Method and system for robust wait time estimation in a multi-skill contact center with abandonments - Patents.com
The system addresses the challenge of inaccurate wait time estimation in contact centers by using historical data and cache-based calculations to predict latency accurately, enhancing customer satisfaction and service quality.
Patent Information
- Application Number
- JP2023527404
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2020-12-08
- Filing Date
- 2021-12-08
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2041-12-08
AI Technical Summary
Contact centers face challenges in accurately estimating wait times for customer interactions due to the multifaceted nature of their environment and limitations in existing statistical approximations, leading to customer dissatisfaction and reduced service quality.
A system and method for estimating latency in contact centers using historical data to calculate latency estimates based on factors like media type, skill requirements, queue position, and agent availability, incorporating a cache for stored interactions and adjusting processing times for more accurate predictions.
The system provides robust and accurate latency estimates, enabling better customer management and resource optimization by predicting wait times with high probability intervals, reducing customer abandonment and improving service quality.
Smart Images

Figure 0007580602000027 
Figure 0007580602000028 
Figure 0007580602000029
Abstract
Description
[Technical field]
[0001] (CROSS REFERENCE TO RELATED APPLICATIONS) This application claims priority to U.S. Provisional Patent Application No. 63 / 122,905, filed December 8, 2020, the contents of which are incorporated herein by reference in their entirety. [Background technology]
[0002] Contact centers typically distribute incoming interactions among available agents who are suitable to handle a particular interaction. However, when all agents are unavailable, the incoming interaction may be placed in one or more queues to await the next available agent. Long wait times may lead to customers abandoning the communication or to a reduced effectiveness of the communication when finally connected with an agent. As a result, excessive wait times may reduce customer satisfaction and reduce service quality to the detriment of call center operations.
[0003] In contact center operations, wait time information for particular interactions facilitates handling of the interactions and promotes higher quality customer service. Various systems and methods for providing wait time information to customers during periods when agents are unavailable suffer from several shortcomings, such as undesirable accuracy, insufficient consideration of customer behavior during wait periods, the multifaceted nature of certain contact center environments, and limitations associated with certain statistical approximations, to name a few. Summary of the Invention
[0004] One embodiment is directed to unique systems, components, and methods for estimating wait times for connecting a customer to an agent. Other embodiments are directed to apparatus, systems, devices, hardware, methods, and combinations thereof for estimating contact center wait times.
[0005] According to one embodiment, a system for estimating a wait time for connecting a customer to an agent may include at least one processor and at least one memory containing a plurality of instructions stored therein that, in response to execution by the at least one processor, cause the system to: receive incoming interaction data from a customer indicating a request for estimating a wait time; in response to receiving the request, retrieve historical data indicative of interaction information for one or more interactions that have already been serviced; and generate a wait time estimate based on the historical data.
[0006] In some embodiments, the interaction data may include multiple route paths, each defined by one or more of: a media type associated with the request, data stored in the queue associated with one or more previous interactions, a language associated with the request, one or more skill requirements associated with the request, the position of the request in the queue, the number of agents available to service the queue, the time of the request, or the time of service of the request.
[0007] In some embodiments, in response to execution by at least one processor, the instructions may cause the system to maintain a waiting cache for each of the plurality of route paths to store interaction data associated with one or more interactions that have not yet been serviced, and to maintain a completed cache for each of the plurality of route paths to store interaction data associated with one or more interactions that have already been serviced.
[0008] In some embodiments, generating a wait time estimate based on historical data may include determining an actual wait time corresponding to one or more interactions already serviced, determining a position in queue for the one or more interactions already serviced, determining a number of available agents serving the one or more interactions already serviced, and calculating an adjusted average handle time corresponding to the one or more interactions already serviced based on the determined actual wait time, the determined position in queue, and the determined number of available agents.
[0009] In some embodiments, generating a wait time estimate based on the historical data may include calculating an average processing time estimator based on one or more of a median of the adjusted average processing time calculations or a mean of the adjusted average processing time calculations.
[0010] In some embodiments, generating a wait time estimate based on historical data may include determining a position in a queue corresponding to the request, determining a number of available agents corresponding to the request, and calculating the wait time estimate based on the average processing time estimate, the determined position in the queue corresponding to the request, and the determined number of available agents corresponding to the request.
[0011] In some embodiments, generating a latency estimate based on historical data may include calculating an estimated interval of latency.
[0012] In some embodiments, calculating the estimated interval of the wait time may include determining a sample number of one or more interactions already serviced, determining a statistical significance level associated with the estimated interval, determining a lower quantile of the estimated interval, determining an upper quantile of the estimated interval, calculating a lower limit of the estimated interval based at least in part on an adjusted average processing time corresponding to the determined lower quantile, and calculating an upper limit of the estimated interval based at least in part on an adjusted average processing time corresponding to the determined upper quantile.
[0013] In some embodiments, in response to execution by at least one processor, the instructions may cause the system to determine whether the number of conversations stored in the wait cache that do not have an actual wait time exceeds a reference value, determine whether a conversation identifier associated with the incoming conversation is already present in the wait cache, and remove the incoming conversation from the wait cache in response to a determination that the number of conversations stored in the wait cache that do not have an actual wait time exceeds the reference value or in response to a determination that the conversation identifier is already present in the wait cache.
[0014] According to another embodiment, one or more non-transitory machine-readable storage media may include a plurality of instructions stored thereon that, in response to execution by at least one processor, cause the system to receive incoming interaction data from a customer indicating a request to estimate a wait time for connecting the customer to an agent, in response to receiving the request, retrieve historical data indicative of interaction information for one or more interactions already serviced, and generate a wait time estimate for connecting the customer to an agent based on the historical data.
[0015] In some embodiments, the interaction data may include multiple route paths, each defined by one or more of: a media type associated with the request, data stored in the queue associated with one or more previous interactions, a language associated with the request, one or more skill requirements associated with the request, the position of the request in the queue, the number of agents available to service the queue, the time of the request, or the time of service of the request.
[0016] In some embodiments, generating a wait time estimate based on historical data may include determining an actual wait time corresponding to one or more interactions already serviced, determining a position in queue for the one or more interactions already serviced, determining a number of available agents serving the one or more interactions already serviced, and calculating an adjusted average handle time corresponding to the one or more interactions already serviced based on the determined actual wait time, the determined position in queue, and the determined number of available agents.
[0017] In some embodiments, generating a wait time estimate based on the historical data may include calculating an average processing time estimator based on one or more of a median of the adjusted average processing time calculations or a mean of the adjusted average processing time calculations.
[0018] In some embodiments, generating a wait time estimate based on historical data may include determining a position in a queue corresponding to the request, determining a number of available agents corresponding to the request, and calculating the wait time estimate based on the average processing time estimate, the determined position in the queue corresponding to the request, and the determined number of available agents corresponding to the request.
[0019] In some embodiments, generating a wait time estimate based on the historical data may include calculating an estimated interval of the wait time, which may include determining a sample number of one or more interactions already serviced, determining a statistical significance level associated with the estimated interval, determining a lower quantile of the estimated interval, determining an upper quantile of the estimated interval, calculating a lower limit of the estimated interval based at least in part on an adjusted average processing time corresponding to the determined lower quantile, and calculating an upper limit of the estimated interval based at least in part on an adjusted average processing time corresponding to the determined upper quantile.
[0020] In some embodiments, in response to execution by at least one processor, the instructions may cause the system to determine whether the number of conversations stored in the wait cache that do not have an actual wait time exceeds a reference value, determine whether a conversation identifier associated with the incoming conversation is already present in the wait cache, and remove the incoming conversation from the wait cache in response to a determination that the number of conversations stored in the wait cache that do not have an actual wait time exceeds the reference value or in response to a determination that the conversation identifier is already present in the wait cache.
[0021] According to yet another embodiment, a method for estimating a contact center wait time for connecting a customer to an agent may include receiving, by a contact center system, incoming interaction data from a customer indicating a request for estimating a wait time; retrieving, by the contact center system, historical data indicating interaction information for one or more interactions already serviced in response to receiving the request; and generating, by the contact center system, a wait time estimate based on the historical data.
[0022] In some embodiments, the interaction data may include multiple route paths, each defined by one or more of: a media type associated with the request, data stored in the queue associated with one or more previous interactions, a language associated with the request, one or more skill requirements associated with the request, the position of the request in the queue, the number of agents available to service the queue, the time of the request, or the time of service of the request.
[0023] In some embodiments, generating a wait time estimate based on the historical data may include: determining, by the contact center system, an actual wait time corresponding to one or more interactions already serviced; determining, by the contact center system, a position in queue for the one or more interactions already serviced; determining, by the contact center system, a number of available agents serving the one or more interactions already serviced; and calculating, by the contact center system, an adjusted average handle time corresponding to the one or more interactions already serviced based on the determined actual wait time, the determined position in queue, and the determined number of available agents.
[0024] In some embodiments, generating a wait time estimate based on the historical data may include: calculating, by the contact center system, an average handle time estimate based on a median of the adjusted average handle time calculations or based on an average of the adjusted average handle time calculations; determining, by the contact center system, a position in a queue corresponding to the request; determining, by the contact center system, a number of available agents corresponding to the request; and calculating, by the contact center system, a wait time estimate based on the average handle time estimate, the determined position in the queue corresponding to the request, and the determined number of available agents corresponding to the request.
[0025] This Summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in limiting the scope of the claimed subject matter. Further embodiments, forms, features, and aspects of the present application will become apparent from the description and figures provided herein. [Brief description of the drawings]
[0026] The concepts described herein are illustrated by way of example, and not by way of limitation, in the accompanying drawings. For simplicity and clarity of illustration, elements illustrated in the figures have not necessarily been drawn to scale. Where considered appropriate, reference labels have been repeated among the figures to indicate corresponding or analogous elements. [Figure 1] FIG. 1 is a simplified flow diagram of at least one embodiment of a method for estimating contact center wait times. [Diagram 2] FIG. 2 is a simplified flow diagram of another embodiment of a method for estimating contact center wait times. [Diagram 3] FIG. 13 is a simplified flow diagram of yet another embodiment of a method for estimating contact center wait times, including calculating a point value for the estimated wait time. [Figure 4] FIG. 4 is a simplified flow diagram of the method of FIG. 3, including calculating an interval of estimated waiting time. [Diagram 5] FIG. 11 is a simplified flow diagram of yet another embodiment of a method for estimating contact center wait times that includes selectively evicting one or more interactions from a wait cache. [Figure 6] FIG. 6 is a simplified flow diagram of the method of FIG. 5, including calculating a point value for an estimated wait time. [Figure 7] FIG. 1 is a simplified flow diagram of at least one embodiment of a method for controlling the number of contact center interactions added to a waiting cache. [Figure 8]1 is a simplified flow diagram of at least one embodiment of a method for evaluating special conditions when estimating contact center wait times. [Figure 9] FIG. 1 is a simplified flow diagram of at least one embodiment of a method for applying control limits when estimating contact center wait times. [Figure 10] FIG. 1 is a simplified block diagram of at least one embodiment of a contact center system. [Figure 11] 1 is a simplified block diagram of at least one embodiment of a computing device. [Figure 12] 1 is a graphical comparison of latency estimation according to at least one method of the present disclosure and at least one conventional method. [Figure 13] 1 is a chart-based depiction of the distribution of wait time estimates according to at least one conventional method. [Figure 14] 1 is a chart-based depiction of a distribution of latency estimates according to at least one method of the present disclosure. [Figure 15] 1 is a boxplot comparison of the accuracy of latency estimation according to at least one method of the present disclosure and at least one conventional method. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0027] While the concepts of the present disclosure are susceptible to various modifications and alternative forms, specific embodiments have been shown by way of example in the drawings and are herein described in detail, It should be understood, however, that there is no intention to limit the concepts of the present disclosure to the particular forms disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives consistent with the scope of the present disclosure and the appended claims.
