Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claim Rejections - 35 USC § 101
Claims 1 – 18 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to an abstract idea without significantly more.
This judicial exception is not integrated into a practical application. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Claims 1 and 10 are directed to the abstract idea of workforce scheduling for deferrable contacts —specifically, the longstanding business practice of measuring how long work items take to resolve, computing averages, estimating total staffing needs for a pending backlog, obtaining supervisory approval, and generating agent schedules.
Step 2A, Prong 1: Claims Directed to an Abstract Idea
A contact center supervisor/manager could perform all of the claimed steps manually or with pen-and-paper:
Claimed Step
Receiving deferrable contacts
Distributing contacts to agents
Measuring time required for each resolution
Computing an average of required times
Receiving a second plurality of contacts
Computing total required time
Displaying total required time to manager
Receiving approval
Generating required staffing
Generating an agent schedule
Publishing the schedule
Human Activity
A supervisor receives a stack of emails,
voicemails, or social media messages
A supervisor assigns work items to agents by hand (manually)
A supervisor uses a stopwatch, watch or timesheet to track resolution times
A supervisor calculates the average with a calculator or mental math
More work items arrive in the inbox
Supervisor multiplies the count of pending items × average time (basic arithmetic)
Supervisor writes the number on a whiteboard or screen, or sends an email to a manager
Manager says “approved” verbally or via email
Supervisor divides total required time by available hours per agent
Supervisor fills in a shift schedule spreadsheet
Supervisor posts the schedule on a bulletin board or sends it to agents
The claims fall within two recognized categories of abstract ideas:
A- Certain Methods of Organizing Human Activity
* Commercial interactions and business relations (workforce management)
* Managing personal behavior or relationships between people (assigning work, scheduling)
* Following rules or instructions (scheduling rules database)
B- Mathematical Concepts
* Computing an average (arithmetic mean)
* Computing a total required time (multiplication)
* These are basic mathematical operations that can be performed mentally
Step 2A, Prong Two (No Practical Application)
The additional elements do not integrate the abstract idea into a practical application. The server, processor, memory, GUI, databases, and networked devices are all conventional, off-the-shelf computing components performing their standard functions. The claims recite no novel algorithm, data structure, or architectural change to the computer itself — the improvement is to the business process, not to computer functionality. The “continuously, in real time” limitation is merely a speed characteristic that any networked computer can achieve, not a technical innovation. The GUI display, approval step, and schedule publishing are mere data output and insignificant extra-solution or post-solution activity.
Step 2B (No Inventive Concept)
Each additional element — the server, network communication, named software services, database, and GUI — is well-understood, routine, and conventional. The ordered combination of collecting data, computing on it, displaying results, receiving approval, generating output, and distributing that output is the normal expected sequence for networked workforce management systems. This does not transform the abstract idea into something “significantly more.”
Claims 2–5 and 11–14 add contact type specifications and skill-based assignment, which further refine the abstract business process rather than the technology.
Claims 6 and 15 add a step of receiving the manager’s estimate first, which is additional human input that falls under organizing human activity.
Claims 7–8 and 16–17 add the use of a large language model (LLM) to generate natural language output, but this uses a generic and available AI tool to format text — applying a known tool to the abstract idea without disclosing specific technical improvements to the LLM itself.
Claims 9 and 18 repeat the same abstract distribution step for a new set of contacts.
Again, the claims recite no specific algorithm — the specification never discloses how staffing is actually generated, saying only “automatically generating a required staffing.” The asserted improvement is at the business level (helping supervisors make better decisions), not at the technical level (improving how the computer itself functions) as required under Alice. The LLM usage is too generic, with no prompt engineering detail, model architecture, or technical integration mechanism claimed.
Claim Rejections - 35 USC § 103
Claims 1 – 5, 9– 14 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Stearns (US 2011/0145032 A1) alone or in view of Fama et al (US 2021/0158267 A1, hereinafter “Fama”).
Regarding claim 1, Stearns discloses a workforce management system for forecasting and scheduling work items that cross intervals, including deferrable contacts work items. Specifically, Stearns teaches:
A system adapted to automatically manage deferrable contacts in a contact center — Stearns explicitly addresses deferrable response work items in a contact center (ACD) (Stearns, ¶¶ [0007]–[0008], [0014], [0056]: “The following section describes an algorithm for a staff requirement calculation for ‘deferred response contacts’ for long average handling time (AHT) work items”).
