Balancing multiple computer models in a call center routing system

Abstract

Systems and methods are disclosed for routing callers to agents in a contact center utilizing a multi-layer processing approach to matching a caller to an agent. A first layer of processing may include two or more different computer models or methods for scoring or determining caller-agent pairs in a routing center. The output of the first layer may be received by a second layer of processing for balancing or weighting the outputs and selecting a final caller-agent match. The two or more methods may include conventional queue based routing, performance based routing, pattern matching algorithms, affinity matching, and the like. The output or scores of the two or more methods may be processed be the second layer of processing to select a caller-agent pair and cause the caller to be routed to a particular agent.

I claim: 
     
       1. A method for routing callers to agents in a call-center routing environment, the method comprising the acts of:
 receiving, by one or more computers, input data associated with callers in a set of callers and agents in a set of agents at a first layer of processing; 
 processing, by the one or more computers communicatively coupled to and configured to operate in the call-center routing environment, the input data associated with the callers in the a set of callers and the agents in the a set of agents using a first measurement algorithm in a pair-wise fashion in the a first layer of processing, to output respective first measurement data for each caller-agent pair of a first set of caller-agent pairs; 
 processing, by the one or more computers, the input data associated with the callers in the set of callers and the agents in the set of agents using a second measurement algorithm in a pair-wise fashion in the first layer of processing, to output respective second measurement data for each caller-agent pair of the first set of caller-agent pairs; 
 receiving, by the one or more computers, the output measurement data from each of the first measurement algorithm and the second measurement algorithm at a second layer of processing, the first measurement data and the second measurement data for each caller-agent pair of the set of caller-agent pairs; 
 processing, by the one or more computers, the caller-agent pair output measurement data from each of the first measurement algorithm and the second measurement algorithm by an first measurement data and the second measurement data for each caller-agent pair of the set of caller-agent pairs by a weighting algorithm to weight the caller-agent pair output first measurement data of one of the measurement algorithms for each caller-agent pair of the set of caller-agent pairs relative to the other of the measurement algorithms second measurement data for each caller-agent pair of the set of caller-agent pairs in the second layer of processing, to output a weighted caller-agent pair output measurement data; 
 selecting, by the one or more computers, a caller-agent pair based at least in part on the weighted caller-agent pair output measurement data; and 
 mapping, by the one or more computers, the caller in the caller-agent pair selected to the agent in the caller-agent pair selected 
 establishing, by the one or more computers, in a switch module of the call-center routing environment, a communication channel between the caller and the agent in the selected agent-pair to optimize performance of the call-center routing environment attributable to the weighted caller-agent pair output measurement data. 
 
     
     
       2. The method of  claim 1 ,
 wherein the processing the input data steps comprises scoring each of the caller-agent pairs using the respective algorithm for the respective processing step,   wherein the output measurement data from the first measurement algorithm comprises a respective score for each of the first set of caller-agent pairs and is based on at least a first data associated with one or both of the set of the callers and the set of the agents; and   wherein the output measurement data from the second measurement algorithm comprises a respective score,   wherein the scores from the first and second measurement algorithms for each of the caller-agent pairs are output to the second layer of processing.   
     
     
       3. The method of  claim 2 , wherein scoring the caller-agent pairs according to the second measurement algorithm is based on at least a second data associated with one or both of the set of callers and the set of agents. 
     
     
       4. The method of  claim 2 , wherein scoring the caller-agent pairs comprises ranking agents based on performance. 
     
     
       5. The method of  claim 2 , wherein scoring the caller-agent pairs comprises determining a suitability score for a desired output performance. 
     
     
       6. The method of  claim 2 , wherein one of the measurement algorithms comprises a multi-data element pattern matching algorithm utilizing caller data associated with multiple callers in the set of callers and agent data associated with multiple agents in the set of agents. 
     
     
       7. The method of  claim 1 , wherein one of the measurement algorithms comprises a performance based matching algorithm based on performance data of at least two of the agents. 
     
     
       8. The method of  claim 1 , wherein one of the measurement algorithms comprises a multi-data element pattern matching algorithm utilizing caller data associated with multiple callers from the set of callers and agent data associated with multiple agents from the set of agents. 
     
     
       9. The method of  claim 1 , wherein one of the measurement algorithms utilizes affinity data associated with one or both of the agents and callers. 
     
     
       10. The method of  claim 1 , wherein at least one of the at least two measurement algorithms comprises a neural network algorithm. 
     