[0028] References herein to "one embodiment," "one embodiment," "exemplary embodiment," and the like indicate that the embodiment being described may include a particular feature, structure, or characteristic, but that all embodiments may or may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. It should be further understood that references to "preferred" components or features may indicate the desirability of a particular component or feature with respect to one embodiment, but that the present disclosure is not so limited with respect to other embodiments that may omit such component or feature. Moreover, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is intended that the implementation of such feature, structure, or characteristic in connection with other embodiments, whether or not expressly described, is within the knowledge of one of ordinary skill in the art. Moreover, particular features, structures, or characteristics may be combined in any suitable combinations and / or subcombinations in various embodiments.
[0029] Additionally, it should be understood that an item in a listing of "at least one of A, B, and C" can mean (A), (B), (C), (A and B), (B and C), (A and C), or (A, B, and C). Similarly, it should be understood that an item in a listing of "at least one of A, B, or C" can mean (A)(B), (C), (A and B), (B and C), (A and C), or (A, B, and C). Further, with respect to the claims, the use of words and phrases such as "a," "an," "at least one," and / or "at least one portion" should not be construed as limiting to only one of such elements, unless specifically stated to the contrary, and the use of phrases such as "at least a portion" and / or "a portion" should be construed to encompass both embodiments including only a portion of such an element and embodiments including the entirety of such an element, unless specifically stated to the contrary.
[0030] The disclosed embodiments may be implemented in hardware, firmware, software, or a combination thereof, in some cases. The disclosed embodiments may also be implemented as instructions executed by or stored on one or more transitory or non-transitory machine-readable (e.g., computer-readable) storage media that may be read and executed by one or more processors. A machine-readable storage medium may be embodied as any storage device, mechanism, or other physical structure for storing or transmitting information in a form readable by a machine (e.g., a volatile or non-volatile memory, a media disk, or other media device).
[0031] In the drawings, some structural or method features may be shown in a specific arrangement and / or ordering. However, it should be understood that such specific arrangement and / or ordering may not be required. Rather, in some embodiments, such features may be arranged in a different manner and / or order than that shown in the illustrative drawings, unless indicated to the contrary. Furthermore, the inclusion of a structural or method feature in a particular figure does not imply that such feature is required in all embodiments, and in some embodiments, may not be included or may be combined with other features.
[0032] The present disclosure provides several unique approaches for accurately and robustly predicting how long an interaction or conversation may wait in a queue before being connected to an agent. As described further below, in at least some embodiments, the systems and methods described herein take into account several relevant factors, such as, for example, the skill requirements of the interaction, the urgency of the interaction (e.g., a voice interaction may be prioritized over an email interaction), interaction synchronicity (e.g., two or more interactions may be handled by one agent), and agent skills (e.g., a multi-skilled agent may be active in multiple queues).
[0033] Current methods and configurations that may be used in Automatic Call Distribution (ACD) systems to provide wait time information to customers during periods of agent unavailability may not provide a desired degree of accuracy and / or precision. In particular, these methods and configurations may fail to consider various factors related to customer-agent interactions that affect wait times. These factors include, but are not limited to, the occurrence of abandonment, the priority / urgency of the interaction, the customer's tolerance for waiting, and the multi-skill nature of contact center environments where agents may handle multiple interactions and be active in multiple queues. Additionally, some methods and configurations may utilize average-based estimates that may not accurately approximate actual wait times.
[0034] Due to the complexity and difficulty of considering individual factors and / or variables, the robust estimation framework presented herein utilizes information from previously serviced interactions combined with information about the current state of the system, as well as a feedback mechanism that learns from changes in system resources and assumptions. It should be appreciated that an estimation framework that provides more accurate wait time predictions (e.g., point value estimations and / or estimation intervals) facilitates reliable decision making and improved responsiveness to customer interactions. Such predictions may also direct preemptive callbacks to customers and the provision of alternative contact channels for handling customer interactions if wait times exceed certain thresholds. In addition, the wait time estimation framework provided by the systems and methods of the present disclosure may enable contact centers to manage agent queue activation with increased reliability among a network of queues to optimize operations toward reducing wait times.
[0035] In at least some embodiments, estimating latency according to the systems and methods disclosed herein includes calculating a point value estimate of latency using parameters extracted through a series of historical latency observations based on previously serviced interactions, and calculating an estimate of latency using an estimation interval. In some embodiments, the latency estimate is provided as an interval T=[t1,t2], where the true / actual value of the latency should fall within the interval T with high or relatively high probability. Additionally, in some embodiments, three types of intervals are considered to determine the interval T, namely, (i) T1=(-∞,t2], (ii) T2=[t1,+∞), and (iii) T3=[t1,t2].
[0036] For purposes of this disclosure, the definitions provided below are applicable to the subsequent discussion of defined terms. · Queue ID - A unique identifier that represents a particular queue. Medium Type - The medium type into which a particular interaction is classified. Conversation ID - A unique identifier that represents the current interaction. Skillset ID - A unique set of skill identifiers associated with the current interaction that serve as requirements for an agent to handle the interaction. Queue Length(n) – The number of interactions currently waiting in the queue. Position in Queue (PiQ) - the position of the current interaction in the queue (equals n+1). Agent / Server Count (S) - An integer value representing the total number of available agents. Available agents are agents who are activated on the queue, available agents can be either Available / Unavailable or Waiting. · Initial Timestamp - A timestamp representing the time when the interaction is put into the cache. Service Timestamp − The timestamp representing when the interaction was answered / serviced by an agent. Actual Wait Time (AWT) - The time (e.g., in seconds) that elapsed between when the EWT calculation is performed and when an agent answers / services the conversation. Note that AWT does not represent the actual total actual wait time of the conversation after it is queued. Estimated Wait Time (EWT) – an estimate of the AWT (e.g., in the form of a point value estimate or an estimated interval). Average Handle Time (AHT) – The average agent service time for handling an interaction. Adjusted AHT - The adjusted handling time calculated for each interaction that takes into account various effects and is used to predict subsequent new estimates.
[0037] It should be understood that PiQ can be determined in different ways depending on the use case, as explained above. When estimating the latency of a conversation, the conversation's current position in the queue is used to determine the PiQ. However, when estimating the latency for a specific media type, PiQ can be determined by looking at queue length(n)+1, as if estimating for a conversation that just entered the queue with position=queue length(n)+1.
[0038] By way of background, some approaches to calculating EWT are based on queuing theory. Typically, it is accepted that the time to service completion (service time) is exponentially distributed with a mean equal to 1 / μ, or 1 / Sμ in the case of multiple servers, where S represents the number of servers and μ represents the rate of service. When all servers are busy, a new interaction must wait in a queue until the interaction currently being handled by the server has been serviced and the n interactions before it have been serviced / answered. Thus, the waiting time of a new interaction can be represented as a sum of n+1 independent and identically distributed (IID) exponential random variables, each with mean 1 / Sμ and an Erlang distribution. For the purposes of this disclosure, EWT can be a random variable describing the waiting time of interest.
[0039] The EWT may be expressed and / or determined according to the mathematical relationship (1) below:
[0040]
number
[0041]
number
[0042] In some cases, it may be desirable to assess the accuracy of the EWT value as measured by a confidence interval, for which an estimated wait time interval may be used to provide a range of values that are likely to contain the actual wait time (AWT) for a particular interaction in a queue.
[0043] For ease of explanation, the waiting time in a queue may be considered as a random variable τ with a certain probability distribution. The interval T should satisfy the condition P(τ∈T)=γ, where γ is expressed as a percentage close to 100%. Often, γ can be written as γ=1-α, where α is close to 0. A typical value of α (sometimes referred to herein as the significance level associated with the estimated waiting time interval) may be 1%, 5%, or 10%, at least in some embodiments.
[0044] Q(p) may also be assumed to be the p-th quantile of τ. Based on this assumption, the interval estimate of EWT can be expressed as: 1. An upper or lower interval given by an upper and lower limit, respectively.
[0045]
number
[0046]
number
[0047] As mentioned above as background to the consideration of relation (1), the original distribution is in Erlangs, so the quantiles cannot be easily stated analytically. Several numerical methods exist for estimating Q(p) in this form, which involve finding the path of the equation F(x)-γ=0, where F(x) is the cumulative probability distribution function.
[0048] The waiting time in the queue may be the sum of the waiting times to the next ready agent in the group, n. If the queue length n is a large number, a normal approximation can be used for the probability distribution of τ. In one approach, the mean and standard deviation values can be taken from an Erlang distribution. In another approach, the mean and standard deviation values can be estimated from a sample of actual data.
[0049] 1, in use, a system may perform a method 100 for estimating contact center wait times. It should be appreciated that in some embodiments, the system may be embodied as a computing device (e.g., computing device 1100 of FIG. 11) and / or a contact center system (e.g., contact center system 1000 of FIG. 10) or systems / devices thereof. It should be appreciated that certain blocks of method 100 are illustrated by way of example, and that such blocks may be combined or divided, added or removed, and / or reordered in whole or in part, depending on the particular embodiment, unless stated to the contrary.
[0050] The illustrative method 100 begins at block 102 where the system receives inbound interaction data from one or more customers. In an illustrative embodiment, the inbound interaction data includes or is otherwise embodied as automatic call distribution (ACD) data. In some examples, the inbound interaction data includes contact volume, average handle time (AHT), full time equivalency (FTE), capture rate, contact handling data by contact type, contact handling data by staff type, and / or other relevant ACD data for a specified interval. Additionally, in some examples, the interaction data includes one or more route paths each defined by one or more of the following: a media type associated with the incoming interaction data, data stored in the queue associated with one or more previous interactions, a language associated with the incoming interaction data, one or more skill requirements associated with the incoming interaction data, a position of the incoming interaction data in the queue, a number of agents available to service the queue, a time of the incoming interaction data, a time of service of the incoming interaction data, or any additional route dimensions. It should be appreciated that, as described below, in at least some embodiments, the incoming interaction data received at block 102 indicates a request to estimate a wait time for connecting the customer to an available agent.
[0051] From block 102, the illustrative method 100 proceeds to block 104, where the system distributes one or more interactions to one or more available agents activated on the queue. As discussed above, the available agents activated on the queue can be either waiting or available. From block 104, the method 100 proceeds to block 106.
[0052] At block 106, the system determines whether all available agents are in use. If the system determines that all available agents are in use, the method 100 proceeds to block 108.
[0053] At block 108, the system places one or more interactions associated with the incoming interaction data in one or more waiting queues. As will become apparent from the discussion below, the one or more interactions placed in the one or more waiting queues may be stored in one or more caches, such as, for example, the caches described below with reference to Figures 5 and 6. In any event, the illustrative method 100 proceeds from block 108 to block 110.
