System for analyzing workflow and detecting inactive operators and methods of using the same

Abstract

Provided is an improved apparatus and method for identifying inefficient workers in a workplace. The apparatus receives voice commands from a worker in the workplace and transforms the voice commands into text commands by converting audio signals associated with the voice commands to one or more points in data that are stored in a database. The apparatus further creates a dialogue-stream including the text commands and a time of receipt of each voice command associated with the respective text command. Further, a workflow record may be created from the dialogue-stream and may be used to generate an efficiency metric for the worker. Both the dialogue-stream and workflow record can be used to monitor the efficiency of the worker and implement effective changes to the workflow process.

We claim: 
     
         1 . A voice controlled apparatus, comprising:
 a processor configured to:
 transform a plurality of voice commands into a plurality of text commands; 
 create a first dialogue-stream based on the plurality of text commands, wherein the first dialogue-stream includes one or more tags that are associated with the plurality of text commands of the first dialogue-stream, and wherein each of the one or more tags of the first dialogue-stream indicates that an associated text command of the first dialogue-stream corresponds to a user-initiated command; 
 recognize the one or more tags in the first dialogue-stream; 
 determine items in the first dialogue-stream that are not associated with the one or more recognized tags; 
 remove the items in the first dialogue-stream; and 
 create a second dialogue-stream based on the removal of the items. 
   
     
     
         2 . The apparatus of  claim 1 , wherein the first dialogue-stream includes the plurality of text commands and a time of receipt of each voice command of the plurality of voice commands associated with the respective text command and is gathered into a single collection organized by the time of receipt. 
     
     
         3 . The apparatus of  claim 1 , wherein the processor is further configured to receive a plurality of manually-inputted button commands, incorporate the plurality of button commands into the plurality of text commands by identifying one or more points in data stored in a database related to each of the plurality of button commands, and incorporate the plurality of text commands from the plurality of button commands into the second dialogue-stream. 
     
     
         4 . The apparatus of  claim 1 , wherein the processor is further configured to provide speech output by converting one or more points in data to one or more audio signals to be converted to the speech output via a speaker. 
     
     
         5 . The apparatus of  claim 1 , wherein the apparatus further comprises a microphone and a speaker. 
     
     
         6 . The apparatus of  claim 1 , wherein the second dialogue-stream is transmitted to a supervisor device. 
     
     
         7 . The apparatus of  claim 1 , wherein the second dialogue-stream is further used to prepare one or more alarms associated with a maximum duration of a particular task of a workflow record that is created from the second dialogue-stream. 
     
     
         8 . The apparatus of  claim 1 , wherein the second dialogue-stream is further used to determine a rate of interaction of a user with the apparatus. 
     
     
         9 . The apparatus of  claim 1 , wherein the second dialogue-stream is further used to identify periods of non-activity of a user with the apparatus. 
     
     
         10 . A method comprising:
 transforming a plurality of voice commands into a plurality of text commands;   creating a first dialogue-stream based on the plurality of text commands, wherein the first dialogue-stream includes one or more tags that are associated with the plurality of text commands of the first dialogue-stream, and wherein each of the one or more tags of the first dialogue-stream indicates that an associated text command of the first dialogue-stream corresponds to a user-initiated command;   recognizing the one or more tags in the first dialogue-stream;   determining items in the first dialogue-stream that are not associated with the one or more recognized tags;   removing the items in the first dialogue-stream; and   creating a second dialogue-stream based on the removal of the items.   
     
     
         11 . The method of  claim 10 , further comprising gathering the plurality of text commands and a time of receipt of each voice command of the plurality of voice commands associated with the respective text command into a single collection organized by the time of receipt. 
     
     
         12 . The method of  claim 10 , wherein the method further comprises:
 determining a rate of interaction of a user by calculating a number of text commands over a period of time in the second dialogue-stream.   
     
     
         13 . The method of  claim 10 , further comprising creating a workflow record for a user associated with the first dialogue-stream and the second dialogue-stream by:
 identifying at least one task associated with any one of the plurality of voice commands from the user based on the text command associated with the respective voice command; and   determining a duration of the at least one task based on a time associated with each text command, wherein the workflow record is used to generate an efficiency metric.   
     
