Systems and methods for allocating resources to various processes in network environments
By aggregating data from disparate sources and constructing an optimization model, the system efficiently allocates resources for radiation oncology planning, addressing inefficiencies and improving treatment timelines and patient care.
Patent Information
- Application Number
- PCT/US2026/027394
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-09
- Filing Date
- 2026-05-08
- Publication Date
- 2026-11-12
AI Technical Summary
Existing systems face challenges in efficiently allocating resources for radiation oncology planning due to fragmented and heterogeneous data across disparate sources, lack of flexibility in handling planner absences, and difficulty in balancing workloads, leading to delays and suboptimal patient care.
A computing system aggregates data from multiple sources to construct an optimization model that allocates resources across time slots, considering planner qualifications, plan complexities, and real-time availability, with features to dynamically adjust schedules and provide actionable feedback.
This approach reduces time-to-treatment, improves scheduling efficiency, and ensures plans are assigned to suitable planners, enhancing overall workflow management and clinical outcomes.
Smart Images

Figure US2026027394_12112026_PF_FP_ABST
Abstract
Description
Atty. Dkt. No.: 115872-3496SYSTEMS AND METHODS FOR ALLOCATING RESOURCES TO VARIOUS PROCESSES IN NETWORK ENVIRONMENTSCROSS REFERENCE TO RELATED APPLICATIONS
[0001] The present application claims priority to and benefit of U. S. Provisional Patent Application No. 63 / 803,360, titled “Systems and Methods for Allocating Resources to Various Processes in Network Environments,” filed May 9, 2025, which is incorporated by reference in its entirety.BACKGROUND
[0002] A computing system may track an amount of resources allocated to various processes.SUMMARY
[0003] Aspects of the present solution are related to systems and methods of allocating resources to process across time slots using data from disparate sources. One or more processors can retrieve, from one or more data sources, a plurality of data objects associated with execution of a plurality of processes on a plurality of instances using a plurality of resources across a plurality of time slots. The plurality of data objects may include a first data object identifying at least a first subset of processes of the plurality of processes unassigned to any of the plurality of time slots, each of the first subset of processes to be executed on a respective first instance of the plurality of instances using at least one of the plurality of resources by at least one of the plurality of time slots, a second data object identifying a subset of resources of the plurality of resources as assigned to a second subset of processes of the plurality of processes, each of the second subset of processes to be executed on a respective second instance of the plurality of instances using one respective resource of the subset of resources, and a third data object identifying, for each respective resource of the plurality of resources, a first subset of time slots of the plurality of time slots during which the respective resource is available for allocation. The one or more processors can identify, from the plurality of data objects, a plurality of parameters defining allocation of the plurality of processes on the plurality of instances with one or more of the plurality of resources across one orAtty. Dkt. No.: 115872-3496more of the plurality of time slots. The one or more processors can generate a model including a first plurality of constraints to define the allocation to satisfy the plurality of parameters. The one or more processors can modify the first plurality of constraints of the model to generate a second plurality of constraints to define the allocation to satisfy a portion of the plurality of parameters. The one or more processors can determine, based on applying the model in accordance with the second plurality of constraints, an allocation schedule defining, for each respective process of the plurality of processes, at least one resource of the plurality of resources in at least one time slot of the plurality of time slots. The one or more processors can provide an output including information based on the allocation schedule.
[0004] In some embodiments, the one or more processors can receive, via an input on a user interface, a second plurality of parameters to redefine the allocation of the plurality of processes on the plurality of instances with one or more of the plurality of resources across one or more of the plurality of time slots, subsequent to providing the output for presentation, modify, using the second plurality of parameters, the second plurality of constraints of the model to generate a third plurality of constraints, and determine, based on applying the model in accordance with the third plurality of constraints, a second allocation schedule defining, for at least one process of the plurality of processes, at least one second resource of the plurality of resources in at least second one time slot of the plurality of time slots. In some embodiments, the one or more processors can retrieve the plurality of data objects including a fourth data object identifying, for each resource of the plurality of resources, a respective capability to handle one or more of a plurality of process types.
[0005] In some embodiments, the one or more processors can provide a user interface to input a second plurality of parameters to define at least one of (i) a first maximum number of processes to be executed over one or more time slots of the plurality of time slots (ii) a second maximum number of processes to be executed by at least one of the plurality of resources over the one or more time slots, (iii) a ramp up period prior to executing at least one of the plurality of processes by at least one of the plurality of resources, or (iv) a score indicating a degree of complexity of at least one of the plurality of processes. The one or more processors can receive, via input on the user interface, the second plurality of parameters to include in a fifth data object of the plurality of data objects. In some embodiments, the one or more processors can modify theAtty. Dkt. No.: 115872-3496first plurality of constraints to generate the second plurality of constraints by changing the second plurality of parameters to set at least one of the first maximum number of processes, the second maximum number of processes, the ramp up period, or the score indicating the degree of complexity of at least one of the plurality of processes.
[0006] In some embodiments, the one or more processors can identify, for each respective resource of the plurality of resources, a second subset of time slots of the plurality of time slots during which the respective resource is unavailable for allocation, using the third data object. In some embodiments, the one or more processors can identify, for each respective process of the plurality of processes, a score indicating a degree of complexity to execute the respective process; and determine the allocation schedule based on the score degree of complexity to execute the respective process of the plurality of processes. In some embodiments, the one or more processors can identify, using the second data object, the subset of resources as including (i) at least one resource associated with a first resource group and a second resource group of a plurality of resource groups and (ii) a remainder of resources associated with the first resource group; retrieve, from the one or more data sources, a fourth data object identifying a second subset of resources as assigned to a third subset of processes of the plurality of processes, each of the third subset of processes to be executed on a respective third instance of the plurality of instances using one respective resource of the second subset of resources, the second subset of resources associated with the second resource group; identify the plurality of parameters from the plurality of data objects and the fourth data object; and determine the allocation schedule for the first resource group and the second resource group. In some embodiments, the one or more processors can generate the model to include the first plurality of constraint comprising at least one first constraint to distribute the allocation across one or more of the plurality of time slots to satisfy the plurality of parameters; modify the first plurality of constraints to generate the second plurality of constraints to include at least one second constraint to distribute the allocation across one or more of the plurality of time slots to satisfy the plurality of parameters; and determine the allocation schedule based on applying the model in accordance with the second plurality of constraints including the at least one second constraint.
[0007] In some embodiments, the one or more processors can provide, for presentation, the output including the information via a user interface including a plurality of user interfaceAtty. Dkt. No.: 115872-3496elements, the plurality of user interface elements including (i) a first user interface element identifying the plurality of resources, (ii) a second user interface element identifying the allocation of the plurality of processes to one or more of the plurality of time slots, and (iii) a third user interface element identifying the plurality of instances. In some embodiments, the one or more processors can generate the output to include a plurality of metrics corresponding to the plurality of resources based on the allocation schedule, each of the plurality of metrics indicating a degree of occupancy of a respective resource of the plurality of resources. In some embodiments, the one or more processors can determine that no allocation schedule satisfies a third plurality of constraints of the model using a second plurality of parameters identified from a second plurality of data objects, and provide a second output to indicate that no allocation schedule is generated in accordance with the third plurality of constraints of the model.
[0008] In some embodiments, at least one of the plurality of processes includes administration of radiation therapy to a subject corresponding to at least one of the plurality of instances during at least one of the plurality of time slots. In some embodiments, the radiation therapy may include at least one of a brachytherapy, a magnetic resonance imaging-guided linear accelerator (MR-LINAC), external beam radiotherapy (EBRT), stereotactic body radio therapy (SBRT), intensity-modulated radiotherapy (IMRT), a volumetric modulated arc therapy (VMAT), or proton beam therapy (PBT). The subject may be diagnosed with a cancer to be administered with the radiotherapy.BRIEF DESCRIPTION OF THE DRAWINGS
[0009] The foregoing and other objects, aspects, features, and advantages of the disclosure will become more apparent and better understood by referring to the following description taken in conjunction with the accompanying drawings, in which:
[0010] FIG. 1 is an example of a block diagram of a system for allocating resources to process across time slots using data from disparate sources, in accordance with an illustrative embodiment.
[0011] FIG. 2 is a block diagram of a process for retrieving a plurality of data objects and identifying a plurality of parameters, in accordance with an illustrative embodiment.Atty. Dkt. No.: 115872-3496
[0012] FIG. 3 is a block diagram of a process for generating a model and modifying constraints of the model, in accordance with an illustrative embodiment.
[0013] FIG. 4 is a block diagram of a process for determining an allocation schedule and providing an output, in accordance with an illustrative embodiment.
[0014] FIG. 5 is a block diagram of a process for determining a second allocation schedule based on a user input, in accordance with an illustrative embodiment.
[0015] FIGs. 6A-6C is a flow diagram of an example implementation for allocating resources to process across time slots using data from disparate sources, in accordance with an illustrative embodiment.
[0016] FIG. 7 is a block diagram of an example environment for allocating resources to process across time slots using data from disparate sources, in accordance with an illustrative embodiment.
[0017] FIGs. 8A-8C is an example screenshot of retrieving a plurality of data objects, in accordance with an illustrative embodiment.
[0018] FIGs. 9A and 9B is an example screenshot of generating a model and modifying constraints of the model, in accordance with an illustrative embodiment.
[0019] FIGs. 9C and 9D are example screenshots of user interfaces to enter seed parameters for generating constraints of the model, in accordance with an illustrative embodiment.
[0020] FIG. 10 is a flow diagram of method for allocating resources to process across time slots using data from disparate sources, in accordance with an illustrative embodiment.
[0021] FIG. 11 is a block diagram of a computing environment according to an example implementation of the present disclosure.DETAILED DESCRIPTION
[0022] Following below are more detailed descriptions of various concepts related to, and embodiments of, systems and methods for allocating resources to process across time slots usingAtty. Dkt. No.: 115872-3496data from disparate sources. It should be appreciated that various concepts introduced above and discussed in greater detail below may be implemented in any of numerous ways, as the disclosed concepts are not limited to any particular manner of implementation. Examples of specific implementations and applications are provided primarily for illustrative purposes.
[0023] Section A describes an overview of techniques for allocating resources.
[0024] Section B describes mathematical formulation related to the techniques for allocating resources.
[0025] Section C describes systems and methods for allocating resources to process across time slots using data from disparate sources.
[0026] Section D describes a network and computing environment.A. OVERVIEW OF TECHNIQUES FOR ALLOCATING RESOURCES
[0027] The resource allocation platform (referred sometimes herein as Radiation Oncology Allocation of Plans (RALP)) aims to make scheduling easier for Treatment Plan Coordinators (TPCs) and more equitable for the planners. The resource allocation platform may use an optimization model for a decision assistance tool for the TPCs who have the ability to alter and review the model-proposed schedule, nothing will happen automatically. The resource allocation platform can be integrated into a workflow management system. When deployed with a workflow management system (e.g., PlanQ), the time-to-treatment for radiation oncology (“RadOnc”) may be reduced as planning runs more efficiently.
[0028] Treatment plan for coordinator (TPC) may refer to any entity that creates the schedules for the planners. In some examples, there may be one TPC for each disease-specific team; the TPCs may be the primary users of the resource allocation platform.
[0029] A planner may refer to those who are specialized dosimetrists or medical physicists. The planner may design specific RadOnc plans for patients receiving radiation therapy. The planners may translate the directions from Radiation Oncologists into instructions for the radiation machines. The planners may be certified for a given disease-site and radiation machine. The planners may be indirect users of the resource allocation platform (e.g., the planners may notAtty. Dkt. No.: 115872-3496see the model or tool, but their workload may be impacted). The planners may be organized into disease-specific teams. Each may handle a set of disease-sites. In some examples, the planners may be organized into teams having different turn around requirements (e.g., one team may handle cases with a very tight turn around).