[0054] At block 110, the system estimates one or more wait times for one or more interactions associated with the incoming interaction data. To do so, in an illustrative embodiment, the system performs blocks 112 and 114. At block 112, the system retrieves historical interaction data that indicates interaction information for one or more interactions that have already been serviced, as described further below with reference to FIG. 3. At block 114, the system generates one or more wait time estimates based on the historical interaction data. In at least some embodiments, the wait time estimates associated with block 114 may be generated according to method 300 of FIGS. 3 and 4. Additionally, in at least some embodiments, the wait time estimation associated with block 114 takes into account one or more of a media type associated with the incoming interaction data, data stored in the queue associated with one or more previous interactions, a language associated with the incoming interaction data, one or more skill requirements associated with the incoming interaction data, a position of the incoming interaction data in the queue, a number of agents available to service the queue, a time of the incoming interaction data, a time of service of the incoming interaction data, or any additional route dimensions. From block 110, the example method 100 proceeds to block 116.
[0055] In block 116, the system provides the wait time estimate generated in block 110 to one or more customers. From block 116, the method 100 returns to block 104.
[0056] Returning to block 106 , if the system determines that all available agents are not in use, the method 100 returns to block 104 .
[0057] Although blocks 102-116 are described relatively serially, it should be understood that various blocks of method 100 may be performed in parallel in some embodiments.
[0058] 2, in use, the system may perform a method 200 of estimating contact center wait times. In view of the discussion below, it should be appreciated that method 200 may be similar to method 100. It should be appreciated that in some embodiments, the system may be embodied as a computing device (e.g., computing device 1100 of FIG. 11) and / or a contact center system (e.g., contact center system 1000 of FIG. 10) or systems / devices thereof. It should also be appreciated that certain blocks of method 200 are illustrated by way of example, and that such blocks may be combined or divided, added or removed, and / or reordered in whole or in part, depending on the particular embodiment, unless stated to the contrary.
[0059] The illustrative method 200 begins at block 202 where the system receives incoming interaction data from one or more customers. The incoming interaction data received at block 202 may be similar to the incoming interaction data received at block 102. The method 200 proceeds from block 202 to block 204.
[0060] In some embodiments, at block 204, the system may retrieve historical interaction data from one or more caches. The historical interaction data retrieved at block 204 may be similar to the historical interaction data retrieved at block 112. In at least some embodiments, at block 204, the historical interaction data may be retrieved from a cache, which will be described below with reference to FIGS. 5 and 6. In either case, to perform block 204, the system illustratively performs block 206. At block 206, the system may obtain information updates from one or more caches based on one or more previously serviced interactions. The dashed illustrations of blocks 204 and 206 in FIG. 2 are intended to convey data flow that is updated based on historical data (i.e., previously serviced interactions) stored in the caches. The illustrative method 200 proceeds from block 204 to block 208.
[0061] At block 208, the system estimates one or more latency periods for one or more interactions associated with the incoming interaction data. The latency period estimation performed at block 208 may be similar to the latency period estimation performed at block 110. The illustrative method 200 proceeds from block 208 to block 210.
[0062] In block 210, the system provides the wait time estimate generated in block 208 to one or more customers. Providing the wait time estimate in block 210 can be similar to providing the wait time estimate in block 116.
[0063] Although blocks 202-210 are described relatively serially, it should be understood that various blocks of method 200 may be performed in parallel in some embodiments.
[0064] 3 and 4, in use, the system may execute a method 300 for estimating contact center wait times. In some embodiments, the method 300 may be implemented to perform blocks 110 and 208. It should be understood that in some embodiments, the system may be embodied as a computing device (e.g., computing device 1100 of FIG. 11) and / or a contact center system (e.g., contact center system 1000 of FIG. 10) or systems / devices thereof. It should be understood that certain blocks of the method 300 are illustrated by way of example, and that such blocks may be combined or divided, added or removed, and / or reordered in whole or in part, depending on the particular embodiment, unless stated to the contrary.
[0065] The illustrative method 300 begins at block 302, where the system retrieves historical interaction data in response to receiving incoming interaction data from one or more customers. The historical interaction data retrieved at block 302 may be similar to the historical interaction data retrieved at block 112. In an illustrative embodiment, to perform block 302, the system performs blocks 304 and 306. At block 304, the system retrieves interaction data for a number of previously serviced recent interactions stored in a cache (e.g., a completed cache, described below with reference to FIG. 6). At block 306, the system retrieves interaction data for each particular one of the previously serviced recent interactions for which interaction data is retrieved at block 304. In some embodiments, each previously serviced recent interaction for which interaction data is retrieved at block 306 may correspond to an interaction in a queue having a non-zero wait time. The method 300 proceeds from block 302 to block 308.
[0066] At block 308, the system determines the AWT for each particular serviced recent interaction for which interaction data was obtained at block 302. Each particular serviced recent interaction for which the AWT is calculated at block 308, and for which other calculations / determinations are made as discussed below, may be referred to herein as the "ith" interaction.
[0067] It should be appreciated that in order to calculate the waiting time estimates according to the above relationships (1) and (2), the states of AHT, PiQ, and S need to be estimated. In some cases, the states of PiQ and S can be relatively easily observed, where PiQ is the depth or length of the queue and S is the number of agents servicing the queue. Due to the transient state of the AHT variables, the variables cannot be known a priori. As a result, some approaches rely on average values of past service times to provide estimates of the AHT variables.
[0068] Of course, in a real contact center setting, average service time is often affected by other factors that may change frequently based on the specific circumstances, such as the incidence of abandonment, agent efficiency, interaction concurrency, and agent availability, etc. Therefore, a more accurate estimate of AHT may be made from recently serviced interactions.
[0069] In an illustrative embodiment, to estimate the AHT as discussed further below, an adjusted AHT (AHT adj ) is collected from the sample corresponding to the i-th most recent interaction that had a nonzero wait time in the queue. For each of these interactions, the actual wait time (AWT) is calculated by the system in block 308, as well as the number of agents and position in the queue when the i-th interaction first enters the queue.
[0070] Making various substitutions in relation (2) above, the non-zero AWT of the i-th interaction (i.e., where i=1,...,n) may be expressed and / or determined according to relation (3) below.
[0071]
number
[0072] The illustrative method 300 proceeds from block 308 to block 310. In block 310, the system determines the position in the queue (PiQ_i) for each i-th interaction. From block 310, the method 300 proceeds to block 312.
[0073] At block 312, the system determines the number of agents available for each i-th interaction (S_i). The method 300 proceeds from block 312 to block 314.
[0074] In block 314, the system calculates the adjusted average handling time for each i-th interaction.
[0075]
number
[0076]
number
[0077]
number
[0078] The illustrative method 300 proceeds from block 314 to block 316. At block 316, the system calculates
[0079]
number
[0080]
number
[0081] When there is a significant number of recently serviced interactions, a step is performed in block 314.
[0082]
number
[0083] The illustrative method 300 proceeds from block 316 to block 318. At block 318, the system calculates an estimated wait time (EWT) point value based at least in part on the AHT^ calculated at block 316. More specifically, at block 318, the system calculates the EWT according to the following mathematical relationship (6):
[0084]
number
[0085]
number
[0086] 4, in an illustrative embodiment, method 300 proceeds from block 318 to block 320 where the system calculates an interval estimate of the latency. To implement block 320, the system illustratively implements blocks 322, 324, 326, 328, 330, and 332, which are described in more detail below. At a high level, it should be appreciated that the implementation of block 320 incorporates various aspects of the point value estimation approach discussed above with reference to blocks 308-318.
[0087] As background to the interval estimate of the waiting time calculated in block 320, the order statistics AdjAHT of a sample of n interactions (k)is adjusted according to the following relationship (7): adj ) can be used as an approximation of
[0088]
number
[0089] For example, if there are 19 observations, the order statistics
[0090]
number
[0091]
number
[0092]
number
[0093] Finally, to provide bilateral spacing,
[0094]
number
[0095]
number
[0096] Returning to illustrative block 320, its implementation by the system includes implementation of blocks 322, 324, 326, 328, 330, 332 as set forth above. In block 322, the system determines a sample number of one or more interactions that have already been serviced. In at least some embodiments, the sample of serviced interactions may be drawn from serviced interactions stored in a completed cache, described below with reference to FIG. 6. In block 324, the system determines a statistical significance level (α) associated with the estimation interval. In block 326, the system determines a lower quantile of the estimation interval. In block 328, the system determines an upper quantile of the estimation interval. In block 330, the system calculates a lower bound of the estimation interval based at least in part on the adjusted average processing time corresponding to the determined lower quantile. In at least some embodiments, the adjusted average processing time corresponding to the determined lower quantile is a variable included in relationship (8).
[0097]
number
[0098]
number
[0099] Although blocks 302-332 are described relatively serially, it should be understood that various blocks of method 300 may be performed in parallel in some embodiments.
[0100] 5 and 6, in use, the system may perform a method 500 for estimating contact center wait times. It should be appreciated that in some embodiments, the system may be embodied as a computing device (e.g., computing device 1100 of FIG. 11) and / or a contact center system (e.g., contact center system 1000 of FIG. 10) or systems / devices thereof. It should be appreciated that certain blocks of method 500 are illustrated by way of example, and that such blocks may be combined or divided, added or removed, and / or reordered in whole or in part, depending on the particular embodiment, unless stated to the contrary.
[0101] At a high level, for each latency estimation request received in method 500 (i.e., at block 502), the system tracks the service time of recently completed requests / serviced interactions in a history table (e.g., a completed cache). It should be appreciated that the history table may be, for example, a database or a temporary caching mechanism. When a new request is received, an estimate for the request is calculated using the history table. Once the request is completed and the interaction is serviced, the measured service time for that interaction is updated in the history table. Not all requests coming through the system are monitored, since many requests are not affected by latency. Such requests do not contribute to the latency prediction of future requests and are not added to the history table. As will become apparent from the discussion below, only requests associated with a latency estimate are included in the history table.
[0102] According to relationships (3)-(6) above, in at least some embodiments, the system stores relevant interaction data from wait time estimation requests to infer future predictions of EWT for each unique queue and media type combination. As noted above, the interaction data may include: (i) the queue and media type of the interaction, (ii) the position in the queue of the interaction, (iii) the number of agents servicing the queue at the time of the request, (iv) the time of the request, and (v) the time of service.
[0103] As discussed further below, in an illustrative embodiment, the system maintains two caches: one cache (i.e., the waiting cache) stores all information regarding estimated wait time requirements for interactions that have not yet been serviced, and the other cache (i.e., the completed cache) stores information about interactions that have already been serviced and are no longer waiting in a queue.
[0104] From the foregoing discussion, it should be appreciated that in order to calculate the adjusted wait time, in at least some embodiments, the actual wait time needs to be determined. More specifically, for each interaction that enters the queue, interaction data corresponding to that interaction will be captured and stored in the wait cache. At the same time, in at least some embodiments, the status of all previous requests will be updated to reflect whether the request has been processed or is still waiting in the queue.
[0105] For each handled interaction, the time of the interaction entering the service (T service ) is recorded, and then the first arrival time at the queue or the time when a waiting time estimation request was received (T request ) to calculate the actual latency (AWT). Thus, the actual latency may be expressed and / or determined according to the following relationship (9):
[0106]
number
[0107] To make a prediction, interaction data is retrieved from the completed cache. Based on the retrieved data, adjusted predictions for those interactions are made.
