Neighborhood search heuristics for selecting hierarchically well‐formulated subsets in polynomial regression

The importance of subset selection in multiple regression has been recognized for more than 40 years and, not surprisingly, a variety of exact and heuristic procedures have been proposed for choosing subsets of variables. In the case of polynomial regression, the subset selection problem is complicated by two issues: (1) the substantial growth in the number of candidate predictors, and (2) the desire to obtain hierarchically well‐formulated subsets that facilitate proper interpretation of the regression parameter estimates. The first of these issues creates the need for heuristic methods that can provide solutions in reasonable computation time; whereas the second requires innovative neighborhood search approaches that accommodate the hierarchical constraints. We developed tabu search and variable neighborhood search heuristics for subset selection in polynomial regression. These heuristics are applied to a classic data set from the literature and, subsequently, evaluated in a simulation study using synthetic data sets. © 2009 Wiley Periodicals, Inc. Naval Research Logistics, 2010     

The effect of nonnormality on variables sampling plans

Acceptance sampling plans based on variables have been in use for many years. Recently, there has been a renewal of interest in these plans, because of the relative efficiencies that they offer with respect to attributes sampling regarding sample size. Furthermore, in situations where acceptable quality levels are very small, and a high level of protection is desired, variables sampling is often much more efficient than attributes sampling. An important disadvantage of variables sampling is that the distribution of the parameter being inspected must be known. Most standard variables sampling plans assume that the distribution of this parameter is normal. This article examines the effect of the normality assumption in variables sampling. Methods to detect departures from normality are reviewed.     

An analysis of constrained robust regression estimators

Multicollinearity and nonnormal errors are problems often encountered in the application of linear regression. Estimators are proposed for dealing with the simultaneous occurrence of both multicollinearity and nonnormality. These estimators are developed by combining biased estimation techniques with certain robust criteria. An iteratively reweighted least-squares procedure is used to compute the estimates. The performance of the combined estimators is studied empirically through Monte Carlo experiments structured according to factorial designs. With respect to a mean-squared-error criterion, the combined estimators are superior to ordinary least-squares, pure biased estimators, and pure robust estimators when multicollinearity and nonnormality are present. The loss in efficiency for the combined estimators relative to least squares is small when these problems do not occur. Some guidelines for the use of these combined estimators are given.     

Multiechelon repairable-item provisioning in a time-varying environment using the randomization technique

Multiechelon repairable-item provisioning systems are considered under a time-varying environment. Such conditions could arise, for example, in a military context where a shift from peacetime operation to wartime operation takes place; or, in a civilian setting where a public transit system decides to increase its hours of operation or frequency of service. Exact Markovian models, incorporating a finite population of repairable components and limited repair capacity (nonample service), are treated, with transient solutions obtained using the randomization technique. The exact models are compared with the approximate Dyna-METRIC model which assumes an infinite population of components and ample repair capacity.     

Interval estimation of a global optimum for large combinatorial problems

Consider an “intractable” optimization problem for which no efficient solution technique exists. Given a systematic procedure for generating independent heuristic solutions, we seek to obtain interval estimates for the globally optimal solution using statistical inference. In previous work, accurate point estimates have been derived. Determining interval estimates, however, is a considerably more difficult task. In this paper, we develop straightforward procedures which compute confidence intervals efficiently in order to evaluate heuristic solutions and assess deviations from optimality. The strategy presented is applicable to a host of combinatorial optimization problems. The assumptions of our model, along with computational experience, are discussed.     

What is the global landpower network and what value might it provide?

US national security guidance, as well as the US Army’s operational experiences since 2001, emphasizes the importance of working closely with partner countries to achieve US strategic objectives. The US Army has introduced the global landpower network (GLN) concept as a means to integrate, sustain and advance the Army’s considerable ongoing efforts to meet US national security guidance. This study develops the GLN concept further, and addresses three questions. What benefits can the GLN provide the Army? What are the essential components of the GLN? What options exist for implementing the GLN concept? By developing the GLN concept, the Army has the opportunity to transition the GLN from an often ad hoc and reactive set of relationships to one that the Army more self-consciously prioritizes and leverages as a resource to meet US strategic objectives.     

Cost allocation rules for elastic single‐attribute situations

Many cooperative games, especially ones stemming from resource pooling in queueing or inventory systems, are based on situations in which each player is associated with a single attribute (a real number representing, say, a demand) and in which the cost to optimally serve any sum of attributes is described by an elastic function (which means that the per‐demand cost is non‐increasing in the total demand served). For this class of situations, we introduce and analyze several cost allocation rules: the proportional rule, the serial cost sharing rule, the benefit‐proportional rule, and various Shapley‐esque rules. We study their appeal with regard to fairness criteria such as coalitional rationality, benefit ordering, and relaxations thereof. After showing the impossibility of combining coalitional rationality and benefit ordering, we show for each of the cost allocation rules which fairness criteria it satisfies. © 2017 Wiley Periodicals, Inc. Naval Research Logistics 64: 271–286, 2017     

Economic Losses From a Fire in a Dense-Packed U.S. Spent Fuel Pool

In 2013, the staff of the U.S. Nuclear Regulatory Commission estimated the reduction of the off-site economic losses from a fire in a drained U.S. spent fuel pool if fuel that had cooled for more than five years were transferred to dry cask storage—an option it called “expedited transfer.” In this article, it is shown that the savings would be much higher than the NRC estimated. Savings increase to about $2 trillion if: losses beyond 50 miles are included; the land-contamination threshold for long-term population relocation is changed to that used for the Chernobyl and Fukushima accidents and recommended by the U.S. Environmental Protection Agency; and, based on the experience of Japan, decontamination of land areas to levels acceptable for population return is assumed to take at least four years. If expedited transfer were implemented, the off-site economic losses would be reduced by about 98%.