[0030] A planning date may refer to a date when the TPC assigns the case the planner and / or run the RALP. The planning date may be when the work of the planner starts on the plan. As used herein, the term “due date” may refer to a date when the planner delivers the plans to the radiation oncologist (e g., RadOnc). This may be two days prior to treatment and / or any time period that allows the planner to approve and / or review the plans with the RadOnc and to get it signed off by a Medical Physicist.
[0031] A workflow management system (e.g., software such as PlanQ) where the TPC and planner management can be stored in a unified manner. As used herein, the term “Eclipse” may refer to a platform where the planners build the plan. As used herein, the term QGenda may refer to a platform that maintains information about time off and different planner schedules. For example, the planners may enter the time-off information in QGenda.
[0032] A process is discussed herein as an overview, while discussed in greater detail below. A patient can be identified as a candidate for radiation therapy. The MD (e.g., RadOnc) of the patient can place a Simulation (“SIM”) order to initiate a SIM Appointment. Then, the patient can go in for the SIM imaging. The SIM appointment can be logged into PlanQ. The TPCs can make sure that the SIM has been completed and that the information on the record matches the SIM order. Then, the TPC can mark the plans as “reviewed.” Once the plans are marked as reviewed, the TPC can assign the plans into the schedule. With the RALP, the planner will be able to click a button to “run RALP” to get recommended assignments. The TPC can review the calendar-week workload and schedule constraints and / or considerations. The TPC can start by assigning the junior planners and then assign those with the soonest start date and / or those with the highest complexity. The TPC can assign the plans to the planner in PlanQ and “submit” the schedule. The plans can be put in the work queue of the planner, which then can complete the plan. Once the plans are completed, the completed plans can be reviewed by a Medical Physicist andAtty. Dkt. No.: 115872-3496the RadOnc. The patient can then attend the treatment appointment where a Radiation Treatment Therapist (RTT) delivers the radiation therapy.
[0033] In some embodiments, the RALP may operate based on ranked rules for optimization. While non-limiting examples are discussed in greater detail below, rule-specific terms are defined herein. As used herein, the term “Tier 1 Planner” may refer to planners who address overflow if a given team is at capacity. These planners may be qualified to work outside of their ‘home’ group, and when assigning plans to these planners, at least 50% of their plans can be from their ‘home’ group. As used herein, the term “Tier 1 Plan” may refer to plans that are eligible to be assigned outside of the ‘home’ group. These plans can be assigned to Tier 1 planners in other groups but can go to a non-Tier 1 planner if staying within its ‘home’ group. As used herein, the term “CWW” (Compressed Work Week) may refer to certain planners who have compressed work weeks, for example, working fewer, longer days. This may not reduce their capacity but impact how days off are handled.
[0034] In this Section, the MR-Linac” may refer to a specialized RadOnc machine that combines radiation with an MR1 machine and allows for real-time adjustments to the radiation therapy. Planners who are certified for MR-Linac plans may have reserved capacity so they are available if an MR-Linac case arises. MR-Linac cases are not assigned by the resource allocation platform. As used herein, the term “Brachy” may refer to a specialized type of radiation. Planners who are Brachy certified may have reserved capacity so they are available if a Brachy case arises. Brachy cases are not assigned by the resource allocation platform. As used herein, the term “Junior Planners” may refer to planners who are still getting credentialed. They may get assigned first and then paired with a senior to make sure they are on track for their training / certification. This may change how complexity is calculated (e.g., score* 1.2 / 2 = new score per person). For the first phase of the resource allocation platform, the TPC may assign these cases first and then run the model for the remaining plans. The ranking of these rules may be based on TPCs.
[0035] Referring to Table I below, discussed herein are how the optimization model of the resource allocation platform can consider constraints, etc. Table I provides an example of the various rules and considerations that can be used in the optimization model when allocating resources. These rules may be ranked based on their importance and impact on the schedulingAtty. Dkt. No.: 115872-3496process. Table T may include details such as the rule number, name, description, rank, pipeline, and implementation status. For example, some rules may ensure that planners are credentialed for the plans they are assigned to, while others may address the handling of planners’ time off and workload distribution. Additionally, specialized rules for planners certified in MR-Linac and Brachy may be included to ensure they have reserved capacity for specific cases.Atty. Dkt. No.: 115872-3496
[0036] While the optimization model can operate based on various rules (e.g., the rules in Table I), users can input their preferences to adjust the planning process. These may include setting a maximum number of plans due per calendar week (e.g., default is 6), limiting the number of plans assigned while a planner is out (e.g., default is 3), selecting specific plans and planners for running the optimization model. In some embodiments, users may choose location preferences, with options including all plans, campus-specific plans, or region-specific plans. The balance function objective can also be adjusted between “count” and “score.” The location approach mapping shown in Table II below provides an example of how different plan sites can be associated with planner sites under either a campus or regional approach. While discussed as a non-limiting example, the optimization model may operate based on various parameters disclosed herein, including but not limited to, data parameters, team parameters, planner parameters, workload parameters, or any other parameters that can be incorporated into plan coordination without departing from scope of the present disclosure.Table IIAtty. Dkt. No.: 115872-3496
[0037] The optimization model may be configured to calculate the average workload and determine the deviation from that average for each planner. The optimization model may minimize the sum of these differences, while handling the workload balance in various manners. In some embodiments, the optimization model may be configured to operate based on the workload (e.g., which can be defined by the number of plans (or caseload) or the total plan complexity score), as a user parameter. For example, a user can select between the number of plans and the total plan complexity score, and the optimization model may minimize the sum of the differences based on the user selection. In some embodiments, the RALP may be implemented in various modes, providing different timeframes (e.g., looking at one calendar week prior, the current calendar week, two calendar weeks in the future, etc.). In other examples, the optimization model may look 45 days forward and back to account for planners’ time off. in some implementations, the optimization model may be configured to send “completed” cases that were due up to two full weeks before today, as well as any “in-progress” cases for assigned cases. The optimization model may also look indefinitely into the future for unassigned plans, which is hard-coded in the back-end and not a user parameter.
[0038] In some instances, there may be situations (referred to as an “Infeasible plan”) where it is not possible to generate a recommended schedule based on the given constraints. In response to the optimization model determining that no allocation schedule is feasible, the optimization model may be configured to provide an output to indicate that no allocation schedule is generated in accordance with given constraints. In such cases, the optimization model may output a message to the user, for example, “RALP schedule not possible,” “parameters too restrictive,” “consider relaxing parameters,” etc. This can thereby ensure that users are informedAtty. Dkt. No.: 115872-3496when the given conditions are too restrictive for the optimization model to generate a feasible schedule, allowing users to make adjustments.B. MATHEMATICAL FORMULATION RELATED TO THE TECHNIQUES FOR ALLOCATING RESOURCES
[0039] The optimization model (e.g., RALP) may be configured to support Treatment Planner Coordinators (TPCs) in allocating the workload of incoming Radiation Oncology (RadOnc) plans to treatment planners. This can balance the planners’ workload and reduce the time to treatment for patients. The formulation may include several sets, such as DATES, which is a set of dates from the planning date to the planning end date, and WEEKS, which is derived from the date range of the planning dates. The formulation may include PLANS, a set of unique identifiers for unassigned plans, and EMPS, a set of unique identifiers for available or selected planners. T1EMPS is a subset of employees who are part of the Tier 1 team, while PE TUPLE is a set of tuples with unique pairs of plan IDs and planner IDs, reflecting credentialed plans where the planner can take the plans and complete it before the due date, considering absences. EMP OOO PLANDATE is a set of planners who are out of the office on the planning date, and ED OOO TUPLE represents the start date of long-term out-of-office (OOO) days for planners.
[0040] Data parameters may include the complexity score (“scoreP”) of each unassigned plan, the patient identifier (“mrnP”) for each unassigned plan, the due date (“dueDateP”) of each unassigned plan, and the number of days of work (“numDaysWorkp”) needed for each unassigned plan. Additional parameters may include the team (“teame”) to which each planner belongs, the total current complexity score (“currentScoree”) of each planner, the current daily average complexity score (“currentDailyAvgScoree”) of each planner, the total number of days (“numDaysWorkede”) each planner has worked, the total number of plans (“currentNumPlansc,d”) currently assigned to each planner, and the number of plans (“currentNumPlansDatee”) per due date that are already assigned for each planner. The target load (“targetLoade”) is the maximum load that can be assigned to each planner, while “consWordDaysAvailP,e” represents the availability of consecutive workdays per planner and plan. Parameters “longOOORampUpStarte” and “longOOORampUpEnde” represent the start and end dates of the ramp-up period after a long OOO, and “lastDayLongOOOe” represents the last day of a long OOO.Atty. Dkt. No.: 115872-3496
[0041] Incorporating the mathematical formulation discussed above, the optimization model may be configured based on such variables as decision variables, constraints, objective functions, etc. discussed below. The decision variables may be defined as:AssignPlanEmployeepe= {JifPlan^can beassigned to employee e,otherwise.
[0042] The constraints may include a plurality of constraints, including logical constraints, hard constraints, specialty planners, vacation rules, patient constraints, etc. For the logical constraints, each plan may be assigned to a planner (Rule: -; Code: 3-5), for example:SecEMPs AssignPlanEmployeep c= 1 Vp G PLANS,while the planner is credentialed for the plan (Rule: 1; Code: 7-11)'.AssignPlanEmployeep, e = 0V(p, e) $ PE_TUPLE.
[0043] For the hard constraints, the number of new plans that can be assigned per day may be limited (Rule: 5; Code: 13-17), for example:^pePLANs AssignPlanEmployeepe< maxNumNewPlans We E EM PS,while the number of plans due per plan due date for each planner may be limited (Rule: 6; Code: 47-59), for example:< maxPlansPerDueDate (VeG EMPS, Wd E DATES')The number of plans due per week for each planner based on their target workload may be limited (Rule: 3; Code: 104-116), for example:2 AssignPlanEmployeep currentNumPlansDatee dp-.week (dueDatep)=w d.week(d)=w< maxPlansPlannerWeek x targetLoade(Vvv 6 WEEK, We E EMPS)Atty. Dkt. No.: 115872-3496
[0044] Next, Sufficient Consecutive Work Days for the Employee Before the Plan’s Due Date can be ensured (Rule: 2d; Code: 118-120), for example:numDaysWorkpAssignPlanEmployee^.e < consWordDaysAvailp.e (Vp 6 PLANS, We E EM PS.currentNumPlansDatee dp-.week (dueDatep)=w d:week(d)=< maxPlansBraquiPlanner (We, w E EWBRAQUITUPLE),< maxPlansMRLinecPlanner (We, w G EW_MRLINEC_TUPLE).A less load may be set for Tier 1 Planners (Rule 10c; Code: 90-102), for example:< maxPlansTierlPlanner (We E TLEMPS. Ww E WEEKS).
[0045] For the vacation rules, if a plan’s due date is after a long vacation, then the optimization model may not assign until 3 days or less before last day of PTO (Rule 7; Code: 19-25), for example:(W (e,ooo) E ED_OOO TUPLE).Atty. Dkt. No.: 115872-3496In some instances, where the planner is in ramp-up period, at most one plan may be assigned (Rule: 8b; Code: 32-36), for example:(V (e, ooo) G ED_OOO_TUPLEfPlans due after OOO before the start of 000 may not be assigned (Rule: 8c; Code: 38-45), for example:AssignPlanEmployeep e= 0p-.d>lastDayLongOOOe Ooo andplanning_date<ooo(V (e, ooo) G ED_000 TUPLE, Vd G DATES).
[0046] For the patient constraints, the optimization model may be configured to support constraints to define binary AssignMRNEmployee variable (e.g., big M formulation left side) (Rule: 4; Code: 123-125), for example:(Vmrn G MRNS, G Ve EMPS).In some instances, the optimization model may be configured to support constraint to define binary AssignMRNEmployee variable (e.g., big M formulation right side) (Rule: 4; Code: 127-130), for example:AssignMRNEmployeemrTl e> AssignPlanEmployeep eif planMRNp= mm(Vp G PLANS e G EMPS).In some embodiments, the optimization model may be configured to ensure that all plans for the same patient go to the same planner (Rule: 4; Code: 132-134), for example:Atty. Dkt. No.: 115872-3496(Vmrn G MRNS).