[0108]
number
[0109] 5, the illustrative method 500 begins at block 502. At block 502, the system receives a request to estimate a wait time for a queued interaction. In some embodiments, the placement of interaction data for a particular interaction in the queue may serve as or otherwise correspond to the wait time estimation request. Method 500 proceeds from block 502 to block 504.
[0110] At block 504, the system stores interaction data associated with the request (i.e., from block 502) in a wait cache. In an illustrative embodiment, to perform block 504, the system performs blocks 506, 508, and 510. At block 506, the system documents the time of storage of the request and associated interaction data in the wait cache to estimate the wait time. At block 508, the system documents the position in the queue of the particular request and its associated interaction data. At block 510, the system documents the number of agents available at the time of the particular request and its associated interactions. Method 500 proceeds from block 504 to block 512.
[0111] At block 512, the system receives status updates for one or more previous requests and associated interactions. In an illustrative embodiment, the updates received at block 512 include updates based on measured service times of serviced interactions. Method 500 proceeds from block 512 to block 514.
[0112] At block 514, the system determines whether the interaction associated with the request received at block 502 has been serviced by one or more agents. In response to the system determining at block 514 that the interaction has been serviced, method 500 proceeds to block 522.
[0113] As shown in FIG. 6, in block 522, the system stores the serviced interaction data in a completed cache. In an illustrative embodiment, to perform block 522, the system performs blocks 524, 526, 528, and 530. In block 524, the system documents the time of request and storage of the serviced interaction in the completed cache. In block 526, the system documents the position in the queue of the serviced interaction. In block 528, the system documents the number of agents available at the time of the serviced interaction. In block 530, the system documents the time of service of the completed interaction. Method 500 proceeds from block 522 to block 532.
[0114] At block 532, based on the information documented at block 522, the system determines the actual wait time of the serviced interaction according to relationship (9). The method 500 proceeds from block 532 to block 534.
[0115] At block 534, in accordance with relationship (4), the system calculates an adjusted average processing time based on the determination made at block 532. From block 534, the method 500 proceeds to block 536.
[0116] At block 536, in accordance with relationship (5), the system calculates an average processing time estimate based on the calculations performed at block 534. From block 536, the method 500 proceeds to block 538.
[0117] In block 538, in accordance with relationship (6), the system calculates a point value of the estimated wait time for the specific interaction data and / or the specific route path based on the calculations performed in block 536.
[0118] Returning to block 514 , if in block 514 the system determines that the interaction associated with the request received in block 502 is not being serviced by one or more agents, the method 500 proceeds to block 516 .
[0119] In block 516, the system determines whether the unserviced interaction has been abandoned. In response to the system determining in block 516 that the unserviced interaction has been abandoned, the method 500 proceeds to block 518.
[0120] At block 518, the system evicts any interactions from the waiting cache that have not yet been serviced.
[0121] Returning to block 516 , if the system determines in block 516 that the unserviced interaction has not been abandoned, the method 500 proceeds to block 520 .
[0122] At block 520, the system determines whether the interaction has an actual latency of less than one second, which may indicate a short abandonment of the interaction. In response to the system determining at block 520 that the interaction has an actual latency of less than one second, the method 500 proceeds to block 518. However, in response to the system determining at block 520 that the interaction does not have an actual latency of less than one second, the method 500 returns to block 504.
[0123] Although blocks 502-538 are described relatively serially, it should be understood that various blocks of method 500 may be performed in parallel in some embodiments.
[0124] 7, in use, the system may perform a method 700 for controlling the number of contact center interactions added to a waiting cache. It should be appreciated that in some embodiments, the system may be embodied as a computing device (e.g., computing device 1100 of FIG. 11) and / or a contact center system (e.g., contact center system 1000 of FIG. 10) or systems / devices thereof. It should be appreciated that certain blocks of method 700 are illustrated by way of example, and that such blocks may be combined or divided, added or removed, and / or reordered in whole or in part, depending on the particular embodiment, unless stated to the contrary.
[0125] Ideally, the cache capacity is infinite to allow all interactions to be captured and subsequently tracked, however, in a real-world contact center environment where resources are limited, controlling the cache capacity to a certain level can be beneficial to minimize resources and reduce customer wait times.
[0126] It should be appreciated that interaction data is stored in the waiting cache based on, among other things, the type of interaction. To maintain prediction accuracy, the system operates to maintain unique interactions in the waiting cache. In doing so, each time a request and its associated interaction data arrives, the system checks whether the interaction is already in the waiting cache to ensure that only unique interactions are added. At the same time, if the cache has reached a maximum capacity threshold, the current request is not added to the waiting cache.
[0127] It should also be appreciated that there are types of interactions that have low actual latency, such as, for example, interactions with AWTs that are less than one second as discussed above with reference to block 520. In addition to these interactions, abandoned interactions (e.g., abandoned interactions discussed above with reference to block 516) and interactions that have flowed out of the waiting cache or have been transferred to another queue or media type are evicted from the waiting cache. Such interactions will not be added to the completed cache even if they enter service. At the end of each latency prediction, the system will acknowledge that the prediction has been made. A separate process will subscribe to and archive these predictions for accuracy analysis.
[0128] 7, the illustrative method 700 begins at block 702 where the system receives incoming interaction data from one or more customers. The incoming interaction data received at block 702 may be similar to the incoming interaction data received at block 102. Method 700 proceeds from block 702 to block 704.
[0129] At block 704, the system determines the number of conversations or interactions in the queue that do not have an actual wait time. The method 700 proceeds from block 704 to block 706.
[0130] At block 706, the system determines whether the number of conversations or interactions determined at block 704 exceeds a criterion value. In some embodiments, the criterion value may define a capacity limit for a wait cache for interactions with low or no actual latency. In response to the system determining at block 706 that the number of conversations or interactions exceeds the criterion value, method 700 proceeds to block 708.
[0131] At block 708, the system leaves the incoming conversation or interaction without adding it to the waiting cache.
[0132] Returning to block 706 , if the system determines in block 706 that the number of conversations or interactions determined in block 704 does not exceed the threshold value, the method 700 proceeds to block 710 .
[0133] At block 710, the system determines whether a conversation ID request has been received for an incoming conversation or interaction. In response to the system determining at block 710 that a conversation ID request has been received, the method 700 proceeds to block 712.
[0134] At block 712, the system determines whether the conversation ID associated with the request received at block 710 is already present in the pending cache. In response to the system determining at block 712 that the conversation ID is already present in the pending cache, method 700 proceeds to block 708.
[0135] Returning to block 710 , if at block 710 the system determines that a conversation ID request has not been received, the method 700 proceeds to block 716 .
[0136] At block 716, the system retrieves the most recent conversation or interaction from the waiting queue. The method 700 proceeds from block 716 to block 714.
[0137] At block 714, the system retains the most recent conversation or interaction from the wait queue and adds the conversation or interaction to a wait cache.
[0138] Returning to block 712 , if the system determines at block 712 that the conversation ID is not already in the waiting cache, the method 700 proceeds to block 714 .
[0139] Although blocks 702-716 are described relatively serially, it should be understood that various blocks of method 700 may be performed in parallel in some embodiments.
[0140] 8, in use, the system may perform a method 800 for estimating contact center wait times. It should be appreciated that in some embodiments, the system may be embodied as a computing device (e.g., computing device 1100 of FIG. 11) and / or a contact center system (e.g., contact center system 1000 of FIG. 10) or systems / devices thereof. It should be appreciated that certain blocks of method 800 are illustrated by way of example, and that such blocks may be combined or divided, added or removed, and / or reordered in whole or in part, depending on the particular embodiment, unless stated to the contrary.
[0141] During each wait time estimation request and subsequent prediction, certain considerations may be applied to special conditions, including, but not limited to: (i) zero estimated wait time (EWT) when there are available agents waiting to service the interaction, (ii) an undefined estimated wait time when the number of agents available to service the interaction is zero, (iii) the use of average handling time (AHT) (i.e., rather than adjusted AHT) to calculate EWT when the system starts, (iv) the use of a default formula to calculate EWT when the done cache is empty or unavailable, and (v) the application of control limits to the estimated wait time calculation during periods of instability.
[0142] It should be appreciated that the completed cache (which stores historical interaction data for previously serviced interactions) is used to improve the accuracy and precision of the EWT prediction, and therefore availability of the completed cache is important. When the completed cache is not available or is completely empty, the EWT prediction may be calculated according to relationship (2). If some of the variables included in relationship (2) also do not exist or are unavailable, the wait time of the current longest waiting interaction will be returned as the prediction.
[0143] The illustrative method 800 begins at block 802 where the system receives incoming interaction data from one or more customers, as shown in Figure 8. The incoming interaction data received at block 802 may be similar to the incoming interaction data received at block 102. The method 800 proceeds from block 802 to block 804.
[0144] At block 804, the system evaluates the presence (or absence) of certain special conditions. To do so, in an illustrative embodiment, the system performs blocks 806, 808, 810, 812, and 814. At block 806, the system evaluates the presence of a special condition where the EWT prediction is zero because a waiting agent is available. At block 808, the system evaluates the presence of a special condition where the EWT prediction is undefined because zero agents are available to service the interaction. At block 810, the system evaluates the presence of a special condition where the average handling time (AHT) is used for EWT prediction at the initial startup of the system. At block 812, the system evaluates the presence of a condition where the default formula (i.e., relationship (2)) is used to make the EWT prediction because the wait cache is empty or unavailable. At block 814, the system evaluates the presence of a condition where control limits are applied to the EWT prediction due to a period of instability. Control limits are discussed in more detail below with reference to FIG. 9.
[0145] Although blocks 802-814 are described relatively serially, it should be understood that various blocks of method 800 may be performed in parallel in some embodiments.
[0146] 9, in use, the system may execute a method 900 of applying control limits when estimating contact center wait times. It should be appreciated that in some embodiments, the system may be embodied as a computing device (e.g., computing device 1100 of FIG. 11) and / or a contact center system (e.g., contact center system 1000 of FIG. 10) or systems / devices thereof. It should be appreciated that certain blocks of method 900 are illustrated by way of example, and that such blocks may be combined or divided, added or removed, and / or reordered in whole or in part, depending on the particular embodiment, unless stated to the contrary.
[0147] When making forecasts, certain situations may arise when the inputs to the formula do not follow general trends. Often, such situations arise when the system is highly chaotic and unstable. In some cases, the situation occurs at the beginning of a shift or during times when staff levels change rapidly. To avoid adversely affecting the forecast results, limits and control limits may be used to provide a more sensible forecast within established statistical ranges.
[0148] Statistical bounds can be formed from the actual wait times (AWT) of past interactions. Given an ordered set of past interaction wait times S{Wait Time}, upper and lower bounds can be derived from the following mathematical relationship:
[0149]
number
[0150]
number
[0151] In some embodiments, in addition to or as an alternative to calculating the control limits based on relationships (10)-(16), the latency of interactions stored in the wait cache may be used as a component of the final EWT prediction. In one example, the EWT prediction (i.e., during periods of instability) may be based on a percentage of the average actual wait time (AWT) obtained from historical data stored in the completed cache and a percentage of the current EWT prediction that corresponds to periods of normal operation. In at least some embodiments, the EWT prediction during periods of instability may be expressed and / or determined according to the following mathematical relationship:
[0152]
number
[0153] The illustrative method 900 begins at block 902, where the system obtains historical actual latency data (e.g., from a completed cache) during a period of instability. The method 900 proceeds from block 902 to block 904.
