Case Study by Mavryck
CANADA Space Agency – Phase 1 Innovation Challenge
Problem Statement
AI solution to improve the efficiency of a maintenance projects for space robotic sub-systems such as cameras, sensors, and mechanism. Predictive slippages (specifically self-diagnosis and cost/risk/schedule failure prediction). Recommendation engine of action needs based on projects performance patterns.
Solution
- The solution used historical data records. Designed auto-diagnostics and early prediction system, both preventive and preparatory actions can be taken. Utilized new digital technologies like ML and BDA to reduce project & maintenance cycles.
- Developed auto-diagnostics and failure prediction based on learning, hypothesis, and analysis using available data of a space robotic system.
- Provide a data classifier and labeling tool which includes preparation and cleansing to ensure the data is correctly formatted.
- Handle imbalanced data.
- Provide a diagnosis to help isolate the nature of slippages & overruns.
- Included information technology security as part of its inherent design.
Value
5% increased
in speed at which the project information review and value of intelligence to make rational decision and adapt productivity & inefficiencies.
Andon
Early Warning Light - prevents avoidable mistakes from occurring.
Consistent
system & user friendly.