[0047] For the objective functions, the total deviations from mean score may be minimized, for example:and then, minimize the maximum deviation from mean score:C. SYSTEMS AND METHODS FOR ALLOC A TING RESOURCES TO PROCESS ACROSS TIME SLOTS USING DATA FROM DISPARATE SOURCES
[0048] In certain settings, a computing system may be used to determine an allocation of resources to processes over extended time periods, such as in the context of radiation oncology planning and treatment. There may be a number of significant challenges to efficiency and effectively allocate resources for such purposes. Radiation oncology planning may be a multi-step clinical workflow by which radiation therapy is configured by multiple entities and delivered to an individual diagnosed with cancer at multiple time points across a duration of time (e.g., 1 week to 3 months). First, the data to be used for determining allocations may be fragmented and heterogeneous, and stored across disparate data sources. Input data may reside across multiple fragmented systems (e.g., electronic medical records, workflow management software, scheduling platforms), with disconnected digital profiles and asynchronous communication between these systems. Second, even if the data are aggregated, the computing system may lack the ability to detect that there is no feasible allocation of resources from the data, wasting compute cycles attempting to solve an unsolvable instance. Third, such systems may have inadequate means to display the allocation results in a manner that enables a user to review and adjust the schedule.
[0049] For example, other techniques for scheduling radiation oncology plans often face significant challenges such as inefficient resource allocation, lack of flexibility in handling plannerAtty. Dkt. No.: 115872-3496absences, and difficulty in balancing workloads among planners. These issues can lead to delays in treatment, increased stress on planners, and suboptimal patient care. The other methods may not account for the varying complexities of different plans or the specific qualifications of planners, resulting in mismatched assignments and potential errors in treatment planning. Additionally, the inability to dynamically adjust schedules based on real-time data further exacerbates these problems, making it difficult to respond to changes in planner availability or patient needs promptly. Challenges such as digitally fragmented data, disconnected patient digital profiles, and asynchronous team communication contribute to these inefficiencies, and no holistic solution has been offered to address them.
[0050] To address these and other technical challenges, presented herein are systems and methods to allocate resources to processes across time slots, aggregate data from disparate data sources. A computing system may ingest data objects from disparate data sources (e.g., electronic medical records, hospital systems, the workflow management system, the planning platform) or from user devices on a computer network. The retrieved data objects may include a variety of information on unassigned processes (e.g., new radiotherapy plans needing assignment), resources already assigned to existing processes (e.g., planners and their current cases), time slots during which each resource is available for allocation, including time-off, ramp-up / ramp-down, and complexity-score information, among others. The computing system may pre-process the data objects to detect anomalies or missing fields and to compute derived values that could potentially impact the allocation. From the aggregated data objects, the computing system may extract parameters that define how processes can be allocated to resources across time slots. The parameters may include identifying time slots during which each resource is unavailable.
[0051] Using the parameters identified from the aggregated data objects, the computing system may construct an optimization model (sometimes called radiation oncology allocation of plans (RALP)). The optimization model may initially include a set of constraints encoding the allocation rules required to satisfy the parameters. These constraints encode, for example, logical rules (each plan assigned to a credentialed planner), hard constraints (limits on plans per day and per week), specialty-planner rules (reserved capacity for MR-Linac and Brachy planners), amp-up rules, patient-consistency rules (all plans for the same patient go to the same planner), cross-team rules, distribution rules, and process-complexity scores rules, among others. The initial set ofAtty. Dkt. No.: 115872-3496constraints may be computationally complex and may involve a significant amount of time (e.g., at least one or more hours) to execute and arrive at an optimal solution.
[0052] To reduce the computational complexity, the computing system may modify the initial set of constraints to generate an altered set of constraints by inputting specific values and relationships into the variables and coefficients of the model. The computing system may modify by, for example, using the maximum number of plans that can be assigned per day, the maximum number of plans due per week, the ramp-up duration, or a complexity score. This step may also reduce the number of constraints and relax the constraints themselves to increase a potential number of feasible allocations under the modified constraints. At this stage, the user may enter additional parameters or modify the constraints of the optimization model. With the relaxation of the constraints, the computing system may use the optimization model to determine an allocation that best optimizes for the constraints of the model.
[0053] The computing system may present the allocation results and related information via a user interface. The user interface may display resources, assignment of processes to time slots, instances, or workload metrics, among others. The presentation of the allocation results may facilitate semi-automated iterative refinement of the allocation. If the optimization model detect no feasible result, the user interface may output an indication of no feasible allocation under the provided constraints together with a cause of the infeasibility. After the initial output is presented, the computing system may receive modified parameters to update the constraints of the optimization model. With the update, the computing system may reapply the model to generate a new allocation result.
[0054] In this manner, the present systems and methods herein can address these challenges by allocating resources to process across time slots using data from disparate sources. The ingestion of data objects from multiple, disparate data sources and normalization of such data may allow for the computing system to operate across digitally fragmented data and disconnected digital profiles that prior techniques did not integrate. The techniques can efficiently allocate resources, taking into account planner qualifications, plan complexities, and real-time data on planner availability. This can ensure that plans are assigned to the most suitable planners, reducing the time-to-treatment for patients and improving overall scheduling efficiency. The model’s abilityAtty. Dkt. No.: 115872-3496to dynamically adjust schedules based on real-time data allows for greater flexibility in handling planner absences and workload balancing. Furthermore, the techniques include features to identify infeasible plans and provide actionable feedback to users, ensuring that constraints are not overly restrictive and that TPCs can manage their teams’ workloads effectively.
[0055] Additionally, the present systems and methods factor in a multitude of data points such as planner certification and types, planner team, workload requirements and planner capacity, patient consistency, time off including ramp-up and ramp-off periods, planner and plan location, and time needed to complete the plan. The relaxation of the model constraints may narrow a very large combinatorial search space to a tractable optimization, The optimization model can thusly solve combinatorial problems more efficiently, reducing the time to find a solution (e.g., from about an hour to approximately 0.4 seconds) and freeing up computing resources (e g., in terms of processor and memory) that otherwise would have been used on less inefficient techniques. Furthermore, when no allocation schedule can satisfy the constraints, the computing system may return an output indicating as such, thereby avoiding wasting computing resources on unsolvable and intractable constraints. This platform can improve radiotherapy workflow management, track the patient radiotherapy journey in a single longitudinal patient record, centralize communication and management, monitor risk events, and efficiently distribute care teams’ workload, leading to improved clinical outcomes on the part of individuals receiving therapy plans.
[0056] FIG. 1 is an example of a block diagram of a system 100 for allocating resources to process across time slots using data from disparate sources, in accordance with an illustrative embodiment. In a brief overview, the system 100 can include at least one data processing system 105, at least one data source 110, at least one computing device 115, and at least one database 125, among others, communicatively coupled via at least one network 120. The data processing system 105 can include at least one data retriever 130, at least one parameter extractor 140, at least one model generator 145, at least one rule applier 150, at least one resource allocator 155, and at least one interface handler 160, among others. The system 100 may be used to implement any of the functionalities as detailed herein in Section A and B (e.g., in conjunction with the RALP optimization model). Each of the components of the system 100 can be implemented using the computing system as described in Section D.Atty. Dkt. No.: 115872-3496
[0057] The data processing system 105 can be any computing device comprising one or more processors coupled with memory and software capable of performing the various processes and tasks described herein. The data processing system 105 can be housed within a computing system (e.g., laptop, PC, smart device, cell phone) or within a server group (e.g., data center, server site, office, etc.). The data processing system 105 can include instructions to determine that a grade satisfies a threshold indicative, generating a flag to add an attribution, and storing the first data structure. The data processing system 105 can be in communication with the data sources 110, the computing device 115, and the database 125, among others, via a network 120.
[0058] The data processing system 105 can include one or more components, subcomponents, modules, or processes, among others, to carry out the various processes and tasks described herein. On the data processing system 105, the data retriever 130 can receive, retrieve, or otherwise obtain a plurality of data objects from the computing device 115. The parameter extractor 140 can identify, select, or otherwise retrieve, from the plurality of data objects (e.g., retrieved by the data retriever 130), a plurality of parameters defining allocation. The model generator 145 can generate, identify, or otherwise determine a model including a first plurality of constraints to define the allocation to satisfy the plurality of parameters. The rule applier 150 can modify, redefine, or otherwise adjust the first plurality of constraints of the model to generate a second plurality of constraints. The resource allocator 155 can determine, select, or otherwise identify an allocation schedule defining at least one resource in at least one time slot. The interface handler 160 can generate, identify, or otherwise provide an output including information based on the allocation schedule.
[0059] The data sources 110A-N (referred to as “data source 110” or “data sources 110”) can be any collection of webservers, Electronic Medical Records (EMR), online sources, computing systems, hospitals, Laboratory systems, or any device or server comprising one or more processors coupled with memory and software, capable of transmitting data associated with patients. The data sources 110 can be associated with an entity (e.g., hospital, physician, doctor, data center) conducting test on patient or providing health records for the patient. The data sources 110 can be in communication with the computing device 115 and the data processing system 105.Atty. Dkt. No.: 115872-3496
[0060] The computing device 115 can be any device comprising one or more processors coupled with memory and software and capable of receiving and displaying the various user interfaces. The computing device 115 can be associated with an entity (e.g., clinician, physician, doctor) interacting with the user interface 116. The computing device 115 can be in communication with the data processing system 105 and the data sources 110 to exchange data. The computing device 115 can include the user interface 116 (referred to as the user interface 116) to display the various data structures and flags described herein.
[0061] The system 100 can be configured to allocate resources to process across time slots using data from disparate sources in administrating radiation therapy to a subject diagnosed with a cancer to be administered with the radiotherapy. In some embodiments, the radiation therapy may include at least one of a brachytherapy, a magnetic resonance imaging-guided linear accelerator (MR-LINAC), external beam radiotherapy (EBRT), stereotactic body radio therapy (SBRT), intensity-modulated radiotherapy (IMRT), a volumetric modulated arc therapy (VMAT), or proton beam therapy (PBT). The cancer may include, for example, any type, such as lung cancer, brain cancer, head and neck cancer, colon cancer, rectal cancer, uterine cancer, endometrial cancer, stomach cancer, ovarian cancer, cervical cancer, bladder cancer, or breast cancer, among others.
[0062] FIG. 2 is a block diagram of a process 200 for retrieving a plurality of data objects and identifying a plurality of parameters, in accordance with an illustrative embodiment. The process 200 can include or correspond to operations performed in the system 100. Under the process 200, the computing device 115 and / or the data sources 110A-N can transmit, send, or otherwise provide data objects 208A-N (referred to as first data objects 208) and data objects 208'A-N (referred to as second data objects 208') to the data retriever 130. The first data objects 208 and the second data objects 208' can be in a plurality of formats (e.g., for each data source 110, computing device 115) to allow the transfer of the first data objects 208 and the second data objects 208' between the components of the system 100 (e.g., computing device 115, each data source 110, data retriever 130, etc.). For example, each of the first data objects 208 and the second data objects 208' can be a primitive data structure (e.g., array), a linear data structure (e.g., linked list, stack, queue), non-linear data structure (e.g., tree, graph, heap), hash-based structure (e.g., hash table, hash map, set), specialized structure (e.g., matric, deque, priority queue), an abstract data type (e.g., list, map, set), or an advanced data structure (e.g., segment tree). Upon receipt byAtty. Dkt. No.: 115872-3496the data processing system 105, the data retriever 130 can generate a notification, message, or indication for the first data objects 208 and / or the second data objects 208', for transmission to the data sources 110 and / or to the computing device 115.
