the system dependency graph in the database for immediate querying. When change requests are received, the machine learning unit continuously processes requests by traversing the pre-built dependency structure, enabling uninterrupted, rapid component recognition without the delays associated with manual expert analysis.
Data Source
AI summary
A method (600) for configuring one or more components of a process plant includes receiving (602) a change request (226) and receiving (604) a system dependency graph (228) corresponding to the process plant. The method (600) further includes selecting (606) a subset of components (230) among the plurality of components based on configuration of the process plant and identifying (608) a subset of nodes (232) among the plurality of nodes by traversing a path in the system dependent graph (228). The method (600) also includes computing (610) an impact parameter (234) value based on a traversed path and computing (612) a plurality of change parameter values (236) based on the traversed path. The method (600) further includes determining (614) an effort estimate (210) based on the impact parameter (234) value and the plurality of change parameter values (236) using a machine learning technique.