USE OF NEURAL NETS IN PREDICTING PERSONNEL ATTRITION

Award Information
Agency: Department of Defense
Branch: Navy
Contract: N/A
Agency Tracking Number: 12599
Amount: $49,900.00
Phase: Phase I
Program: SBIR
Awards Year: 1990
Solicitation Year: N/A
Solicitation Topic Code: N/A
Solicitation Number: N/A
Small Business Information
Intech Inc
7 Devon St, Robbinsville, NJ, 08691
DUNS: N/A
HUBZone Owned: N
Woman Owned: N
Socially and Economically Disadvantaged: N
Principal Investigator
 David Anderson
 (609) 426-0446
Business Contact
Phone: () -
Research Institution
N/A
Abstract
THE APPLICATION OF NEURAL NETWORK TECHNIQUES TO PREDICTING PERSONNEL ATTRITION CAN HAVE SEVERAL ADVANTAGES OVER CONVENTIONAL APPROACHES SUCH AS TIME SERIES METHODS AND ECONOMETRIC MODELS. THE POWER OF THE NEURAL NETWORK APPROACH HAS BEEN DEMONSTRATED BY THEIR ABILITY TO REPRESENT NONLINEAR MAPPING NETWORKS, AND TO SUCCESSFULLY MANAGE MULTIVARIATE DATA. THIS PHASE I PROGRAM WILL IDENTIFY AND CHARACTERIZE THE VARIABLES THAT HAVE THE STRONGEST INFLUENCE ON AN INDIVIDUALS LIKELIHOOD OF ATTRITION. SECONDLY, A NEURAL NETWORK BASED MODEL WILL BE DEVELOPED UTILIZING THE SELECTED INPUT VARIABLES TO PREDICT PERSONNEL ATTRITION RATES. FINALLY, A THOROUGH COMPARISON BETWEEN THE DEVELOPED MODEL AND EXISTING METHODS WILL BE PRESENTED. THIS PROJECT IS EXPECTED TO PROVIDE A SOLID FOUNDATION FOR A PHASE II DEVELOPMENT OF A COMPLETE MANPOWER PLANNING TOOL. WITH THE EXPECTED INCREASE IN PERFORMANCE FROM THE PROPOSED METHOD, FOLLOW FUNDING IS FULLY EXPECTED.

* information listed above is at the time of submission.

Agency Micro-sites

SBA logo
Department of Agriculture logo
Department of Commerce logo
Department of Defense logo
Department of Education logo
Department of Energy logo
Department of Health and Human Services logo
Department of Homeland Security logo
Department of Transportation logo
Environmental Protection Agency logo
National Aeronautics and Space Administration logo
National Science Foundation logo
US Flag An Official Website of the United States Government