[0063] The plurality of data objects (e.g., the first data objects 208, the second data objects 208', etc.) may be associated with execution of a plurality of processes on a plurality of instances using a plurality of resources across a plurality of time slots. As used herein, the term “processes” may refer to plans for administrating radiation therapy to the plurality of instances. As used herein, the term “instance” may refer to a subject, a patient, or other entities diagnosed with a cancer to be administered with radiotherapy. For example, at least one of the plurality of processes may include administration of radiation therapy to a subj ect corresponding to at least one of the plurality of instances during at least one of the plurality of time slots (e g., times such as days, hours, etc. over a given time window such as a week, a month, etc.). As used herein, the term “resource” may refer to a planner, a radiologist, or otherwise other entities configured to plan and / or administrate the radiotherapy.
[0064] The first data objects 208 may identify at least a first subset of processes (e.g., processes 281A-N, referred to as first processes 281) of a plurality of processes. The first processes 281 may be processes that are unassigned to any of a plurality of time slots (e.g., time slots 282A-N, referred to as time slots 282). For example, the first data objects 208 may contain data on unassigned radiotherapy plans. The first data objects 208 may include and / or identify the time slots 282 to which no process has been assigned. The time slots 282 may have a plurality of time slots such as days, hours, etc. over a given time window such as a week, a month, etc. In some embodiments, the time slots 282 may identify time period during which the processes 281 should be executed. For example, the time slots 282 may identify respective due dates of the processes 281.
[0065] The second data objects 208' may identify a subset of resources (e.g., resources 283A-N, referred to as resources 283) of a plurality of resources as assigned to a second subset of processes (e.g., processes 281'A-N, referred to as second processes 281') of the plurality of processes. For example, the second data objects 208' may contain data on planners, radiologists, and their assignments. Each of the second processes 281 may be a process to be executed on aAtty. Dkt. No.: 115872-3496respective instance of a plurality of instances using at least one of the resources 283. In some embodiments, the second data object 208' may identify at least one of the resources 283 as associated with multiple resource groups (e.g., a first resource group or a second resource group). For instance, at least one of the planners corresponding to the resource 283 may belong to multiple teams of planners. In some embodiments, the processes 281' and / or the resources 283 may identify scheduled time slots during which the processes 281' are executed such that the scheduled time slots identify availability of the resources 283. In some embodiments, each of the second processes 281' may be a process to be executed on a respective first instance using one of the resources 283 by one of the time slots 282. For example, when the time slot 282A is taken by the resource 283 A to execute the process 281' A, in a downstream process, any of the processes 281 may be prevented from being assigned to the time slot 282A. Each of the processes 281' may be a process to be executed on a respective second instance of the plurality of instances using one respective resource of the resources 283.
[0066] In some embodiments, the computing device 115 and / or the data sources 110A-N can transmit, send, or otherwise provide one or more third data objects to the data retriever 130. The third data objects may be in a similar format and / or a similar data structure as the first data objects 208 and the second data objects 208'. The third data objects may identify a first subset of the time slots 282 (e.g., for each of the resources 283) during which the respective resource of the resources 283 is available for allocation. For example, the third data objects may contain data on the availability of the planners, radiologists, etc., such as time off information, ramp-up and ramp-off periods, planner locations versus plan locations, complexity scores, etc.
[0067] In some embodiments,, the computing device 115 and / or the data sources 110A-N can transmit, send, or otherwise provide one or more fourth data objects to the data retriever 130. The fourth data objects may be in a similar format and / or a similar data structure as the first data objects 208 and the second data objects 208'. The fourth data objects may identify a respective capability of a resource of the resources 283 to handle one or more of a plurality of process types. For example, the fourth data objects may contain data on credential information of planners, radiologists, etc. In some embodiments, the fourth data objects may identify the respective capability for each of the resources 283. In some embodiments, the fourth data object may include a score indicate a degree of complexity of each process 280 or 281’. The degree of complexityAtty. Dkt. No.: 115872-3496may correspond to a level of difficulty or workload metric to execute the process 280 or 280’. For instance, the score may be predefined via a mapping specifying scores for various types of processes. The mapping may be pre-configured by an administrator of the data processing system 105 (or another user) and maintained on the database 125. The data retriever 130 can receive, obtain, or otherwise retrieve the plurality of data objects (e.g., the first data objects 208, the second data objects 208', the third data objects, the fourth data objects, etc.) from the computing device 115 and / or the data sources 110.
[0068] In some embodiments, the data retriever 130 can determine or identify, using the second data object 208', the subset of resources 283 as including at least one resource 283 associated with the first resource group and the second resource group of a plurality of resource groups. Each resource group may correspond to a team of planners, including senior and junior planners. In some embodiments, the data retriever 130 can identify a remainder of resources 283 as associated with the first resource group. With the identification, the data retriever 130 can receive, obtain, or otherwise retrieve at least one additional data object corresponding to the other, second resource group from the one or more data sources 110 or the computing device 115. The fourth data object may be of a similar form as the second data object 208', but for the second resource group. For example, the fourth data object may identify a subset of resources (e.g., resources 283A-N, referred to as resources 283) of a plurality of resources as assigned to a second subset of processes (e.g., processes 281') of the plurality of processes. For example, the fourth data objects may contain data on planners, radiologists of the second team and their assignments. Each of the second processes may be a process to be executed on a respective instance of a plurality of instances using at least one of the resources. The processes 281' and / or the resources 283 may identify scheduled time slots during which the processes 281' are executed such that the scheduled time slots identify availability of the resources 283 (e.g., planners in the second resource group). In some embodiments, each of the second processes 281' may be a process to be executed on a respective first instance using one of the resources 283 by one of the time slots 282. Each of the processes 281' may be a process to be executed on a respective second instance of the plurality of instances using one respective resource of the resources 283 in the second resource group.
[0069] The parameter extractor 140 can determine or identify parameters 240 defining one or more allocations 241A-N (referred to as allocations 241) from the data objects. TheAtty. Dkt. No.: 115872-3496parameters 240 may define the processes 281 on the plurality of instances with one or more of the resources 283 across one or more of the time slots 282 (e.g., the first subset of the time slots 282). In some embodiments, the parameter extractor 140 can identify the user parameters such as a plan date, a due date, etc. and / or the user parameters discussed in Section B, based on the processes 281, the instances, the resources 283, and the time slots 282 (e.g., the first subset of the time slots 282 during which the respective resource of the resources 283 is available for allocation).
[0070] The parameter extractor 140 can extract the parameters 240 from the plurality of objects (e.g., retrieved by the data retriever 130). In some embodiments, the parameter extractor 140 can identify, select, extract, or otherwise retrieve, for each respective resource of the resources 283, a second subset of time slots of the time slots 282 during which the respective resource is unavailable for allocation. For example, the parameter extractor 140 can identify the second subset of time slots of the time slots 282, based on the first data objects 208 and the second data objects 208'. In some embodiments, the parameter extractor 140 can identify the second subset of time slots of the time slots 282 using the third data object (e.g., containing data on the availability of the planners, radiologists, etc.) and / or the fourth data object (e.g., the credential information), among others. In some embodiments, the parameter extractor 140 can extract the parameters 240 across multiple resource groups (e.g., the first resource group and the second resource group), when at least one resource is identified as associated with the multiple resource groups.
[0071] FIG. 3 is a block diagram of a process 300 for generating a model and modifying constraints of the model, in accordance with an illustrative embodiment. The process 300 can include or correspond to operations performed in the system 100. Under the process 300, the model generator 145 can generate, identify, or otherwise determine at least one model 345 including a first plurality of constraints (referred to as first constraints 351) to define the allocation to satisfy the parameters 240. In some embodiments, the model generator 145 may create the model 345 for an individual resource group using the parameters 240 for the resource group. In some embodiments, the model generator 145 may create the model 345 for multiple resource groups using the parameters 240 for multiple resource groups (e.g., when at least one resource is identified as associated with the multiple resource groups).Atty. Dkt. No.: 115872-3496
[0072] The model 345 may be or include a mathematical model, defined by the first constraints 351 and determined based on the parameters 240 and / or the allocations 241. The model 345 may be stored and maintained using one or more data structures (e.g., linked lists, table, arrays, matrices, binary trees, heaps, hash tables, queues, or stacks). For example, the first constraints 351 may be one or more variables, one or more coefficients, and / or one or more mathematical relationships therebetween, etc. The model 345 may be or include a mathematical equation, including a relationship between the parameters 240 such as the rules discussed in Section A, the user parameters discussed in Section B, etc., which can be defined by the first constraints 351. In some examples, the first constraints 351 may define the model 345 by incorporating various rules as variables and / or relationships therebetween. The first constraints 351 may define the model 345 by incorporating, for example, logical constraints, hard constraints, specialty planners, vacation rules, patient constraints, etc. (discussed in Section B). For example, the first constraints 351 may define a rule that the number of new processes that can be assigned per day may be limited, and incorporate into the model 345. Likewise, the first constraints 351 may define a rule that the number of processes due per week for each resource based on their target workload may be limited, and incorporate into the model 345. In some embodiments, at least one of the first constraints 351 may be to distribute (or more relatively distribute or de-cluster) the allocations of the resources 283 across one or more time slots 282 while satisfying the parameters 240. For instance, the constraint may define the distribution of the allocation as a function of at least one of a number of processes allocated to each resource (e g., the resource 283) of a respective score indicating a degree of complexity of each process (e.g., the process 281 or 281’). The constraint may function as a reward to relatively even distribution as the function of the number of processes and the complexity scores, and may conversely penalize relatively concentrated distribution as the function of the number of processes and the complexity scores.
[0073] The rule applier 150 can modify, redefine, or otherwise adjust the first constraints 351 of the model 345 to generate a second plurality of constraints (referred to as second constraints 352; the model 345 with the second constraints 352 is referred to as a model 345'). The modification may reduce the number of constraints to increase feasible number of allocations in a controlled manner (e.g., as defined in Sections A and B). In some examples, the first constraints 351 may be variables and / or coefficients defining the rules discussed above, and the second constraints 352 may be specific values and / or data for the variables and / or coefficients definingAtty. Dkt. No.: 115872-3496the rules. In some implementations, the second constraints 352 of the model 345' can define the allocations 241 to satisfy a portion of the parameters 240. The second constraints 352 may be variables and / or coefficients modified based on the parameters 240 (and / or the allocations 241). That is, the first constraints 351 can be modified based on the parameters 240 (and / or the allocations 241) to generate the second constraints 352.
[0074] The rule applier 150 can input, insert, or otherwise provide a set of specific values into the variables and / or coefficients of the first constraints 351 to generate the second model 345' in accordance with the second constraints 352. For example, the first model 345 may include one or more algebraic equations that include, as a variable, a first number of new processes that can be assigned per day, among others. The rule applier 150 can modify the model 345 to generate the second model 345', by inputting, inserting or otherwise providing, to the first model 345, a specific number (e.g., 1, 3, 5, etc.) for the first number. In some embodiments, at least one of the second constraints 352 may be to distribute (or more relatively distribute or de-cluster) the allocations of the resources 283 across one or more time slots 282 while satisfying the parameters 240. For instance, the constraint may define the distribution of the allocation as a function of at least one of: a number of processes allocated to each resource (e.g., the resource 283) of a respective score indicating a degree of complexity of each process (e.g., the process 281 or 281’). The constraint may function as a reward to relatively even distribution as the function of the number of processes and the complexity scores, and may conversely penalize relatively concentrated distribution as the function of the number of processes and the complexity scores.
[0075] In some embodiments, the user interface 116 can input second parameters to define various user-specific data. For example, the second parameters may define at least one of (i) a first maximum number of processes to be executed over one or more of the time slots 282, (ii) a second maximum number of processes to be executed by at least one of the resources 283 over the one or more time slots, (iii) a ramp up period prior to executing at least one of the processes by at least one of the resources 283, or (iv) a score indicating a degree of complexity (or difficulty) of at least one process (process 281 or 281'). In these examples, the data retriever 130 can retrieve the userspecific data as a fifth data object. For example, the data retriever 130 can receive, via input on the user interface 116, the second parameters including the fifth data object. The parameter extractorAtty. Dkt. No.: 115872-3496140 can extract the parameters 240 based on the fifth data object such that the model 345 can be generated based on the fifth data object.