     
       11. The method of  claim 1 , wherein the second layer of processing comprises a neural network algorithm. 
     
     
       12. The method of  claim 1 , further comprising:
 providing an electronic interface, by the one or more computers, to change the weighting of the output measurement data from one of the measurement algorithms relative to the other of the measurement algorithms in the second layer of processing; and   generating, by the one or more computers, display data for a display interface, with the display data comprising an estimated effect of the change of the weighting on one or more selected from the group of cost, revenue generation, and customer satisfaction.   
     
     
       13. The method of  claim 1 , wherein the caller-agent pair output measurement data from the first measurement algorithm and from the second measurement algorithm are weighted equally. 
     
     
       14. The method of  claim 1 , wherein the caller-agent pair output measurement data from the first measurement algorithm and from the second measurement algorithm are weighted unequally. 
     
     
       15. A system for routing callers to agents in a call-center routing environment, comprising:
 one or more computers configured with computer-readable program code to perform, when executed, the steps:   receiving, by the one or more computers, input data associated with callers in a set of callers and agents in a set of agents at a first layer of processing;   processing, by the one or more computers, the input data associated with the callers in the set of callers and the agents in the set of agents using a first measurement algorithm in a pair-wise fashion in the first layer of processing, to output respective measurement data for each of a first set of caller-agent pairs;   processing, by the one or more computers, the input data associated with the callers in the set of callers and the agents in the set of agents using a second measurement algorithm in a pair-wise fashion in the first layer of processing, to output respective measurement data for each of the first set of caller-agent pairs;   receiving, by the one or more computers, the output measurement data from each of the first measurement algorithm and the second measurement algorithm at a second layer of processing;   processing, by the one or more computers, the caller-agent pair output measurement data from each of the first measurement algorithm and the second measurement algorithm by an algorithm to weight the caller-agent pair output measurement data of one of the measurement algorithms relative to the other of the measurement algorithms in the second layer of processing;   selecting, by the one or more computers, a caller-agent pair based at least in part on weighted caller-agent pair output data; and   mapping, by the one or more computers, the caller in the caller-agent pair selected to the agent in the caller-agent pair selected.   
     
     
       16. The system of  claim 15 ,
 wherein the program code for processing the input data steps comprises program code for scoring each of the caller-agent pairs using the respective algorithm for the respective processing step,   wherein the output measurement data from the first measurement algorithm comprises a respective score for each of the first set of caller-agent pairs and is based on at least a first data associated with one or both of the set of the callers and the set of the agents; and   wherein the output measurement data from the second measurement algorithm comprises a respective score for each of the first set of caller-agent pairs,   wherein the scores from the first and second measurement algorithms are output to the second layer of processing.   
     
     
       17. The system of  claim 16 , wherein scoring the caller-agent pairs according to the second measurement algorithm is based on at least a second data associated with one or both of the set of callers and the set of agents. 
     
     
       18. The system of  claim 16 , wherein scoring the caller-agent pairs comprises ranking agents based on performance. 
     
     
       19. The system of  claim 16 , wherein scoring the caller-agent pairs comprises determining a suitability score for a desired output performance. 
     
     
       20. The system of  claim 16 , wherein one of the measurement algorithms comprises a multi-data element pattern matching algorithm utilizing caller data associated with multiple callers in the set of callers and agent data associated with multiple agents in the set of agents. 
     
     
       21. The system of  claim 15 , wherein one of the measurement algorithms comprises a performance based matching algorithm based on performance data of at least two of the agents. 
     
     
       22. The system of  claim 15 , wherein one of the measurement algorithms comprises a pattern matching algorithm utilizing caller data associated with multiple callers from the set of callers and agent data associated with multiple agents from the set of agents. 
     
     
       23. The system of  claim 15 , wherein one of the measurement algorithms utilizes affinity data associated with one or both of the agents and callers. 
     
     
       24. The system of  claim 15 , wherein at least one of the at least two measurement algorithms comprises a neural network algorithm. 
     
     
       25. The system of  claim 15 , wherein the second layer of processing comprises a neural network algorithm. 
     
     
       26. The system of  claim 15 , further comprising the one or more computers configured with program code to perform the steps:
 providing an electronic interface, by the one or more computers, to change the weighting of the output measurement data from one of the measurement algorithms relative to the other of the measurement algorithms in the second layer of processing; and   generating, by the one or more computers, display data for a display interface, with the display data comprising an estimated effect of the change of the weighting on one or more selected from the group of cost, revenue generation, and customer satisfaction.   
     