     
         14 . The method of  claim 13 , wherein identifying the at least one task associated with any one of the plurality of voice commands from the user based on the text command associated with the respective voice command comprises:
 determining a current state of a user device associated with the user;   determining a next state of the user device associated with the user;   identifying a trigger to move to the next state of the user device; and   adding the at least one task to the workflow record after identifying the trigger,   wherein the current state is determined by the earliest received voice command and associated text command or a voice command received from the user device that has not been associated with at least one task, the next state is determined by identifying a task intended to follow the current state from a database, and the trigger comprises a dialogue of interest related to the next state.   
     
     
         15 . The method of  claim 13 , wherein determining the duration of the at least one task comprises identifying a text command or a voice command associated with a start of the at least one task in a database, identifying a text command or a voice command associated with an end of the at least one task in the database and determining a time between the start of the at least one task and the end of the at least one task. 
     
     
         16 . The method of  claim 13 , wherein the method further comprises:
 determining an average duration of a plurality of tasks associated with the plurality of voice commands by averaging the duration of each task over a stastically significant number of repetitions of the task.   
     
     
         17 . The method of  claim 13 , wherein the method further comprises:
 creating at least one alarm associated with the at least one task, wherein the alarm is configured to notify a supervisor device that the user has exceeded a maximum duration of the at least one task by causing an indication to be transmitted to the supervisor device when the duration of the at least one task exceeds the maximum duration of the at least one task in the workflow record.   
     
     
         18 . The method of  claim 13 , wherein the method further comprises:
 causing the workflow record for the particular user to be transmitted to a supervisor device, wherein the supervisor device is controlled by a supervisor associated with the user.   
     
     
         19 . The method of  claim 13 , wherein the method further comprises:
 receiving a plurality of dialogue-streams from a plurality of user devices, wherein each dialogue-stream comprises a plurality of text commands and a time associated with each text command, the plurality of text commands formed by transforming a plurality of voice commands into the plurality of text commands by converting audio signals associated with the plurality of voice commands to one or more points in data that are stored in the database or another database;   creating a plurality of workflow records for the plurality of users by identifying at least one task for at least one voice command in each of the plurality of workflow records and determining a duration of each task of each workflow record of the plurality of workflow records; and   comparing the duration of each task of each workflow record of the plurality of workflow records.   
     
     
         20 . The method of  claim 19 , wherein the method further comprises:
 filtering the plurality of workflow records for a single task; and   determining at least one of an average duration of the single task over the plurality of workflow records, a minimum duration of the single task over the plurality of workflow records, or a maximum duration of the single task over the plurality of workflow records.   
     
     
         21 . The method of  claim 20 , wherein the method further comprises:
 creating at least one alarm associated with the single task, wherein the alarm is configured to notify a supervisor device that at least one user has exceeded a set duration of the single task by causing an indication to be transmitted to the supervisor device when the duration of the single task exceeds the set duration in the workflow record of the at least one user.   
     
     
         22 . The method of  claim 21 , wherein the method further comprises causing the at least one alarm to be transmitted to the user device. 
     
     
         23 . The method of  claim 21 , wherein the set duration is determined by calculating an average duration of the single task as performed over a stastically significant number of repetitions of the single task. 
     
     
         24 . The method of  claim 13 , wherein the method further comprises identifying a period of non-activity of the user by identifying a period of time where a rate of interaction decreases to a defined value and causing an indication to be transmitted to a supervisor device notifying of the occurrence of the period of non-activity. 
     
     
         25 . The method of  claim 13 , further comprising initiating an action to improve the workflow record and incorporating at least one note to the workflow record recording the initiated action. 
     
     
         26 . The method of  claim 25 , wherein the action to improve the workflow record comprises causing an indication to be transmitted to a user device operated by the user, the indication indicating that the user is inactive. 
     
     
         27 . The method of  claim 25 , wherein the action to improve the workflow record comprises requesting additional materials, workers, or a combination thereof to perform a task by causing an indication to be transmitted to a server, the indication indicating that additional materials, workers, or a combination thereof are needed to perform the task. 
     
     
         28 . The method of  claim 13 , wherein the workflow record comprises a plurality of tasks associated with the plurality of text commands and each of the plurality of tasks is associated with a duration of the respective task.