[0154] At block 904, the system determines control limits for the EWT forecast. In at least some embodiments, the control limits are determined according to relationships (10)-(15). In any event, in an illustrative embodiment, to perform block 904, the system performs blocks 906, 908, 910, 912, 914, and 916. At block 906, the system determines a median value based on the ordered set of previously serviced interactions (e.g., based on relationship (10)). At block 908, the system determines a lower quantile value for the set of previously serviced interactions (e.g., based on relationship (11)). At block 910, the system determines an upper quantile value for the set of previously serviced interactions (e.g., based on relationship (12)). At block 912, the system determines an interquartile range (IQR) for the set of previously serviced interactions (e.g., based on relationship (13)). At block 914, the system calculates a lower control limit for the set of previously serviced interactions (e.g., based on relationship (14)). At block 916, the system calculates an upper control limit for the set of previously serviced interactions (e.g., based on relationship (15)). Method 900 proceeds from block 904 to block 918.
[0155] At block 918, the system applies the control limits determined at block 904 to the EWT prediction. More specifically, in at least some embodiments, at block 918, the system limits the EWT prediction according to the control limits determined at block 904 (e.g., based on relationship (16)). Method 900 proceeds from block 918 to block 920.
[0156] At block 920, the system adjusts the EWT calculation based on the average actual latency and the current EWT prediction corresponding to periods of normal operation. In some embodiments, at block 920, the system adjusts the latency estimate (EWT instability ) In addition, in some embodiments, block 920 may be performed by the system in lieu of and / or as an alternative to performing blocks 904 and 918. Further, in some embodiments, block 920 may be omitted from method 900.
[0157] Although blocks 902-920 are described relatively serially, it should be understood that various blocks of method 900 may be performed in parallel in some embodiments.
[0158] In certain scenarios, the assumptions underlying the position in queue, agent pool, and position in adjusted AHT, as well as other characteristics of the interaction may change and thus result in a different EWT forecast than the initial forecast. One such scenario may be a change in staffing scenario where the service time of the interaction is long, or during a high load and / or understaffing scenario. Another scenario may be when the eligibility conditions for an interaction change due to changes in skill set, language, priority, etc.
[0159] In such scenarios, periodic wait time estimation updates can be provided to queued interactions using their updated position in the queue as well as the pool of available agents and the dynamically changing AHT estimator (AHT^) to provide an accurate EWT prediction. Another way to initially provide a more accurate EWT prediction is to anticipate the changes and incorporate the changed parameters into the initial wait time prediction. In the case of changing staffing levels (shifts), if agent shifts are known a priori, staffing information can be incorporated by recursively adjusting the number of agents with respect to the estimated wait times until they reach an equilibrium.
[0160] 10, there is shown a simplified block diagram of at least one embodiment of a communications infrastructure and / or contact center system that may be used in conjunction with one or more of the embodiments described herein. The contact center system 1000 may be embodied as any system capable of providing contact center services (e.g., call center services, chat center services, SMS center services, etc.) to end users and performing the functions otherwise described herein. The illustrative contact center system 1000 includes customer devices 1005, a network 1010, a switch / media gateway 1012, a call controller 1014, an interactive media response (IMR) server 1016, a routing server 1018, a storage device 1020, a statistics server 1026, agent devices 1030A, 1030B, 1030C, a media server 1034, a knowledge management server 1036, a knowledge system 1038, a chat server 1040, a web server 1042, an interaction (iXn) server 1044, a universal contact server 1046, a reporting server 1048, a media services server 1049, and an analytics module 1050.The illustrative embodiment of FIG. 10 includes one customer device 1005, one network 1010, one switch / media gateway 1012, one call controller 1014, one IMR server 1016, one routing server 1018, one storage device 1020, one statistics server 1026, one media server 1034, one knowledge management server 1036, one knowledge system 1038, one chat server 1040, one iXn server 1044, one universal contact server 1046, one reporting server 1048, one media services server 1049, and one analytics module 1050. Although only one customer device 1005 is shown, the contact center system 1000, in other embodiments, may include a plurality of customer devices 1005, a network 1010, a switch / media gateway 1012, a call controller 1014, an IMR server 1016, a routing server 1018, a storage device 1020, a statistics server 1026, a media server 1034, a knowledge management server 1036, a knowledge system 1038, a chat server 1040, an iXn server 1044, a universal contact server 1046, a reporting server 1048, a media services server 1049, and / or an analytics module 1050. Additionally, in some embodiments, one or more of the components described herein may be excluded from the system 1000, and one or more of the components described as being separate may form part of another component and / or one or more of the components described as forming part of another component may be separate.
[0161] It should be understood that, as used herein, the term "contact center system" is used to refer to the system depicted in Figure 10 and / or its components, while the term "contact center" is used more generally to refer to contact center systems, the customer service providers that operate these systems, and / or the organizations or companies associated therewith. Thus, unless specifically limited otherwise, the term "contact center" refers generally to a contact center system (such as contact center system 1000), associated customer service providers (such as a particular customer service provider that provides customer service through contact center system 1000), and the organization or company on behalf of which customer service is provided.
[0162] By way of background, customer service providers may offer many types of services through contact centers. Such contact centers may be staffed with employees or customer service agents (or simply "agents") who act as an interface between a company, enterprise, government agency, or organization (hereinafter interchangeably referred to as an "organization" or "enterprise") and people, such as users, individuals, or customers (hereinafter interchangeably referred to as an "individual" or "customer"). For example, contact center agents may assist customers in making purchasing decisions, placing orders, or resolving issues with products or services already received. Within a contact center, such interactions between contact center agents and external entities or customers may occur over a variety of communication channels, such as, for example, voice (e.g., telephone calls or voice over IP, i.e., VoIP calls), video (e.g., video conferencing), text (e.g., email and text chat), screen sharing, co-browsing, and / or other communication channels.
[0163] Operationally, contact centers generally strive to provide quality service to customers while minimizing costs. For example, one way that contact centers operate is to handle all customer interactions with live agents. While this approach may be fully successful from a service quality perspective, it would likely be prohibitively expensive due to the high cost of agent labor. For this reason, most contact centers utilize some level of automated processes, such as, for example, interactive voice response (IVR) systems, interactive media response (IMR) systems, Internet robots, or "bots," automated chat modules, or "chatbots," and / or other automated processes, in place of live agents. In many cases, this has proven to be a successful strategy, as automated processes can be highly efficient at handling certain types of interactions and effective at reducing the need for live agents. Such automation allows contact centers to target the use of human agents to more difficult customer interactions, while the automated processes handle the more repetitive or routine tasks. Furthermore, automated processes can be structured in a way that optimizes efficiency and promotes repeatability. Although a human agent, i.e., a live agent, may forget to answer a particular question or follow up on a particular detail, such mistakes are typically avoided through the use of automated processes. While customer service providers increasingly rely on automated processes to interact with customers, the use of such technology by customers remains far less developed. Thus, on the contact center side of the interaction, IVR systems, IMR systems, and / or bots are used to automate parts of the interaction, while actions on the customer's side remain manually performed by the customer.
[0164] It should be appreciated that the contact center system 1000 may be used by a customer service provider to provide various types of services to customers. For example, the contact center system 1000 may be used to engage and manage interactions in which automated processes (or bots) or human agents communicate with customers. As should be appreciated, the contact center system 1000 may be an in-house facility of a business or enterprise for performing sales and customer service functions related to products and services available through the enterprise. In another embodiment, the contact center system 1000 may be operated by a third-party service provider contracted to provide services on behalf of another organization. Additionally, the contact center system 1000 may be deployed on equipment dedicated to the enterprise or third-party service provider and / or in a remote computing environment, such as, for example, a private or public cloud environment with infrastructure to support multiple contact centers for multiple enterprises. The contact center system 1000 may include software applications or programs that may be executed on-premise or remotely, or any combination thereof. Additionally, it should be appreciated that various components of the contact center system 1000 may be distributed across various geographic locations and are not necessarily contained in a single location or computing environment.
[0165] Further, unless specifically limited otherwise, it should be understood that any of the computing elements described herein may be implemented within a cloud-based or cloud computing environment. As used herein and further described below with respect to computing device 1100, "cloud computing" or simply "cloud" is defined as a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned via virtualization, released with minimal management effort or service provider interaction, and then scaled accordingly. Cloud computing can be configured with a variety of characteristics (e.g., on-demand self-service, wide area network access, resource pooling, rapid elasticity, measurable services, etc.), service models (e.g., Software as a Service ("SaaS"), Platform as a Service ("PaaS"), Infrastructure as a Service ("IaaS"), and deployment models (e.g., private cloud, community cloud, public cloud, etc.). Cloud execution models, often referred to as "serverless architectures," generally involve a service provider dynamically managing the allocation and provisioning of remote servers to achieve desired functions.
[0166] It should be understood that any of the computer-implemented components, modules, or servers described in connection with Figure 10 may be implemented via one or more types of computing devices, such as, for example, computing device 1100 of Figure 11. As will be appreciated, the contact center system 1000 generally manages resources (e.g., personnel, computers, telecommunications equipment, etc.) to enable delivery of services via telephone, email, chat, or other communication mechanisms. Such services may vary depending on the type of contact center and may include, for example, customer service, help desk functions, emergency response, telemarketing, order taking, and / or other features.
[0167] A customer desiring to receive service from the contact center system 1000 may initiate inbound communications (e.g., telephone calls, emails, chats, etc.) to the contact center system 1000 via a customer device 1005. While FIG. 10 shows one such customer device, namely, the customer device 1005, it should be understood that there may be any number of customer devices 1005. The customer device 1005 may be, for example, a communication device such as a telephone, a smartphone, a computer, a tablet, or a laptop. In accordance with the functionality described herein, a customer may generally use the customer device 1005 to initiate, manage, and conduct communications with the contact center system 1000, such as telephone calls, emails, chats, text messages, web browsing sessions, and other multimedia transactions.
[0168] Inbound and outbound communications from and to the customer device 1005 may typically traverse the network 1010, with the nature of the network depending on the type of customer device and form of communication being used. By way of example, the network 1010 may include telephone, cellular, and / or data service communication networks. The network 1010 may be a private or public switched telephone network (PSTN), a local area network (LAN), a private wide area network (WAN), and / or a public WAN such as the Internet. Additionally, the network 1010 may include a wireless carrier network, including a code division multiple access (CDMA) network, a global system for mobile communications (GSM) network, or any wireless network / technology conventional in the art, including, but not limited to, 3G, 4G, LTE, 5G, etc.
[0169] The switch / media gateway 1012 may be coupled to the network 1010 for receiving and transmitting telephone calls between the customer and the contact center system 1000. The switch / media gateway 1012 may include a telephone switch or a communication switch configured to act as a central switch for agent-level routing within the center. The switch may be a hardware switching system or may be implemented via software. For example, the switch 1012 may include an automatic call distributor, a private branch exchange (PBX), an IP-based software switch, and / or any other switch having dedicated hardware and software configured to receive Internet-sourced and / or telephone network-sourced interactions from the customer and route these interactions, for example, to one of the agent devices 1030. Thus, in general, the switch / media gateway 1012 establishes a voice connection between the customer and the agent by establishing a connection between the customer device 1005 and the agent device 1030.