     
       27. The system of  claim 15 , wherein the caller-agent pair output measurement data from the first measurement algorithm and from the second measurement algorithm are weighted equally. 
     
     
       28. The system of  claim 15 , wherein the caller-agent pair output measurement data from the first measurement algorithm and from the second measurement algorithm are weighted unequally. 
     
     
       29. A non-transitory computer readable storage medium comprising computer readable program code for carrying out, when executed by one or more computers, the steps:
 receiving, by the one or more computers, input data associated with callers in a set of callers and agents in a set of agents at a first layer of processing;   processing, by the one or more computers, the input data associated with the callers in the set of callers and the agents in the set of agents using a first measurement algorithm in a pair-wise fashion in the first layer of processing, to output respective measurement data for each of a first set of caller-agent pairs;   processing, by the one or more computers, the input data associated with the callers in the set of callers and the agents in the set of agents using a second measurement algorithm in a pair-wise fashion in the first layer of processing, to output respective measurement data for each of the first set of caller-agent pairs;   receiving, by the one or more computers, the output measurement data from each of the first measurement algorithm and the second measurement algorithm at a second layer of processing;   processing, by the one or more computers, the caller-agent pair output measurement data from each of the first measurement algorithm and the second measurement algorithm by an algorithm to weight the caller-agent pair output measurement data of one of the measurement algorithms relative to the other of the measurement algorithms in the second layer of processing;   selecting, by the one or more computers, a caller-agent pair based at least in part on weighted caller-agent pair output data; and   mapping, by the one or more computers, the caller in the caller-agent pair selected to the agent in the caller-agent pair selected.   
     
     
       30. The computer readable storage medium of  claim 29 ,
 wherein the program code for processing the input data steps comprises program code for scoring each of the caller-agent pairs using the respective algorithm for the respective processing step,   wherein the output measurement data from the first measurement algorithm comprises a respective score for each of the first set of caller-agent pairs and is based on at least a first data associated with one or both of the set of the callers and the set of the agents; and   wherein the output measurement data from the second measurement algorithm comprises a respective score for each of the first set of caller-agent pairs,   wherein the scores from the first and second measurement algorithms are output to the second layer of processing.   
     
     
       31. The computer readable storage medium of  claim 30 , wherein scoring the caller-agent pairs according to the second measurement algorithm is based on at least a second data associated with one or both of the set of callers and the set of agents. 
     
     
       32. The computer readable storage medium of  claim 30 , wherein scoring the caller-agent pairs comprises ranking agents based on performance. 
     
     
       33. The computer readable storage medium of  claim 30 , wherein scoring the caller-agent pairs comprises determining a suitability score for a desired output performance. 
     
     
       34. The computer readable storage medium of  claim 30 , wherein one of the measurement algorithms comprises a multi-data element pattern matching algorithm utilizing caller data associated with multiple callers in the set of callers and agent data associated with multiple agents in the set of agents. 
     
     
       35. The computer readable storage medium of  claim 29 , wherein one of the measurement algorithms comprises a performance based matching algorithm based on performance data of at least two of the agents. 
     
     
       36. The computer readable storage medium of  claim 29 , wherein one of the measurement algorithms comprises a pattern matching algorithm utilizing caller data associated with multiple callers from the set of callers and agent data associated with multiple agents from the set of agents. 
     
     
       37. The computer readable storage medium of  claim 29 , wherein one of the measurement algorithms utilizes affinity data associated with one or both of the agents and callers. 
     
     
       38. The computer readable storage medium of  claim 29 , wherein at least one of the at least two measurement algorithms comprises a neural network algorithm. 
     
     
       39. The computer readable storage medium of  claim 29 , wherein the second layer of processing comprises a neural network algorithm. 
     
     
       40. The computer readable storage medium of  claim 29 , further comprising program code to perform the steps:
 providing an electronic interface, by the one or more computers, to change the weighting of the output measurement data from one of the measurement algorithms relative to the other of the measurement algorithms in the second layer of processing; and   generating, by the one or more computers, display data for a display interface, with the display data comprising an estimated effect of the change of the weighting on one or more selected from the group of cost, revenue generation, and customer satisfaction.   
     
     
       41. The computer readable storage medium of  claim 29 , wherein the caller-agent pair output measurement data from the first measurement algorithm and from the second measurement algorithm are weighted equally. 
     
     
       42. The computer readable storage medium of  claim 29 , wherein the caller-agent pair output measurement data from the first measurement algorithm and from the second measurement algorithm are weighted unequally. 
     
