|
|
|
|
MOR Journal Abstracts
|
 | Number 1 |
| Number 2 |
| Number 3 |
| Number 4 |
Modelling
The Rural Infantry Battle: Overall Structure And A Basic Representation Of The
Approach Battle (L. R. Speight and D. Rowland)
The
ultimate aim of this paper and its successors is to improve the manner in which
the infantry battle is commonly represented in our combat models. In order to do
this it first sets out a generic structure for the infantry battle, based on the
examination of literally hundreds of historical examples involving many
different armies and different theatres of war. It then gives an account of the
computer modelling of one part of this structure, the approach battle. This
simulation depicts the likely performance of trained troops in different types
of rural terrain in benign conditions, when they are not subject to any
realistic lethal threat. It relies heavily on the evidence gathered in actual
instrumented tactical trials. Great care has been taken to build in abstract
representations of the key processes, such as terrain screening, target
acquisition and aiming performance, which appear to have had the greatest impact
on eventual trial outcomes. The model outputs so obtained are consistent with
these trial outcomes. The model and results produced here are of interest for
two reasons. Firstly, and most importantly, they form a foundation upon which
other parts of the infantry battle structure can be constructed, and on which
the likely effects of genuine lethal threats can be imposed. Secondly, although
these extra effects have not yet been added, the results show that formulations
in line with Lanchester’s ‘Square Law’, commonly used in combat models as
a basis for calculating expected attrition rates in direct fire engagements, are
very unlikely to be consistent with real life outcomes. (Pg. 5)
A
Meta-Model Architecture for Fusing Battlefield Information (Patrick J. Driscoll and Steven J. Henderson)
Central
to successful U.S. force transformation is the goal of imposing net-centricity
across a host of operational domains. For sensor-based intelligence processing,
a handful of network learning systems exist that can potentially aid
decision-making in this environment. Steve Henderson and Pat Driscoll developed
a model that unifies these systems under a common architecture, and demonstrate
its ability to accurately estimate a force's unknown operational state by
logically fusing battlefield data. The results lead to a natural means of
assessing information advantage based on the degree of imbalance between
competing intelligence resource capabilities. (Pg. 27)
Automatic
Target Recognition System Evaluation Using Decision Analysis Techniques (Brian
Bassham, Kenneth W. Bauer and J.O. Miller)
In an
effort to assist Air Combat Command (ACC) in its efforts to procure effective
Automatic Target Recognition systems, a methodology is developed to account for
both the world-views of evaluators and warfighters. The method involves the
development of a two-pronged decision analysis model that maps ATR MOPs
(Measures of Performance) into value. This is a direct mapping for the
Evaluator. However, the Warfighter
thinks more in terms of MOEs (Measures of Effectiveness).
To incorporate the Warfighter perspective, a combat model was exercised
in a designed experiment to produce a response surface that could serve a
surrogate and intermediate mapping from MOP to MOE.
Two new methods of performing sensitivity analysis were developed.
Finally, the decision procedure is modified to account for the stochastic
nature of MOPs using the multinomial selection procedure. (Pg. 49)
Materiel
Readiness And The Operational Propulsion Plant Exam (OPPE) (Robert R.
Read and Lyn R. Whitaker)
The
paper treats two issues: an operational one and a methodological one.
The operational part deals with the uncovering of a well hidden signal.
The temporal nearness of the Operational Propulsion Plant Exam (OPPE) has an
effect upon the number of materiel casualty reports endured by a ship. This
number increases as the time of the exam approaches and declines monotonically
as the time of the exam recedes into the past. The common explanation is that
readiness resources are diverted in order to prepare for the exam, and that a
recovery period is needed afterward. Elementary statistical methods do not seem
adequate to capture this general effect, but the available data can be treated
using a log linear model from the categorical data analysis system of models. An
acceptable fit is found for the available data structure. The use of the time
from the exam factor is necessary in order to obtain an acceptable
representation.
The
methodological part deals with the use of the off-the-shelf software in order to
fit multivariate multinomial models to the data.
