Open Access. Powered by Scholars. Published by Universities.®

Physical Sciences and Mathematics Commons

Open Access. Powered by Scholars. Published by Universities.®

Articles 1 - 3 of 3

Full-Text Articles in Physical Sciences and Mathematics

A Heuristic Algorithm For The Multi-Terminal Vehicle Dispatch Problem Using The Modified Sweep Algorithm, Jerry George Johnson Jan 1975

A Heuristic Algorithm For The Multi-Terminal Vehicle Dispatch Problem Using The Modified Sweep Algorithm, Jerry George Johnson

Doctoral Dissertations

"A heuristic algorithm is developed for solving the multi-terminal vehicle dispatch problem. The demand points are first assigned to terminals using a scheme that minimizes the cost of each assignment. Single-terminal vehicle dispatch problems are then solved at each terminal using a heuristic procedure, called the Sweep Algorithm. A sequence of trial reassignments are then attempted in order to obtain further improvements in the overall solution. Eleven multi-terminal vehicle dispatch problems are presented and their solutions, derived by this procedure, are given.

Several changes are also proposed for the Sweep Algorithm in order to improve the solutions obtained and to …


A Study Of Some Life Testing Distributions, Robert Marvin Smith Jan 1975

A Study Of Some Life Testing Distributions, Robert Marvin Smith

Doctoral Dissertations

"This dissertation is presented in publication form and consists of three articles. The first article considers a new life-testing distribution which has the property of possessing a failure rate function which can be U-shaped or exponentially increasing depending on the value of the shape parameter. In addition, this article considers generalized least squares type estimators for location-scale distributions, then uses these estimators in the analysis of the exponential-power distribution. Tables are provided for which inferences on the location and scale parameters and on the reliability can be obtained. The second article gives relationships between a goodness-of-fit statistic based on a …


Mixed Nonderivative Algorithms For Unconstrained Optimization, Roger Ellis Lessman Jan 1975

Mixed Nonderivative Algorithms For Unconstrained Optimization, Roger Ellis Lessman

Doctoral Dissertations

"A general technique is developed to restart nonderivative algorithms in unconstrained optimization. Application of the technique is shown to result in mixed algorithms which are considerably more robust than their component procedures. A general mixed algorithm is developed and its convergence is demonstrated. A uniform computational comparison is given for the new mixed algorithms and for a collection of procedures from the literature"--Abstract, page ii.