A workforce management server having a processor and a non-transitory computer readable medium operably coupled thereto — Stearns discloses a central processing computer 120 executing workforce management system 126 implemented as software (Stearns, ¶¶ [0021]–[0022], [0086]: “the algorithm is implemented in software executing in one or more server machines”; FIG. 1).
The server being in electronic communication with a manager computing device and a plurality of agent computing devices — Stearns discloses central processing computer 120 (“server”) coupled via network 114 to agent workstations 130 and supervisor workstations 140 (Stearns, ¶ [0021]: “The central processing computer 120 is also coupled via the network 114 to one or more agent workstations 130 and to one or more supervisor workstations 140”).
The server being in electronic communication with an intraday manager database — Stearns discloses database 122 coupled to central processing computer 120 storing per-interval performance data including contacts received, handle time, workload, and staffing requirements (Stearns, ¶ [0021]: storing data “preferably but not limited to 15 or 30 minute intervals or real time”).
With the workforce management server, over a first period of time: receiving a first plurality of deferrable contacts — Stearns teaches receiving work items (including deferrable contacts such as emails) per interval via ACD 110 (“contact center”) and multimedia server 112 (Stearns, ¶¶ [0017], [0021], [0056];
Automatically distributing the deferrable contacts of the first plurality of deferrable contacts to a plurality of agents for resolution via the plurality of agent computing devices — Stearns teaches ACD 110 and multimedia server 112 routing contacts to agents at their workstations (Stearns, ¶ [0021]: “agents 132 access, via the telephone (not shown) and agent workstation 130, the ACD 110, multimedia server 112, or other contact servers (not shown) to aid in contact resolution”). FIG. 5 shows agents 1–20 assigned work items.
Measuring a respective time required for each resolution — Stearns teaches the ACD collecting performance data including “the amount of time an agent spends on incoming contacts (‘talk time’), the amount of time an agent spends after the talk time for administrative duties related to contact resolution (‘work time’), the amount of time an agent spends in talk time and work time (‘total time’)” (Stearns, ¶ [0021]).
Computing an average of the respective required times — Stearns teaches computing “Forecast AHT (work minutes)” (i.e., average handling time) from historical data for use in forecasting and planning (Stearns, ¶¶ [0029]–[0030], [0057]; FIG. 4, Column E: “Forecast AHT”).
The claimed GUI is inherent (or at least obvious) in agent workstations 130 and supervisor workstations 140 (Fig. 1).
Continuously, in real time, with the workforce management server: receiving a second plurality of deferrable contacts — Stearns teaches that work items arrive continuously across successive intervals, with each interval receiving new deferrable work items (Stearns, FIG. 4, Column B showing work items received in each successive interval 800, 830, 900, etc.; ¶ [0020]: “the disclosed subject matter relates to forecasting and scheduling in any environment in which contacts (or, more generally, ‘work items’) cross over or bridge multiple ‘intervals’”). Also, in [0021], “…. stores the information in a database 122 coupled to the central processing computer 120, preferably but not limited to 15 or 30 minute intervals or real time”.
Based on the average of the respective required times, computing a total required time to resolve all of the deferrable contacts of the second plurality of deferrable contacts — Stearns teaches computing the “Workload” for each interval as the sum of (contacts × work time per contact) across all active work items in the interval. Column M of FIG. 4 represents “the total workload of active work items in the interval” which is a total time value computed based on the forecast AHT (Stearns, ¶ [0031]: “Workload = (contactsReceived of this interval × intervalWorkTime of this interval) + contactsCarriedOver for each prior interval × intervalWorkTime for each prior interval”).
With the intraday manager service, automatically generating a required staffing — Stearns teaches automatically computing staffing requirements per interval: “staffRequired = (totalContactsInProgress)” (Stearns, ¶¶ [0073], [0045]; Column N of FIG. 4: “total staff required, as expressed in FTE, in the interval”).
With the scheduler service, the required staffing and the scheduling rules database, automatically generating an agent schedule — Stearns teaches a schedule generation process that uses the computed staffing requirements along with scheduling work rules to produce agent schedules: “the schedule generation process must be aware of contacts connected to an agent over a long duration” and accounts for “agent scheduling work rules, and other scheduling parameters, such as agent load balancing” (Stearns, ¶¶ [0025]–[0026]; FIG. 5 showing the schedule output).
