Selected sensitivity and constraint studies for a finite population queuing system deployed to meet a demand

Abstract

The purpose of this thesis is to study the sensitivity of certain parameters in a conceptual system containing a finite population of units deployed to meet a demand. Selected constraints on the design variables are also treated. The measure of effectiveness in this system design is the total annual equivalent cost. This cost is the summation of the total population unit costs, depot and intermediate repair facility costs, training and initial support costs, recurring support costs, and cost of not meeting a demand on the system, Three separate areas are addressed in this study. The establishment of a standard system design, the study of the variation in the cost as compared to the standard design, and the effects of constraints on the system cost compared to the cost of the standard design. A special computer algorithm was developed to perform a multiple search through the design variables. This search yields the minimum total system cost and a resultant system configuration to be specified by the designer which is optimum. The algorithm was then used to analyze the effects on the minimum total system cost of variation in system parameters and constraints on design variables.