This class of models cannot be fitted directly, but the goal can be
accomplished with careful use of the Poisson family capability that is available
in the common software systems. This use is explained in general terms and in
the context of the problem at hand. There are two aspects to be watched when
doing this. First, the point estimation of the parameters necessarily requires
the inclusion of some special terms. Second there are described adjustments to
the standard (Poisson based) output that allow one to test the nested candidate
models after the deviances are appropriately modified. (Pg. 67)
An
Adaptive Tabu Search Approach for 2–Dimensional Orthogonal Packing Problems (John
Harwig, J.W. Barnes and James T Moore)
From a military perspective, fighter aircraft and warships self deploy. Everything else is packed into some type of container and shipped. This paper presents an adaptive tabu search procedure, which can be useful in modeling various types of military packing problems. It details a new and superior method for solving the two-dimensional orthogonal bin packing problem, Adaptive Tabu Search for Bin Packing (ATS-BP). This procedure takes advantage of move history and problem structure to dynamically change search neighborhoods. ATS-BP reduced the average lower bound gap for a common test set, totaling five-hundred problems, by twenty-five percent from the previous best. (Pg. 5)
Engineering
Index: An Engineering Certification/Qualification Metric (Jane M Booker,
Timothy Ross, Michael Hamada, Brian Reardon, Ron Dolin, Cheryll L. Faust and
Lorenzo Najera)
From a
military perspective, fighter aircraft and warships self deploy.
Everything else is packed into some type of container and shipped.
This paper presents an adaptive tabu search procedure, which can be
useful in modeling various types of military packing problems.
It details a new and superior method for solving the two-dimensional
orthogonal bin packing problem, Adaptive Tabu Search for Bin Packing (ATS-BP).
This procedure takes advantage of move history and problem structure to
dynamically change search neighborhoods. ATS-BP
reduced the average lower bound gap for a common test set, totaling five-hundred
problems, by twenty-five percent from the previous best. (Pg. 27)
Fitting
Data Using the Ramberg-Schmeiser Distribution (RSD) (Roy
E. Rice, FS, P.E.)
Whether
we analysts are taking sample data to make inferences about a population
parameter or we are trying to determine an underlying probability distribution
to enable us to develop algorithms for random number generators in a computer
program, we are frequently faced with fitting a distribution to sample data.
Four parameter distributions such as the Tukey family have been used for
many years. Roy Rice details the
Ramberg-Schmeiser Distribution (RSD), shows its derivation, and applies it to
two sample data sets. Its ease of
use and straightforward application make it a powerful tool in fitting
distributions. (Pg. 45)
Military
Operations Research Society (MORS) Oral History Project Interview of Jack
Borsting, FS (Bob Sheldon, FS
and Michael Garrambone)
No
Executive Summary (Pg. 57)
Mine Countermeasures (MCM) Tactical Decision Aids (TDAs), A Historical Review (George Pollitt)
The first automated Mine Countermeasures (MCM) Tactical Decision Aids (TDAs) for the U.S. Navy were developed in 1972-73 to support Operation End Sweep, the clearance of Haiphong and 25 other North Vietnamese ports and waterways. These crude stand-alone TDAs became unsupportable in the late 1970s, but were followed by the main-frame Mine Warfare Simulator in the early 1980s and the portable MCM Commander's Tactical Decision Aid (MCM-CTDA) in the mid-1980s. The MCM-CTDA was used to evaluate and analyze MCM exercises with emphasis on exercise validation of experimental tactics. It was last used to evaluate Desert Storm clean up and related operations in the Arabian Gulf. The modern Mine Warfare and Environmental Decision Aids Library (MEDAL), which began development in the early 1990s, continues to grow as the standard MCM TDA in use by the Navy. (Pg. 7)
Beyond COGNIT: Application of Markov Chains to Naval Minesweeping Theory (Michael McCurdy)
This paper presents an introduction to modeling naval influence minesweeping, specifically against mines equipped with shipcounters. A “many-sweeper model” is presented, in which minesweeper casualties are replaced instantaneously and without limit on the number of replacement sweepers. Three metrics are developed: countermeasures effort, expected casualties, and clearance level. Sample results from the model are presented, as well as results from COGNIT, an optimizing tactical decision aid based on the many-sweeper model. COGNIT optimizes one of the metrics subject to constraints on the other two. The many-sweeper model is then reformulated as a non-homogeneous Markov chain, and extensions of the Markov chain approach to modeling naval mine counter-measures in general are discussed. (Pg. 19)
Estimating Risk to Transiting Ships Due to Multiple Threat Mine Types (W. Reynolds Monach and Joni E. Baker)
