Quad Infotech developed a scrap optimization program which optimizes the scrap recipe that is used by scrap yard operators to build buckets. The program shows the lowest cost recipe for multiple grades given information such as scrap cost, scrap inventory, chemistry specifications and production requirements. This will reduce scrap costs for the melt shop as they will know which scraps can be used for the lowest cost recipe and can purchase those scraps ahead of time. This program is intended to be used as a planning tool to show the best recipe to use based on a forecasted inventory (for the upcoming week for example).
The scrap optimization program takes several inputs such as scrap properties and recipe properties. Scrap properties include the scrap cost, scrap density, scrap inventory (predicted) and scrap chemistry. Recipe properties include the number of grades/recipes for which to optimize, the heat size, the number of buckets, the production quantity and the chemistry specifications for each grade.
Additional constraints can also be specified per bucket, scrap or layer such as weight, volume and layer order. The program uses linear optimization algorithms to identify the lowest cost recipe and shows how much of which scrap to put in which layer.