SBIR-STTR Award

The capability to routinely collect, assess, format, predict, and track readiness, performance, and proficiency data from live aircraft, instrumented ranges, and distributed mission operations simulation environments is represents a unique and critical capability for the Air Force. Lumir Research Institute proposes to build the Predicting, Analyzing, and Tracking Training Readiness and Needs (PATTRN) tool, a software suite that will provide access to performance data from various environments regardless of the native format. PATTRN will collect data from various environments, translate the data from its native format into a common format, store the data, routinely assess and track readiness and predict future readiness or future training proficiency fall offs. PATTRN will enhance the capabilities of existing data processing tools by providing access to data from a wide variety of environments, and in a wide variety of formats, along with linking raw data to performance measurement and readiness models. The ultimate goal of PATTRN is to provide a data framework that is both site- and protocol-independent, thus enabling readiness and future proficiency assessment across environments. PATTRN will not only enable longitudinal studies of performance across a wide variety of environments, but will also contribute to the ongoing efforts to achieve greater interoperability.

Benefit:
The proposed PATTRN system will provide the following

Benefits:
Data translation capability from esoteric data formats to a common data format. Interoperability with existing data processing tools. Standardized means of tagging data across discrete environments. Routine assessment of trainees""proficiency across multiple environments. Routine performance measurement evaluations across multiple environments. Predicting future training proficiency falloffs. The proposed PATTRN system has the following

Potential Commercial Applications:
The system architecture will be applicable in other domains where multiple independent data formats exist (e.g., Navy). The capability to predict future proficiency gaps will be applicable to industries where the time required for a human to complete a routine task (such as UPS loading a truck, or an auto mechanic changing a transmission) are dependent upon the frequency at which the task is performed. A common data format is the gateway by which existing commercial systems may share data with DoD systems.

Lumir is developing the Predicting, Analyzing, and Tracking Training Readiness and Needs (PATTRN) system, a comprehensive multi-format capable data collection, processing, tagging, and storage system which enables users to predict the most effective future training situation(s) for a trainee based on their past performance and their current and future proficiency profile. PATTRN will be capable of identifying ‘which’ skills are decaying and ‘when’ they will fall below an acceptable threshold, and informing ‘how’ scenarios and training exercises can be best constructed and ‘where’ the most efficient training can occur. The PATTRN system is designed and architected for implicit integration with enterprise systems such as AFRL’s ongoing Learning Management System Advanced Technology Development (LMS ATD) project to maximize the benefits derived from existing systems. An underlying component of PATTRN is the Common Data Acquisition system (CoDA), which is a completed Lumir Phase II SBIR that is currently being used within the AFRL DMO testbed. Lumir will leverage CoDA’s extensive capabilities to process, translate, tag, and store data from multiple native formats as the backbone of the PATTRN system. Advanced query capabilities will also provide the ability to examine scenario objects such as the effectiveness of formations and tactics against opposing forces.

Benefit:
At the conclusion of Phase II PATTRN will be a powerful data processing and storage system that will provide much-needed capabilities that will increase training effectiveness while at the same time decreasing training costs. PATTRN will provide the following

Benefits:
• Integration of previously isolated multiple data sources that will provide a robust, detailed representation of pilot proficiency. • Advanced query capabilities that will enable users to address higher order research questions and adapt to changing needs, priorities, and capabilities. • Detailed mappings of component metrics to a comprehensive set of skills that will enable automated compilation, analysis, and representation. • Predictive modeling of skill decay that will inform instructors of current and future training needs, facilitating scenario selection, scheduling, and allocation of training resources for individual pilots as well as pilot groups. • Measures of latency training that will increase the accuracy of skill decay models as well as inform the extent to which non-formal training activities impact skill decay and guide latency training based on pilot needs. • Mapping of training activities to proficiency constructs which will enable PATTRN to recommend training scenarios and activities based on the intersection of individual or group needs. PATTRNs capabilities also position it to be successfully applied in commercial settings. While its expansion into other areas of the U.S. Armed forces is a natural next step, being especially ready to be applied to multiple platforms in the USAF, PATTRN also fits the needs of education, law enforcement, and public safety. In education, changes in teacher licensure requirements indicate the educational system is moving towards a more intentional, high-stakes training system for teachers, and a more authentic, longitudinal growth model for students. Both of these cases fit into PATTRN’s ability to predict when skills need to be trained and represent levels of proficiency from multiple sources. In law enforcement and public safety, missteps and misconduct cost governments hundreds of millions of dollars, if not billions of dollars annually, and a large portion of this can be reasonably attributed to skill decay. PATTRNs ability to help individual manage the changing risks associated with skill decay can improve effectiveness and significantly reduce costs.

Keywords:
Skill Decay, Resource Management, Training capabilities, Learning Management System, Pilot Proficiency, Training Needs, Predictive Modeling, Performance measurement