[0170] As further shown, the switch / media gateway 1012 may be coupled to a call controller 1014 that serves as an adapter or interface between, for example, the switch and other routing, monitoring, and communication processing components of the contact center system 1000. The call controller 1014 may be configured to process PSTN calls, VoIP calls, and / or other types of calls. For example, the call controller 1014 may include computer-telephone integration (CTI) software for interfacing with the switch / media gateway and other components. The call controller 1014 may include a session initiation protocol (SIP) server for processing SIP calls. The call controller 1014 may also extract data regarding incoming interactions, such as a customer's phone number, IP address, or email address, and then communicate these to other contact center components in processing the interaction.
[0171] The interactive media response (IMR) server 1016 may be configured to enable self-help or virtual assistant functions. Specifically, the IMR server 1016 may be similar to an interactive voice response (IVR) server, except that the IMR server 1016 is not limited to voice and may also cover a variety of media channels. In an example illustrating voice, the IMR server 1016 may be configured with IMR scripts to query customers for their needs. For example, a bank contact center may instruct customers via an IMR script to "press 1" if they want to look up their account balance. Through ongoing interaction with the IMR server 1016, customers may receive service without needing to speak with an agent. The IMR server 1016 may also be configured to ascertain the reason why a customer is contacting the contact center so that the communication may be routed to the appropriate resource. IMR configuration may be implemented through the use of self-service and / or assisted service tools, including web-based tools for developing IVR and routing applications that run within the contact center environment (e.g., Genesys® Designer).
[0172] The routing server 1018 may function to route incoming interactions. For example, once it is determined that an inbound communication should be handled by a human agent, functionality within the routing server 1018 may select the most appropriate agent and route the communication to that agent. This agent selection may be based on which available agent is best suited to handle the communication. More specifically, the selection of the appropriate agent may be based on a routing strategy or algorithm implemented by the routing server 1018. In doing so, the routing server 1018 may query data related to the incoming interaction, such as data related to the particular customer, the available agents, and the type of interaction, which may be stored in a particular database, as described herein. Once an agent is selected, the routing server 1018 may interact with the call controller 1014 to route (i.e., connect) the incoming interaction to a corresponding agent device 1030. As part of this connection, information about the customer may be provided to the selected agent via their agent device 1030. This information is intended to enhance the services that the agents can provide to the customer.
[0173] It should be appreciated that the contact center system 1000 may include one or more mass storage devices (generally represented by the storage device 1020) for storing data in one or more databases related to the contact center's functions. For example, the storage device 1020 may store customer data maintained in a customer database. Such customer data may include, for example, customer profiles, contact information, service level agreements (SLAs), and interaction history (e.g., details of previous interactions with a particular customer, including the nature of the previous interactions, disposition data, wait times, handling times, and actions taken by the contact center to resolve the customer's issues). As another example, the storage device 1020 may store agent data in an agent database. The agent data maintained by the contact center system 1000 may include, for example, agent availability and agent profiles, schedules, skills, handling times, and / or other related data. As another example, the storage device 1020 may store interaction data in an interaction database. The interaction data may include, for example, data relating to a number of past interactions between a customer and the contact center. More generally, unless otherwise specifically specified, it should be understood that the storage device 1020 may include databases and / or be configured to store data relating to any of the types of information described herein, and that these databases and / or data are accessible to other modules or servers of the contact center system 1000 in a manner that facilitates the functions described herein. For example, a server or module of the contact center system 1000 may query such databases to retrieve data stored therein or transmit data to a database for storage. The storage device 1020 may take the form of, for example, any conventional storage medium and may be housed locally or operated from a remote location.By way of example, the database may be a Cassandra database, a NoSQL database, or a SQL database, and may be managed by a database management system such as Oracle, IBM DB2, Microsoft SQL Server, Microsoft Access, PostgreSQL, or the like.
[0174] The statistics server 1026 may be configured to record and aggregate data related to the performance and operational aspects of the contact center system 1000. Such information may be compiled by the statistics server 1026 and made available to other servers and modules, such as the reporting server 1048, which may then use the data to generate reports used to manage operational aspects of the contact center and to perform automated actions in accordance with the functionality described herein. Such data may relate to the status of contact center resources, such as average wait times, abandon rates, agent occupancy, and others as may be required by the functionality described herein.
[0175] The agent devices 1030 of the contact center system 1000 may be communication devices configured to interact with various components and modules of the contact center system 1000 in a manner to facilitate the functionality described herein. For example, the agent devices 1030 may include telephones adapted for regular telephone calls or VoIP calls. The agent devices 1030 may further include computing devices configured to communicate with the servers of the contact center system 1000, perform data processing associated with operations, and interface with customers via voice, chat, email, and other multimedia communication mechanisms in accordance with the functionality described herein. Although FIG. 10 shows three such agent devices 1030, namely, agent devices 1030A, 1030B, and 1030C, it should be understood that in a particular embodiment, there may be any number of agent devices 1030.
[0176] The multimedia / social media server 1034 may be configured to facilitate media interactions (other than voice) with the customer device 1005 and / or the server 1042. Such media interactions may relate to, for example, email, voicemail, chat, video, text messaging, web, social media, co-browsing, etc. The multimedia / social media server 1034 may take the form of any IP router conventional in the art having dedicated hardware and software for receiving, processing, and forwarding multimedia events and communications.
[0177] The knowledge management server 1036 may be configured to facilitate interactions between customers and the knowledge system 1038. Generally, the knowledge system 1038 may be a computer system capable of receiving questions or queries and providing answers accordingly. The knowledge system 1038 may be included as part of the contact center system 1000 or may be operated remotely by a third party. The knowledge system 1038 may include an artificial intelligence computer system capable of answering questions posed in natural language by retrieving information from sources such as encyclopedias, dictionaries, newswire articles, literary works, or other documents submitted to the knowledge system 1038 as reference material. As an example, the knowledge system 1038 may be embodied as an IBM Watson or similar system.
[0178] The chat server 1040 may be configured to conduct, orchestrate, and manage electronic chat communications with customers. In general, the chat server 1040 is configured to implement and maintain chat conversations and generate chat transcripts. Such chat communications may be conducted by the chat server 1040 in such a manner that customers communicate with automated chatbots, human agents, or both. In an exemplary embodiment, the chat server 1040 may act as a chat orchestration server that dispatches chat conversations between chatbots and available human agents. In such a case, the processing logic of the chat server 1040 may be rule driven to leverage intelligent workload distribution among available chat resources. The chat server 1040 may further implement, manage, and facilitate user interfaces (UIs) associated with the chat functionality, including UIs generated on either the customer device 1005 or the agent device 1030. The chat server 1040 may be configured to transfer chats between automated and human sources within a single chat session with a particular customer, such that the chat session may be transferred from a chatbot to a human agent or from a human agent to a chatbot. The chat server 1040 may also be coupled to the knowledge management server 1036 and the knowledge system 1038 to receive suggestions and answers to inquiries posed by the customer during the chat, such that links to related articles may be provided.
[0179] A web server 1042 may be included to provide site hosts for various social interaction sites to which customers subscribe, such as Facebook, Twitter, Instagram, etc. Although depicted as part of the contact center system 1000, it should be understood that the web server 1042 may be provided by a third party and / or maintained remotely. The web server 1042 may also provide web pages of businesses or organizations supported by the contact center system 1000. For example, customers may browse web pages to receive information about products and services of a particular business. Within such business web pages, mechanisms may be provided for initiating interactions with the contact center system 1000, for example, via web chat, voice, or email. One example of such a mechanism is a widget that may be deployed on a web page or website hosted on the web server 1042. As used herein, a widget refers to a user interface component that performs a particular function. In some implementations, a widget may include a graphical user interface control that may be overlaid on a web page displayed to customers over the Internet. A widget may include buttons or other controls that show information in a window or text box, or allow a user to access a particular function, such as sharing or opening a file, or initiating a communication. In some implementations, a widget includes a user interface component that has a portable portion of code that can be installed and executed within a separate web page without compilation. Some widgets may include corresponding or additional user interfaces and may be configured to access various local resources (e.g., calendar or contact information on the customer device) or remote resources over a network (e.g., instant messaging, email, or social networking updates).
[0180] The interaction (iXn) server 1044 may be configured to manage the contact center's deferrable activities and the routing of the activities to human agents for completion. As used herein, deferrable activities include back office work that may be performed offline, such as responding to emails, attending training, and other activities that do not require real-time communication with customers. As an example, the interaction (iXn) server 1044 may be configured to interact with the routing server 1018 to select an appropriate agent to handle each of the deferrable activities. Once assigned to a particular agent, the deferrable activities are pushed to that agent, such that the deferrable activities are displayed on the selected agent's agent device 1030. The deferrable activities may be displayed in a work bin as a task for the selected agent to complete. The functionality of the work bin may be implemented via any conventional data structure, such as, for example, a linked list, an array, and / or other suitable data structures. Each of the agent devices 1030 may include a work bin. As an example, a work bin may be maintained in a buffer memory of the corresponding agent device 1030 .
[0181] A universal contact server (UCS) 1046 may be configured to retrieve information stored in a customer database and / or transmit information to a customer database for storage in the customer database. For example, UCS 1046 may be utilized as part of a chat function to facilitate maintaining a history of how chats with particular customers were handled, which may then be used as a reference for how future chat communications should be handled. More generally, UCS 1046 may be configured to facilitate maintaining a history of customer preferences, such as preferred media channels and best times to contact. To do this, UCS 1046 may be configured to identify data related to each customer's interaction history, such as, for example, data regarding comments from agents, customer communication history, and the like. Each of these data types may then be stored in a customer database or other module and retrieved as needed by the functionality described herein.
[0182] The reporting server 1048 may be configured to generate reports from data compiled and aggregated by the statistics server 1026 or other sources. Such reports may include near real-time or historical reports and may relate to the status of contact center resources and performance characteristics, such as, for example, average wait times, abandon rates, and / or agent occupancy. The reports may be generated automatically or in response to a specific request from a requestor (e.g., an agent, an administrator, a contact center application, etc.). The reports may then be used to manage the operation of the contact center in accordance with the functionality described herein.
[0183] The media services server 1049 may be configured to provide audio and / or video services to support contact center functions, which may include, according to functionality described herein, IVR or IMR system prompts (e.g., playing audio files), music on hold, voicemail / single-party recording, multi-party recording (e.g., of audio and / or video calls), voice recognition, dual tone multi frequency (DTMF) recognition, fax, audio and video transcoding, secure real-time transport protocol (SRTP), audio conferencing, video conferencing, coaching (e.g., support for a coach to eavesdrop on an interaction between a customer and an agent and for a coach to provide comments to an agent without the customer hearing the comments), call analysis, keyword spotting, and / or other related functions.
[0184] The analytics module 1050 may be configured to provide a system and method for performing analytics on data received from multiple different data sources, as may be required by the functionality described herein. According to an example embodiment, the analytics module 1050 may also generate, update, train, and modify predictors or models based on collected data, such as, for example, customer data, agent data, and interaction data. The models may include customer or agent behavior models. The behavior models may be used to predict, for example, customer or agent behavior in various situations, thereby enabling embodiments of the technology described herein to adjust interactions based on such predictions or allocate resources in preparation for predicted characteristics of future interactions, thereby improving overall contact center performance and customer experience. Although the analytics module is described as being part of a contact center, it will be understood that such behavior models may also be implemented in customer systems (or, as also used herein, the “customer side” of an interaction) and used to the benefit of the customer.