Stearns does not explicitly disclose:
(i) Receiving an approval from the manager computing device before proceeding; and
(ii) Publishing the agent schedule to the plurality of agent computing devices via the GUI.
On one hand, receiving an approval from the manager of the Contact Center for scheduling agents is obvious, well-known and needed. Also, publishing the agent schedule to the agent’s GUI (workstation 130) is extremely obvious as agents need to know their work schedule. That is, the differences between Stearns and claim 1 are obvious and required/needed. Schedules cannot be organized without the manager’s approval and the agents need to know their work schedule. Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the current claimed invention to utilize the above old, well-known and needed features in Stearns to provide a functional, efficient and organized ACD (call center).
On the other hand:
Fama discloses an automated WFM system for a contact center and teaches a forecaster service, a scheduler service, and the use of output device 161 such as display (“GUI”) and agent’s desktop 120 (“GUI”). Fama discloses a configuration engine 170 comprising forecast engine 210 and schedule engine 220 as distinct functional modules operating within the WFM system (Fama, ¶ [0023]; FIG. 2). The system communicates with administrator 290 through electronic communications (Fama, ¶¶ [0027], [0035]).
Receiving an approval from the manager computing device — Fama teaches that “the schedule engine 220 may prompt the administrator 290 to review the generated schedule 221. The administrator 290 can reject the schedule 221, alter or modify the schedule 221, or allow the schedule 221 to be released without modification” (Fama, ¶ [0035, 0047]; FIG. 4, steps 425–430). The administrator determines “whether or not to release the generated schedule” (Fama, ¶ [0063]).
With the GUI, publishing the agent schedule to the plurality of agent computing devices — Fama teaches that upon approval, “the schedule engine 220 may release the schedule 221 to one or more of the administrator 290 and/or the agents 120” (Fama, ¶ [0043]; FIG. 4, step 420; FIG. 5, step 545: “Release forecast and schedule”).
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the workforce management system for deferrable work items taught by Stearns to incorporate the above features in order to ensure quality control through administrator approval before schedules are released to agents. Stearns already teaches computing staffing requirements and generating schedules; adding a structured approval and release workflow would predictably improve the reliability and quality control of the scheduling output.
The references are in the same field of endeavor (contact center workforce management systems), address the same general problem (efficiently staffing a contact center to handle incoming contacts), and the combination yields no more than predictable results: a WFM system that (1) receives and distributes deferrable contacts, (2) measures handle times and computes workload, (3) displays workload metrics to a supervisor via a GUI for approval, and (4) generates and publishes agent schedules upon receiving approval. See KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 416 (2007).
Claim 10 is rejected for the same reasons as claim 1.
Claim 2 recites: “…. wherein the first plurality of deferrable contacts and the second plurality of deferrable contacts are of a plurality of contact types.”
Stearns teaches that the contact center handles multiple contact types (CTs). Stearns discloses that the multimedia server 112 handles faxes, e-mail, voice mail, web requests, web call-back requests, web chats, web voice calls, web video calls, and the like” (Stearns, ¶ [0021]). Stearns further teaches that the WFM system forecasts and schedules for each contact type separately: “a CT is the type of contact, such as voice call, email, fax, voice mail, web requests, web call-back requests, web chats, web voice calls, web video calls, and the like” (Stearns, ¶ [0021]). The deferrable work items described in Stearns’ second embodiment (FIG. 4) may encompass multiple such contact types, as the back office handles “processing medical claims, processing mortgage applications, processing prescriptions, processing product support calls, and the like” (Stearns, ¶ [0017]).
Claim 3 recites: “….. wherein the plurality of contact types comprises at least two of email contacts, voicemail contacts, social media contacts, or other contacts.”
Stearns teaches that the multimedia server 112 handles multiple deferrable contact types including “e-mail, voice mail, web requests, web call-back requests, web chats, web voice calls, web video calls, and the like” (Stearns, ¶ [0021]). At minimum, Stearns discloses email contacts and voicemail contacts as types of deferrable contacts handled by the contact center.
Claim 4 recites: “… wherein the respective required time for each resolution depends on a contact type of the respective deferrable contact, wherein the average of the respective times comprises a separate average for each contact type.”
