Universal task management system, method and product for automatically managing remote workers, including automatically training the workers

• US 6,938,048 B1
• Filed: 11/14/2001
• Issued: 08/30/2005
• Est. Priority Date: 11/14/2001
• Status: Active Grant

• Current Assignee: Elance Incorporation
• Original Assignee: QGENISYS, INC.
• Inventors: Kellner, Brian L., Ganora, Victor, Kotsines, Thomas A., Jilk, David J.
• Primary Examiner(s): Ali, Mohammad
• Application Number:
  US10/001,947
• Time in Patent Office: 1,385 Days
• Field of Search: 707 1- 10, 707100-1041, 707200-205, 705/11, 705 7- 9, 379/265.01
• US Class Current: N/A
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/063112 Skill-based matching of a person or a group to a task
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A computer implemented method, a system, and a software product to automatically manage one or more human workers carrying out a process of manipulating source data provided to produce result data. The process including a set of one or more task steps each having an input and resulting in a task result. The computer implemented method includes receiving units of source data from a customer. For each unit of source data and each task step for the unit of source data, the method includes dispatching the task step and its corresponding input unit to a worker and, after the worker carries out the dispatched task step on the input unit, receiving the task result corresponding to the dispatched task step and input unit from the worker. Each worker is certified to have one or more task skills, wherein each task step requires a corresponding task skill, and wherein the dispatching of any task step occurs automatically substantially without human management to a worker who is certified to have the corresponding task skill of the task step. The method further includes automatically training the workers at one or more task skills with no human management. A version of the method further includes generating the result data for the unit of source data from one or more of the task results corresponding to the task steps of the set and sending the result data for the unit of source data to the customer.

Claims

1. A system for automatically managing a plurality of remote workers carrying out a variety of jobs for one or more customers, each job including a process of a set of one or more task steps and a set of associated source data units, the system connected to a network, each worker able to communicate with the system using a worker terminal connectable to network, the system comprising:

a storage subsystem containing a database storing information on each remote worker including one or more task skills of the worker that define the types of task steps the worker is certified to carry out, information on one or more customers, and information on each process, the process information including the customer of the process, the order of carrying out the task steps of the process, how the input for each task step is obtained from the results of prior task steps in the process, and any pre-processing and post-processing required;

a task data structure to store tasks to be completed, each task defined by a task step and a unit of input for the task step; and

a data store for storing input and output information for the tasks;

a mechanism coupled to the storage subsystem to accept units of source data from the customers;

a pre-processor coupled to the storage subsystem to carry out any defined pre-processing for the accepted source data;

a mechanism coupled to the network to accept requests from one or more of the remote workers for tasks;

a task dispatcher coupled to the storage subsystem and to the network for dispatching a task from the task data structure to a remote worker requesting tasks, the dispatching according to one or more task dispatch rules, wherein the dispatch rules define one or more task skills which must be in the database information for the remote worker prior to the dispatching of the task to the remote worker;

a task submission unit coupled to the storage subsystem to receive the task results from the remote workers for the task dispatched to the workers;

a post-processor coupled to the storage subsystem to carry out any defined post-processing of the task results corresponding to the tasks of a process for a unit of source data to produce result data for the unit of source data;

a mechanism coupled to the storage subsystem to send the result data to the customers;

a capacity manager coupled to the storage subsystem to manage the capacity of the system based on task load information on the tasks in the task data structure, on the available workers, and on the available worker task skills; and

a training unit coupled to the network and to the storage subsystem to automatically train workers at one or more task skills according to related training scenarios selected based on the one or more task skills defined by the dispatch rules, such that the training unit trains workers automatically substantially without human management.

2. A system as recited in claim 1, wherein the variety of jobs include a plurality of members of the set consisting of:

data entry, telesales, voice transcription, translation, image categorization, sales lead incubation, auditing, repair of documents after OCR, photo retouching, paralegal processes, call center quality assurance, and editorial work.

3. A carrier medium carrying computer readable code segments to instruct one or more processors of a processing system to carry out a method of automatically managing a plurality of remote workers carrying out a variety of jobs for one or more customers, each job including a process of a set of one or more task steps and a set of units of source data, the medium comprising:

one or more code segments to instruct the one or more processors to store in a database information on each remote worker including one or more task skills of the worker that define the types of task steps the worker is certified to carry out;

one or more code segments to instruct the one or more processors to store in the database information on each process, including the customer of the process, the order of carrying out the task steps of the process, how the input for each task step is obtained from the results of prior task steps in the process, and any pre-processing and post-processing required;

one or more code segments to instruct the one or more processors to accept units of source data from the customers;

one or more code segments to instruct the one or more processors to store in a task data structure information on tasks to be completed, each task defined by a task step and a unit of input for the task step;

one or more code segments to instruct the one or more processors to dispatch, upon receiving a task request from a remote worker, a task from the stored tasks to be completed to the remote worker according to one or more task dispatch rules, wherein the dispatch rules define one or more task skills required of the remote worker to receive the dispatched task;

one or more code segments to instruct the one or more processors to accept task results from the remote workers for the tasks dispatched to the workers;

one or more code segments to instinct the one or more processors to carry out any defined post-processing of the task results corresponding to the tasks of a process for a unit of source data to produce result data for the unit of source data;

one or more code segments to instruct the one or more processors to manage the capacity of the system based on information about the stored tasks;

one or more code segments to instruct the one or more processors to send the result data to the customers; and

one or more code segments to instruct the one or more processors to automatically train workers at one or more task skills according to one or more related training scenarios selected based on the one or more task skills defined in the dispatch rules and associated with the dispatched task, such that the training of the workers occurs substantially without human management.