A Planning System for Facilities and Resources in Distribution Networks

A system for planning facilities and resources in distribution networks is presented. It was applied at Cahill May Roberts, one of Ireland's largest pharmaceutical companies, having a turnover of $35 million and employing over 300 people.The planning system, a three stage stochastic programming model, is capable of evaluating alternative management strategies in the face of fluctuating costs and population movements over an extended time horizon. Apart from defining unique territories for the company's distribution centres and optimal customer servicing schedules within those territories, the planning system evaluated alternative locations for distribution centres under the twin uncertain ties of cost and demand.The planning model contained approximately 2,000 variables and 300 equations, whilst the undecomposed route planning model encompassed approximately 300 towns/villages and 1,200 customers Both models, therefore required substantial data inputs as successive management objectives and assumptions were examined. To facilitate this data preparation and to achieve solutions in a realistic time, a special matrix generator was developed which had the capability of producing data in the required format and inputing directly into both planning models. The generator controls the entire optimizing process and permits a three hundred-told reduction in data preparation. In effect the generator permits an optimization procedure to be used in a simulation mode.The model achieved savings in delivery and transport costs of 23.3% and 20% respectively, and increased customer service levels by 60%.