Stearns teaches that the average handling time (AHT) is computed separately for each contact type. Stearns discloses that the central processing computer receives “the average handling time (‘AHT’) of each CT” (Stearns, ¶ [0021]). The forecast AHT (Column E of FIG. 4) is generated per contact type using “prior art algorithms or custom forecast models” (Stearns, ¶ [0064]). Because the AHT differs by contact type (e.g., processing an email takes a different amount of time than processing a voicemail), the respective time required for each resolution depends on the contact type, and the average is maintained separately for each contact type.
Claim 5 recites: “… wherein each agent computing device of the plurality of agent computing devices is associated with an agent of the plurality of agents, wherein the agent schedule comprises a separate schedule for each agent of the plurality of agents, and wherein the separate schedule for each agent is based on a skill of the respective agent with regard to at least one contact type of the plurality of contact types.”
Stearns teaches that each agent workstation 130 is associated with a respective agent 132 (Stearns, ¶ [0021]; FIG. 1 showing multiple agent workstations 130 paired with agents 132). The scheduling algorithm produces individual agent schedules: FIGS. 3 and 5 show per-agent assignments (Agent 1 through Agent 20), each assigned specific work items based on their scheduling and availability. Stearns further teaches that the WFM system provides “skill planning and scheduling” (Stearns, ¶ [0022]).
Fama additionally teaches that agents are assigned to queues based on skills: “there may be queues associated with a language (e.g., English or Chinese), topic (e.g., technical support or billing), or a particular country of origin” and agents 120 “may be further organized into one or more teams…based on a variety of factors including, but not limited to, skills, location, experience, assigned queues” (Fama, ¶¶ [0018]–[0019]). The schedule engine 220 generates a schedule that “may assign agents 120 to queues 125 for each interval that the schedule 221 covers based on the forecast 211 such that one or more service level goals associated with each queue 125 are met” (Fama, ¶ [0038]).
Claim 9 recites: “… wherein the instructions further comprise automatically distributing the deferrable contacts of the second plurality of contacts to the plurality of agents for resolution via the plurality of agent computing devices.”
Stearns teaches that the ACD 110 and multimedia server 112 continuously distribute incoming work items (including deferrable contacts) to agents at their workstations (Stearns, ¶ [0021]). The scheduling algorithm of Stearns assigns work items arriving in each interval to agents: “Additional agents are scheduled for new work items that are forecasted to arrive in an interval in addition to the agents that may already be connected to handle a work item” (Stearns, ¶ [0008]). FIG. 5 shows agents being assigned new work items in successive intervals (e.g., work items 41–50 distributed to agents in the 1100 interval). The automatic distribution applies to contacts received in all intervals, including the second plurality of deferrable contacts received in subsequent intervals.
Claims 10 – 14 are rejected for the same reasons as claims 1 – 5.
Claim 18 is rejected for the same reasons as claims 9.
Claims 6 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Stearns and Fama, as applied against claims 1 and 10 above, and further in view of Kosiba (20020184069 A1).
Claim 6 recites: “… wherein displaying the total required time to the manager computing device involves first receiving a manager estimate of the total required time” (which is not explicitly taught by Stearns or Fama).
Kosiba discloses a graphical user interface (GUI 900, [0062], Fig. 9) that displays computed workforce planning data—including staffing requirements, workload metrics, staffing levels, and performance outputs—to a workforce planner/manager on a computing device. ([0095]).
Specifically, the GUI includes “analysis fields 910” that display “full-time staff fields 935, part time staff fields 940, staff total fields 945, calls fields 950, and performance fields 955.” ([0119], [0121]. The “effective staff” field 972 and associated workload calculations are displayed to the user ([0123], [0124], [0125], representing computed staffing/time requirements presented via the GUI to the manager’s computing device. Additionally, the system “immediately” displays performance outputs when inputs change: “the user may freely change the full time staff fields 935, the part time staff fields 940, the schedule adherence field 971, and the calls fields 950 and see immediately the corresponding performance outputs in fields 955.” See “staff required to achieve that performance” ([0132], [0133], [0144], [0157], [0163]).
More particularly, Kosiba teaches that user-provided estimates serve as input before the system generates and displays computed results. The system “is able to work ‘in reverse’ by changing the service level goal (field 976) and/or average speed of answer (field 977) and looking up the staff required to achieve that performance in the planning and analysis database” (Kosiba, [0133]. Kosiba further teaches that the system allows management to understand “the effects on contact center performance of changes in staff, budgets, call handling, and service goals” (]0057]. The method may include using the predictions to generate multiple performance scenarios and executing comparisons among the multiple performance scenarios to evaluate alternative resource levels and process parameters [0027], [0028].
