FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
(See section 10, rule 13)
1. Title of the invention: PREDICTING PERFORMANCE OF A CALL CENTER
2. Applicants)
NAME NATIONALITY ADDRESS
TATA CONSULTANCY Nirmal Building, 9th Floor, Nariman Point,
TATA CONSULTANCY
SERVICES LIMITED Indian Maharashtra 400021
India
3. Preamble to the description
COMPLETE SPECIFICATION
The following specification particularly describes the invention and the manner in which it
is to be performed.

TECHNICAL FIELD
[0001] The present subject matter relates, in general, to computing systems and, in
particular, to systems and methods for predicting performance of a call center.
BACKGROUND
[0002] Certain organizations, such as call centers or business process outsourcing
(BPO) units, henceforth collectively referred to as call centers, are dedicated for the purpose of receiving, forwarding and catering to a large number of requests, usually by telephone, henceforth referred to as call requests. The incoming call requests may pertain to support for products or information inquiries from consumers, whereas the outgoing call requests are usually for activities, such as telemarketing, debt collection, etc. Conventionally, a call center is operated through an extensive workspace, for call center agents, with work stations, that usually include a computing device for each agent, a communication device such as a headset or telecom unit connected to a telecom switch, and one or more supervisor stations. Further, a call center may either operate independently or may be networked with other call centers usually over a communication network.
[0003] Some call centers also have a specialized division known as contact center to
handle postal letters, faxes, live chat, instant messaging and e-mails. The contact center manages all the communications of the organization(s) to which the call center caters to. The contact center also routes and tracks communication to the intended recipient, manages contact information and sometimes even caters to the organization(s)' customer relationship management (CRM) process.
SUMMARY
[0004] This summary is provided to introduce concepts related to systems and methods
for predicting performance of a call center and the concepts are further described below in the detailed description. This summary is neither intended to identify essential features of the claimed subject matter nor is it intended for use in determining or limiting the scope of the claimed subject matter.

[0005] In one implementation, the method of predicting performance in a call center
comprises receiving details pertaining to at least one resource of the call center so as to generate at least one resource model representative of the at least one resource based on the received details. The method further comprises receiving at least one workflow parameter associated with the at least one resource model and generating at least one report indicative of the quality of service provided by the call center based on the at least one workflow parameter.
BRIEF DESCRIPTION OF THE DRAWINGS
[0006] The detailed description is described with reference to the accompanying figures.
In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the drawings to reference like features and components.
[0007] Figure 1 illustrates network environment implementing a call center performance
prediction system, in accordance with an embodiment of the present subject matter.
[0008] Figure 2 illustrates a simulated model of a call center as generated by the call
center performance prediction system, in accordance with an implementation of the present subject matter.
[0009] Figure 3 illustrates a method of predicting performance in a call center, in
accordance with an implementation of the present subject matter.
DETAILED DESCRIPTION
[00010] Systems and methods for predicting performance of a call center are described
herein. The systems and methods can be implemented in a variety of computing systems. The computing systems that can implement the described method(s) include, but are not restricted to, mainframe computers, workstations, personal computers, desktop computers, minicomputers, servers, multiprocessor systems, laptops, mobile computing devices and the like.

[00011] A call center, including a business process outsourcing (BPO) organization, is an independent organization or a department of an organization to administer incoming call requests, such as product support or information inquiries from consumers. Outgoing calls made from a call center are usually for telemarketing, debt collection, etc. In addition, a call center may also handle collective handling of letters, faxes, live chat, and e-mails pertaining to one or more organizations.
[00012] Typically, customers and consumers initiate the incoming call requests to the call center through a public switched telephone network (PSTN). The PSTN is often connected to a private automatic branch exchange (PABX) of the call center though one or more trunk lines. If at the time of arrival of an incoming request by a customer all trunk lines are in use, i.e., busy, the call may be blocked, else the call is usually answered by an Interactive Voice Response (IVR) unit. The IVR unit is implemented through a computing device pre-configured to interact with customers through the use of voice and keyboard inputs. In certain cases, the customers may be able to obtain the requisite service through the IVR unit.
[00013] In other cases, the calls are routed from the IVR unit to an agent based on various call routing parameters such as category of the incoming call request, skill set and expertise of the agent. If a suitable agent is not available for servicing the incoming call request, the call request is placed in a queue.
[00014] Generally, in call centers, the agents are categorized according to their skill set into various groups indicative of the categories of calls the agents can cater to. In such cases, skill-based routing (SBR), which involve defining rules in an automatic call distributor (ACD), route incoming call requests to the agents. Since the call centers usually provide support to customers and consumers of various time zones, the agents in the call center usually work in shifts such as morning shift, evening shift, and night shift. A call center supervisor is usually responsible for staffing of the call center and to ensure that the operations of the call center meet the pre-specified quality standards.
[00015] The operations of a call center are measured based on various performance metrics such as blockage index indicative of the percentage of incoming call requests which