[0076] Using the input parameters, the rule applier 150 can modify the first constraints 351 to generate the second constraints 352 by changing the parameters 240 and / or adding the second parameters to adjust the model 345 (thereby generating the second model 345'). For example, the second parameters may be or include at least one of the first maximum number of processes, the second maximum number of processes, the ramp up period, or the score indicating the degree of complexity, and the model generator 145 can generate the model 345 in accordance with the first constraints 351 that are based on these second parameters. The rule applier 150 can then input specific values and / or relationships for the first constraints 351, in addition to the parameters 240, thereby generating the model 345' in accordance with the second constraints 352.
[0077] FIG. 4 is a block diagram of a process 400 for determining an allocation schedule and providing an output, in accordance with an illustrative embodiment. The process 400 can include or correspond to operations performed in the system 100. Under the process 400, the resource allocator 155 can determine, select, or otherwise identify an allocation schedule 455. The resource allocator 155 can generate the allocation schedule 455 based on applying, relaxing, or otherwise optimizing the model 345' (which is in accordance with the second constraints 352). In some embodiments, the model generator 145 may create the allocation schedule 455 for an individual resource group using the model 345' for the resource group. In some embodiments, the model generator 145 may create the allocation schedule 455 for multiple resource groups using the model 345' for multiple resource groups (e.g., when at least one resource is identified as associated with the multiple resource groups).
[0078] The model 345' may allow for generation of a plurality of potential allocation schedules, each including processes 281" A-N (referred to as processes 281") to be executed by respective resources 283' on respective ones of the time slots 282. For example, a first potential allocation schedule includes the process 281 " A to be executed by the resource 283'B on the time slot 282B, and the process 281" B to be executed by the resource 283'A on the time slot 282A; while a second potential allocation schedule includes the process 281 " A to be executed by the resource 283'A on the time slot 282A, and the process 281 " B to be executed by the resource 283'BAtty. Dkt. No.: 115872-3496on the time slot 282B. As such, the number of potential allocation schedules under the model 345' may be high, and / or not all of the potential allocation schedules provides an optimal allocation schedule. The resource allocator 155 can optimize the model 345' by identifying, from the plurality of potential allocation schedules, the allocation schedule 455 that optimizes the model 345'. For example, a third potential allocation schedule may include the process 281 " A to be executed by the resource 283 'A on the time slot 282A, and the process 281" B to be executed by the resource 283'B on the time slot 282B, and the resource allocator 155 may identify the third potential allocation schedule as the allocations schedule 455, as shown in FIG. 4.
[0079] In some embodiments, the resource allocator 155 can optimize the model 345' by identifying and / or minimizing a cost function of the model 345'. The resource allocator 155 can identify the allocation schedule 344 that minimizes the cost function to optimize the model 345'. In some examples, the cost function may be associated with a balance across the resources 283, and the resource allocator 155 can optimize the balance. For example, the second constraints 352 include maximum numbers of processes to be executed by the resources 283. The resource allocator 155 can calculate deviation from the maximum number of processes for each of the resources 283, and determine a sum of the deviation for the resources 283. The resource allocator 155 can determine the allocation schedule 455 by minimizing the sum of deviation. In some other examples, the cost function may be associated with average workload. The resource allocator 155 can minimize the cost function for balance by calculating the average workload and determining the deviation from that average for each of the resources 283, aiming to minimize the sum of these differences. In some embodiments, in optimizing the model 345', the resource allocator 155 may use the constraint 352 to distribute (or more relatively distribute or de-cluster) the allocations of the resources 283 across one or more time slots 282. For example, the resource allocator 155 may use the constraint 352 to reward balanced distributions of allocations as the function of the number of processes and the complexity scores. On other hand, the resource allocator 155 may use the constraint 352 to penalize concentrated distributions as the function of the number of processes and the complexity scores. In general, candidate allocations with many processes that are concentrated in time for a given resource may be less likely to be generated from optimizing the model 345'.Atty. Dkt. No.: 115872-3496
[0080] The allocation schedule 455 can identify or include a definition of allocations of resources 283' and time slots 282' to processes 281". The allocations of resources 283' and time slots 282' to the processes 281" may be in accordance with the constraints 352 of the model 345’. In some embodiments, the allocation schedule 455 may define or identify, for each of the processes 281", at least one of the resources 283' in at least one of the time slots 282. In some embodiments, the allocation schedule 455 may define or identify, for each of the resources 283', at least one process 281 during in at least one time slot 282. In some embodiments, the allocation schedule 455 may define or identify, for each time slot 282, at least one of the resources 283 to execute the at least one process 281. For example, the allocation schedule 455 may define that a particular radiologist is to oversee a specified patient during a given time window on a day of the week. The allocation schedule 455 can be stored and maintained using one or more data structures, (e.g., linked lists, table, arrays, matrices, binary trees, heaps, hash tables, queues, or stacks) or data files (e g., a comma-separate value (CSV) file, extensible markup language (XML) file, or structured query language (SQL) file), among others.
[0081] The interface handler 160 can generate, identify, or otherwise provide an output 460. The output 460 can include information based on the allocation schedule 455. The output 460 can be a document, a dataset, a paper, a collection of strings, or any other medium that provides detailed information associated with allocation of the resources 283' for the processes 281". The output 460 can include the information in a human readable format for a planner, a radiologist, or a subject. The information of the output 460 can include a plurality of elements representing the allocation schedule 455. For example, a first element can include a first allocation (e.g., the first process 281" A assigned to the first resource 283'A in the first time slot 282), a second element can include a second allocation (e.g., the second process 281 " B assigned to the second resource 283'B in the second time slot 282), etc.
[0082] In some embodiments, the output 460 may include a plurality of metrics. For example, the metrics may be workload metrics. Each metric can indicate or identify a degree of occupancy of a respective resource of the resources 283". The interface handler 160 can determine and / or include the workload metrics after the allocation schedule 455 is created, for example, based on the created allocation schedule 455. In some embodiments, the workload metrics can be determined prior to creating the allocation schedule 455. For example, the resource allocator 155Atty. Dkt. No.: 115872-3496can generate the workload metrics while generating the allocation schedule 455. In some embodiments, the resource allocator 155 can determine the workload metrics including an occupied rate of a particular planner / radiologist over a given time period, and generate the allocation schedule 455 based on the workload metrics. In some embodiments, the resource allocator 155 may calculate the workload metric based on a combination (e.g., summation or multiplication) of the score indicating the degree of complexity and the number of processes assigned to a given resource.
[0083] In some embodiments, the interface handler 160 can display, provide, or otherwise present the output 460 identifying the allocation schedule 455 via the user interface 116. The interface handler 160 can include instructions to generate, render, or otherwise provide the user interface 116 to display the output 460 and / or information associated with the output 460. For example, the interface handler 160 can transmit the output 460 within a message (e.g., notification, alerts, logs, data sets, triggers, email, text message, among other methods to transmit the output 460). From here, the user interface 1160 can be configured to display the output 460 and / or associated information. The user interface 116 may include a plurality of user interface elements, each of which can represent one or more values of the allocation schedule 455. For example, the user interface elements may include (i) a first user interface element identifying the resources 283 ", (ii) a second user interface element identifying the allocation of the processes 281 " to one or more of the time slots 282, and (iii) a third user interface element identifying a plurality of instances.
[0084] In some examples, where no schedule is allowed based on the model 345' (e.g., no resource 283' is available for the processes 281 " in any of the time slots 282), the interface handler 160 can include instructions to generate, render, or otherwise provide the user interface 116 to generate and / or display an error message. The error message may be, for example, “Infeasible plan: Schedule Not Possible.” Furthermore, the user interface 116 can display an analysis report indicating the cause of the infeasible plan and / or solutions. For example, the user interface 116 can generate a message, “the user parameters and / or the rules are too restrictive.” In some embodiments, the resource allocator 155 can determine that no allocation schedule satisfies the second constraints 352 of the model 345' using one or more parameters identified from the second data objects 208' (e.g., availability of planners, radiologists, their assignments, etc ). Based on thisAtty. Dkt. No.: 115872-3496determination, the interface handler 160 can provide a second output to indicate that no allocation schedule is generated in accordance with the constraints 352 of the model 345'.
[0085] FIG. 5 is a block diagram of a process 500 for determining a second allocation schedule based on a user input, in accordance with an illustrative embodiment. The process 500 can include or correspond to operations performed in the system 100. Under the process 500, the interface handler 160 can receive (e.g., via an input on the user interface 116) second parameters 540. For example, the interface handler 160 can receive the second parameters 540, subsequent to providing the output 460, to determine a second allocation schedule 455'. The second parameters 540 may include and / or define a plurality of allocations 541A-541N (referred to as allocations 541). Each allocation 541 may allocate at least one plurality of processes on a plurality of instances with one or more of the resources 283 across the time slots 282.
[0086] Upon receiving the second parameters 540, the rule applier 150 can generate third constraints 553 using the second parameters 540. In some embodiments, the rule applier 150 can modify the second constraints 352 to generate the third constraints 554 based on the second parameters 540. The rule applier 150 and / or the model generator 145 can modify and / or generate a model 345" based on the third constraints 554. The rule applier 150 can input, insert, or otherwise update a set of specific values corresponding to the second constraints 352 to generate the model 345" in accordance with the third constraints 553. For example, the first model 345 may include one or more algebraic equations that include, as a variable, a first number of new processes that can be assigned per day, among others. The rule applier 150 can modify the model 345 to generate the second model 345', by inputting, inserting or otherwise providing, to the first model 345, a specific number (e.g., 1, 3, 5, etc.) for the first number. Then, the rule applier 150 can further modify the model 345' to generate the third model 345", by updating the specific number and / or the constraints 553. For example, a first set of values can be input (e.g., as the second constraints 352) to the first model 345 to generate the second model 345', and then an updated set of values can be input (e.g., as the third constraints 553) to the second model 345' to generate the third model 345".
[0087] The resource allocator 155 can determine, based on applying the model 345" (in accordance with the third constraints 553), an allocation schedule 455'. The allocation scheduleAtty. Dkt. No.: 115872-3496455' can determine and / or define, for at least one process of the plurality of processes, at least one second resource of the plurality of resources in at least second one time slot of the plurality of time slots. In some embodiments, the allocation schedule 455' may be an updated version of the allocation schedule 455, updated based on the model 345" in accordance with the third constraints 553. For example, with the allocation schedule 455 having the process 281 " A to be executed by the resource 283' A on the time slot 282A and the process 281" B to be executed by the resource 283 'B on the time slot 282B, the resource allocator 155 can generate the allocation schedule 455' having the process 281" A to be executed by the resource 283'A on the time slot 282A and the process 281" B to be executed by the resource 283'A on the time slot 282B, for example, based on the third constraints 553 indicating that the resource 283'B is not available.
[0088] In this manner, the techniques disclosed herein allow for solving complex combinatorial problems in cost efficient manners by reducing computation resources required therefor (e.g., consuming a smaller amount of computing / memory resources). By modifying and relaxing the constraints, the combinatorial allocation problems may be solved in a more efficient manner by reducing the computing and memory resources to generate a feasible allocation schedule. The model may avoid enumerating every potential allocation across many resources and time slots, and instead may optimize for a function (e.g., workload or process-complexity) and output an allocation schedule that best satisfies the constraints. This model can reduce the time to find the optimal allocation schedule from approximately over hours to approximately less than a second. This represents a multiple-orders-of-magnitude reduction in latency for the same class of problem.
[0089] Along with these computing improvements, scheduling and workflow optimization in radiotherapy can be significantly enhanced while computer functionalities are improved. The techniques disclosed herein can streamline the optimization process but also empower more adaptive, data-driven decision-making in clinical environments. In the radiotherapy industry, these technological advances translate to tangible improvements in care delivery. Planners can now be matched more effectively to cases based on complexity, team roles, workload capacity, patient consistency, geographic location, and time-off schedules (e.g., including ramp-up and ramp-down periods). This results in more efficient plan assignments, reduced time-to-treatment, and improved scheduling overall. The techniques also centralizeAtty. Dkt. No.: 115872-3496communication, tracks patient journeys in a unified longitudinal record, monitors risk events, and supports workload management for care teams. These enhancements can increase efficiency in resource allocation clinical outcomes, ultimately improving patient care and reducing stress on planning teams.