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the current invention to utilize the teachings of Kosiba so that an estimate of the total required time is provided to the manager in order to enable the manager to plan accordingly based on estimated needs.
Claim 15 is rejected for the same reason as claim 6.
Claims 7 – 8 and 16 – 17 are rejected under 35 U.S.C. 103 as being unpatentable over Stearns, Fama, as applied against claims 1 and 10 above, and further in view of Wicaksono et al. (US 2026/0075142 A1, hereinafter “Wicaksono”).
Claim 7 recites: “…. wherein displaying the total required time to the manager computing device includes generating and displaying a natural language output related to the total required time” (which is not explicitly suggested by Stearns or Fama).
On one hand, the use of natural language has become well-known in the art and the advantages of using it are also well-known.
On the other hand, Wicaksono teaches generating and displaying natural language outputs related to workload metrics. Specifically, Wicaksono teaches that the system processes analysis data using “at least one large language model to generate a set of insights associated with the metric data of the contact center” (Wicaksono, Abstract, ¶ [0004]) and displays those insights in natural language on the GUI. For example, the workload insights displayed to the supervisor include the following natural language text:
This natural language output is “related to the total required time” because it directly addresses workload metrics including volume, AHT (which determines total required time), and staffing sufficiency. The Wicaksono system further generates natural language prescriptions including: “Schedule an additional 5 agents at 9:00 am and 10:00 am, and 1 agent at 1:00 pm to handle the increased workload” (Wicaksono, ¶ [0116]) — which is a natural language expression of the staffing/time requirement.
Claim 8 recites: “…. wherein generating and displaying the natural language output involves: sending a prompt to a large language model (LLM); receiving a natural language output from the LLM; and displaying the natural language output” (which is not explicitly suggested by Stearns or Fama).
The use of LLM is also well-known in the art.
Wicaksono explicitly discloses an LLM Engine 314 that receives analysis data and generates natural language summaries (Wicaksono, ¶¶ [0086]–[0087]; FIG. 3, element 314). Specifically:
Sending a prompt to a large language model (LLM): Wicaksono teaches sending structured prompts to one or more LLMs.
The LLM generates outputs including diagnosis, prognosis, and prescriptive sections in natural language (Wicaksono, ¶¶ [0086], [0092]–[0093]). The natural language insights are displayed on the graphical user interface as shown in FIG. 4, element 404.
For claims 7- 8, it would have been obvious to one of ordinary skill in the art before the effective filing date of the current invention to utilize the above features taught by Wicaksono since these well-known and available features (“natural language” , “LLM”) have many known advantages and provide added convenience, efficiency and ease of operation.
Claims 16-17 are rejected for the same reasons as claims 7-8.
Response to Arguments
Applicant's arguments filed 9/23/2026 have been fully considered but they are not persuasive.
Rejection under 35 U.S.C. § 101
I. Step 2A, Prong One — The Claims Recite a Judicial Exception
Applicant argues (on page 3 of the remarks) that the claims are not directed to an abstract idea because they recite a “particularized computer-implemented workflow” involving an intraday manager service, scheduler service, forecaster service, and graphical user interface, and that the claimed operations “cannot practically be performed in the human mind.”
Applicant’s arguments are not persuasive.
First, the recitation of named software components (an intraday manager service, scheduler service, forecaster service, and GUI) does not remove the claims from the realm of abstract ideas. Under the 2019 PEG and MPEP § 2106.04, the proper inquiry is whether the claim recites a judicial exception, not whether the exception is couched in functional or system-level language. When the additional software labels are stripped away, the claimed steps- receiving contacts, distributing contacts to agents, measuring the time required for each resolution, computing an average of those times, computing a total required time, displaying that time, receiving an approval, generating staffing, generating a schedule, and publishing the schedule - recite both:
(a) Certain methods of organizing human activity, specifically managing workforce activity and the commercial/managerial practice of assigning work and scheduling personnel; and
(b) Mathematical concepts, specifically computing an average and computing a total (i.e., multiplying an average handling time by a number of pending contacts).
These categories are expressly identified as abstract ideas in MPEP § 2106.04(a)(2).