would be dropped due to insufficient resources such as agents, computing systems, trunk lines, etc. Other performance metrics include abandon rate indicative of the percentage of incoming call requests which are abandoned by the customer while waiting to be serviced; average speed of service indicative of the average time, say in seconds, taken to service an incoming call request; service level index indicative of the percentage of incoming call services within a specified time interval.
[00016] Furthermore, various other performance parameters are also used to determine the quality of operation of a call center. These performance parameters include agent occupancy index indicative of the ratio of actual time spent by an agent in processing incoming call requests to actual time interval during which the agent is idle; schedule adherence index indicative of the degree to which the specified working hours are actually utilized by the agents; staff shrinkage index indicative of the duration of time interval for which an agent is unavailable for servicing incoming call requests due to training, breaks, etc.; cost per call indicative of the resources consumed for servicing an incoming call request. Examples of such resources include, but are not limited to, cost of manpower including the agents, cost of facilities of the call center, etc.
[00017] Conventionally, a specialized branch of mathematics, queuing theory, is used to model the operations of the call center. In such models, the call center is modeled as a queuing network. Such systems use a numerical value followed by "Erlangs" to represent the average number of concurrent calls carried out by the service-providing elements, wherein the average is calculated over some reasonable period of time, e.g., 15 minutes. This measure is especially used where short spurts of demand of resources, as usually seen in the operation of the call center, are very common.
[00018] Other conventionally known systems may be based on Erlang B Model, employing an exponentially distributed patience time interval of the customer is associated with each incoming call request. Each incoming call request is associated with an exponentially distributed patience time indicative of the time interval for which a customer would wait for service. The said technique also includes an assumption that each incoming call request would be offered a waiting time, i.e., the time that the customer generating an

incoming call request would have to wait given that the customer's patience is infinite. If the wait time exceeds the customer's patience time, the incoming call request would be abandoned by the customer else the customer waits to be serviced.
[00019] Other conventional systems compute the probability, of a customer generating an incoming request, having to wait in a queuing model. In said model, an infinite number of sources, mimicking customers generating incoming call requests, jointly generate traffic in form of incoming call requests for the agents of the call center. Hence, if all the agents are busy when an incoming call request arrives from a source, the incoming call request is placed in a queue. Thus, it is possible for an unlimited number of incoming call requests to be held in the queue simultaneously. This model, referred to as the Erlang C model, facilitates determination of the number of agents needed to staff a call center, for a specified or desired probability of queuing based on various assumptions, for example the incoming call request traffic can be modeled by a Poisson process, the waiting times can be depicted by a negative exponential distribution. As the Poisson process is a stochastic process, in the above model each source generates traffic or operates independent of the operations of the other sources.
[00020] The conventional methods of simulating the operation of call centers also assume that the call requests follow a first in first out (FIFO) model which does not have the provision of prioritizing an incoming call request. Further, the expected time to service an incoming call request is different for each type of call and is also dependent on the skill set, experience, etc., of the agent which is not taken into account in the conventional methods.
[00021] The conventional methods of analyzing the operations of a call center also do not have the provision for call routing based on skills required to service an incoming call request, which is very common in the practical scenario, as not all agents can cater to all types of incoming call requests. Some call centers may have dedicated agents for each category of incoming call requests. Further, certain agents have multiple skills set so as to handle multiple categories of incoming call requests.
[00022] Also the conventional methods of analyzing the operations of a call center assume that no incoming call requests are dropped and there are no repeated attempts to

connect to a call center. Hence the staffing requirements generated by conventional techniques are inaccurate and do not provide a clear indication of how agents should be allocated work in the call center so as to meet the desired quality standards.
[00023] The present subject matter discloses systems and methods for predicting performance of a call center so as to assist the call center supervisor in staffing the call center and in ensuring that the operations of the call center meet the pre-specified quality standards. In one implementation, the method comprises of generating a performance report indicative of the time taken to service an incoming call request for each category of incoming call request, utilization of each agent, queue sizes for each category of call, etc.
[00024] In one embodiment, the proposed method facilitates an user, such as the call center supervisor, to generate a modeling and simulation environment of the call center in which the user defines various resource models for the simulation such as the model of the customers, the model of an interactive voice response unit, henceforth referred to as the IVR unit, the model of an automatic call distributor (ACD), the model of a private automatic branch exchange (PABX), the model of the agents, and so on. The user may define various resource attributes for each resource model, indicative of the behavior or the characteristics of the resource depicted by the resource model. The user may further defines workflow parameters, indicative of the processes followed in the call center, using connectors and assigns values to each resource model. Based on the simulated model of the call center, user may generate various test scenarios and - optimize the operations
[00025] By varying the resource attributes associated with the resource models or the workflow parameters, the user may simulate the operations of the call center for various test scenarios and predict performance of the call center. Based on the simulation, various reports indicative of the operations of the call center may be generated. The reports may be indicative of the time taken to service an incoming call requests for each category of incoming call requests, utilization of each agent, queue sizes for each category of incoming call requests, etc. Further, the reports may also include a staffing index so as to optimize the staffing, work allocation, work timings, scheduling of training sessions of agents, etc.