     
       43. A method for handling contacts and agents in a contact center system comprising:
 determining, by at least one computer processor communicatively coupled to and configured to operate in the contact center system, a first ability of a first agent to influence outcomes of contact-agent interactions, wherein the first ability of the first agent is determined based at least in part upon:
 processing, by the at least one computer processor, input data associated with contacts in a set of contacts and agents in a set of agents using a first measurement algorithm in a pair-wise fashion in a first layer of processing, to output first measurement data for each contact-agent pair of a set of contact-agent pairs; 
 processing, by the at least one computer processor, the input data associated with the contacts in the set of contacts and the agents in the set of agents using a second measurement algorithm in a pair-wise fashion in the first layer of processing, to output second measurement data for each contact-agent pair of the set of contact-agent pairs; 
 receiving, by the at least one computer processor, at a second layer of processing, the first measurement data and the second measurement data for each contact-agent pair of the set of contact-agent pairs; 
 processing, by the at least one computer processor, the first measurement data and the second measurement data for each contact-agent pair of the set of contact-agent pairs by a weighting algorithm to weight the first measurement data for each contact-agent pair of the set of contact-agent pairs relative to the second measurement data for each contact-agent pair of the set of contact-agent pairs in the second layer of processing, to output a weighted contact-agent pair output measurement data; and 
 selecting, by the at least one computer processor, a contact-agent pair based at least in part on the weighted contact-agent pair output measurement data; 
   matching, by the at least one computer processor, the first agent with a first contact based at least in part on the first ability of the first agent; and   establishing, by the at least one computer processor, in a switch module of the contact center system, a communication channel between the first agent and the first contact to optimize performance of the contact center system attributable to the first ability of the first agent.   
     
     
       44. The method of claim 43, wherein determining the first ability comprises indicating that the first agent is unlikely to influence an outcome of an agent-contact interaction regardless of contact selection, and wherein the first contact has a low likelihood of being influenced. 
     
     
       45. The method of claim 43, further comprising:
 determining, by the at least one computer processor, a first likelihood of the first contact to be influenced during agent-contact interactions, wherein matching the first agent with the first contact is based in part on the first likelihood.   
     
     
       46. The method of claim 45, wherein determining the first likelihood comprises indicating that the first contact is likely to have a successful interaction regardless of agent selection, wherein the first agent is determined to have a low ability to influence an outcome of an agent-contact interaction. 
     
     
       47. The method of claim 45, wherein determining the first likelihood comprises indicating that the first contact is likely to have an unsuccessful interaction regardless of agent selection, and wherein the first agent is determined to have a low ability to influence an outcome of an agent-contact interaction. 
     
     
       48. The method of claim 45, wherein determining the first likelihood comprises indicating that the first contact is likely to be influenced based on agent selection, and wherein the first agent is determined to have a high ability to influence an agent-contact outcome of an interaction. 
     
     
       49. The method of claim 45, wherein determining the first likelihood comprises analyzing caller affinity data of the first contact. 
     
     
       50. A system for handling contacts and agents in a contact center system comprising:
 at least one computer processor communicatively coupled to and configured to operate in the contact center system, wherein the at least one computer processor is further configured to:
 determine a first ability of a first agent to influence outcomes of agent-contact interactions, wherein the first ability of the first agent is determined based at least in part upon:
 processing, by the at least one computer processor, input data associated with contacts in a set of contacts and agents in a set of agents using a first measurement algorithm in a pair-wise fashion in a first layer of processing, to output first measurement data for each contact-agent pair of a set of contact-agent pairs; 
 processing, by the at least one computer processor, the input data associated with the contacts in the set of contacts and the agents in the set of agents using a second measurement algorithm in a pair-wise fashion in the first layer of processing, to output second measurement data for each contact-agent pair of the set of contact-agent pairs; 
 receiving, by the at least one computer processor, at a second layer of processing, the first measurement data and the second measurement data for each contact-agent pair of the set of contact-agent pairs; 
 processing, by the at least one computer processor, the first measurement data and the second measurement data for each contact-agent pair of the set of contact-agent pairs by a weighting algorithm to weight the first measurement data for each contact-agent pair of the set of contact-agent pairs relative to the second measurement data for each contact-agent pair of the set of contact-agent pairs in the second layer of processing, to output a weighted contact-agent pair output measurement data; and 
 selecting, by the at least one computer processor, a contact-agent pair based at least in part on the weighted contact-agent pair output measurement data; 
 
 match the first agent with a first contact based at least in part on the first ability of the first agent; and 
 establish, in a switch module of the contact center system, a communication channel between the first agent and the first contact to optimize performance of the contact center system attributable to the first ability of the first agent. 
   
