Predictive Analytics for Optimal Repair Determination

Air Logistics and Engineering (ALAE) Solutions is under contract to develop, test and evaluate a Predictive Maintenance methodology for critical shop systems supported by a novel Digital Twin framework seeking to optimize system design and maintenance policies.

Current maintenance practices are reactive and lack the tools to further improve the assets readiness.  ALAE Solutions will build upon a successful Phase I effort to develop, test and evaluate novel Predictive Maintenance and Digital Twin technologies to ascertain that component inspection, repair and quality control processes are operating reliably, efficiently and expeditiously.

We introduce a dual approach to Predictive Maintenance (PM) addressing the on-process maintenance of critical manufacturing systems in real time while a second approach builds on probabilistic design methods and relies on novel reliability analysis and life cycle management tools to deliver optimum component design and maintenance policies.

Innovative aspects include:

• Data mining and deep learning for feature extraction/selection
• An integrated fault diagnosis and prognosis architecture with performance guarantees.
• Usage-based long-term prognostic algorithms seek an estimate of the component’s End of Life (EoL).
• Condition Based Maintenance and other AR/VR maintenance training aids.

An intelligent Digital Twin framework – an interplay between the physical and virtual worlds – supports the Predictive Maintenance functions and allows for expedient design, testing and evaluation of “smart” manufacturing technologies.