[00026] While aspects of described systems and methods of predicting performance of
a call center can be implemented in any number of different computing systems, environments, and/or configurations, the embodiments are described in the context of the following exemplary system(s).
[00027] Figure 1 illustrates a network environment 100 implementing a call center
performance prediction system 102, henceforth referred to as the CCPP system 102, configured to simulate the operations of a call center so as to assist a user, such as a call center supervisor, in scheduling the call center agents according to an embodiment of the present subject matter. In one implementation, the CCPP system 102 may be included within an existing information technology infrastructure of the call center. The CCPP system 102 may be implemented in a variety of computing systems such as a laptop computer, a desktop computer, a notebook, a workstation, a mainframe computer, a server, a network server and the like. It will be understood that the CCPP system 102 may be accessed by users, i.e. the call center supervisors, call center planners, stakeholders of the call center, through one or more client devices 104 or applications residing on client devices 104. Examples of the client devices 104 include, but are not limited to, a portable computer 104-1, a personal digital assistant 104-2, a handheld device 104-3, and a workstation 104-N. As shown in the figure, such client devices 104 are communicatively coupled to the CCPP system 102 through a network 106 for facilitating one or more users.
[00028] The network 106 may be a wireless network, wired network or a combination
thereof. The network 106 can be implemented as one of the different types of networks, such as intranet, local area network (LAN), wide area network (WAN), the internet, and such. The network 106 may either be a.dedicated network or a shared network, which represents an association of the different types of networks that use a variety of protocols, for example, Hypertext Transfer Protocol (HTTP), Transmission Control Protocol/Internet Protocol (TCP/IP), Wireless Application Protocol (WAP), etc., to communicate with each other. Further the network 106 may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, etc.

[00029] In one implementation, the CCPP system 102 includes a processor(s) 108,
interface(s) 110, and a memory 112 coupled to the processor(s) 108. The processor(s) 108 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor(s) 108 are configured to fetch and execute computer-readable instructions stored in the memory 112.
[00030] The interface(s) 110 may include a variety of software and hardware
interfaces, for example, a web interface, a graphical user interface, etc., allowing the CCPP system 102 to interact with the client devices 104. Further, the interface(s) 110 may enable the CCPP system 102 to communicate with other computing devices, such as web servers and external data servers (not shown in figure). The interface(s) 110 can facilitate multiple communications within a wide variety of networks and protocol types, including wired networks, for example LAN, cable, etc., and wireless networks such as WLAN, cellular, or satellite. The interface(s) 110 may include one or more ports for connecting a number of devices to each other or to another server.
[00031] The memory 112 can include any computer-readable medium known in the art
including, for example, volatile memory (e.g., RAM), and/or non-volatile memory (e.g., EPROM, flash memory, etc.). In one embodiment, the memory 112 includes module(s) 114 and data 116. The modules 114 further include a resource management module 118, a workflow management module 120, a report generation module 122, and other module(s) 124. It will be appreciated that such modules may be represented as a single module or a combination of different modules. Additionally, the memory 112 further includes data 116 that serves, amongst other things, as a repository for storing data fetched, processed, received and generated by one or more of the modules 114. The data 116 includes, for example, operational data 126, workflow data 128, and other data 130. In one embodiment, the operational data 126, the workflow data 128, and the other data 130, may be stored in the memory 112 in the form of data structures. Additionally, the aforementioned data can be organized using data models, such as relational or hierarchical data models.

