On the convergence of stochastic dual dynamic programming and related methods

We discuss the almost-sure convergence of a broad class of sampling algorithms for multistage stochastic linear programs. We provide a convergence proof based on the finiteness of the set of distinct cut coefficients. This differs from existing published proofs in that it does not require a restrictive assumption.     

Stochastic programming for nurse assignment

We present a brief overview of four phases of nurse planning. For the last phase, which assigns nurses to patients, a stochastic integer programming model is developed. A Benders’ decomposition approach is proposed to solve this problem, and a greedy algorithm is employed to solve the recourse subproblem. To improve the efficiency of the algorithm, we introduce sets of valid inequalities to strengthen a relaxed master problem. Computational results are provided based upon data from Baylor Regional Medical Center in Grapevine, Texas. Finally, areas of future research are discussed.     

Structured minimal-memory inexact quasi-Newton method and secant preconditioners for augmented Lagrangian optimization

Augmented Lagrangian methods for large-scale optimization usually require efficient algorithms for minimization with box constraints. On the other hand, active-set box-constraint methods employ unconstrained optimization algorithms for minimization inside the faces of the box. Several approaches may be employed for computing internal search directions in the large-scale case. In this paper a minimal-memory quasi-Newton approach with secant preconditioners is proposed, taking into account the structure of Augmented Lagrangians that come from the popular Powell–Hestenes–Rockafellar scheme. A combined algorithm, that uses the quasi-Newton formula or a truncated-Newton procedure, depending on the presence of active constraints in the penalty-Lagrangian function, is also suggested. Numerical experiments using the Cute collection are presented.     

一般整数规划中有效不等式的同步升维

有效不等式在整数规划的定界研究中具有重要的意义.研究了一般整数规划问题的有效不等式的升维方法,引入超加性函数给出同步升维的条件,并给出有效不等式的同步升维的具体方法,算例表明本文提出的方法是有效的.     

A sequential method for a class of pseudoconcave fractional problems

The aim of the paper is to maximize a pseudoconcave function which is the sum of a linear and a linear fractional function subject to linear constraints. Theoretical properties of the problem are first established and then a sequential method based on a simplex-like procedure is suggested.      

The method of feasible directions for mathematical programming problems with preconvex constraints

The convergence of the method of feasible directions is proved for the case of the smooth objective function and a constraint in the form of the difference of convex sets (the so-called preconvex set). It is shown that the method converges to the set of stationary points, which generally is narrower than the corresponding set in the case of a smooth function and smooth constraints. The scheme of the proof is similar to that proposed earlier by Karmanov.     

Stability of multiobjective NLP problems with fuzzy parameters in the objectives and constraints functions

This paper deals with the stability of multiobjective nonlinear programming problems with fuzzy parameters in the objectives and constraints functions. These fuzzy parameters are characterized by fuzzy numbers. The existing results concerning the qualitative analysis of the notions (solvability set, stability sets of the first kind and of the second kind) in parametric nonlinear programming problems are reformulated to study the stability of multiobjective nonlinear programming problems under the concept of α-pareto optimality. An algorithm for obtaining any subset of the parametric space which has the same corresponding α-pareto optimal solution is also presented. An illustrative example is given to clarify the obtained results.     

Transcendental geometric programs

This paper treats a class of posynomial-like functions whose variables may appear also as exponents or in logarithms. It is shown that the resulting programs, called transcendental geometric programs, retain many useful properties of ordinary geometric programs, although the new class of problems need not have unique minima and cannot, in general, be transformed into convex programs. A duality theory, analogous to geometric programming duality, is formulated under somewhat more restrictive conditions. The dual constraints are not all linear, but the notion ofdegrees of difficulty is maintained in its geometric programming sense. One formulation of the dual program is shown to be a generalization of the chemical equilibrium problem where correction factors are added to account for nonideality. Some of the computational difficulties in solving transcendental programs are discussed briefly.This research was partially supported by the National Institute of Health Grant No. GM-14789; Office of Naval Research under Contract No. N00014-75-C-0276; National Science Foundation Grant No. MPS-71-03341 A03; and the US Atomic Energy Commission Contract No. AT(04-3)-326 PA #18.     

On the number of registers needed to evaluate arithmetic expressions

The question of how many temporary storage registers are needed to evaluate compiled arithmetic and masking expressions is discussed. It is assumed that any combination of left-to-right, right-to-left, top-to-bottom, and bottom-to-top techniques may be used to evaluate an expression, but that no factoring or re-arranging of the expression may occur. On this basis, the maximum number of registers needed to evaluate nonparenthesized expressions isN+1, withN the number of dyadic operator precedence levels. For parenthesized expressions with a maximum ofK nested parenthetical subexpressions, the maximum number of registers needed is (K+1)N+1.     

An application of Neustadt's abstract maximum principle to probability kinematics

An important class of problems in philosophy can be formulated as mathematical programming problems in an infinite-dimensional vector space. One such problem is that of probability kinematics: the study of how an individual ought to adjust his degree-of-belief function in response to new information. Much work has recently been done to establish maximum principles for these generalized programming problems (Refs. 3–4). Perhaps, the most general treatment of the problem presented to date is that by Neustadt (Ref. 1). In this paper, the problem of probability kinematics is formulated as a generalized mathematical programming problem and necessary conditions for the optimal revised degree-of-belief function are derived from an abstract maximum principle contained in Neustadt's paper.This work was supported by the National Research Council of Canada.The author is grateful to G. J. Lastman and J. A. Baker of the University of Waterloo for numerous suggestions made for improvement of this paper. The problem of probability kinematics was brought to the author's attention by W. L. Harper of the University of Western Ontario.