Research for Computational Methods

Syndicate content
Wartime Medical Requirements Models: A Comparison of MPM, MEPES, and LPX-MED
October 1, 1996
The end of the Cold War has led to a major reexamination of the requirements fir medical personnel and other resources necessary to care for Department of Defense (DoD) beneficiaries. Despite large numbers of peacetime beneficiaries, the primary need for medical resources remains the wartime mission. During the Cold War, the requirement for wartime medical providers was high and could easily justify large numbers of active duty personnel. Today's defense guidance is based on fighting two major regional contingencies (MRCs), and the required number of medical providers has fallen sharply. Determining the number and types of medical resources needed to treat casualties of future conflicts is an important and complicated issue. To help shed light on the process of determining wartime medical requirements, N-931 asked CNA to examine the methods and models that are currently in use or may be used in the near future. The current process is in flux. The models used today are being changed. The use of one model was discontinued recently when the Joint Staff changed to a new command-and-control system. The Joint Staff has proposed a replacement, but the new model is still being developed and there is some concern about its adequacy in determining service medical requirements. In this memorandum, we examine the current and proposed theater-level requirements models.
Read More | Download Report
Land Warfare and Complexity, Part I: Mathematical Background and Technical Sourcebook
July 1, 1996
The purpose of this paper is to provide the theoretical framework and mathematical background necessary to understand and discuss the various ideas of nonlinear dynamics and complex system theory to plant seeds for a later, more detailed discussion (provided in Part II of this report) of how these ideas might apply to land warfare issues. This paper is also intended to be a general technical sourcebook of information. The main idea put forth in this paper is that significant new insights into the fundamental processes of land warfare can be obtained by viewing land warfare as a complex adaptive system. That is to say, by viewing a military 'conflict' as a nonlinear dynamical system composed of many interacting semi-autonomous and hierarchically organized agents continuously adapting to a changing environment. See also CRM 96-68.
Read More | Download Report
Some New Estimates of the Navy's Indirect Manning Costs
December 1, 1995
How does a change in the manning of ships and squadrons at sea affect the Navy's shore-based manning? This question, while hardly new, has arisen recently in several different contexts. One involves cost-effectivness analyses of arsenal ships - which require relatively small crews - as alternatives to traditional surface combatants. The purpose of this paper is to provide a set of empirical estimates of the response of ashore manning to changes in manning of ships and squadrons - hereafter called afloat manning - based on the most recent time-series information available. Over the past six or seven years, the drawdowns in budgets, force structure, and manning have been substantial. Inclusion of that experience in the database from which cost-estimating relationships are developed is essential to the validity of the relationships for use in assessing the cost consequences of decisions presently or soon to be at hand. The analytical construct adopted here is a model that posits delayed adjustment of shore manning to changes in afloat manning.
Read More | Download Report
Correct and Incorrect Error Specifications in Statistical Cost Models
June 1, 1993
This paper reports on a Monte Carlo study of the accuracy of each estimation method under correct and incorrect error specifications. The multi-predictor model that serves as the basis of the study is representative of models with which cost analysts frequently deal. Results suggest that ordinary least squares, applied to the logs of the variables, may be the preferred method under either specification. This finding, however, is applicable strictly to the issue of parameter estimation. Accuracy of cost prediction requires further analysis.
Read More | Download Report
Computing AFQT Scores From Historical Data
August 1, 1992
The Armed Forces Qualification Test (AFQT) serves as an initial screen for military selection. AFQT scores must accurately reflect the quality of recruits who enter the services and must maintain the same meaning and interpretation over time. Unfortunately, several changes in the AFQT and other problems make such historical computations and score interpretations difficult. This research memorandum details these past problems and presents solutions that will allow for the proper computation of current AFQT scores from historical databases.
Read More | Download Report
An Assessment of Surrogates for Hands-On Tests: Selection Standards and Training Needs
July 1, 1990
A surrogate measure is an expedient indicator of job proficiency that is reliable and valid and that yields results similar to a Hands On Performance Test (HOPT). HOPTs are the benchmark by which surrogates should be compared but HOPTs are too expensive, dangerous, or time intensive to use for most purposes. This paper uses Job Performance Measurement (JPM) data to analyze the usefulness of six surrogates (proficiency marks, training grade point average, conduct marks, job knowledge tests, video firing tasks, and supervisor ratings) for infantry tests. The paper addresses how useful each proxy would be for setting infantry enlistment standards and assessing Marine Corps training needs.
Read More | Download Report
Costs of Improving Recruit Aptitudes: A Joint Product Approach
July 1, 1990
This research memorandum presents and applies a methodology for estimating the cost of recruiting individuals with alternative distributions of Armed Forces Qualification Test (AFQT) scores. The methodology takes account of the key institutional features of the recruiting process, including recruiter time allocation and procedural guidelines. The method is used to estimate the costs of different recruit-aptitude distributions, using data on applicants and accessions for all of the services.
Read More | Download Report
Approximations for Three Statistical Functions
April 1, 1990
Statistical work connected with the computerized adaptive testing version of the Armed Services Vocational Aptitude Battery often involves the density and cumulative functions of the normal distribution, and the logistic function. In long, iterative calculations, computation of these functions can be time consuming. This research memorandum presents piecewise linear and cubic approximations for the normal cumulative distribution function, the logistic function, and the normal density function. The approximations are as accurate as library functions, and much faster in computing speed.
Read More | Download Report
A Redundant Linear Constraint Elimination Technique
April 1, 1990
In linear programming, the goal is to solve for an M-dimensional vector, x, that minimizes (or maximizes) a linear 'cost' function, and which satisfies certain linear constraints. This paper concentrates specifically on these linear constraints, provides techniques for simplifying the linear programming problem by eliminating unneeded constraints, and presents a technique for identifying and eliminating redundant constraints.
Read More | Download Report
Cost Minimization and Uncertainty in Defense Manufacturing
March 1, 1990
This paper presents a mathematical model for two purposes. First, it extends the theoretical analysis of annual contracting to the more general case of uncertainty as to the magnitude of year-two procurement, as opposed simply to its cancellation. The second is to assign parameter values having at least some empirical grounding to the Utgoff-Thaler model, thereby providing a basis for assessing quantitatively the consequences of the theoretical results for both annual and multiyear contracting.
Read More | Download Report