[00032] In one implementation, the resource management module 118 is configured to
receive the various details and resource details pertaining to the call center from the user such as the call center supervisor. The various details may include the category of resources such as manpower, information technology infrastructure, communication network infrastructure; the type of entity in each category of resources, such as agents, customers, communication systems and computing systems, of the call center. In said implementation, the resource management module 118 is configured to generate one or more resource models based on the received details. For example, the resource management module 118 may generate a first resource model to represent the customers, a second resource model indicative of the agents, a third resource model for the communication network infrastructure of the call center and so on. In another implementation, the resource management module 118 may generate a separate resource model for each entity of each category of the resources. For example, in said implementation, the resource management module 118 may generate a first resource model for an IVR unit of the communication network infrastructure, a second resource model for a PABX of the communication network infrastructure, a third resource model for trunk lines of the communication network infrastructure and so on.
[00033] The resource management module 118 is further configured to prompt the user
to enter various resource attributes to the generated resource models. For example, the user may define various customer resource attributes, such as patience time, time interval between redials, which control the behavior of the resource model depicting the customers. In another example, various agent resource attributes, such as efficiency factor, skill set index, experience index, proficiency index, may control or influence the behavior of the resource model depicting the agents.
[00034] In one implementation, the workflow management module 120 is configured to
receive the workflow parameters from the user. The workflow parameters may be indicative of a series or a sequence of connected or related activities to accomplish a task, in this case service an incoming call request. The workflow parameters may depict a sequence of actions, which may be assigned to a single agent or a team of agent based the skill sets, expertise, and proficiency of the agent or the team of agents. In said implementation the workflow

management module 120 analyzes the workflow parameters so as to generate a simulated model of operations of the call center. The simulated model of the call center may have various resources allocated for various categories of work. The simulated model would also represent the business process adopted by the call center, a systematic organization of resources, defined roles of agents, path of flow of the incoming call request, etc. Yet in another implementation, the workflow parameters may be used to depict the organizational structure of the call center, such as functions, teams, projects, policies and hierarchies.
[00035] In one implementation, the report generation module 122 is configured to
utilize the simulated model of the call center to predict the actual operation and performance of the call center. In one implementation, the report generation module 122 may include various other sub modules (not shown in figure), such as random number generator, event scheduler, router, to facilitate the simulation. The simulation would be indicative of the actual operation of the call center.
[00036] For example, the report generation module 122 may trigger the resource model
representing customers to generate a specified number of call requests in various distribution patterns. In one implementation, the specified number and the distribution may be entered by the user, say in form of a file such as spreadsheet, flat file, data file. In another implementation, the report generation module 122 may be configured to use any standard distribution models, such as Bernoulli, Beta, Erlang, Exponential, Gamma, Geometric, Hotset, Hyper-exponential, Hyper-geometric and Zipf as the distribution pattern of incoming call requests. Further, as indicated by the user-specified resource attribute, the resource model representing the trunk lines of the call center have a definite capacity, i.e. the resource model representing the trunk lines may handle a specified number of incoming call requests. If the number of incoming call requests exceeds the capacity of the resource model representing the trunk lines, the excess number of incoming call requests may be abandoned or dropped. The simulation process would be described in greater detail in conjunction with Figure 2.
[00037] The report generation module 122 is configured to simulate the operation of
the call center based on the resource models generated by the user, the resource attributes associated with the resource models and the workflow parameters. The report generation

module 122 may be further configured to retrieve various operational constraints from the operational data 126 such as maximum working hours per agent as is allowed by the law, breaks taken by the agents, training provided to the agent and account for the same during the simulation of the operation of the call center.
[00038] Based on the simulation, the report generation module 122 is configured to
generate a comprehensive report indicative of the time taken to service an incoming call requests for each category of incoming call requests, utilization of each agent, queue sizes for each category of incoming call requests, etc. The report also provides details about blocked calls, abandoned calls and wait times due to busy agents. The statistics include mean as well as percentile values for various handle and wait times.
[00039] In one implementation, the report generation module 122 may be configured to
generate suggestions for the call center supervisor on how the call center operations may be changed so as to comply with predefined quality standards such as globally recognized performance management frameworks for the call centers including standards prescribed by Customer Operations Performance Center Incorporation. (COPC). For example, the report generation module 122 may suggest changes in resource models or resource attributes or workflow parameters so as to comply with COPC-2000 Customer Service Provider (CSP) standards.
[00040] In another implementation, the CCPP system 102 may be communicatively
coupled to a computing system, such as a server 132, of the call center so as to retrieve the real-time or live data of operations of the call center. Based on the retrieved data, the CCPP system 102 may be configured to analyze and generate suggestions for the call center supervisor, indication actions which may enhance the quality of service provided by the CCPP system 102. Thus the CCPP system 102 facilitates simulation of a call center so as to predict the performance of the call center under various test conditions so as to enhance the efficiency of operations of the call center and to meet pre-defined quality standards.
[00041] Figure 2 illustrates a simulated model 200 of the call center as generated by the
resource generation module 122 of the CCPP system 102, in accordance with one