The Mine Warfare and Environmental Decision Aids Library (MEDAL) is the U.S. Navy’s system for planning and evaluating mine warfare operations. One important component of MEDAL is the Advanced Risk Evaluation Module (AREM), which uses information concerning the exact locations and types of threat mines that have been located to evaluate the risk posed to transiting ships by threat mines which may remain undetected in the area. Early Simple Initial Threat (SIT) algorithms produced an estimate of risk due to only a single mine type in one tactical segment. The authors discuss their approach (currently used in AREM) for extending this algorithm to the more general case, in which multiple threat mine types may be located in multiple consecutive tactical segments. Their algorithm employs a Bayesian technique, and much of the discussion therefore centers on the nontrivial problem of choosing an appropriate prior distribution on the number and types of mines present. (Pg. 35)
Estimating the Probability of Landmine Contamination (LTC Stephen R. Riese, Donald E. Brown and Yacov Y. Haimes)
This paper introduces a new approach to forecasting landmine contamination in war-torn areas. The Probability of Mine (PoM) forecast incorporates information not used in traditional predictive pattern analysis, and provides more accurate estimates to support decision-making in the allocation of scarce demining resources, relocation of refugees, or planning of peacekeeping operations. That new information is spatial feature data that helps drive human behavior, in this case the choice of where to place landmines. Data gathered by the U.S. Army in Bosnia in 1995 and 1996 provides real world data to test and evaluate the model. The PoM forecast is built upon an empirical Bayesian prediction model that uses examples of areas known to be mined, examples of areas known to be free of mines, feature data (e.g., location of roads, vegetation, fighting lines), and an initial estimate of the overall mine density within the region. Results demonstrate that the PoM model is able to make accurate probabilistic forecasts on the presence of landmines, as well as provide measures of forecast quality that address both the model’s ability to account for uncertainty and the model’s predictive power. (Pg. 49)
Katz
Distributions and Minefield Clearance (Alan
Washburn)
This paper deals with the encoding of necessarily uncertain information about the number of mines initially present when clearing a minefield. A particular “Katz” class of probability distributions is found to be convenient, as well as flexible enough to encompass the kinds of uncertainty usually encountered. The Katz class is closed under many of the usual operations involved in clearing a minefield, which permits mine clearance to be regarded as a sequential process where the output of one stage is the input to the next. The Katz class also permits a simple, analytic measure of minefield threat (SIT) that can be the basis of tactical optimization. (Pg. 63)
Probability
Modeling of Autonomous Unmanned Combat Aerial Vehicles (Moshe
Kress, Dipl.-Ing. Arne Baggesen and Eylam Gofer)
Advances in sensors and command, control, communications, computers, intelligence, surveillance and reconnaissance (C4ISR) technologies, coupled with operational needs, like the war against terror, have led in recent years to the development of a new class of weapon systems called Unmanned Combat Aerial Vehicles, or in short – UCAVs. Autonomous UCAVs combine a unique set of capabilities in one platform; they have an eye that senses the area and gathers target information, a brain that processes this information, wings that move the UCAV around and keep it aloft and a fist, in a form of a warhead. This paper addresses several design and operational issues related to the employment of UCAVs. In particular, we study tradeoffs among properties related to the eye, brain, wing and fist such as detection, situational awareness, memory, coordination, vulnerability and lethality. (Pg. 5)
Distorted
Risk Measures with Application to Military Capability Shortfalls (Edwin
J. Offut, Jeffrey P. Kharoufeh and Richard F. Deckro)
In
today’s environment of transformation, budget restrictions, asymmetric
conflict, and evolving technologies, it is essential that the risks associated
with military capability shortfalls are correctly modeled and evaluated,
especially when low-likelihood events result in potentially catastrophic losses.
This study focuses on selecting an appropriate distortion function and
associated parameters to account for rare but catastrophic events that may
result from shortfalls in military capabilities.
Using a notional example, we illustrate how our approach might be applied
within the context of resource allocation. This
work was selected as Best Working Group Paper in WG 21, Readiness, at the 73rd
MORS Symposium. (Pg. 25)
Estimating
Total Program Cost of a Long-Term, High-Technology, High-Risk Project with Task
Durations and Costs That May Increase Over Time (Gerald
G. Brown, Maj Roger T. Grose and Robert A. Koyak)
The U.S.
Army’s Future Combat Systems (FCS) exemplifies the challenges of scheduling
large-scale military acquisitions. The
Cost Analysis Improvement Group (CAIG) in the Program Analysis and Evaluation
(PA&E) branch of OSD wanted to compare three different schedule plans for
FCS. Brown, Grose, and Koyak develop an innovative application of
integer programming, combined with simulation, which brings greater realism into
analyzing alternate scheduling plans. Using
FCS to demonstrate their approach, the authors show how useful comparisons can
be made taking into account uncertainty in task durations and budget constraints
over the planning cycle of the project. (Pg. 41)
Military
Operations Research Society (MORS) Oral History Project Interview of Gregory S.
Parnell, FS (Bob Sheldon, FS)
No Executive Summary (Pg. 63)
Some
of our links are in portable document format (pdf); they will be indicated by
this symbol
|
