Showing papers on "Bounding overwatch published in 1991"
TL;DR: A program called POINCARE is described that analyzes one-parameter autonomous planar systems at the level of experts through a combination of theoretical dynamics, numerical simulation, and geometric reasoning.
62 citations
Journal Article•
24 citations
12 citations
TL;DR: In this article, a review of lower bounding methodologies for lot-sizing production/inventory models is performed for both dynamic discrete time demand and continuous review, constant demand situations, where lower bounds are an essential ingredient in the evaluation of heuristics when optima are unknown or too computationally complex to comfortably evaluate.
Abstract: A review is undertaken of lower bounding methodologies for lot-sizing production/inventory models. This is done for both dynamic discrete time demand and continuous review, constant demand situations. The view is taken that lower bounds are an essential ingredient in the evaluation of heuristics when optima are unknown or too computationally complex to comfortably evaluate. Several directions of future research are suggested.
11 citations
11 citations
TL;DR: In this article, the problem of parameter bounding in the case of linear systems with bounded errors in variables is addressed, and the necessary and sufficient condition for a concise description of the feasible parameter set is given.
10 citations
TL;DR: New results on optimal outer bounding ellipsoids or boxes which, together with some extension of previous work on optimal inner bounds, represent a complete set of results on the problem.
10 citations
08 Apr 1991
TL;DR: In this article, an analytical method is presented for bounding the peak transient response of a structure to an arbitrary, dynamic input force vector of given total magnitude, which does not require either a simulation or a numerical search to bound the worst-case response.
Abstract: An analytical method is presented for bounding the peak transient response of a structure to an arbitrary, dynamic input force vector of given total magnitude. Since the method is based on the linear operator norm for the multivariable impulse response of the structural system, it does not require either a simulation or a numerical search to bound the worst-case response. Instead, the bound is computed directly from the structural eigenmodes. The method reported in this paper is a generalization of existing bounding methods for single-input/single-output methods to the more practical case of vector-valued forces and vector-valued responses. In addition, the paper presents a method for computing one of the possible input force vector tune histories that will excite a structural response equal to the bounded value. This worst-case input force is proposed as a candidate for force time histories to be used in structural qualification testing.
2 citations
TL;DR: An upper bounding algorithm is applied to the problem of the elastoplastic analysis of structures expressed as linear complementarity problems and the time consuming procedure of solving large quadratic optimization problems can be avoided.
2 citations
TL;DR: Procedures using interval analysis for computing guaranteed bounds on the solution set provides a means for doing a sensitivity analysis or simply bounding the effect of errors in data.
Abstract: Consider a global optimization problem in which the objective function and/or the constraints are expressed in terms of parameters. Suppose we wish to know the set of global solutions as the parameters vary over given intervals. In this paper we discuss procedures using interval analysis for computing guaranteed bounds on the solution set. This provides a means for doing a sensitivity analysis or simply bounding the effect of errors in data.
2 citations
Journal Article•
11 Jun 1991
TL;DR: The authors give a brief overview of the problem of set-membership estimation in adaptive signal processing as there seems to be a resurgence of interest in these types of estimation schemes as they feature 100% confidence regions for parameter estimates.
Abstract: The authors give a brief overview of the problem of set-membership estimation in adaptive signal processing. Recently, there seems to be a resurgence of interest in these types of estimation schemes as they feature 100% confidence regions for parameter estimates. Some of these recursive schemes also feature selective updating. More recently, the systolic implementation and lattice structure of some particular schemes have also been investigated. Exact polytope bounding, orthotope bounding, and the OBE algorithms are discussed. >