embodiment of the present subject matter. In one implementation, the simulated model 200 includes resource models in form of a plurality of incoming call request generators 202-1, 202-2, 202-3, ... , 202-N, henceforth collectively referred to as the ICR generators 202 and singularly referred to as the ICR generator 202. The resource management module 118 facilitates the ICR generators 202 to generate incoming call requests based on any standard distribution such as Bernoulli, Beta, Erlang, Exponential, Gamma, Geometric, Hotset, Hyper-exponential, Hyper-geometric and Zipf. The ICR generators 202 may also be configured to generate incoming call requests based on actual data provided by the user. The actual data may be provided by the user in form of a spreadsheet, data file, XML file, CSV file, flat file, etc. In one implementation, the ICR generator 202 may be configured to transform the generated distribution based on one or more transformation modes. For example, in one implementation, the ICR generator 202 may operate in a scale transformation mode, wherein the generated number of incoming call requests may be multiplied by a user defined scaling factor. The scaling factor may be any integer or fraction. In another implementation, the ICR generator 202 may operate in a translation transformation mode, wherein a user defined amount of translation may be added to the generated number of incoming call requests. In yet another implementation, the ICR generator 202 may operate in a truncate transformation mode, wherein the generated number of incoming call requests may be truncated if the generated number of incoming call requests is outside a user defined limit. In another example, the ICR generator 202 may operate in a bound transformation mode, wherein the generated number of incoming call requests may be regenerated if the generated number of incoming call requests is outside a user defined limit.
[00042] In one implementation, the resource attributes associated with the ICR
generators 202 may include arrival time interval indicative of the time interval between arrival of incoming call requests; percentage of high priority call requests indicative of the percentage of high priority incoming call requests or incoming call requests generated by high priority customers. The other resource attributes may include retrial percentage indicative of the percentage of customers, i.e. the ICR generators 202, who would regenerate the incoming call request if the previous incoming call request did not get serviced, patience time interval

indicative of the time interval a customer would wait for being serviced before abandoning the call.
[00043] The incoming call requests generated by the ICR generators 202 are forwarded
trough resource models representing trunk lines as indicated by arrows 204-1, 204-2, ..., 204-N, henceforth collectively referred to as the trunk lines 204 and singularly referred to as the trunk line 204. It is well known by those acquainted with the art that each call center has a specified and limited number of trunk lines 204. The trunk lines 204 are communicatively coupled to a resource model depicting private automatic branch exchange 206, henceforth referred to as the PABX 206. In one implementation the communicative coupling of the trunk line 204 with the PABX 206 is defined by the user using the workflow parameters.
[00044] If all the trunk lines 204 are in use, i.e. busy, the incoming call request is
abandoned or dropped. In the simulated model 200, the abandoned or dropped incoming call requests are moved, as shown by arrow marked 208, to a resource model depicting a call sink 210. The call sink 210 may be modeled as a data repository which stores the details pertaining to the dropped or abandoned incoming call requests. If any of the trunk lines are free, the incoming call request is received at the PABX 206 and, as indicated by arrow marked 212, is forwarded to a resource model depicting an Interactive Voice Response Unit 214, henceforth referred to as the IVR unit 214. The IVR unit 214 is usually a computing device that is configured to interact with humans through the use of voice and dual tone multi frequency (DTMF) keypad inputs. In one implementation, the IVR unit 214 may be configured to interact with the customers generating the incoming call request based on speech recognition. Further, based on the resource attributes associated with the IVR unit 214, the IVR unit 214 may interact with the customers to service the incoming call request using prerecorded or dynamically generated audio to direct customers. As mentioned earlier, in certain cases, the incoming call request may be completely serviced at the IVR unit 214. In such cases, the completed incoming call requests are moved, as shown by arrow marked 216, to the call sink 210. In one example, the service time of the IVR Unit 214, indicative of the distribution of service time or time spent by the ICR generator 202 at the IVR unit 214 before being

forwarded to an agent, is taken as a performance parameter to determine the quality of service offered by the call center to customers generating incoming call requests.
[00045] If the incoming call request could not be serviced completely at the IVR unit
216, the same may be forwarded, as indicated by arrow marked 218, to a resource model depicting an automatic call distributor unit 220, henceforth referred to as the ACD unit 220. In one implementation, the ACD unit 220 is configured to distribute incoming call requests to a specific group of computing or communication devices used by the agents. In another implementation, the ACD unit 220 may be incorporated with existing computer telephony integration (CTI) system of the call center. In said implementations, the ACD unit 220 may be a resource model, which comprises of hardware for the communicating with the computing or communication devices used by the agents, network switches, communication wires, and routing rules defining the basis of routing. The ACD unit 220 forwards the call request to queues associated with a particular team of agents. The basis for routing the call request may be based on the skill required for the call and the priority of the call. Each team of agents may have as many queues as the number of skills handled by the team.
[00046] From the various queues the calls are assigned to an agent who is free and also
has the requisite skill by a job assignor (JA) unit 226. In one embodiment, the JA unit 226 may be configured to route the incoming call requests based on a most idle technique. In the most-idle technique, an incoming call request is routed to the agent, who has the requisite skill set to service the incoming call request and has been idle, i.e. not serviced an incoming call request, for the longest time interval. The most-idle technique ensures balanced utilization of all the agents including the multi skilled agents. In another implementation, the JA unit 226 may be configured to route the incoming call requests based on round robin technique. In the round robin technique, the incoming call requests are allocated to agents, who have the requisite skill set to service the incoming call requests, in a round robin manner. The round robin technique ensures balanced distribution of incoming call requests to agents in terms of number of incoming call requests serviced.
[00047] In another implementation, the JA unit 226 may be configured to route the
incoming call requests based on lowest cost technique. In the lowest cost technique, an

