ILLUMINATION WORKS, LLC — Department of Defense SBIR Phase I: AF192-001

Description

Information captured in field maintenance records provides vital data used by numerous communities in planning, logistics, and specifically, condition-based maintenance (CBM) and other maintenance-related disciplines. Today’s data capture process can be labor intensive, prone to errors, and difficult to assure a high level of data quality given the vast amount of enumerated codes, parts, actions, and maintenance types coupled with the heavy demands and time constraints on Air Force maintainers. Easing the data capture process reduces the time maintainers spend performing non-maintenance actions. Mobile technologies paired with machine learning offer an excellent pathway towards simplifying data capture, improving data quality, and exposing the vast amount of data available through online sources. Providing maintainers with mobile technology allows them to interact with these technologies as part of the maintenance workflow, creating a seamless way to enter data, examine schematics, and query manuals. Mobile platforms have numerous software tools resident on devices that can be utilized to improve data collection. By leveraging speech-to-text software and a machine learning pipeline, a maintainer can dictate narratives into a device’s software, and their notes will be transcribed to text with relevant data elements extracted and the resulting dataset recorded to the maintenance record.