Patent Issued for Systems And Methods For Multi-Resource Scheduling (USPTO 10,997,530)

2021 MAY 14 (NewsRx) -- By a News Reporter-Staff News Editor at Hospital & Nursing Home Daily -- From Alexandria, Virginia, NewsRx journalists report that a patent by the inventors Bollapragada, Srinivas (Schenectady, NY); Johnson, Christopher Donald (Clifton Park, NY); Thomas, Bex (Ballston Lake, NY); Chace, Dianne (Charlotte, VT), filed on June 29, 2017, was published online on May 17, 2021.

The patent’s assignee for patent number 10,997,530 is General Electric Company (Schenectady, New York, United States).

News editors obtained the following quote from the background information supplied by the inventors: “The invention relates generally to process management systems, and more particularly to scheduling systems in the clinical setting, such as healthcare delivery institutions or hospitals.

“Healthcare delivery institutions are business systems that can be designed and operated to achieve their stated missions robustly. As is the case with other business systems such as those designed to provide services and manufactured goods, there are benefits to managing variation such that the stake-holders within these business systems can focus more fully on the value added core processes that achieve the stated mission and less on activity responding to variations such as emergency procedures, regular medical interventions, delays, accelerations, backups, underutilized assets, unplanned overtime by staff and stock outs of material, equipment, people and space that are impacted in the course of delivering healthcare.

“Currently clinical process decisions have historically relied on the art of understanding symptoms and diagnosing causality much in alignment with the practice of the medical diagnosis arts. In an ever-evolving environment, judgment and experientially-developed mental models are utilized by the healthcare providers to utilize the information currently at hand to make decisions. Presented with similar data, the decision made from one caregiver to another typically exhibits a variation. Presented with partial information, which is the byproduct of being organized in functional departments, specialties, roles and by the nature of having partial and/or current or dated information availability on hand--clinical process decisions vary widely and typically are locally focused for lack of a systems view upstream and downstream of the decision point.

“As a hospital processes care plans on an increasing patient load, these variations in medical condition and selected treatment plans perturbs the schedules of doctors, nurses and assets such as rooms and equipment. If there is protective capacity in these schedules and staff, the providers of care can manage variation while maintaining care quality. When randomness and interdependencies exceed the ability to serve, care providers are forced to make choices amongst poor alternative options; someone or something is going to be bottlenecked or overextended. Delays, queues, overtime, burnout and emotional decision making characterize systems that are over-taxed or beyond their ability to perform.

“Where information systems exist, they are simply informational in nature. Examples include scheduled rooms, people, materials and equipment. Recent advances in locating devices such as those utilizing radio-frequency identification (RFID) technology to report a location of a tagged asset are utilized, yet personnel are loath to be tracked by wearing a device. RFID devices are not pervasive, and the systems that monitor them are not fully integrated with the requisite adjacent systems that gather contextual information. The current art is not predictive, probabilistic nor necessarily systemic. For example, knowing the location of an asset is desirable but knowing its anticipated need and interdependencies is required to make a decision to use a located asset actionable. The information required for such a decision comes from a plurality of adjacent health information systems and must have an ability to play forward into the future.

“Today, current procedure duration and room status is provided without any proactive or forward-looking capability. Schedules are produced before a day’s activities commence. Process status is displayed along with trending and, often, alarm functionality should a process variable trip a threshold set point. Today, processes are planned for a given volume; when that volume is exceeded or processes have sufficient variation to overtax their capability, scheduling and recovery are reduced to manual triage and experience to sort out. Typically, queues, delay, overtime and cancellation result; there is no proactive decision support to dynamically reschedule people or physical assets or supplies.

“Radiology Information Systems (RIS) and other clinical information systems are in wide use in the healthcare industry to manage radiology departments in hospitals and independent radiology clinics. These systems typically incorporate functionality to schedule patients on radiology equipment such as computed tomography (CT) and magnetic resonance imaging (MRI) machines. However, radiology exams also require a numbers of other resources such as technicians, nurses, radiologists, anesthesiologists and other equipment such as portable ultra sound and X-ray machines. In general, these resources are not scheduled and are assumed to be available during the times when the exams are scheduled. However, this is not always true and leads to delays in completing the scheduled exams.”