incoming call request is routed to an agent, who has the requisite skill to service the incoming call request and has minimum associated cost for example in terms of wages. In another example, the JA unit 226 may be configured to route the incoming call requests based on minimum skill technique. In said technique, an incoming call request is allocated to agent, who has the requisite skill set to service the incoming call request, but has the minimum number of skills. The minimum skill technique preserves availability of agents with more skills and thus reduces the number of incoming call requests that are waiting for availability of agents with requisite skill.
[00048] In yet another implementation, the JA unit 226 may be configured to route the
call request based on the efficiency of the agent. Thus, an incoming call request is assigned to an agent who free and also is most efficient in the skill required by the call. Further, the JA unit 226 may be configured to route incoming call requests based mixed technique, i.e. more than one technique. For example, the JA unit 226 may be configured to route incoming call requests based on the most idle technique till a predefined percentage, say 75%, of the agents are busy, i.e. servicing incoming call requests, and if there are more incoming call requests then the JA unit 226 may be configured to route incoming call requests based on the maximum efficiency technique.
[00049] If the waiting time for servicing the incoming call request at the skill based
queues exceeds the patience time of the customer, the incoming call request may be deemed as abandoned or dropped and as indicated by arrow marked 224, would be moved to the call sink 210. If the incoming call is not abandoned, the incoming call request is forwarded to JA unit 226. The JA unit 226 forwards the incoming call request, as indicated by arrow marked 228-1, 228-2, ..., 228-N to one or more resource models representing the agents 230-1, 230-2, ...230-N, of the call center, henceforth referred to as agents 230. The basis for allocation has been described in the previous sections. It should be appreciated by those skilled in the art that agents 230 include individual agents 230 as well as a team or group of agents 230. Further based on the resource attributes assigned to agents 230, each agent may have a different work shift. A work shift may be understood to be different if at least one of the parameters

associated with the work shift, such as start time of the work shift, end time of the work shift, break time, duration of the break, is different.
[00050] The simulated model 200 may be used by the user to generate reports
indicative of what-if analysis based on variation of resource models or variation of resource attributes associated with the resource model or variation of workflow parameters. The reports may be further indicative of the various performance metrics associated with the call center, such as blockage index, abandon rate, average speed of service, service level index, agent occupancy index, schedule adherence index, cost per call, etc. The user may also vary the resource models and the resource attributes associated with the resource models and monitor the changes in the various performance metrics. For example, the user may monitor how the performance metrics change by having dedicated team of agents for each skill as compared to having a team with multi skilled agents; varying the number of multi skilled agents; assigning a dedicated team to handle high priority calls; varying the work shifts of the agents and so on. Thus, the simulated model 200 of the call center facilitates the user to optimize the various factors which contributes to the quality of service provided by the call center. It should be appreciated by those skilled in the art that the concepts of transformation explained in the context of generation of ICR requests may also be applied to other parameters. For example, the values of time spend at the IVR unit, time taken to service an incoming call request, patience time of a customer may also be transformed based on one or more transformation modes.
[00051] Figure 3 illustrates a method 300 of simulation in order to predict the
performance of a call center, in accordance with an implementation of the present subject matter. The exemplary method may be described in the general context of computer executable instructions. Generally, computer executable instructions can include routines, programs, objects, components, data structures, procedures, modules, functions, and the like that perform particular functions or implement particular abstract data types. The method 300 may also be practiced in a distributed computing environment where functions are performed by remote processing devices that are linked through a communication network. In a

