Inventory mirroring

A method of determining an inventory mirroring plan for a set of distinct items in a fulfillment network, the fulfillment network comprising a plurality of distribution centers, the method being implemented at least in part via execution of computer instructions configured to run at one or more processors and configured to be stored at one or more non-transitory memory storage systems, the method comprising:

• determining via the one or more processors, for each distinct item (i) of the set of distinct items and for each demand zone (z) of a set of demand zones, a location-specific demand (di,z), wherein the each distinct item (i) of the set of distinct items comprises high-velocity stock keeping units (“SKUs”) with at least a wide geo-demand spread to be stocked at the plurality of distribution centers within the fulfillment network, wherein the high-velocity SKUs comprises a SKU having a sales volume higher than most other SKUs carried by the fulfillment network, and wherein the wide geo-demand spread comprises the SKU having a geo-demand spread wider than most other SKUs carried by the fulfillment network;
• determining via the one or more processors, for each of a number of clusters (k) ranging from 1 to a predetermined maximum number of clusters (K), a k-cluster profile that partitions the plurality of distribution centers in the fulfillment network into k distribution center clusters;
• determining via the one or more processors, for each of the number of clusters (k) ranging from 1 to the predetermined maximum number of clusters (K) and for each demand zone (z) of the set of demand zones, a closest distribution center cluster of the k distribution center clusters that is nearest to a first demand zone (z) of the set of demand zones;
• determining via the one or more processors, for each of the number of clusters (k) ranging from 1 to the predetermined maximum number of clusters (K) and for each demand zone (z) of the set of demand zones, an average zone distance (l(k)(z)) from the first demand zone (z) of the set of demand zones to the closest distribution center cluster;
• determining, by solving an integer programming formulation, via the one or more processors a solution value of the number of clusters (k) for the each distinct item (i) of the set of distinct items that minimizes a sum of a total shipping cost (ci(k)) of the each distinct item (i) of the set of distinct items, subject to a total distinct item capacity (M) of the plurality of distribution centers in the fulfillment network, wherein:
• c
• i
• (
• k
• )
• =
• ∑
• z
• ⁢
• d
• i
• ,
• z
• ⁢
• c
• ⁡
• (
• l
• k
• ⁡
• (
• z
• )
• ,
• w
• i
• )
• ,
• c(lk(z), wi) is a unit shipping cost of the each distinct item (i) based on the average zone distance (l(k)(z)) for each demand zone (z) of the set of demand zones and a shipping weight (wi) of the each distinct item (i), wherein the integer programming formulation is solved using the one or more processors by:
• using a binary variable to determine whether to mirror the each distinct item (i) of the set of distinct items in the k distribution center clusters;
• determining a size (N) of the set of distinct items;
• capping a maximum number of slots available in the k distribution center clusters;
• generating a shipping cost resulting from stocking the each distinct item (i) of the set of distinct items at the k distribution center clusters;
• and
• applying the shipping cost to minimize the total shipping cost for the each distinct item (i) of the set of distinct items;
• and
• stocking the each distinct item (i) in a particular number of distribution centers of the plurality of distribution centers based at least in part on one or more of the solution values.