Analysis of some two-characteristic Markov-type manpower flow models with retraining applications.

Abstract

A discrete state, discrete time Markov-type manpower flow model having a two-dimensional state space (i.e., a two characteristic model) is analyzed. The probabilistic properties of the model and the equations of stocks and flows are developed. A new method of representing the stocks as a sum of steady-state and transient components is presented. Two specific applications of the model to multi-grade hierarchical systems in which the dimensions of the state space are (grade, length of service) and (grade, time in grade) are analyzed in detail. The problem of combining states across the second dimension of the state space is studied and methods are derived for the (usual) case where the states are not lumpable. Finally, some applications of the model with state space (grade, skill group) to retraining problems between various skill categories are presented.