Processing plant location studies II: Policy alternatives for New Zealand wool selling centres

Abstract

The nature of the general facility location problem has been discussed previously by McCarthy, Rodgers & Higham and a method of solution outlined. The present paper aims to extend the solution technique so that a number of low cost solutions with differing spatial characteristics are generated. This permits entrepreneurs or policy makers to choose among the alternatives. The facility location problem can be stated as the determination of the number, size and location of a certain type of facility in order to satisfy the demand for the services provided; the "goodness" of any solution, relative to others, is measured in terms of some effectiveness criterion. For example one problem reported extensively in the agricultural economics literature is the determination of the number, size and location of agricultural commodity processing plants (freezing works, wool selling centres, grain storage depots), so that the sum of total transport (assembly plus distribution) costs and total processing costs are minimised. Another general type of problem receiving attention in applied operations research relates to the number, size and location of public emergency services (for example ambulance and hospital services). Here, some measure of cost effectiveness is optimised Empirical studies include those of Fitzsimmons, Savas and Volz. This paper is concerned primarily with processing and has two major sections. The first section outlines methodology and discusses Monte Carlo and "forcing" extensions of standard solutions. The second section applies such procedures to the New Zealand wool selling centres' location problem.