This paper contains a description of a model and solution procedure for a class of resource allocation problems. The work was motivated by the task of allocating energy resources to areas of consumption in the copper processing industry, but is a generalization of the single-source transportation model and the generalized assignment model. A heuristic is presented which, for typical situations, finds solutions to within .1 to.2% of a theoretical lower bound. Computer time and storage requirements for this procedure are negligible. Conservative estimates indicate that savings greater than three quarters of a million dollars per year can be realized in one firm by using this procedure. It also provides accountability of energy usage and costs as well as giving management the ability to analyze a variety of “What-if” situations such as changes in price, availabilities of fuels, new contractual opportunities, and new regulatory policies.