Second, Applicant’s assertion that the claimed steps “cannot practically be performed in the human mind” mischaracterizes the analysis. The steps of measuring how long work takes, averaging those times, estimating total staffing need, presenting that estimate for approval, and assigning personnel are the very type of managerial and mathematical activities historically performed by human supervisors, with or without pen and paper. As set forth in the prior Office Action, each claimed step maps directly to an activity a contact center supervisor could perform manually. That a general-purpose computer performs these steps more quickly or continuously does not render the underlying concept non-abstract. See Intellectual Ventures I LLC v. Capital One Bank, 792 F.3d 1363 (Fed. Cir. 2015); MPEP § 2106.04(a)(2)(III).
Accordingly, the claims recite an abstract idea under Step 2A, Prong One.
II. Step 2A, Prong Two — The Judicial Exception Is Not Integrated Into a Practical Application
Applicant argues (page 4-5) that any abstract idea is integrated into a practical application because the computed information is used to “drive subsequent machine operations,” including staffing generation, schedule generation, and schedule publication, and that the invention improves workforce-management operations.
Applicant’s arguments are not persuasive.
The additional elements recited in the claims—a workforce management server, a processor, a non-transitory computer readable medium, a graphical user interface, databases, and networked agent/manager computing devices—are recited at a high level of generality and perform only their well-understood, conventional functions of receiving, storing, processing, displaying, and transmitting data. Using generic computer components to collect data, perform calculations on that data, display results, and transmit outputs does not integrate the abstract idea into a practical application. See MPEP § 2106.05(f); Alice Corp. v. CLS Bank Int’l, 573 U.S. 208 (2014).
Applicant’s contention that the claims “drive subsequent machine operations” does not establish a technical improvement. Generating staffing, generating a schedule, and publishing a schedule are themselves part of the abstract idea of managing and scheduling a workforce; they are not improvements to the functioning of a computer or to any technology. The claims do not recite any specific algorithm, data structure, or technical mechanism by which staffing or schedules are generated. Indeed, the Specification describes this operation only in result-oriented terms (e.g., “automatically generating a required staffing”) without disclosing any particular technical implementation. An improvement to a business process—here, assisting supervisors in determining staffing for deferrable contacts—is not an improvement to technology. See Enfish, LLC v. Microsoft Corp., 822 F.3d 1327 (Fed. Cir. 2016); MPEP § 2106.05(a).
The “continuously, in real time” limitation likewise fails to integrate the exception into a practical application. Performing an abstract idea faster or on a continuous basis using generic computing capability is not a technological improvement. See FairWarning IP, LLC v. Iatric Sys., Inc., 839 F.3d 1089 (Fed. Cir. 2016).
The remaining recited actions—displaying the total required time, receiving an approval, and publishing the schedule—constitute insignificant extra-solution activity (data output and data transmission) and the mere receipt of a user input, none of which impose meaningful limits on the judicial exception. See MPEP § 2106.05(g).
Accordingly, the judicial exception is not integrated into a practical application.
III. Step 2B — The Claims Do Not Recite Significantly More
Applicant argues (page 5) that the claims recite “significantly more” than the abstract idea by virtue of a specific ordered combination of elements, including automatically distributing contacts, measuring resolution times, computing historical handling information, continuously receiving contacts in real time, generating staffing, generating schedules, and publishing schedules.
Applicant’s arguments are not persuasive.
Considered individually, each additional element—the server, processor, memory, GUI, databases, and networked devices—is a generic computer component performing routine and conventional functions of receiving, storing, processing, displaying, and transmitting data. Such generic computer implementation cannot provide an inventive concept. See Alice, 573 U.S. at 223–26; MPEP § 2106.05(d).
Considered as an ordered combination, the claimed sequence—collecting data, performing mathematical operations on that data, displaying the result, receiving approval, generating an output, and transmitting that output - is the normal and expected sequence of operations for a networked workforce-management system. The ordered combination does not add any element or interrelationship beyond what would be expected when automating the abstract managerial/mathematical process on a general-purpose computer. The claims therefore do not transform the abstract idea into patent-eligible subject matter.
Applicant’s reliance on the Specification to establish an improvement is unavailing because the Specification characterizes the asserted benefit in terms of improving the supervisor’s decision-making and forecasting accuracy - an improvement to the abstract business process itself - rather than identifying any improvement to computer functionality or to any other technology.
