| 主权项 |
1. In an inventory control system having a database including a product identifier corresponding to each of a plurality of commodity items, a method of updating a delivery quantity in the database of each of the commodity items, comprising:
identifying, for each of a plurality of items represented in an inventory database, a field for a product identifier and a field denoting a safety stock for the item; determining, for each of the product identifiers, a product velocity field indicative of increased product replenishment demands based on a sales volume; determining, based on the velocity field, whether to compute a safety stock, the safety stock computation further comprising: identifying each SKU of a plurality of stock SKUs as a unique combination of an item at a location; identifying a customer service level target defined as a percentage of the stock SKUs available at a particular time, at one or multiple locations, regularly available from the managed inventory, and the customer service level target further indicative of an expected percentage of the stock SKUs for which at least one item is in stock; identifying, in a retail business, a natural pattern defining a sales cycle, the sales cycle being a week and the sales periods being days of the week (DOW), with the sales cycle defining a sequence of the sales periods; computing, based on a statistical inference from previous sales periods in the sales cycle, a forecast bias and random forecast error for each stock SKU sold from the inventory, for each day of the week (DOW), the statistical inference based on a sales history including at least 4 previous sales cycles of corresponding sales periods defined by a similar day of the days of the week; computing, prior to placing an order, a safety stock quantity for each day of a scheduled order arrival, according to its DOW, based on the standard deviation of the computed random forecast error for the days of the week spanning a variable forecast interval (VFI), multiplied by a Z factor based on the customer service level target, and subtracting the computed forecast bias for each day of a variable order interval (VOI), computing the safety stock further comprising: computing the VFI based on a sum of the lead time and VOI, expressed as specific days of the week within the VFI, where the lead time comprises the days of the week from order placement to order arrival, and the VOI comprises the specific days of the week in an order interval from order arrival until the arrival of the next successive order; aggregating the sums of the squares of the random forecast error, for the days of the week defining the VFI period taken together, for all the weeks of the sales history, and then calculating the average mean squared error, by dividing the sum of squares by the number of historic observations, and then calculating the standard deviation by taking the square root of the aggregated sums; maintaining, based on the computed safety stock quantity, for each day of the week of a successive sales cycle, a stock level of each SKU at the lowest possible level while maintaining the target level of non-zero inventory of a percentage of the stock SKUs based on the customer service target percentage; rendering an order quantity for each scheduled order arrival day that is based on the safety stock so calculated for that order arrival day, summing forecast error over the VFI, in a non-transitory medium of expression for initiating inventory replenishment; and updating the computed safety stock for each item in the inventory database. |