[0090] FIGs. 6A-6C is a flow diagram of an example implementation 600 for allocating resources to process across time slots using data from disparate sources, in accordance with an illustrative embodiment. The implementation 600 can include or correspond to operations performed in the system 100. The implementation 600 is a non-limiting example, and one or more operations and / or processes can be added, omitted, or modified, without departing from scope of the present disclosure. Under the process 621, the system 100 (e.g., the data retriever 130) can retrieve, from disparate sources, various data including process data 611, resource data 612, schedule data 613, credential data 614, and parameter data 615. The process data 611 may include plan information, such as the processes 281, the processes 281', etc. of FIG. 2. For example, the process data 611 includes data on unassigned plans, plans associated with the resources 283, etc. The data retriever 130 can retrieve inputs, such as plan identifiers, Team, due dates, status of the plans, etc., from the process data 611. The resource data 612 may include resource information such as the resources 283 of FIG. 2. For example, the resource data 612 includes data on planners, radiologists, and / or their assignments. The data retriever 130 can retrieve planner identifiers, Team, pre-existing assignments, etc. from the resource data 612. The schedule data 613 may include information on staff schedules such as the resources 283 of FIG. 2. For example, the schedule data 613 includes data on availability of planners, radiologists, etc. The data retriever 130 can retrieve planned days out, completed days out, CWW days, etc. from the schedule data 613. The credential data 614 may include credential information such as the fourth data objects discussed with respect to FIG. 2. For example, the credential data 614 includes data on credentials of planners, radiologists, etc. The data retriever 130 can retrieve planner identifiers, plan identifiers (e g., for credentialed pairs) from the credential data 614. The parameter data 615 may include various parameters such as the parameters of FIG. 2, etc. For example, the parameter data 615 includes the user parameters (e.g., discussed in Section B), etc. The data retriever 130 can retrieve plan identifiers, Team, due dates, status of the plans, etc. from the parameter data 615.Atty. Dkt. No.: 115872-3496
[0091] In some embodiments, under the process 621, the system 100 (e.g., the data retriever 130, etc.) can pre-process the input data (e.g., the process data 611, the resource data 612, the schedule data 613, the credential data 614, the parameter data 615, etc.). In some examples, the system 100 can read the input data to identify anomalies or missing fields. In some embodiments, under the process 622, the system 100 (e.g., the data retriever 130, etc.) can pre-process absence data. For example, the system 100 can read planned absence and run calculations to determine long vacations, their impact on each plan to be assigned, etc. In some embodiments, under the process 621, the system 100 (e g., the parameter extractor 140) can generate instance data, which may be or include the parameters 240 including the allocations 241. The instance data may be in a dictionary form or a data structure based on the input data (e.g., the process data 611, the resource data 612, the schedule data 613, the credential data 614, the parameter data 615, etc.). Under the process 624, the system 100 (e.g., the model generator 145) can generate a model (e.g., the model 345).
[0092] The system 100 can build a shell of a mathematical model based on the input data and / or the parameters. For example, the system 100 can generate the shell that incorporates and / or translates the input data, the rules, the parameters, etc. into algebraic equations. In some embodiments, under the process 625, the system 100 (e.g., the rule applier 150) can instantiate the model. For example, the system 100 can use the data, generated and / or obtained through the processes 621 to 623, to fill the model. In some embodiments, the system 100 (e g., the resource allocator 155) can generate an allocation schedule (e.g., the allocation schedule 455) based on applying and / or optimizing the model. In some embodiments, under the process 627, the system 100 (e.g., the interface handler 160) can generate an output (e.g., the output 460). The output may include assignments 631 (e.g., from an optimized version of the model) of new treatment plans to planners and / or radiologists. In some examples, the system 100 can generate a planner output 632 and a schedule output 633. The planner output 632 may include workload metrics (e.g., the workload metric discussed with respect to FIG. 4) for the existing and / or new assignments. For example, the workload metrics may include the workload of the planners and / or radiologists identified in the assignments 631. The schedule output 633 may include a schedule view of the existing and / or new plans assigned to the planners and / or radiologists. Based on the planner output 632 and / or the schedule output 633, for example, the workload metrics, the system 100 canAtty. Dkt. No.: 115872-3496generate a second model (e.g., the model 345") to generate a second allocation schedule (e.g., the allocation schedule 455').
[0093] FIG. 7 is a block diagram of an example environment 700 for allocating resources to process across time slots using data from disparate sources, in accordance with an illustrative embodiment. The environment 700 can include or correspond to operations performed in the system 100. The environment 700 is a non-limiting example, and one or more operations and / or processes can be added, omitted, or modified, without departing from scope of the present disclosure.
[0094] The system 100 can be configured to allocate resources 783 for execution of processes 781. Each of the processes 781 may have different requirements (e.g., due dates, status, etc.), while each of a plurality of teams (e.g., Teams A, B, C, D, and E as shown) may have different capabilities to execute the processes 781 and / or each of the planners within the plurality of teams may have different qualifications (e.g., credentials) and availabilities (e.g., absences, workload balancing, etc.). For example, Teams A, B, C, D, and E may be disease-specific teams for different diseases. The system 100 can efficiently allocate the resources 783 to execute the processes 781 across a plurality of time slots. As discussed above, the system 100 can retrieve such requirements as data objects from disparate data sources, and generate a model that incorporates the requirements. By optimizing the model, the system 100 can determine an allocation schedule in which the resources 783 are allocated to the processes 781 across the plurality of time slots. This can ensure that the processes 781 are assigned to the suitable resources 783, reducing the time-to-treatment for patients and improving overall scheduling efficiency. The model’s ability to dynamically adjust schedules based on real-time data allows for greater flexibility in handling various requirements such as planner absences and workload balancing.
[0095] FIGs. 8A-8C is an example implementation 800 of retrieving a plurality of data objects, in accordance with an illustrative embodiment. In some embodiments, the implementation 800 may be displayed on the user interface 116. The implementation 800 can include or correspond to operations performed in the system 100. In the implementation 800, the system 100 (e.g., the data retriever 130) can retrieve a plurality of data objects associated with execution of a plurality of processes 881 on a plurality of instances 890 using a plurality of resources 883 across a pluralityAtty. Dkt. No.: 115872-3496of time slots 882. The data objects retrieved by the system 100 may include a first data object identifying the plurality of processes 881 unassigned to any of the plurality of time slots 882. Each of the plurality of processes 881 is to be executed on a respective one of the plurality of instances 890 using at least one of the plurality of resources 883 by at least one of the plurality of time slots 882. As shown, the data object may include various information associated with the plurality of processes 881, such as dates, patient names, plan sites, due dates, Team, treatment organizations, tier levels, etc. The data objects retrieved by the system 100 may include a second data object identifying the plurality of resources 883 as assigned to the plurality of processes 88 T. In some embodiments, the data objects may include various information associated with the plurality of resources 883, such as resource names, resource types (e.g., planners or radiologists), resource campus or location, workloads, Team, and resource availabilities (e.g., day off, the processes 88 T already assigned, etc ). In some embodiments, the data objects retrieved by the system 100 may include a third data object identifying available time slots 895 of the plurality of time slots 882 during which the resources 883 are available for allocation.
[0096] FIGs. 9A and 9B is an example implementation 900 of generating a model and modifying constraints of the model, in accordance with an illustrative embodiment. In some embodiments, the implementation 900 may be displayed on the user interface 116. The implementation 900 can include or correspond to operations performed in the system 100. In the implementation 900, the system 100 (e.g., the model generator 145, etc.) can generate a model 945 including a first plurality of constraints 951A-N (referred to as first constraints 951), based on parameters extracted from the data objects in the implementation 800. For example, the system 100 (e.g., the parameter extractor 140) can generate the model 945 based on parameters identified from the data objects that define allocation of the processes 881 on the instances 890 with one or more of the resources 883 across one or more of the time slots 882. The model 945 can include the first constraints 951, such as a number of maximum sites after PTO, a number of maximum sites per day, a number of maximum sites per week, a regional preference, etc. In some embodiments, the model 945 can include planner constraints associated with the resources 883 (e.g., whether to opt out a certain planner, workloads, etc.).
[0097] In the implementation 900, the system 100 (e.g., the rule applier 150, etc.) can modify the first constraints 951 to generate the model 945 in accordance with a second pluralityAtty. Dkt. No.: 115872-3496of constraints 952A-N. The system 100 can input certain values, relationships, and / or preferences into the model 945 so as to modify the model 945. For example, the system 100 can input “3” for the second constraint 952A to be applied to the first constraint 951 A, input “2” for the second constraint 952B to be applied to the first constraint 95 IB, input “6” for the second constraint 952N to be applied to the first constraint 95 IN, etc. The system 100 can select “None,” “Regional,” or “Campus” (e.g., as one of the second constraints 952) for the regional preference (e.g., one of the first constraints 951). The system 100 can determine an allocation schedule based on the model 945 in accordance with the second constraints 952.
[0098] Referring now to FIG. 9C, depicted is a screenshot of a user interface 970 to enter seed parameters for generating constraints of the model. The user interface 970 may be presented via a computing device (e.g., via computing device 115 and user interface 116) for receiving user-entered seed parameters used to instantiate or modify constraints of the optimization model (e.g., model 345). The user interface 970 may include one or more input elements to specify constraint values that define allocation of processes to resources across time slots, such as maximum workload thresholds (e.g., maximum processes per day, per week, or after out-of-office periods), planner / resource inclusion or exclusion selections, and preference settings (e.g., campus-based or region-based allocation). Seed parameters may be received via the user interface and may be used to derive the parameters into a data object or apply the parameters to modify the model constraints (e.g., from first constraints 951 to second constraints 952) in generating the allocation schedule.
[0099] Referring now to FIG. 9D, depicted is a screenshot of a user interface 980 to enter seed parameters for generating constraints of the model. The user interface 980 may be presented via a computing device (e.g., via computing device 115 and user interface 116) for receiving user-entered seed parameters used to instantiate or modify constraints of the optimization model (e.g., model 345). The user interface 980 may include one or more interface elements to specify constraint values define allocation of processes to resources across time slots, such as a maximum number of new sites, a maximum score (e.g., process-complexity score) for new sites, a maximum number of sites after leave, a maximum sites per day, a maximum score per day, a maximum number of sites, a maximum number sites per week, a maximum score per week, planner / resource inclusion or exclusion selections, and preference settings (e g., campus-based or region-based allocation), among others. The parameters may be received via the user interface and may be usedAtty. Dkt. No.: 115872-3496to derive the parameters into a data object or apply the parameters to modify the model constraints (e g., from first constraints 951 to second constraints 952) in generating the allocation schedule.
