US 12,254,429 B2
Generating status of construction site based on hierarchical modeling that standardizes physical relationships of elements of a structure
Arunava Saha, Dublin, CA (US); Dobromir Voyager Montauk, Redwood City, CA (US); Richard William Turner, Ann Arbor, MI (US); and Matthew Paul Orban, San Bruno, CA (US)
Assigned to Doxel, Inc., Palo Alto, CA (US)
Filed by DOXEL, INC., Palo Alto, CA (US)
Filed on May 31, 2022, as Appl. No. 17/804,758.
Claims priority of provisional application 63/194,873, filed on May 28, 2021.
Prior Publication US 2022/0383231 A1, Dec. 1, 2022
Int. Cl. G06Q 10/06 (2023.01); G06Q 10/0631 (2023.01); G06Q 30/0283 (2023.01); G06Q 50/08 (2012.01)
CPC G06Q 10/063114 (2013.01) [G06Q 30/0283 (2013.01); G06Q 50/08 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method of evaluating progress of a structure undergoing construction in near-real time, the method comprising:
receiving building information modeling (BIM) data in a non-standardized format for a set of structures undergoing construction,
wherein the BIM data for each structure includes digital drawings of elements of the structure;
receiving scheduling data associated with construction of each structure in the set of structures,
wherein the scheduling data for each structure includes an expected progress of construction activities at different points in time;
accessing a database storing construction data that associates multiple elements of construction projects and associated construction activities in a hierarchical configuration,
wherein the hierarchical configuration defines branches that each have one or more types of elements and construction activities including a group and a system, a subsystem, a component, or a subcomponent at respective levels, and
wherein the hierarchical configuration defines unidirectional paths between the levels of the branches;
generating, based on the BIM data and the scheduling data, a machine learning (ML) model that standardizes how particular elements of a particular structure and construction activities for the particular structure relate to the multiple elements and the associated construction activities in the hierarchical configuration,
wherein the ML model is trained based on a training dataset including multiple mappings of elements of structures other than the particular structure or construction activities for the structures to the multiple elements or the associated construction activities in the hierarchical configuration;
receiving updated construction data regarding an ongoing construction of the particular structure,
wherein the updated construction data indicates current states of the particular elements and construction activities for the particular structure, and
wherein the updated construction data regarding the ongoing construction of the particular structure is in the non-standardized format;
using the ML model, trained based on the training dataset including multiple mappings of elements of structures other than the particular structure or construction activities for the structures to the multiple elements or the associated construction activities in the hierarchical configuration, generate a mapping between the updated construction data regarding the ongoing construction of the particular structure and the hierarchical configuration,
wherein the construction data indicates whether the particular elements are installed, and
wherein the hierarchical configuration is standardized;
based on the mapping generated by the ML model, determining a status of the construction of the particular structure,
wherein the status of the construction includes an estimate of completion of the particular structure;
obtaining the mapping between the updated construction data regarding the ongoing construction of the particular structure and the hierarchical configuration including an incorrect mapping; and
retraining the ML model using the mapping between the updated construction data regarding the ongoing construction of the particular structure and the hierarchical configuration.