Research for Probabilities
December 1, 1990
Two forms, each containing 35 verbal and 30 mathematics items, have been developed for a new Enlistment Screening Test (EST) to predict Armed Forces Qualification Test (AFQT) scores of military applicants. These forms were constructed in two stages from items in discontinued versions of the Defense Department's test batteries. The first stage was to develop overlength forms from the available item pool. This research memorandum describes the second stage: constructing final forms by selecting items from the overlength forms. Item selection was based on the correlation of the item with AFQT, in a subsample of applicants with AFQT percentiles between 21 and 65. For each EST form, the AFQT score was predicted from the total score on the final EST items. The results were used to calculate expectancy tables which, for any given EST score, provide probabilities of exceeding the specified AFQT cutoffs. These probabilities are reported in tables.
Read More
| Download Report
November 1, 1990
This research memorandum presents a means of calculating the expected length of 'awaiting-parts time' of weapon replaceable assemblies, which are needed to calculate aircraft readiness in a multi-indenture, readiness-based sparing model. The method presented is appropriate for any other level of indenture. Although full scale application of the model is not feasible at this time, the model aids in comparing and evaluating existing models.
Read More
| Download Report
December 1, 1987
Using a sample of Naval Reserve recruiters for its analysis, this research memorandum estimates the effects of recruiting-duty experience and other personal characteristics on recruiter productivity. A Poisson probability model and multinomial model are estimated. The results are intended for use in calculating the number of recruiters required to meet Selected Reserve recruiting goals.
Read More
| Download Report
November 1, 1987
When an aircraft is being repaired or inspected and a certain part is found to have failed, the part is replaced immediately, if a spare part is available. This research memorandum describes a queueing model used to analyze a sparing decision for a part with general probabilistic demand. It also describes an extension of the model to include discriminating treatment of the repair and resupply pipelines. The final section applies the model to an example.
Read More
| Download Report
October 1, 1987
The probability of survival can be estimated using a parametric survival model in which a single regression equation is used to summarize the data. For some data, the hazard rate at particular intervals of time differs markedly from the rest of the hazard function. This sharp rise in the hazard rate violates the requirements of the common parametric survival models, and indicates that some alternative modeling approach should be considered. This research contribution outlines a methodology for estimating parametric survival models given non-smooth survival data. As an example of this methodology the procedure is used to estimate the survival pattern of Marine Corps Selected Reservists.
Read More
| Download Report
July 1, 1987
Many machines are used intermittently, that is, repeatedly turned on and off during normal operation. Eventually the machine fails. This research memorandum presents methodology for analyzing failures of machines that are repeatedly turned on and off. Because a machine can fail both when it is on and off, different parametric models for failure are used for each of these periods. An important issue addressed for such machines is how the intermittent use itself affects failure. Because the models can predict the chance of failure under different usage patterns, less harmful usage patterns can be recommended. As an example, the models are applied to a radar system, and both the immediate and cumulative effects of on-off cycling are demonstrated.
Read More
| Download Report
September 1, 1986
This research memorandum examines the effects of three factors on attrition from the Navy within two and six months of shipping for recruits who entered in FY 1985. These three factors are participation in the Delayed Entry Program (DEP), month of shipment, and enlistment program. The results indicate that DEP recruits have lower attrition than direct shippers, that attrition among direct shippers is higher in months with high accession rates, and that attrition rates vary by enlistment program.
Read More
| Download Report
August 1, 1986
This paper reports on the construction of an interactive, historical database for recruit survival probabilities. This FY 1978-1984 base includes accession and monthly survival information for non-prior service recruits by accession program, gender, shipment mode, educational category, mental group, and age. An important finding is that for recruits who are otherwise identical, survival is worse if they are shipped within the month they signed their initial contract.
Read More
February 1, 1981
This paper defines path integrals in phase space without using a time-division approach followed by a limiting process, thereby generalizing a similar procedure used in configuration space.
February 1, 1981
This paper presents the results of a study with two goals: (1) to develop a model of losses during the first year of service that could be used to evaluate recruiting policy changes and improve the screening of applicants for enlistment; and, (2) to develop a model for evaluating the productivity of Navy Recruiting Districts that could be used in setting quotas, allocating canvassers, and assessing recruiting performance.
Read More
| Download Report
Follow CNA on
© 2013 CNA. All Rights Reserved.
4825 Mark Center Drive, Alexandria, VA 22311 | 703.824.2000 | Directions | Site Map