[0100] FIG. 10 is a flow diagram of method 1000 for allocating resources to process across time slots using data from disparate sources, in accordance with an illustrative embodiment. The method 1000 can be implemented or performed by any components detailed herein, such as system 100 or system 1100. A computing system can retrieve a plurality of data objects associated with execution of a plurality of processes on a plurality of instances using a plurality of resources across a plurality of time slots (1005). The computing system can identify a plurality of parameters defining allocation of the plurality of processes on the plurality of instances with one or more of the plurality of resources across one or more of the plurality of time slots (1010). The computing system can generate a model comprising a first plurality of constraints to define the allocation to satisfy the plurality of parameters (1015). The computing system can modify the first plurality of constraints of the model to generate a second plurality of constraints to define the allocation to satisfy a portion of the plurality of parameters (1020). The computing system can determine at least one resource of the plurality of resources in at least one time slot of the plurality of time slots (1025). The computing system can provide an output including information based on the allocation schedule (1030).D. NETWORK AND COMPUTING ENVIRONMENT
[0101] Various operations described herein can be implemented on computer systems. FIG. 11 shows a simplified block diagram of a representative server system 1100, computing system 1114, and network 1126 usable to implement certain embodiments of the present disclosure. In various embodiments, server system 1100 or similar systems can implement services or servers described herein or portions thereof. Computing system 1114 or similar systems can implement clients described herein. The system 100 described herein can be similar to the server system 1100. Server system 1100 can have a modular design that incorporates a number of modules 1102 (e.g., blades in a blade server embodiment); while two modules 1102 are shown, any number can be provided. Each module 1102 can include processing unit(s) 1104 and local storage 1106.Atty. Dkt. No.: 115872-3496
[0102] Processing unit(s) 1104 can include a single processor, which can have one or more cores, or multiple processors. In some embodiments, processing unit(s) 1104 can include a general-purpose primary processor as well as one or more special-purpose co-processors such as graphics processors, digital signal processors, or the like. In some embodiments, some, or all processing units 1104 can be implemented using customized circuits, such as application specific integrated circuits (ASICs) or field programmable gate arrays (FPGAs). In some embodiments, such integrated circuits execute instructions that are stored on the circuit itself. In other embodiments, processing unit(s) 1104 can execute instructions stored in local storage 1106. Any type of processors in any combination can be included in processing unit(s) 1104.
[0103] Local storage 1106 can include volatile storage media (e.g., DRAM, SRAM, SDRAM, or the like) and / or non-volatile storage media (e g., magnetic, or optical disk, flash memory, or the like). Storage media incorporated in local storage 1106 can be fixed, removable, or upgradeable as desired. Local storage 1106 can be physically or logically divided into various subunits such as a system memory, a read-only memory (ROM), and a permanent storage device. The system memory can be a read-and-write memory device or a volatile read-and-write memory, such as dynamic random-access memory. The system memory can store some or all of the instructions and data that processing unit(s) 1104 need at runtime. The ROM can store static data and instructions that are needed by processing unit(s) 1104. The permanent storage device can be a non-volatile read-and-write memory device that can store instructions and data even when module 1102 is powered down. The term “storage medium” as used herein includes any medium in which data can be stored indefinitely (subject to overwriting, electrical disturbance, power loss, or the like) and does not include carrier waves and transitory electronic signals propagating wirelessly or over wired connections.
[0104] In some embodiments, local storage 1106 can store one or more software programs to be executed by processing unit(s) 1104, such as an operating system and / or programs implementing various server functions such as functions of the system 100 or any other system described herein, or any other server(s) associated with system 100 or any other system described herein.Atty. Dkt. No.: 115872-3496
[0105] Software” refers generally to sequences of instructions that, when executed by processing unit(s) 1104, cause server system 1100 (or portions thereof) to perform various operations, thus defining one or more specific machine embodiments that execute and perform the operations of the software programs. The instructions can be stored as firmware residing in readonly memory and / or program code stored in non-volatile storage media that can be read into volatile working memory for execution by processing unit(s) 1104. Software can be implemented as a single program or a collection of separate programs or program modules that interact as desired. From local storage 1106 (or non-local storage described below), processing unit(s) 1104 can retrieve program instructions to execute and data to process in order to execute various operations described above.
[0106] In some server systems 1100, multiple modules 1102 can be interconnected via a bus or other interconnect 1108, forming a local area network that supports communication between modules 1102 and other components of server system 1100. Interconnect 1108 can be implemented using various technologies, including server racks, hubs, routers, etc.
[0107] A wide area network (WAN) interface 1110 can provide data communication capability between the local area network (e.g., through the interconnect 1108) and the network 1126, such as the Internet. Other technologies can be used to communicatively couple the server system 1100 with the network 1126, including wired (e g., Ethernet, IEEE 602.3 standards) and / or wireless technologies (e.g., Wi-Fi, IEEE 602.11 standards).
[0108] In some embodiments, local storage 1106 is intended to provide working memory for processing unit(s) 1104, providing fast access to programs and / or data to be processed while reducing traffic on interconnect 1108. Storage for larger quantities of data can be provided on the local area network by one or more mass storage 1112 that can be connected to interconnect 1108. Mass storage 1112 can be based on magnetic, optical, semiconductor, or other data storage media. Direct attached storage, storage area networks, network-attached storage, and the like can be used. Any data stores or other collections of data described herein as being produced, consumed, or maintained by a service or server can be stored in mass storage 1112. In some embodiments, additional data storage resources may be accessible via WAN interface 1110 (potentially with increased latency).Atty. Dkt. No.: 115872-3496
[0109] Server system 1100 can operate in response to requests received via WAN interface 1110. For example, one of modules 1102 can implement a supervisory function and assign discrete tasks to other modules 1102 in response to received requests. Work allocation techniques can be used. As requests are processed, results can be returned to the requester via WAN interface 1110. Such operation can generally be automated. Further, in some embodiments, WAN interface 1110 can connect multiple server systems 1100 to each other, providing scalable systems capable of managing high volumes of activity. Other techniques for managing server systems and server farms (collections of server systems that cooperate) can be used, including dynamic resource allocation and reallocation.
[0110] Server system 1100 can interact with various user-owned or user-operated devices via a wide-area network such as the Internet. An example of a user-operated device is shown in FIG. 11 as computing system 1114. Computing system 1114 can be implemented, for example, as a consumer device such as a smartphone, other mobile phone, tablet computer, wearable computing device (e.g., smart watch, eyeglasses), desktop computer, laptop computer, and so on.
[0111] For example, computing system 1114 can communicate via WAN interface 1110. Computing system 1114 can include computer components such as processing unit(s) 1116, storage device 1118, network interface 1120, user input device 1122, and user output device 1124. Computing system 1114 can be a computing device implemented in a variety of form factors, such as a desktop computer, laptop computer, tablet computer, smartphone, other mobile computing device, wearable computing device, or the like.
[0112] Processing unit 1116 and storage device 1118 can be similar to processing unit(s) 1104 and local storage 1106 described above. Suitable devices can be selected based on the demands to be placed on computing system 1114. For example, computing system 1114 can be implemented as a “thin” client with limited processing capability or as a high-powered computing device. Computing system 1114 can be provisioned with program code executable by processing unit(s) 1116 to enable various interactions with server system 1100.
[0113] Network interface 1120 can provide a connection to the network 1126, such as a wide area network (e.g., the Internet) to which WAN interface 1110 of server system 1100 is also connected. In various embodiments, network interface 1120 can include a wired interface (e.g.,Atty. Dkt. No.: 115872-3496Ethernet) and / or a wireless interface implementing various RF data communication standards such as Wi-Fi, Bluetooth, or cellular data network standards (e.g., 3G, 4G, LTE, etc ).
[0114] User input device 1122 can include any device (or devices) via which a user can provide signals to computing system 1114; computing system 1114 can interpret the signals as indicative of particular user requests or information. In various embodiments, user input device 1122 can include any or all of a keyboard, touch pad, touch screen, mouse, or other pointing device, scroll wheel, click wheel, dial, button, switch, keypad, microphone, and so on.
[0115] User output device 1124 can include any device via which computing system 1114 can provide information to a user. For example, user output device 1124 can include display-to-display images generated by or delivered to computing system 1114. The display can incorporate various image generation technologies, e.g., a liquid crystal display (LCD), light-emitting diode (LED) display including organic light-emitting diodes (OLED), projection system, cathode ray tube (CRT), or the like, together with supporting electronics (e.g., digital-to-analog or analog-to-digital converters, signal processors, or the like). Some embodiments can include a device such as a touchscreen that function as both input and output device. In some embodiments, other user output devices 1124 can be provided in addition to or instead of a display. Examples include indicator lights, speakers, tactile “display” devices, printers, and so on.
[0116] Some embodiments include electronic components, such as microprocessors, storage, and memory that store computer program instructions in a computer readable storage medium. Many of the features described in this specification can be implemented as processes that are specified as a set of program instructions encoded on a computer readable storage medium. When one or more processing units execute these program instructions, they cause the processing unit(s) to perform various operations indicated in the program instructions. Examples of program instructions or computer code include machine code, such as is produced by a compiler, and files including higher-level code that are executed by a computer, an electronic component, or a microprocessor using an interpreter. Through suitable programming, processing unit(s) 1104 and 1116 can provide various functionality for server system 1100 and computing system 1114, including any of the functionality described herein as being performed by a server or client, or other functionality.Atty. Dkt. No.: 115872-3496
[0117] It will be appreciated that server system 1100 and client computing system 1114 are illustrative and that variations and modifications are possible. Computer systems used in connection with embodiments of the present disclosure can have other capabilities not specifically described here. Further, while server system 1100 and client computing system 1114 are described with reference to particular blocks, it is to be understood that these blocks are defined for convenience of description and are not intended to imply a particular physical arrangement of component parts. For instance, different blocks can be, but need not be, located in the same facility, in the same server rack, or on the same motherboard. Further, the blocks need not correspond to physically distinct components. Blocks can be configured to perform various operations, e.g., by programming a processor or providing appropriate control circuitry, and various blocks might or might not be reconfigurable depending on how the initial configuration is obtained. Embodiments of the present disclosure can be realized in a variety of apparatus including electronic devices implemented using any combination of circuitry and software.
[0118] While the disclosure has been described with respect to specific embodiments, one skilled in the art will recognize that numerous modifications are possible. Embodiments of the disclosure can be realized using a variety of computer systems and communication technologies, including but not limited to specific examples described herein. Embodiments of the present disclosure can be realized using any combination of dedicated components and / or programmable processors and / or other programmable devices. The various processes described herein can be implemented on the same processor or different processors in any combination. Where components are described as being configured to perform certain operations, such configuration can be accomplished, e.g., by designing electronic circuits to perform the operation, by programming programmable electronic circuits (such as microprocessors) to perform the operation, or any combination thereof. Further, while the embodiments described above may reference specific hardware and software components, those skilled in the art will appreciate that different combinations of hardware and / or software components may also be used and that particular operations described as being implemented in hardware might also be implemented in software or vice versa.
[0119] Computer programs incorporating various features of the present disclosure may be encoded and stored on various computer readable storage media; suitable media includeAtty. Dkt. No.: 115872-3496magnetic disk or tape, optical storage media such as compact disk (CD) or digital versatile disk (DVD), flash memory, and other non-transitory media. Computer readable media encoded with the program code may be packaged with a compatible electronic device, or the program code may be provided separately from electronic devices (e.g., via Internet download or as a separately packaged computer-readable storage medium).
[0120] Thus, although the disclosure has been described with respect to specific embodiments, it will be appreciated that the disclosure is intended to cover all modifications and equivalents within the scope of the following claims.
Claims
Atty. Dkt. No.: 115872-3496WHAT IS CLAIMED IS:
1. A method of allocating resources to process across time slots using data from disparate sources, comprising:retrieving, by one or more processors, from one or more data sources, a plurality of data objects associated with execution of a plurality of processes on a plurality of instances using a plurality of resources across a plurality of time slots, the plurality of data objects comprising:a first data object identifying at least a first subset of processes of the plurality of processes unassigned to any of the plurality of time slots, each of the first subset of processes to be executed on a respective first instance of the plurality of instances using at least one of the plurality of resources by at least one of the plurality of time slots;a second data object identifying a subset of resources of the plurality of resources as assigned to a second subset of processes of the plurality of processes, each of the second subset of processes to be executed on a respective second instance of the plurality of instances using one respective resource of the subset of resources;a third data object identifying, for each respective resource of the plurality of resources, a first subset of time slots of the plurality of time slots during which the respective resource is available for allocation;identifying, by the one or more processors, from the plurality of data objects, a plurality of parameters defining allocation of the plurality of processes on the plurality of instances with one or more of the plurality of resources across one or more of the plurality of time slots;generating, by the one or more processors, a model comprising a first plurality of constraints to define the allocation to satisfy the plurality of parameters;modifying, by the one or more processors, the first plurality of constraints of the model to generate a second plurality of constraints to define the allocation to satisfy a portion of the plurality of parameters;determining, by the one or more processors, based on applying the model in accordance with the second plurality of constraints, an allocation schedule defining, for each respective process of the plurality of processes, at least one resource of the plurality of resources in at least one time slot of the plurality of time slots; andproviding, by the one or more processors, an output including information based on the allocation schedule.Atty. Dkt. No.: 115872-34962. The method of claim 1, further comprising:receiving, by the one or more processors, via an input on a user interface, a second plurality of parameters to redefine the allocation of the plurality of processes on the plurality of instances with one or more of the plurality of resources across one or more of the plurality of time slots, subsequent to providing the output for presentation;modifying, by the one or more processors, using the second plurality of parameters, the second plurality of constraints of the model to generate a third plurality of constraints; and determining, by the one or more processors, based on applying the model in accordance with the third plurality of constraints, a second allocation schedule defining, for at least one process of the plurality of processes, at least one second resource of the plurality of resources in at least second one time slot of the plurality of time slots.