distributed computing environment, computer executable instructions may be located in both local and remote computer storage media, including memory storage devices.
[00052] The order in which the method 300 is described is not intended to be construed
as a limitation, and any number of the described method blocks can be combined in any order to implement the method 300, or alternate methods. Additionally, individual blocks may be deleted from the method 300 without departing from the spirit and scope of the subject matter described herein. Furthermore, the method 300 can be implemented in any suitable hardware, software, firmware, or combination thereof. The method 300 described herein is with reference to the CCPP system 102 and in the context of predicting performance of a call center; however, the method can be implemented in other similar systems albeit and in other organizations with a few variations as will be understood by a person skilled in the art.
[00053] At block 302, various details pertaining to resource(s) of a call center are
received from a user. In one example, the resource management module 118 is configured to receive details pertaining to various resource(s) of the call center. The resources of the call center include man power, such as the agents, the communication network entities, such as the trunk lines 204, the PABX 206, the IVR unit 214, the ACD unit 220; and other computing and communication systems used in the call center. The details associated with the resources may include capacity of each resource, number of entities in each resource, number of input and/ or output ports in each resource, etc.
[00054] As illustrated in block 304, one or more resource models are generated to
represent the resources of the call center. In one implementation, the resource management module 118 is configured to generate resource models of the resources of the call center based on the details provided by the user. The resource models represent each resource of the call center in a simulated model of the call center, such as the simulated model 200.
[00055] As depicted in block 306, various resource attributes are assigned to the
generated resource models. In one implementation, the resource management module 118 is configured to prompt the user for assigning resource attributes with the resource models. The resource attributes control the behavior of the resource models. For example, the resource

attributes associated with the resource model depicting the agents 230 may include their efficiency factor, skill set, work shifts, etc. Table 1 depicts an illustrative input that may be provided by the user to indicate the work shifts of the agents 230 belonging to a team, say team A.

Team Schedule
Employee Numbers Start Time Start of Break End of Break End Time Days of Week
1-5 05:30 09:30 10:30 14:30 2-6
6-9 09:30 13:00 14:00 18:30 2-6
10 14:00 17:00 18:00 23:00 2-6
11-12 05:30 09:30 10:30 14:30 2-6
13-16 09:30 13:00 14:00 18:30 2-6
17-18 14:00 17:00 18:00 23:00 2-6
Table-1
[00056] As shown in block 308, the work flow parameters are received. In one
implementation, the workflow management module 120 is configured to receive the workflow parameters from the user. The workflow parameters define the operational aspects of the call center. The workflow parameters define how the incoming call requests would be serviced in the call center. Table 2 shows an illustration of how the workflow parameters may be defined by the user. It should be appreciated that the values given in Table 2 are merely illustrative and for better understanding of the present subject matter and the same should not be construed as a limitation.

Workflow Parameter Parameter Value Distribution Parameters
Arrival Distribution Exponential
% Retrials 70
Time between Retrials Uniform 5,20
Patience Time Exponential 45
% High Priority customers 15

IVR Time
Exponential 30
Handle Time LogNormal 105,25
Table-2
[00057] As depicted in block 310, the operations of the call center are simulated based
on the resource models, the resource attributes assigned with the resource models and the
workflow parameters. In one implementation, the report generation module 122 is configured
to simulate the operations of the call center based on the resource models, the resource
attributes assigned with the resource models and the workflow parameters.
[00058] As illustrated in block 312, various reports may be generated by the simulator
of the operations of the call center. In one implementation the report generation module 122 may be configured to generate various reports indicative of the time taken to service an incoming call requests for each category of incoming call requests, utilization of each agent, queue sizes for each category of incoming call requests, etc. These reports may be used by the user, i.e. the call center supervisor, to optimize the staffing, work allocation, work timings, scheduling of training sessions of agents, etc. Table 3 and Table 4 show parts of an illustrative report generated by the report generation module 122.

Report for Team A
Response Time Stats Time in Seconds
Mean : 68.00
95.0% Confidence +/- 2.69
Minimum : 52.38
Maximum : 90.0
Standard Deviation : 13.49
Variance : 182.05
Count : 30974
Throughput : 0.02354
Report for End to E Ind Time
Turnaround Time Stats Time in Seconds
Mean : 83.00

95.0% Confidence +/- 2.69
Minimum : 67.38
Maximum : 105.0
Standard Deviation : 13.49
Variance : 182.05
Count : 30974
Throughput : 0.02354
Table -3

Detailed Report for Team A
Agent Number % Busy Cost
1 67.17 4500
2 72.3 5000
3 72.41 6500
4 71 5250
5 70.91 4750
6 68.36 5500
7 68.44 4750
Table -4
[00059] Thus the method 300 helps in predicting the performance of the call center.
Further the reports generated by the call center assist the call center supervisor to take appropriate actions to enhance the quality of service of the call center. Although embodiments for predicting performance of a call center have been described in language specific to structural features and/or methods, it is to be understood that the invention is not necessarily limited to the specific features or methods described. Rather, the specific features and methods are disclosed as exemplary embodiments for methods and systems for predicting the performance of a call center.