IV. Dependent Claims
With respect to dependent claims 7–8 and 16–17, Applicant argues that the use of a large language model to generate natural-language output adds significantly more. This argument is not persuasive. The claims recite the LLM at a high level of generality - sending a prompt, receiving a natural-language output, and displaying that output - which amounts to applying a known, generic tool to the abstract idea and to insignificant extra-solution output activity. The claims do not recite any specific prompt architecture, model configuration, training, or technical integration that improves the functioning of the computer or the LLM itself. Applying generic artificial-intelligence tools to an abstract idea does not confer eligibility. See MPEP §§ 2106.05(a), (f).
The remaining dependent claims (2–6, 9–14, 15, 18) further narrow the abstract idea by specifying contact types, skill-based assignment, additional human input (receiving a manager estimate), and repetition of the distribution step. These limitations refine the abstract idea and/or add insignificant extra-solution activity, and therefore do not render the claims patent eligible.
For at least the foregoing reasons, the claims (1) recite a judicial exception under Step 2A, Prong One; (2) do not integrate that exception into a practical application under Step 2A, Prong Two; and (3) do not recite significantly more than the exception under Step 2B. Accordingly, the rejection of claims 1–18 under 35 U.S.C. § 101 is maintained.
Rejections under 35 U.S.C. § 103
Claims 1–5, 9–14, and 18 (Stearns in view of Fama)
1) Applicant argues (page 6) that Stearns does not disclose using historical resolution times to determine average handling times, and from those averages a total required time to resolve a subsequently received set of deferrable contacts.
The arguments are not convincing because the claimed “average of the respective required times” and “total required time” read directly on Stearns’ average handle time and workload computations. Stearns ¶ [0021] discloses collecting “the amount of time an agent spends in talk time and work time (‘total time’)” and “the average handling time (‘AHT’) of each CT.” Stearns FIG. 4, Column E recites “Forecast AHT (work minutes),” and Stearns ¶ [0031] computes “workload = (contacts Received of this interval × interval Work Time of this interval) + contacts Carried Over for each prior interval × interval Work Time for each prior interval,” which Stearns ¶ [0031] describes as “the total workload of active work items in the interval.” The claims do not recite any particular manner of measuring or averaging that distinguishes over these disclosures.
2) Applicant argues that the references do not address the specific problem of deferrable (asynchronous) contact handling.
The arguments are not convincing because Stearns expressly discloses a deferrable embodiment, see ¶ [0082]. Stearns ¶ [0056] states: “The following section describes an algorithm for a staff requirement calculation for ‘deferred response contacts’ for long average handling time (AHT) work items.” Stearns FIG. 4 is titled, in relevant part, “WFM for long AHT: Deferrable Response Work Items,” and Stearns ¶ [0008] discloses scheduling additional agents for “new work items that are forecasted to arrive in an interval.”
3) Applicant argues (page 7 and 8) that the references lack the claimed continuous loop in which newly arriving deferrable contacts drive re-computation.
The arguments are not convincing because Stearns discloses ongoing, interval-by-interval processing of work items that arrive and carry over across intervals. Stearns ¶ [0020] states the subject matter “relates to forecasting and scheduling in any environment in which contacts (or, more generally, ‘work items’) cross over or bridge multiple ‘intervals,’” and Stearns FIG. 4, Column B shows work items received in successive intervals (e.g., 800, 830, 900). The recited “continuously, in real time” is recited at a high level of generality and does not distinguish over Stearns’ repeated interval processing.
4) Applicant argues that Fama merely adds a generic approval workflow and does not cure the deficiencies of Stearns.
The arguments are not convincing because Fama is relied upon specifically for the manager-facing display, approval, and schedule-publishing limitations, which it expressly discloses. Fama ¶ [0047] states: “the schedule engine 220 may prompt the administrator 290 to review the generated schedule 221. The administrator 290 can reject the schedule 221, alter or modify the schedule 221, or allow the schedule 221 to be released without modification.” Fama ¶ [0043] states that upon approval “the schedule engine 220 may release the schedule 221 to one or more of the administrator 290 and/or the agents 120,” and Fama FIG. 5, step 545 recites “Release forecast and schedule.” Stearns separately supplies the staffing and schedule-generation steps (Stearns, FIG. 4 Column N; Stearns ¶¶ [0025]–[0026], FIG. 5).
5) Applicant argues (page 7) that combining Stearns and Fama is based on impermissible hindsight.
The arguments are not convincing because both references are in the same field of contact-center workforce management and address the same problem of staffing to meet service objectives. The Office Action articulated a rational basis - adding administrator review/approval before release to provide quality control over the staffing output -which relies on knowledge within the level of ordinary skill rather than Applicant’s disclosure. See KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398 (2007).