3. The method of claim 1, wherein retrieving the plurality of data objects further comprises retrieving the plurality of data objects comprising a fourth data object identifying, for each resource of the plurality of resources, a respective capability to handle one or more of a plurality of process types.
4. The method of claim 1, further comprising providing, by the one or more processors, a user interface to input a second plurality of parameters to define at least one of: (i) a first maximum number of processes to be executed over one or more time slots of the plurality of time slots (ii) a second maximum number of processes to be executed by at least one of the plurality of resources over the one or more time slots, (iii) a ramp up period prior to executing at least one of the plurality of processes by at least one of the plurality of resources, or (iv) a score indicating a degree of complexity of at least one of the plurality of processes, andwherein retrieving the plurality of data objects further comprises receiving, via input on the user interface, the second plurality of parameters to include in a fifth data object of the plurality of data objects.
5. The method of claim 4, wherein modifying the first plurality of constraints further comprises modifying the first plurality of constraints to generate the second plurality ofAtty. Dkt. No.: 115872-3496constraints by changing the second plurality of parameters to set at least one of the first maximum number of processes, the second maximum number of processes, or ramp up period, or the score indicating the degree of complexity of at least one of the plurality of processes.
6. The method of claim 1, wherein identifying the plurality of parameters further comprises identifying, for each respective resource of the plurality of resources, a second subset of time slots of the plurality of time slots during which the respective resource is unavailable for allocation, using the third data object.
7. The method of claim 1, wherein identifying the plurality of parameters further comprises identifying, for each respective process of the plurality of processes, a score indicating a degree of complexity to execute the respective process,wherein determining the allocation schedule further comprises determining the allocation schedule based on the score degree of complexity to execute the respective process of the plurality of processes.
8. The method of claim 1, further comprising:identifying, by the one or more processors, using the second data object, the subset of resources as including (i) at least one resource associated with a first resource group and a second resource group of a plurality of resource groups and (ii) a remainder of resources associated with the first resource group; andretrieving, by the one or more processors, from the one or more data sources, a fourth data object identifying a second subset of resources as assigned to a third subset of processes of the plurality of processes, each of the third subset of processes to be executed on a respective third instance of the plurality of instances using one respective resource of the second subset of resources, the second subset of resources associated with the second resource group;wherein identifying the plurality of parameters further comprises identifying the plurality of parameters from the plurality of data objects and the fourth data object,wherein determining the allocation schedule further comprises determining the allocation schedule for the first resource group and the second resource group.Atty. Dkt. No.: 115872-34969. The method of claim 1, wherein generating the model further comprises generating the model to include the first plurality of constraint comprising at least one first constraint to distribute the allocation across one or more of the plurality of time slots to satisfy the plurality of parameters,wherein modifying the first plurality of constraints further comprises modifying the first plurality of constraints to generate the second plurality of constraints to include at least one second constraint to distribute the allocation across one or more of the plurality of time slots to satisfy the plurality of parameters,wherein determining the allocation schedule further comprises determining the allocation schedule based on applying the model in accordance with the second plurality of constraints including the at least one second constraint.
10. The method of claim 1, wherein providing the output further comprises providing, for presentation, the output including the information via a user interface comprising a plurality of user interface elements, the plurality of user interface elements comprising (i) a first user interface element identifying the plurality of resources, (ii) a second user interface element identifying the allocation of the plurality of processes to one or more of the plurality of time slots, and (iii) a third user interface element identifying the plurality of instances.
11. The method of claim 1, wherein providing the output further comprises generating the output to include a plurality of metrics corresponding to the plurality of resources based on the allocation schedule, each of the plurality of metrics indicating a degree of occupancy of a respective resource of the plurality of resources.
12. The method of claim 1, further comprising:determining, by the one or more processors, that no allocation schedule satisfies a third plurality of constraints of the model using a second plurality of parameters identified from a second plurality of data objects, andproviding, by the one or more processors, a second output to indicate that no allocation schedule is generated in accordance with the third plurality of constraints of the model.Atty. Dkt. No.: 115872-349613. The method of claim 1, wherein at least one of the plurality of processes comprises administration of radiation therapy to a subject corresponding to at least one of the plurality of instances during at least one of the plurality of time slots,wherein the radiation therapy comprises at least one of a brachytherapy, a magnetic resonance imaging-guided linear accelerator (MR-LINAC), external beam radiotherapy (EBRT), stereotactic body radio therapy (SBRT), intensity -modulated radiotherapy (IMRT), a volumetric modulated arc therapy (VMAT), or proton beam therapy (PBT), andwherein the subject is diagnosed with a cancer to be administered with the radiotherapy.
14. A system for allocating resources to processes across time slots using data from disparate sources, comprising:one or more processors coupled with memory, configured to:retrieve, from one or more data sources, a plurality of data objects associated with execution of a plurality of processes on a plurality of instances using a plurality of resources across a plurality of time slots, the plurality of data objects comprising:a first data object identifying at least a first subset of processes of the plurality of processes unassigned to any of the plurality of time slots, each of the first subset of processes to be executed on a respective first instance of the plurality of instances using at least one of the plurality of resources by at least one of the plurality of time slots;a second data object identifying a subset of resources of the plurality of resources as assigned to a second subset of processes of the plurality of processes, each of the second subset of processes to be executed on a respective second instance of the plurality of instances using one respective resource of the subset of resources;a third data object identifying, for each respective resource of the plurality of resources, a first subset of time slots of the plurality of time slots during which the respective resource is available for allocation;identify, from the plurality of data objects, a plurality of parameters defining allocation of the plurality of processes on the plurality of instances with one or more of the plurality of resources across one or more of the plurality of time slots;generate a model comprising a first plurality of constraints to define the allocation to satisfy the plurality of parameters;Atty. Dkt. No.: 115872-3496modify the first plurality of constraints of the model to generate a second plurality of constraints to define the allocation to satisfy a portion of the plurality of parameters;determine, based on applying the model in accordance with the second plurality of constraints, an allocation schedule defining, for each respective process of the plurality of processes, at least one resource of the plurality of resources in at least one time slot of the plurality of time slots; andprovide an output including information based on the allocation schedule.
15. The system of claim 14, wherein the one or more processors are configured to:receive, via an input on a user interface, a second plurality of parameters to redefine the allocation of the plurality of processes on the plurality of instances with one or more of the plurality of resources across one or more of the plurality of time slots, subsequent to providing the output for presentation;modify, using the second plurality of parameters, the second plurality of constraints of the model to generate a third plurality of constraints; anddetermine, based on applying the model in accordance with the third plurality of constraints, a second allocation schedule defining, for at least one process of the plurality of processes, at least one second resource of the plurality of resources in at least second one time slot of the plurality of time slots.
16. The system of claim 14, wherein the one or more processors, to retrieve the plurality of data objects, are configured to retrieve the plurality of data objects comprising a fourth data object identifying, for each resource of the plurality of resources, a respective capability to handle one or more of a plurality of process types.
17. The system of claim 14, wherein the one or more processors are configured to:provide a user interface to input a second plurality of parameters to define at least one of: (i) a first maximum number of processes to be executed over one or more time slots of the plurality of time slots (ii) a second maximum number of processes to be executed by at least one of the plurality of resources over the one or more time slots, (iii) a ramp up period prior toAtty. Dkt. No.: 115872-3496executing at least one of the plurality of processes by at least one of the plurality of resources, or (iv) a score indicating a degree of complexity of at least one of the plurality of processes, and receive, via input on the user interface, the second plurality of parameters to include in a fifth data object of the plurality of data objects.
18. The system of claim 17, wherein the one or more processors, to modify the first plurality of constraints, are configured to modify the first plurality of constraints to generate the second plurality of constraints by changing the second plurality of parameters to set at least one of the first maximum number of processes, the second maximum number of processes, the ramp up period, or the score indicating the degree of complexity of at least one of the plurality of processes.
19. The system of claim 14, wherein the one or more processors, to identify the plurality of parameters, are configured to identify, for each respective resource of the plurality of resources, a second subset of time slots of the plurality of time slots during which the respective resource is unavailable for allocation, using the third data object.
20. The system of claim 14, wherein the one or more processors are further configured to: identify, for each respective process of the plurality of processes, a score indicating a degree of complexity to execute the respective process;determine the allocation schedule based on the score degree of complexity to execute the respective process of the plurality of processes.
21. The system of claim 14, wherein the one or more processors are further configured to: identify, using the second data object, the subset of resources as including (i) at least one resource associated with a first resource group and a second resource group of a plurality of resource groups and (ii) a remainder of resources associated with the first resource group;retrieve, from the one or more data sources, a fourth data object identifying a second subset of resources as assigned to a third subset of processes of the plurality of processes, each of the third subset of processes to be executed on a respective third instance of the plurality ofAtty. Dkt. No.: 115872-3496instances using one respective resource of the second subset of resources, the second subset of resources associated with the second resource group;identify the plurality of parameters from the plurality of data objects and the fourth data object; anddetermine the allocation schedule for the first resource group and the second resource group.
22. The system of claim 14, wherein the one or more processors are configured to:generate the model to include the first plurality of constraint comprising at least one first constraint to distribute the allocation across one or more of the plurality of time slots to satisfy the plurality of parameters;modify the first plurality of constraints to generate the second plurality of constraints to include at least one second constraint to distribute the allocation across one or more of the plurality of time slots to satisfy the plurality of parameters; anddetermine the allocation schedule based on applying the model in accordance with the second plurality of constraints including the at least one second constraint.
23. The system of claim 14, wherein the one or more processors, to provide the output, are configured to provide, for presentation, the output including the information via a user interface comprising a plurality of user interface elements, the plurality of user interface elements comprising (i) a first user interface element identifying the plurality of resources, (ii) a second user interface element identifying the allocation of the plurality of processes to one or more of the plurality of time slots, and (iii) a third user interface element identifying the plurality of instances.
24. The system of claim 14, wherein the one or more processors, to provide the output, are configured to generate the output to include a plurality of metrics corresponding to the plurality of resources based on the allocation schedule, each of the plurality of metrics indicating a degree of occupancy of a respective resource of the plurality of resources.
25. The system of claim 14, wherein the one or more processors are configured to:Atty. Dkt. No.: 115872-3496determine that no allocation schedule satisfies a third plurality of constraints of the model using a second plurality of parameters identified from a second plurality of data objects, and provide a second output to indicate that no allocation schedule is generated in accordance with the third plurality of constraints of the model.
26. The system of claim 14, wherein at least one of the plurality of processes comprises administration of radiation therapy to a subject corresponding to at least one of the plurality of instances during at least one of the plurality of time slots,wherein the radiation therapy comprises at least one of a brachytherapy, a magnetic resonance imaging-guided linear accelerator (MR-LINAC), external beam radiotherapy (EBRT), stereotactic body radio therapy (SBRT), intensity-modulated radiotherapy (IMRT), a volumetric modulated arc therapy (VMAT), or proton beam therapy (PBT), andwherein the subject is diagnosed with a cancer to be administered with the radiotherapy.