[0185] According to an example embodiment, the analytics module 1050 may have access to data stored in the storage device 1020, including a customer database and an agent database. The analytics module 1050 may also have access to an interaction database that stores data related to interactions and interaction content (e.g., transcripts of interactions and events detected therein), interaction metadata (e.g., customer identifiers, agent identifiers, medium of the interaction, length of the interaction, interaction start and end times, departments, tagged categories), and application settings (e.g., interaction paths through the contact center). Additionally, the analytics module 1050 may be configured to search the data stored in the storage device 1020 for use in developing and training algorithms and models, for example, by applying machine learning techniques.
[0186] One or more of the included models may be configured to predict customer or agent behavior and / or aspects related to contact center operation and performance. Additionally, one or more of the models may be used for natural language processing, including, for example, intent recognition, and the like. The models may be developed based on known first principles equations describing the system, data resulting in empirical models, or a combination of known first principles equations and data. In developing models for use in the present embodiments, it may generally be preferred to build empirical models based on collected and stored data, since first principles equations are often not available or easily derived. In order to adequately capture the relationships between manipulated / disturbance variables and controlled variables of a complex system, it may be preferred in some embodiments for the models to be nonlinear. This is because nonlinear models may exhibit curvilinear relationships, rather than linear relationships, between manipulated / disturbance variables and controlled variables, which are common for complex systems such as those discussed herein. Considering the aforementioned requirements, machine learning or neural network-based approaches may be preferred embodiments for implementing the models. For example, neural networks may be developed based on empirical data using advanced regression algorithms.
[0187] The analysis module 1050 may further include an optimizer. As will be appreciated, an optimizer may be used to minimize a "cost function" objective to which a set of constraints are applied, where the cost function is a mathematical expression of a desired objective or system behavior. Because the model may be nonlinear, the optimizer may be a nonlinear programming optimizer. However, it is contemplated that the techniques described herein may be implemented using a variety of different types of optimization approaches, individually or in combination, including, but not limited to, linear programming, quadratic programming, mixed integer nonlinear programming, stochastic programming, global nonlinear programming, genetic algorithms, particle / swarm techniques, and the like.
[0188] According to some embodiments, the model and optimizer may be used together in an optimization system. For example, the analytics module 1050 may utilize the optimization system as part of an optimization process in which aspects of the contact center's performance and operations are optimized or at least enhanced. This may include, for example, features related to customer experience, agent experience, interaction routing, natural language processing, intent recognition, or other functionality related to automated processes.
[0189] The various components, modules, and / or servers of FIG. 10 (as well as other figures contained herein) may each include one or more processors that execute computer program instructions and interact with other system components to perform the various functions described herein. Such computer program instructions may be stored in memory implemented using standard memory devices such as, for example, random-access memory (RAM), or on other non-transitory computer-readable media such as, for example, CD-ROM, flash drive, etc. Although the functionality of each of the servers is described as being provided by a particular server, those skilled in the art should understand that the functionality of the various servers may be combined or consolidated into a single server, or the functionality of a particular server may be distributed across one or more other servers without departing from the scope of the technology described herein. Additionally, the terms "interaction" and "communication" are used interchangeably and generally refer to any real-time and non-real-time interaction using any communication channel, including, but not limited to, telephone calls (PSTN or VoIP calls), email, vmail, video, chat, screen sharing, text messages, social media messages, WebRTC calls, and the like. Access to and control of the components of the contact center system 1000 may be affected through user interfaces (UIs) that may be generated on the customer device 1005 and / or the agent device 1030. As previously mentioned, the contact center system 1000 may be operated as a hybrid system in which some or all components are hosted remotely, such as in a cloud-based or cloud computing environment. It should be understood that each of the devices of the contact center system 1000 may be embodied as part of, include, or form one or more computing devices similar to the computing device 1100 described below with reference to FIG. 11.
[0190] 11, a simplified block diagram of at least one embodiment of a computing device 1100 is shown. The illustrative computing device 1100 depicts at least one embodiment of each of the computing devices, systems, servicers, controllers, switches, gateways, engines, modules, and / or computing components (e.g., which may be collectively referred to interchangeably as computing devices, servers, or modules for brevity herein) described herein. For example, the various computing devices may be processes or threads running on one or more processors of one or more computing devices 1100 that may be executing computer program instructions and interacting with other system modules to perform various functions described herein. Unless otherwise specifically limited, functionality described in conjunction with multiple computing devices may be integrated into a single computing device, or various functionality described in conjunction with a single computing device may be distributed across several computing devices. Further, with respect to a computing system described herein, such as the contact center system 1000 of FIG. 10, the various servers and computer devices of the system may be located on a local computing device 1100 (e.g., on-site in the same physical location as the contact center agents), a remote computing device 1100 (e.g., off-site, i.e., in a cloud-based environment, or in a cloud computing environment, e.g., in a remote data center connected via a network), or some combination thereof.In some embodiments, functionality provided by a server located on an off-site computing device may be accessed and provided via a virtual private network (VPN) as if such server were on-site, or the functionality may be provided using software as a service (SaaS) accessed over the Internet using various protocols, e.g., by exchanging data via extensible markup language (XML), JSON, and / or the functionality may be otherwise accessed / utilized.
[0191] In some embodiments, the computing device 1100 may be embodied as a server, a desktop computer, a laptop computer, a tablet computer, a notebook, a netbook, an Ultrabook™, a mobile phone, a mobile computing device, a smartphone, a wearable computing device, a personal digital assistant, an Internet of Things (IoT) device, a processing system, a wireless access point, a router, a gateway, and / or any other computing, processing device, and / or communications device capable of performing the functions described herein.
[0192] The computing device 1100 includes a processing device 1102 that executes algorithms and / or processes data in accordance with operational logic 1108, an input / output device 1104 that enables communication between the computing device 1100 and one or more external devices 1110, and a memory 1106 that stores data received from the external device 1110 via, for example, the input / output device 1104.
[0193] The input / output devices 1104 enable the computing device 1100 to communicate with external devices 1110. For example, the input / output devices 1104 may include a transceiver, a network adapter, a network card, an interface, one or more communication ports (e.g., a USB port, a serial port, a parallel port, an analog port, a digital port, VGA, DVI, HDMI, FireWire, CAT5, or any other type of communication port or interface), and / or other communication circuitry. The communication circuitry of the computing device 1100 may be configured to perform such communication using any one or more communication technologies (e.g., wireless or wired communication) and associated protocols (e.g., Ethernet, Bluetooth, Wi-Fi, WiMAX, etc.) depending on the particular computing device 1100. The input / output devices 1104 may include hardware, software, and / or firmware suitable for implementing the techniques described herein.
[0194] The external device 1110 may be any type of device that allows data to be input or output from the computing device 1100. For example, in various embodiments, the external device 1110 may be embodied as one or more of the devices / systems described herein and / or portions thereof. Further, in some embodiments, the external device 1110 may be embodied as another computing device, a switch, a diagnostic tool, a controller, a printer, a display, an alarm, a peripheral device (e.g., a keyboard, a mouse, a touch screen display, etc.), and / or any other computing, processing, and / or communication device capable of performing the functions described herein. Further, it should be understood that in some embodiments, the external device 1110 may be integrated into the computing device 1100.
[0195] The processing device 1102 may be embodied as any type of processor capable of performing the functions described herein. In particular, the processing device 1102 may be embodied as one or more single-core or multi-core processors, microcontrollers, or other processors or processing / control circuits. For example, in some embodiments, the processing device 1102 may include or be embodied as an arithmetic logic unit (ALU), a central processing unit (CPU), a digital signal processor (DSP), a graphics processing unit (GPU), a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), and / or another suitable processor. The processing device 1102 may be of a programmable type, a dedicated hardwired state machine, or a combination thereof. A processing device 1102 having multiple processing units may utilize distributed processing, pipelined processing, and / or parallel processing in various embodiments. Further, the processing device 1102 may be dedicated to performing only the operations described herein, or may be utilized in one or more additional applications. In an illustrative embodiment, the processing device 1102 is programmable and executes algorithms and / or processes data according to operating logic 1108 defined by programming instructions (e.g., software or firmware) stored in memory 1106. Additionally or alternatively, the operating logic 1108 of the processing device 1102 may be defined at least in part by hardwired logic or other hardware. Further, the processing device 1102 may include one or more components of any type suitable for processing signals received from the input / output device 1104 or from other components or devices and providing a desired output signal.Such components may include digital circuitry, analog circuitry, or a combination thereof.
[0196] The memory 1106 may be one or more types of non-transitory computer-readable media, such as solid-state memory, electromagnetic memory, optical memory, or combinations thereof. Additionally, the memory 1106 may be volatile and / or non-volatile, and in some embodiments, some or all of the memory 1106 may be of a portable type, such as a disk, tape, memory stick, cartridge, and / or other suitable portable memory. In operation, the memory 1106 may store various data and software used during operation of the computing device 1100, such as an operating system, applications, programs, libraries, and drivers. It should be appreciated that the memory 1106 may store data that is manipulated by the operating logic 1108 of the processing device 1102, such as, for example, data representing signals received from and / or transmitted to the input / output devices 1104, in addition to or instead of storing programming instructions that define the operating logic 1108. As shown in FIG. 11, the memory 1106 may be included in and / or coupled to the processing device 1102, depending on the particular embodiment. For example, in some embodiments, the processing device 1102, the memory 1106, and / or other components of the computing device 1100 may form part of a system-on-a-chip (SoC) and be integrated into a single integrated circuit chip.
[0197] In some embodiments, various components of computing device 1100 (e.g., processing device 1102 and memory 1106) may be communicatively coupled via an input / output subsystem, which may be embodied as circuits and / or components for facilitating input / output operations with processing device 1102, memory 1106, and other components of computing device 1100. For example, the input / output subsystem may be embodied as or may otherwise include a memory controller hub, an input / output control hub, firmware devices, communications links (i.e., point-to-point links, bus links, wires, cables, light guides, printed circuit board traces, etc.), and / or other components and subsystems for facilitating input / output operations.
[0198] The computing device 1100 may, in other embodiments, include other or additional components, such as those typically found in a typical computing device (e.g., various input / output devices and / or other components). It should be further understood that one or more of the components of the computing device 1100 described herein may be distributed across multiple computing devices. In other words, the techniques described herein may be employed by a computing system including one or more computing devices. In addition, while only a single processing device 1102, I / O device 1104, and memory 1106 are illustratively shown in FIG. 11, it should be understood that a particular computing device 1100 may, in other embodiments, include multiple processing devices 1102, I / O devices 1104, and / or memory 1106. Furthermore, in some embodiments, two or more external devices 1110 may communicate with the computing device 1100.