     
     
       51. The system of claim 50, wherein determining the first ability comprises indicating that the first agent is unlikely to influence an outcome of an agent-contact interaction regardless of contact selection, and wherein the first contact has a low likelihood of being influenced. 
     
     
       52. The system of claim 50, wherein the at least one computer processor is further configured to:
 determine a first likelihood of the first contact to be influenced during agent-contact interactions,   wherein matching the first agent with the first contact based in part on the first likelihood.   
     
     
       53. The system of claim 52, wherein determining the first likelihood comprises indicating that the first contact is likely to have a successful interaction regardless of agent selection, and wherein the first agent is determined to have a low ability to influence an outcome of an agent-contact interaction. 
     
     
       54. The system of claim 52, wherein determining the first likelihood comprises indicating that the first contact is likely to have an unsuccessful interaction regardless of agent selection, and wherein the first agent is determined to have a low ability to influence an outcome of an agent-contact interaction. 
     
     
       55. The system of claim 52, wherein determining the first likelihood comprises indicating that the first contact is likely to be influenced based on agent selection, and wherein the first agent is determined to have a high ability to influence an outcome of an agent-contact interaction. 
     
     
       56. The system of claim 52, wherein determining the first likelihood comprises analyzing caller affinity data of the first contact. 
     
     
       57. An article of manufacture for handling contacts and agents in a contact center system comprising:
 a non-transitory processor readable medium; and   instructions stored on the medium;   wherein the instructions are configured to be readable from the medium by at least one computer processor communicatively coupled to and configured to operate in the contact center system and thereby cause the at least one computer processor to operate so as to:
 determine a first ability of a first agent to influence outcomes of agent-contact interactions, wherein the first ability of the first agent is determined based at least in part upon:
 processing, by the at least one computer processor, input data associated with contacts in a set of contacts and agents in a set of agents using a first measurement algorithm in a pair-wise fashion in a first layer of processing, to output first measurement data for each contact-agent pair of a set of contact-agent pairs; 
 processing, by the at least one computer processor, the input data associated with the contacts in the set of contacts and the agents in the set of agents using a second measurement algorithm in a pair-wise fashion in the first layer of processing, to output second measurement data for each contact-agent pair of the set of contact-agent pairs; 
 receiving, by the at least one computer processor, at a second layer of processing, the first measurement data and the second measurement data for each contact-agent pair of the set of contact-agent pairs; 
 processing, by the at least one computer processor, the first measurement data and the second measurement data for each contact-agent pair of the set of contact-agent pairs by a weighting algorithm to weight the first measurement data for each contact-agent pair of the set of contact-agent pairs relative to the second measurement data for each contact-agent pair of the set of contact-agent pairs in the second layer of processing, to output a weighted contact-agent pair output measurement data; and 
 selecting, by the at least one computer processor, a contact-agent pair based at least in part on the weighted contact-agent pair output measurement data; 
 
 match the first agent with a first contact based at least in part on the first ability of the first agent; and 
 establish, in a switch module of the contact center system, a communication channel between the first agent and the first contact to optimize performance of the contact center system attributable to the first ability of the first agent. 
   
     
     
       58. The article of manufacture of claim 57, wherein determining the first ability comprises indicating that the first agent is unlikely to influence an outcome of an agent-contact interaction regardless of contact selection, and wherein the first contact has a low likelihood of being influenced. 
     
     
       59. The article of manufacture of claim 57, wherein the at least one computer processor is caused to operate further so as to:
 determine a first likelihood of the first contact to be influenced during agent-contact interactions,   wherein matching the first agent with the first contact based in part on the first likelihood.   
     
     
       60. The article of manufacture of claim 59, wherein determining the first likelihood comprises indicating that the first contact is likely to have a successful interaction regardless of agent selection, and wherein the first agent is determined to have a low ability to influence an outcome of an agent-contact interaction. 
     
     
       61. The article of manufacture of claim 59, wherein determining the first likelihood comprises indicating that the first contact is likely to have an unsuccessful interaction regardless of agent selection, and wherein the first agent is determined to have a low ability to influence an outcome of an agent-contact interaction. 
     
     
       62. The article of manufacture of claim 59, wherein determining the first likelihood comprises indicating that the first contact is likely to be influenced based on agent selection, and wherein the first agent is determined to have a high ability to influence an outcome of an agent-contact interaction.