US 11,657,347 B2
Systems and methods for optimization of pick walks
Seon Ki Kim, Dublin, CA (US); Aditya Arcot Srinivasan, Sunnyvale, CA (US); and Mingang Fu, Palo Alto, CA (US)
Assigned to WALMART APOLLO, LLC, Bentonville, AR (US)
Filed by Walmart Apollo, LLC, Bentonville, AR (US)
Filed on Jan. 31, 2020, as Appl. No. 16/778,355.
Prior Publication US 2021/0241197 A1, Aug. 5, 2021
Int. Cl. G06Q 10/06 (2012.01); G06Q 10/04 (2012.01); G06Q 10/0631 (2023.01); G06Q 10/0875 (2023.01); G06Q 30/0601 (2023.01); G06Q 10/0637 (2023.01); G06Q 10/10 (2023.01)
CPC G06Q 10/06316 (2013.01) [G06Q 10/04 (2013.01); G06Q 10/0637 (2013.01); G06Q 10/0875 (2013.01); G06Q 10/10 (2013.01); G06Q 30/0635 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method, initialized by a computer, for improving performance of a neighborhood search algorithm used for processing and displaying a plurality of items to be picked at a warehouse, the method comprising:
approximating, by a greedy tote reduction algorithm executed by the computer, an original solution, using shorter and less intensive data processes than a traditional ant colony optimization algorithm;
selecting, via the greedy tote reduction algorithm within an infeasible totes loop search, the plurality of items that have a highest volume and a highest weight within a maximum weight and a maximum volume restriction for each of a plurality of respective totes as determined by an exogenous system;
iteratively executing a large-scale neighborhood search algorithm to achieve local improvements that outperform the neighborhood search beyond a local minimum, by searching on a neighborhood of the original solution, a swap within combinations of the plurality of items swapped between the plurality of respective totes until a maximum number of iterations is reached, wherein each of the plurality of respective totes adhere to: the maximum weight, the maximum volume restriction, a maximum tote capacity of a trolley corresponding to the warehouse, or different temperatures requirements corresponding to packing each of the plurality of items;
executing, the computer, a randomized tote local search loop by iteratively searching for candidate solutions within a search space;
applying, by the computer, local changes among the candidate solutions until an optimal solution of picklists for the plurality of totes is found;
executing, by the computer, a minimum trolley loop algorithm on the optimal solution by combining a shortest picklist with at least one other picklist of a plurality of picklists, as swapped; and
displaying, to a picker on an interface of a computing device, turn-by-turn directions of the combined picklists within a pick walk overlaid on a map of the warehouse.