	US 12,014,298 B2
	Automatically scheduling and route planning for service providers
Abhishek Mishra, Bengaluru (IN); Sunil Kumar Potnuru, Bengaluru (IN); Nimish Kumar, Bangalore (IN); Paulami Chaudhuri, Bangalore (IN); Ashish Gupta, Bangalore (IN); Noyle Christopher, Bentonville, AR (US); Lauren Jean Shores, Springfield, IL (US); Rahul Verma, Bangalore (IN); Hema Vaishanav, Bengaluru (IN); Abhishek Ray Chaudhury, Bangalore (IN); and Himanshu Singh, Kanpur (IN)
Assigned to WALMART APOLLO, LLC, Bentonville, AR (US)
Filed by Walmart Apollo, LLC, Bentonville, AR (US)
Filed on Dec. 29, 2021, as Appl. No. 17/564,934.
Claims priority of provisional application 63/194,587, filed on May 28, 2021.
Prior Publication US 2022/0383230 A1, Dec. 1, 2022
Int. Cl. G06Q 10/06 (2023.01); G06Q 10/0631 (2023.01); G06Q 10/0633 (2023.01)

CPC G06Q 10/063112 (2013.01) [G06Q 10/063116 (2013.01); G06Q 10/0633 (2013.01)]

20 Claims

1. A system comprising:

one or more processors; and

one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, perform:

training a machine learning algorithm to determine an estimated duration of a new work order, based on (a) historical input data for the machine learning algorithm and (b) historical output data for the machine learning algorithm, wherein:

the historical input data comprise a respective trade or a respective problem code of each of one or more fulfilled work orders;

determining, in real-time, one or more work orders for a service provider based at least in part on a respective duration of each of the one or more work orders, comprising:

selecting, in real-time, each of the one or more work orders from unprocessed work orders stored in a database; and

determining, in real-time by the machine learning algorithm, as trained, the respective duration of the each of the one or more work orders;

determining, in real-time, an optimal service route for the one or more work orders by calculating a score for the optimal service route by a mathematical formula associated with the one or more work orders, wherein:

the mathematical formula is Σ_i=0ⁿΣ_{j=1, i≠j}ⁿ⁺¹(e_ij*w_ij), wherein:

n comprises a count of the one or more work orders in the optimal service route;

b_ijcomprises an indication of whether a j^thwork order (WO_j) of the one or more work orders is scheduled immediately after and adjacent to an i^thWO (WO_i) of the one or more work orders in the optimal service route, wherein:

when WO_jis scheduled immediately after and adjacent to WO_j, b_ijis equal to 1;

when WO_jis not scheduled immediately after or adjacent to WO_j, b_ijis equal to 0;

∀i∈{0, n};

∀j∈{1, n+1};

WO₀comprises a virtual work order for a starting location of the service provider; and

WO_n+1comprises a virtual work order for an ending location of the service provider; and

e_ijcomprises a priority score for b_ijdetermined based on e_ij=Σ_k=1^mw_jk*f_jk/s_ij+c, wherein:

m comprises a count of one or more weighted factors associated with WO_j;

W_jkcomprises a respective weight for a k^thfactor of the one or more weighted factors associated with WO_jand determined by a second machine learning model;

f_jkcomprises a respective value for the k^thfactor of the one or more weighted factors associated with WO_j;

s_ijcomprises a respective normalized drive time value between a respective service location of WO_iand the respective service location of WO_j, and

C comprises a constant number;

updating, in real-time, an available time slot in a work schedule of the service provider based on the optimal service route; and

transmitting, in real-time through a computer network, the work schedule updated with the optimal service route to be displayed on a user interface executed on a device of the service provider.