A Feasibility Study of the Collection of Unscheduled Maintenance Data Using Statistical Sampling Techniques.
Abstract
This thesis explores the feasibility of using statistical sampling techniques in lieu of a census to collect Air Force maintenance (MDC) data. A practical sampling methodology is identified and the sample size required to collect data with a specified degree of statistical precision is illustrated. The variable cost of MDC data collection and processing is also identified. Using the F-16A Fire Control System on the aircraft at one base as an example, the potential cost and effort savings resulting from sampling are evaluated. The sampling concept is based on a simple random sample of aircraft, by serial number, with full data collected on all aircraft in the sample. The sampling plan is designed to estimate the base level monthly total unscheduled maintenance manhours at the two digit work unit code level, with 10 percent relative precision and 90 percent confidence. The methodology used to estimate the variable cost of collecting and processing MDC data records is limited to base and Air Force Logistics Command (AFLC) levels. Base level costs considered are the opportunity cost of a maintenance technician's time to enter one MDC record into an automated system terminal, and the cost of computer processing and transmission of data to AFLC. AFLC costs considered are the machine time charges assessed against the DO56 Product Performance System. In the single system studied, the variability in monthly unscheduled manhours per aircraft was found to be high.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1985
- Accession Number
- ADA161448
Entities
People
- Robert A. Heinlein
Organizations
- Air Force Institute of Technology