I/We claim:
1. A computer implemented method for predicting performance of a call center, the
method comprising:
receiving details pertaining to resources of the call center, wherein the resources comprise at least one agent and a call request of the call center;
generating a plurality of resource models representative of the received details;
receiving at least one workflow parameter associated with the plurality of resource models; and
generating at least one report indicative of a quality of service provided by the call center based on the at least one workflow parameter.
2. The computer implemented method as claimed in claim 1, wherein the resources of the call center further comprise at least one of a private automatic branch exchange (PABX), an interactive voice response (IVR) unit, a mono-skilled agent and a multi skilled agent.
3. The computer implemented method as claimed in claim 2, wherein each agent of the mono-skilled agents and the multi skilled agents is associated with an efficiency factor indicative of a skill set and expertise of the agent.
4. The computer implemented method as claimed in claim 1, wherein the resources of the call centre further comprises an automatic call distributor (ACD) and skill based queues (SBQ).
5. The computer implemented method as claimed in claim 1, wherein the resources of the call center further comprises a job assignor.
6. The computer implemented method as claimed in claim 5, wherein the job assignor (JA) routs the plurality of incoming call requests based in part on at least one of a most

idle technique, round robin technique, lowest cost technique, maximum efficiency technique and minimum skill technique.
7. The computer implemented method as claimed in claim 1, wherein the method further comprises generating a simulated model (200) of the call center.
8. The computer implemented method as claimed in claim 7, wherein the method further comprises generating a plurality of incoming call requests based on number of calls expected in a specified time interval.
9. The computer implemented method as claimed in claim 8, wherein the method further comprises generating the plurality of incoming call requests based on at least one of a Bernoulli distribution model, a Beta distribution model, an Erlang distribution model, an Exponential distribution model, a Gamma distribution model, a Geometric distribution model, a Hotset distribution model, a Hyper-exponential distribution model, a Hyper-geometric distribution model and a Zipf distribution model.
10. The computer implemented method as claimed in claim 8, wherein the method further comprises assigning a priority to at least one of the plurality of incoming call requests based on at least one of an origin of the incoming call request and a category of the incoming call request.
11. The computer implemented method as claimed in claim 7, wherein the method further comprises assigning a priority to at least one of the plurality of incoming call requests based on at least one of an origin of the incoming call request and a category of the incoming call request.

12. The computer implemented method as claimed in claim 7, wherein the method further
comprises assigning a fixed number of trunk lines, wherein the fixed number of trunk
lines are indicative of a maximum number of concurrent calls.
13. The computer implemented method as claimed in claim 7, wherein the method further comprises assigning distribution of time that an incoming call request spends in IVR before the incoming call request is forwarded to an agent.
14. The computer implemented method as claimed in claim 7, wherein the method further comprises of the ACD distributing the calls to different teams of agents based on call type and priority
15. The computer implemented method as claimed in claim 7, wherein the at least one of the plurality of resource models is representative of the agents of the call center, wherein the agents are at least one of a mono skilled, multi skilled and a part of a dedicated team of the call center.
16. The computer implemented method as claimed in claim 7, wherein the method comprises generating at least one of a patience time, wait time and service time based on at least one of a Bernoulli distribution model, a Beta distribution model, an Erlang distribution model, an Exponential distribution model, a Gamma distribution model, a Geometric distribution model, a Hotset distribution model, a Hyper-exponential distribution model, a Hyper-geometric distribution model and a Zipf distribution model.
17. The computer implemented method as claimed in claim 7, wherein the method comprises generating at least one of a patience time, wait time and service time based on data provided by a user in form of a file.

18. A call center performance prediction system center(CCPP) system (102) comprising:
a processor(108); and
a memory (114) coupled to the processor (108), the memory (114) comprising
a resource generation module (118) configured to
- receive details pertaining to resources of a call center;
- generate at least one resource model representative of the resources of the call center;
receive at least one resource attribute associated with the at least one resource model, indicative of the characteristics of the at least one resource attribute;
a workflow generation module (120) configured to receive at least one
workflow parameter associated with the at least one resource model; and
a report generation module (122) configured to
generate a simulated model (200) of the call center so as to simulate
operations of the call center.
19. The CCPP system (102) as claimed in claim 18 wherein the simulated model (200) of the call center comprises at least one resource model representing at least one of a private automatic branch exchange (PABX), an interactive voice response (IVR) unit and a multi skilled agent.
20. The CCPP system (102) as claimed in claim 18, wherein report generation module (122) is further configured to generate at least one report, wherein the at least one report is further indicative of the time taken to service an incoming call requests for each category of incoming call requests, utilization of each agent of the call center, queue sizes for the each category of incoming call requests.
21. A computer-readable medium having embodied thereon a computer program for executing a method comprising:
receiving details pertaining to at least one resource of the call center;

generating at least one resource model representative of the at least one resource based on the received details;
receiving at least one of a resource attribute and a workflow parameter associated with the at least one resource model; and
generating a simulated model (200) of the call center so as to simulate the operations of the call center based on the at least one of a resource attribute and a workflow parameter.