US 11,681,963 B2
Method and system for optimization of task management issue planning
Salwa Husam Alamir, Bournemouth (GB); Alberto Pozanco, Madrid (ES); Sameena Shah, Scarsdale, NY (US); Daniele Magazzeni, London (GB); Daniel Borrajo, Pozuelo de Alarcon (ES); Parisa Zehtabi, London (GB); Rui Manuel Ramos Teixeira da Silva, London (GB); and Maria Manuela Veloso, New York, NY (US)
Assigned to JPMORGAN CHASE BANK, N.A., New York, NY (US)
Filed by JPMorgan Chase Bank, N.A., New York, NY (US)
Filed on Apr. 5, 2021, as Appl. No. 17/222,424.
Prior Publication US 2022/0318712 A1, Oct. 6, 2022
Int. Cl. G06Q 10/0631 (2023.01); G06Q 10/10 (2023.01)
CPC G06Q 10/06316 (2013.01) [G06Q 10/06312 (2013.01); G06Q 10/103 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A method for optimizing personnel utilization, the method being implemented by at least one processor, the method comprising:
identifying and acquiring, using a machine learning algorithm executed by the at least one processor, a skill set of a plurality of persons to assemble a project team for a first project that has not been completed;
identifying a plurality of previous tasks performed in a plurality of previous projects;
clustering, via the machine learning algorithm executed by the at least one processor, the plurality of previous tasks into a set of task classes, the clustering performed by creating a vector representation of the plurality of previous tasks based on a Term-Frequency-Inverse Document Frequency (TD-IDF) operation performed on technical details corresponding to each of the plurality of previous tasks;
extracting task clusters, from the vector representations of the plurality of previous tasks and using the machine learning algorithm, via Latent Dirichlet Allocation (LDA) topic modelling, and wherein the LDA depends on a target parameter that specifies a number of clusters, and wherein a parameter value of each project that leads to highest cross-validated coherence score is set as the target parameter;
training the LDA model via the plurality of previous projects for clustering of a new task;
accessing, by the at least one processor, first task management planning information that relates to the first project that has not been completed, wherein the first task management planning information includes:
identification of a plurality of tasks to be performed in the first project,
determination of planned order for execution of the plurality of tasks to be performed,
determination of assignments of the plurality of persons in the project team to the plurality of tasks based on the planned order for execution, and
determination of time allocations for the plurality of tasks to be performed by the plurality of persons in the project team for performing the plurality of tasks;
retrieving, by the at least one processor and from a server via a network, historical task management information that relates to at least one project that has been completed;
first adjusting, by the at least one processor and via the machine learning algorithm, at least a first portion of the first task management planning information based on the retrieved historical task management information;
tracking execution status of the first project by collecting data that relates to actual order in which the plurality of tasks are performed, actual persons that have performed the plurality of tasks, and actual amount of time that have performed on the plurality of tasks;
tracking, via a network environment, availability information of the plurality of persons assigned to the plurality of tasks of the first project;
dynamically second adjusting, by the at least one processor and via the machine learning algorithm, at least a second portion of the first task management planning information based on the tracking of the execution status and the availability information; and
updating the machine learning algorithm based on the second adjusting via a continuous unsupervised machine learning for outputting a more accurate project plan for subsequent projects.