6) Applicant argues (as to the dependent claims) that the references fail to teach contact types, skill-based assignment, and distribution of a second plurality of contacts.
The arguments are not convincing because these features are expressly disclosed. Contact types: Stearns ¶ [0021] multimedia server 112 such as email, chat, voice; Fama ¶ [0018] (queues by language/topic). Skill-based assignment: Stearns ¶ [0022] (“skill planning and scheduling”); Fama ¶ [0038] (assigning agents to queues to meet service level goals) and ¶ [0019] (agents organized by “skills, location, experience, assigned queues”). Distribution of a second plurality: Stearns ¶ [0008] (additional/new work items in subsequent intervals) and FIG. 5 (assignment of later-arriving work items).
To the extent Applicant attacks the references individually, nonobviousness cannot be shown by attacking references separately where the rejection is based on their combination. See In re Keller, 642 F.2d 413 (CCPA 1981); In re Merck & Co., 800 F.2d 1091 (Fed. Cir. 1986).
II. Claims 6 and 15 (Stearns in view of Fama, further in view of Kosiba)
1) Applicant argues that Kosiba does not disclose first receiving a manager estimate of the total required time before displaying a system-generated total required time, and that Kosiba’s general planning inputs are not equivalent to the claimed manager estimate.
The arguments are not convincing because Kosiba discloses a planning interface in which a user first enters planning values and the system then computes and displays corresponding staffing/performance results, including a reverse mode in which the user supplies targets and the system returns required staff. Kosiba ¶ [0133] states the system “is able to work ‘in reverse’ by changing the service level goal (field 976) and/or average speed of answer (field 977) and looking up the staff required to achieve that performance in the planning and analysis database.” Kosiba ¶ [0132] further states that the user “may freely change the full time staff fields … and see immediately the corresponding performance outputs in fields 955.” Entering such user-supplied values before the system displays computed requirements corresponds to the claimed step of first receiving a manager estimate, and the claims recite no particular comparison or processing of the estimate that distinguishes over Kosiba.
It would have been obvious to incorporate Kosiba’s user-estimate-then-display approach into the Stearns/Fama system to allow a manager to compare an estimate against system-generated requirements, yielding predictable results.
III. Claims 7–8 and 16–17 (Stearns in view of Fama, further in view of Wicaksono)
1) Applicant argues that Wicaksono’s LLM-generated insights are not specifically tied to a total required time derived from historical handling-time of deferred contacts.
The arguments are not convincing because Wicaksono expressly uses large language models to generate natural-language staffing/workload output, and the claims recite the LLM functionality generically without a limitation tying the output to deferred-contact historical handling time. Wicaksono ¶ [0004] discloses processing data “using at least one large language model to generate a set of insights associated with the metric data of the contact center and at least one possible action to improve a condition,” and Wicaksono ¶ [0116] provides an LLM-generated prescription reciting “Schedule an additional 5 agents at 9:00 am and 10:00 am, and 1 agent at 1:00 pm to handle the increased workload.” Applicant’s asserted distinction is not commensurate with the claim scope.
2) Applicant argues that neither Stearns nor Fama uses an LLM, so combining them with Wicaksono is impermissible hindsight.
The arguments are not convincing because Wicaksono is directed to the same field of contact-center workforce analytics and discloses the recognized benefit of presenting staffing/workload information to users in natural language via an LLM. Wicaksono ¶ [0086] discloses an “LLM Engine 314” that generates natural-language summaries, and Wicaksono ¶¶ [0090]–[0093] disclose sending a prompt to the LLM, receiving the natural-language output, and displaying it (FIG. 4, element 404). This provides a rational basis to combine Wicaksono’s LLM-based narrative output with the staffing/workload values of Stearns and Fama, and the claims recite no specific prompt construction, model configuration, or technical integration that distinguishes over Wicaksono.
For at least the foregoing reasons, and based on the specific teachings of Stearns, Fama, Kosiba, and Wicaksono set forth above, Applicant’s arguments do not overcome the prior-art rejections, and the rejections under 35 U.S.C. § 103 are maintained.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AHMAD F. MATAR whose telephone number is (571)272-7488. The examiner can normally be reached M-F 9 - 5:30.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/AHMAD F. MATAR/Supervisory Patent Examiner, Art Unit 2693