[0199] The computing device 1100 may be one of multiple devices connected by a network or connected to other systems / resources via a network. The network may be embodied as any one or more types of communication networks capable of facilitating communication between various devices communicatively connected via the network. Thus, the network may include one or more networks, routers, switches, access points, hubs, computers, client devices, endpoints, and / or other intervening network devices. For example, the network may be embodied as or otherwise include one or more cellular networks, telephone networks, local or wide area networks, publicly available global networks (e.g., the Internet), ad-hoc networks, short range communication links, or combinations thereof. In some embodiments, the network may include circuit-switched voice or data networks, packet-switched voice or data networks, and / or any other network capable of carrying voice and / or data. In particular, in some embodiments, the network may include an Internet Protocol (IP)-based network and / or an asynchronous transfer mode (ATM)-based network. In some embodiments, the network may handle voice traffic (e.g., via a Voice over IP (VOIP) network), web traffic, and / or other network traffic, depending on the particular embodiment and / or the devices of the system that communicate with each other.In various embodiments, the networks may include analog or digital wired and wireless networks (e.g., IEEE 802.11 networks, Public Switched Telephone Networks (PSTN), Integrated Services Digital Networks (ISDN), and Digital Subscriber Lines (xDSL)), Third Generation (3G) mobile networks, Fourth Generation (4G) mobile networks, Fifth Generation (5G) mobile networks, wired Ethernet networks, private networks (e.g., intranets, etc.), radio, television, cable, satellite, and / or any other distribution or tunneling mechanism for carrying data, or any suitable combination of such networks. It should be understood that various devices / systems may communicate with each other over different networks depending on the source and / or destination device / system.
[0200] It should be understood that the computing device 1100 may communicate with other computing devices 1100 via any type of gateway or tunneling protocol, such as Secure Socket Layer or Transport Layer Security. The network interface may include a built-in network adapter, such as a network interface card, suitable for interfacing the computing device to any type of network capable of performing the operations described herein. Furthermore, the network environment may be a virtual network environment in which various network components are virtualized. For example, the various machines may be virtual machines implemented as software-based computers running on a physical machine. The virtual machines may share the same operating system, or in other embodiments, different operating systems may run on each virtual machine instance. For example, a "hypervisor" type of virtualization is used in which multiple virtual machines run on the same host physical machine, each functioning as if it had its own dedicated box. Other types of virtualization may be employed in other embodiments, such as networks (e.g., via software-defined networking) or functions (e.g., via network function virtualization).
[0201] Thereby, one or more of the computing devices 1100 described herein may be embodied as or form a part of one or more cloud-based systems. In a cloud-based embodiment, the cloud-based system may be embodied, for example, as a server-ambiguous computing solution that executes instructions on demand, executes instructions only when prompted by a particular activity / trigger, and does not consume computing resources when not in use. That is, the system may be embodied as a virtual computing environment that exists "on" the computing system (e.g., a distributed network of devices) where various virtual functions (e.g., Lambda functions, Azure functions, Google Cloud functions, and / or other suitable virtual functions) may be executed corresponding to the functionality of the system described herein. For example, when an event occurs (e.g., data is transferred to the system for processing), the virtual computing environment may be communicated (e.g., via a request to the virtual computing environment's API), which may then route the request to the correct virtual function (e.g., a particular server-ambiguous computing resource) based on a set of rules. Thus, when a request for transmission of data is made by a user (eg, via an appropriate user interface to the system), an appropriate virtual function may be executed to perform the action before deleting the instance of the virtual function.
[0202] Referring now to FIG. 12, the robust estimation system was tested using observations from 73 queues from a contact center setting such that wait time estimation for 338,084 conversations was performed within a 7-day period. The actual wait times were recorded as a baseline to compare the estimates made using the industry standard formula with those made using the robust estimation system. The comparison depicted in FIG. 12 is an aggregated time-average comparison. It was found that the robust estimation system tracks the actual average wait time better, while the industry standard formula underestimates the actual average wait time most of the time.
[0203] 13 and 14, distribution bar graphs of both the industry standard and robust estimation prediction errors are shown bucketing the results into five buckets (below -75%, -25% to -75%, -25% to +25%, +25% to 75%, and above 75%). As shown in Figs. 13 and 14, the robust estimation system produces an accuracy distribution that is much more centered with 46% of the observations falling between -25% and +25% accuracy, while only 34% of the industry standard estimations are observed to fall in the same accuracy band. Furthermore, the median percentage error of the robust estimation system is only off by -2% compared to the industry standard of -8%.
[0204] Referring now to FIG. 15, this figure shows the measures of central tendency of both approaches in a boxplot, which further confirms that the robust estimation system is more accurate and precise in estimating waiting times compared to the industry standard approach.
Claims
1. 1. A system for estimating a wait time for connecting a customer to an agent, comprising: At least one processor; At least one memory containing a plurality of instructions stored therein, the instructions, upon execution by the at least one processor, providing the system with: receiving incoming interaction data from the customer indicating a request to estimate the wait time; responsive to receiving the request, retrieving historical data indicative of interaction information for one or more interactions already serviced; generating a latency estimate based on the historical data; and determining whether the number of conversation identifiers stored in the wait cache that have no actual wait time exceeds a reference value; and in response to determining that a number of conversation identifiers stored in the wait cache that have no actual wait time exceeds the reference value, removing the conversation identifier from the wait cache.
2. 2. The system of claim 1, wherein the interaction data includes a plurality of route paths, each defined by one or more of: a media type associated with the request, data stored in a queue associated with one or more previous interactions, a language associated with the request, one or more skill requirements associated with the request, a position of the request in the queue, a number of agents available to service the queue, a time of the request, or a time of service of the request.
3. In response to execution by the at least one processor, the instructions further provide the system with: maintaining the waiting cache for each of the plurality of route paths to store interaction data associated with one or more interactions that have not yet been serviced; and maintaining a completed cache for each of the plurality of route paths to store interaction data associated with one or more interactions that have already been serviced.
4. Generating the latency estimate based on the historical data includes: determining an actual wait time corresponding to one or more interactions already serviced; determining a position in the queue of one or more interactions that have already been serviced; determining a number of available agents to serve one or more interactions already being serviced; and calculating an adjusted average handle time corresponding to one or more interactions already serviced based on the determined actual wait time, the determined position in queue, and the determined number of available agents.
5. 5. The system of claim 4, wherein generating the wait time estimate based on the historical data includes calculating an average processing time estimator based on one or more of a median of adjusted average processing time calculations or a mean of adjusted average processing time calculations.
6. Generating the latency estimate based on the historical data includes: determining a position in a queue corresponding to the request; determining a number of available agents to serve the request; and calculating the wait time estimate based on the average processing time estimate, the determined position in a queue corresponding to the request, and the determined number of available agents corresponding to the request.
7. The system of claim 1 , wherein generating the latency estimate based on the historical data comprises calculating an estimated interval of latency.
8. Calculating the estimated interval of waiting time determining a sample number of one or more interactions that have already been serviced; determining a statistical significance level associated with the estimation interval; determining a lower quantile of the estimation interval; determining an upper quantile of the estimation interval; calculating the lower bound of the estimation interval based at least in part on an adjusted average processing time corresponding to the determined lower quantile; and calculating the upper limit of the estimation interval based at least in part on an adjusted average processing time corresponding to the determined upper quantile.
9. One or more non-transitory machine-readable storage media including a plurality of instructions stored thereon, the instructions being responsive to execution by at least one processor to cause a system to: receiving incoming interaction data from a customer indicating a request to estimate a wait time for connecting the customer to an agent; responsive to receiving the request, retrieving historical data indicative of interaction information for one or more interactions already serviced; generating a wait time estimate based on the historical data for connecting the customer to the agent; determining whether the number of conversation identifiers stored in the wait cache that have no actual wait time exceeds a reference value; and in response to determining that the number of the conversation identifiers stored in the wait cache that do not have an actual wait time exceeds the reference value, removing the conversation identifier from the wait cache.
10. 10. The one or more non-transitory machine-readable storage media of claim 9, wherein the interaction data includes a plurality of route paths each defined by one or more of: a media type associated with the request, data stored in a queue associated with one or more previous interactions, a language associated with the request, one or more skill requirements associated with the request, a position of the request in the queue, a number of agents available to service the queue, a time of the request, or a time of service of the request.
11. Generating the latency estimate based on the historical data includes: determining an actual wait time corresponding to one or more interactions already serviced; determining a position in the queue of one or more interactions that have already been serviced; determining a number of available agents to serve one or more interactions already being serviced; and calculating an adjusted average handling time corresponding to one or more interactions already serviced based on the determined actual wait time, the determined position in queue, and the determined number of available agents.
12. 12. The one or more non-transitory machine-readable storage media of claim 11, wherein generating the wait time estimate based on the historical data comprises calculating an average processing time estimator based on one or more of a median of adjusted average processing time calculations or a mean of adjusted average processing time calculations.
13. Generating the latency estimate based on the historical data includes: determining a position in a queue corresponding to the request; determining a number of available agents to serve the request; and calculating the wait time estimate based on the average processing time estimate, the determined position in a queue corresponding to the request, and the determined number of available agents corresponding to the request.
14. Generating the latency estimate based on the historical data includes calculating an estimated interval of latency, the calculating the estimated interval of latency comprising: determining a sample number of one or more interactions that have already been serviced; determining a statistical significance level associated with the estimation interval; determining a lower quantile of the estimation interval; determining an upper quantile of the estimation interval; calculating the lower bound of the estimation interval based at least in part on an adjusted average processing time corresponding to the determined lower quantile; and calculating the upper limit of the estimation interval based at least in part on an adjusted average processing time that corresponds to the determined upper quantile.
15. 1. A method for estimating contact center wait time for connecting a customer to an agent, the method comprising: receiving, by a contact center system, incoming interaction data from the customer indicating a request to estimate the wait time; retrieving, by the contact center system, in response to receiving the request, historical data indicative of interaction information for one or more interactions already serviced; generating, by the contact center system, a wait time estimate based on the historical data; determining, by the contact center system, whether a number of conversation identifiers stored in a wait cache that do not have an actual wait time exceeds a threshold value; and in response to determining that the number of the conversation identifiers stored in the wait cache that do not have an actual wait time exceeds the reference value, removing the conversation identifier from the wait cache by the contact center system.
16. 16. The method of claim 15, wherein the interaction data includes a plurality of route paths, each defined by one or more of: a media type associated with the request, data stored in a queue associated with one or more previous interactions, a language associated with the request, one or more skill requirements associated with the request, a position of the request in the queue, a number of agents available to service the queue, a time of the request, or a time of service of the request.
17. Generating the latency estimate based on the historical data includes: determining an actual wait time corresponding to one or more interactions already serviced by the contact center system; determining a position in a queue of one or more interactions already being serviced by the contact center system; determining, by the contact center system, a number of available agents to serve one or more interactions already being serviced; and calculating, by the contact center system, an adjusted average handle time corresponding to one or more interactions already serviced based on the determined actual wait time, the determined position in queue, and the determined number of available agents.
18. Generating the latency estimate based on the historical data includes: calculating, by the contact center system, an average handle time estimate based on a median of the adjusted average handle time calculations or based on a mean of the adjusted average handle time calculations; determining, by the contact center system, a position in a queue corresponding to the request; determining, by the contact center system, a number of agents available to service the request; and calculating, by the contact center system, the wait time estimate based on the average handling time estimate, the determined position in a queue serving the request, and the determined number of available agents serving the request.
Citation Information
Patent Citations
Communication terminal, exchange, telephone network, and connection assurance method in the telephone network
JP2000349907A
Automatic voice answering device
JP2011061410A
Telephone exchange apparatus
JP2011061661A
