Systems and methods for determining pack allocations

A computer-implemented system, comprising:

• one or more data processors;
• and
• one or more non transitory computer-readable storage media containing instructions operable to cause the one or more processors to perform operations including:
• receiving demand profile information corresponding to each store in a group of stores, wherein the demand profile information indicates anticipated demand for each of multiple varieties of a product at each store;
• receiving mismatch cost data for each store in the group, wherein the mismatch cost data includes:
• a set of overdelivery costs for each store, each set of overdelivery costs including an overdelivery cost for each variety of the product, wherein each overdelivery cost is a marginal cost incurred when more than a demanded amount of the respective variety is shipped to the respective store;
• and
• a set of underdelivery costs for each store, each set of underdelivery costs including an underdelivery cost for each variety of the product, wherein each underdelivery cost is a marginal cost incurred when less than a demanded amount of the respective variety is shipped to the respective store;
• receiving one or more constraints;
• with respect to each store in the group, defining an initial pack configuration, wherein defining includes using the demand profile information, the mismatch cost data, and the one or more constraints, and, wherein the pack configuration facilitates lowest total mismatch costs for a corresponding store;
• planning shipments of the product to the stores, wherein planning includes selecting at least one, but less than all of the initial pack configurations, wherein selecting the at least one of the initial pack configurations includes using a pack elimination process that includes:
• accessing information representing a constraint that imposes a maximum number of distinct pack configurations with respect to each of the shipments;
• forming multiple initial clusters of the stores by grouping each of the stores with at least one other of the stores based on similarities between characteristics of the stores;
• with respect to each of the initial clusters of the stores:
• selecting a representative pack configuration from amongst the initial pack configurations defined with respect to the stores of the initial cluster;
• grouping at least two of the initial clusters based on similarities between the representative pack configurations selected with respect to the initial clusters;
• eliminating at least one of the representative pack configurations selected with respect to the grouped initial clusters;
• and
• determining that the representative pack configurations not eliminated by the pack elimination process amount to not more than the maximum number of distinct pack configurations;
• and
• specifying that the planned shipments be made using the representative packs not eliminated by the pack elimination process.

A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, the computer-program product including instructions operable to cause a data processing apparatus to perform operations including:

• receiving demand profile information corresponding to each store in a group of stores, wherein the demand profile information indicates anticipated demand for each of multiple varieties of a product at each store;
• receiving mismatch cost data for each store in the group, wherein the mismatch cost data includes:
• a set of overdelivery costs for each store, each set of overdelivery costs including an overdelivery cost for each variety of the product, wherein each overdelivery cost is a marginal cost incurred when more than a demanded amount of the respective variety is shipped to the respective store;
• and
• a set of underdelivery costs for each store, each set of underdelivery costs including an underdelivery cost for each variety of the product, wherein each underdelivery cost is a marginal cost incurred when less than a demanded amount of the respective variety is shipped to the respective store;
• receiving one or more constraints;
• with respect to each store in the group, defining an initial pack configuration, wherein defining includes using the demand profile information, the mismatch cost data, and the one or more constraints, and, wherein the pack configuration facilitates lowest total mismatch costs for a corresponding store;
• planning shipments of the product to the stores, wherein planning includes selecting at least one, but less than all of the initial pack configurations, wherein selecting the at least one of the initial pack configurations includes using a pack elimination process that includes:
• accessing information representing a constraint that imposes a maximum number of distinct pack configurations with respect to each of the shipments;
• forming multiple initial clusters of the stores by grouping each of the stores with at least one other of the stores based on similarities between characteristics of the stores;
• with respect to each of the initial clusters of the stores:
• selecting a representative pack configuration from amongst the initial pack configurations defined with respect to the stores of the initial cluster;
• grouping at least two of the initial clusters based on similarities between the representative pack configurations selected with respect to the initial clusters;
• eliminating at least one of the representative pack configurations selected with respect to the grouped initial clusters;
• and
• determining that the representative pack configurations not eliminated by the pack elimination process amount to not more than the maximum number of distinct pack configurations;
• and
• specifying that the planned shipments be made using the representative pack configurations not eliminated by the pack elimination process.