Computing Aviation Sparing Policies: Solving a Large Nonlinear Integer Program

Deployed US Navy aircraft carriers must stock a large number of spare parts to support the various types of aircraft embarked on the ship. The sparing policy determines the spares that will be stocked on the ship to keep the embarked aircraft ready to fly. Given a fleet of ten or more aircraft carriers and a cost of approximately 50 million dollars per carrier plus the cost of spares maintained in warehouses in the United States, the sparing problem constitutes a significant portion of the Navy’s resources. The objective of this work is to find a minimum-cost sparing policy that meets the readiness requirements of the embarked aircraft. This is a very large, nonlinear, integer optimization problem. The cost function is piecewise linear and convex while the constraint mapping is highly nonlinear. The distinguishing characteristics of this problem from an optimization viewpoint are that a large number of decision variables are required to be integer and that the nonlinear constraint functions are essentially “black box” functions; that is, they are very difficult (and expensive) to evaluate and their derivatives are not available. Moreover, they are not convex. Integer programming problems with a large number of variables are difficult to solve in general and most successful approaches to solving nonlinear integer problems have involved linear approximation and relaxation techniques that, because of the complexity of the constraint functions, are inappropriate for attacking this problem. We instead employ a pattern search method to each iteration of an interior point-type algorithm to solve the relaxed version of the problem. From the solution found by the pattern search on each interior point iteration, we begin another pattern search on the integer lattice to find a good integer solution. The best integer solution found across all interations is returned as the optimal solution. The pattern searches are distributed across a local area network of non-dedicated, heterogeneous computers in an office environment, thus, drastically reducing the time required to find the solution.     

Random orderings of the integers and card shuffling

In this paper we study random orderings of the integers with a certain invariance property. We describe all such orders in a simple way. We define and represent random shuffles of a countable set of labels and then give an interpretation of these orders in terms of a class of generalized riffle shuffles.     

A 24-GHz Patch Array with a Power Amplifier/Low-Noise Amplifier MMIC

This paper presents an active patch array designed at 24 GHz. It can be used as a front-end component for a phased array. A series resonant array structure is chosen which is compact and easy excite. With 5 elements, the array proved a 12-dB antenna gain. A power amplifier and a low noise amplifier are designed on a single GaAs chip (PALNA). Bias switch is used in the PALNA, which greatly reduces the switch loss in a transceiver and increases the efficiency. 20-dB small signal gain is achieved in both power amplifier and low noise amplifier. The active patch array is built by the combination of the patch array and PALNA. The measured active gain of this antenna is 35-dB for the PA mode and 31-dB for the LNA mode. This active patch array can obtain an EIRP of 34 dBm with a total radiated power of 22dBm and a maximum PAE of 32%. To check the noise performance, we applied sources at both normal temperature and 77K (liquid nitrogen) and extracted the noise figure (3.5 dB) of the active antenna by the Y factor method. The results proved that the active antenna is working efficiently as both a transmitting and receiving antenna.     

Geometry optimization in Cartesian coordinates: Constrained optimization

An efficient algorithm for constrained geometry optimization in Cartesian coordinates is presented. It incorporates mode-following techniques within both the classical method of Lagrange multipliers and the penalty function method. Both constrained minima and transition states can be located and, unlike the standard Z-matrix using internal coordinates, the desired constraints do not have to be satisfied in the initial structure. The algorithm is as efficient as a Z-matrix optimization while presenting several additional advantages.     

Electrostatic potentials of molecules and prediction of the lewis acid-base properties of solvents,part I. General considerations

Summary Electrostatic potentials in the vicinity of isolated molecules of typical nonaqueous solvents were computed in the MNDO approximation. It was found that the calculated dipole moments are in good agreement with experimental values. However, it turned out that in many cases electrostatic interactions of ions with solvent molecules cannot be described in the frame of ion-dipole interactions. Thus, two computable parameters,UN andUP describing short-range electrostatic interactions of a solvent molecule with a cation or an anion of radiusR were proposed. The dependence ofUN andUP uponR is analyzed in the context of ion-solvent interactions.

---

Elektrostatische Potentiale von Molekülen und Voraussage von Lewis-Säure-Base-Eigenschaften von Lösungsmitteln, 1. Mitt.: Allgemeine Grundlagen

Zusammenfassung Es wurden die elektrostatischen Potentiale in der Umgebung von isolierten Molekülen typischer nichtwäßriger Lösungsmittel in der MNDO-Näherung berechnet. Die berechneten Dipolmomente waren in guter Übereinstimmung mit den experimentellen Werten. Es wurde jedoch beobachtet, daß in vielen Fällen die elektrostatischen Wechselwirkungen von Ionen mit Lösungsmittelmolekülen nicht im Rahmen von Ionen-Dipol-Wechselwirkungen beschrieben werden können. Dafür wurden zwei berechenbare ParameterUN undUP zur Beschreibung von elektrostatischen Wechselwirkungen im Nahbereich eines Lösungsmittelmoleküls mit einem Kationen- oder AnionenradiusR vorgeschlagen. Die Abhängigkeit vonUN undUP vonR wurde im Zusammenhang von Ionen-Solvens-Wechselwirkungen untersucht.

     

Regularity and other properties of absolutely continuous invariant measures for the quadratic family

In the current paper we study in more detail some properties of the absolutely continuous invariant measures constructed in the course of the proof of Jakobson's Theorem. In particular, we show that the density of the invariant measure is continuous at Misiurewicz points. From this we deduce that the Lyapunov exponent is also continuous at these points (our considerations apply just to the parameters constructed in the proof of Jakobson's Theorem). Other properties, like the positivity of the Lyapunov exponent, uniqueness of the absolutely continuous invariant measure and exactness of the corresponding dynamical system, are also proved.This paper was written during the author's stay at the IAS while supported by NSF grant DMS-860 1978     

Grow-up rate of solutions for a supercritical semilinear diffusion equation

We study solutions of the Cauchy problem for a supercritical semilinear parabolic equation which converge to a singular steady state from below as t→∞. We show that the grow-up rate of such solutions depends on the spatial decay rate of initial data.     

Structural properties and tractability results for linear synteny

The syntenic distance between two species is the minimum number of fusions, fissions, and translocations required to transform one genome into the other. The linear syntenic distance, a restricted form of this model, has been shown to be close to the syntenic distance. Both models are computationally difficult to compute and have resisted efficient approximation algorithms with non-trivial performance guarantees. In this paper, we prove that many useful properties of syntenic distance carry over to linear syntenic distance. We also give a reduction from the general linear synteny problem to the question of whether a given instance can be solved using the maximum possible number of translocations. Our main contribution is an algorithm exactly computing linear syntenic distance in nested instances of the problem. This is the first polynomial time algorithm exactly solving linear synteny for a non-trivial class of instances. It is based on a novel connection between the syntenic distance and a scheduling problem that has been studied in the operations research literature.