As a supplement to the background information on this patent, NewsRx correspondents also obtained the inventors’ summary information for this patent: “Certain examples systems and methods for multi-resource scheduling to schedule resources involved in an exam. Certain examples enable both automatic and manual scheduling and/or rescheduling of inpatient and outpatient appointments.

“Certain examples provide a multi-resource scheduler system for a clinical enterprise. The system includes a processor connected to a memory, wherein the processor is programmed to implement the system. A scheduler engine is to generate a schedule for a clinical facility involving multiple tasks and using multiple resources. The scheduler engine is to identify a slot for a task defined by a scheduled task duration and one or more resources. The task includes a plurality of sub-tasks, and each sub-task has a sub-task duration utilizing one or more of the one or more resources. Each sub-task is to be performed consecutively based on resource constraints. The scheduler engine is to identify and select a time slot for the task based on resource availability, the plurality of sub-tasks in the task, and a duration associated with each sub-task. Resource availability information is obtained from a clinical information system. Each resource is scheduled only for one or more sub-tasks in which the resource is involved. A scheduler interface is to display and facilitate access to the schedule including the task and the plurality of sub-tasks.

“Certain examples provide a tangible computer-readable storage medium including a set of instructions for execution on a computer. The set of instructions, when executed, implementing a multi-resource clinical scheduler. The scheduler includes a scheduler engine to generate a schedule for a clinical facility involving multiple tasks and using multiple resources. The scheduler engine is to identify a slot for a task defined by a scheduled task duration and one or more resources. The task includes a plurality of sub-tasks. Each sub-task has a sub-task duration utilizing one or more of the one or more resources. Each sub-task is to be performed consecutively based on resource constraints. The scheduler engine is to identify and select a time slot for the task based on resource availability, the plurality of sub-tasks in the task, and a duration associated with each sub-task. Resource availability information is obtained from a clinical information system. Each resource is scheduled only for one or more sub-tasks in which the resource is involved. The system includes a scheduler interface to display and facilitate access to the schedule including the task and the plurality of sub-tasks.

“Certain examples provide a computer-implemented method for scheduling of clinical tasks involving multiple sub-tasks and multiple resources in a clinical enterprise. The method includes identifying a slot for a task defined by a task duration and one or more resources, the task including a plurality of sub-tasks, each sub-task having a sub-task duration utilizing one or more of the one or more resources, wherein each sub-task to be performed consecutively based on resource constraints; selecting a time slot for the task based on resource availability, the plurality of sub-tasks in the task, and a duration associated with each sub-task, wherein resource availability information is obtained from a clinical information system, and wherein each resource is scheduled only for one or more sub-tasks in which the resource is involved; displaying the schedule including the task and the plurality of sub-tasks; and facilitating access to view and modify the schedule.”

The claims supplied by the inventors are:

“The invention claimed is:

“1. A multi-resource scheduling apparatus comprising: a processor to implement a scheduler engine in conjunction with one or more clinical systems, wherein the scheduler engine comprises at least programming instructions which, in response to execution by the processor, cause the processor to: enable the one or more clinical systems to operate with the scheduler engine in an analytical mode and an operating mode, wherein, when in the analytical mode: the scheduler engine dynamically calculates one or more binding constraints on the one or more clinical systems for scheduling, wherein the calculation of the one or more binding constraints is based on one or more event inputs received from the one or more clinical systems, the event inputs comprising at least a sequence of one or more tasks and sub-tasks, and associated resource locations for the one or more tasks and sub-tasks; when in the operating mode: the scheduler engine configures and transmits a schedule to one or more scheduler interfaces of the one or more clinical systems, wherein the schedule is based on the one or more binding constraints calculated in the analytical mode, and wherein the scheduler engine monitors, via a computer network, execution of the schedule by the one or more clinical systems, wherein the monitoring comprises tracking the resource locations during execution of the sequence of the one or more tasks and sub-tasks; wherein the scheduler engine dynamically switches between the analytical mode and the operating mode based at least in part on a probabilistic determination of delay associated with the schedule, wherein the probabilistic determination of delay is associated with a schedule risk and the scheduler engine continually calculates the probabilistic determination of delay based on the monitoring, the scheduler engine triggered to switch from the operating mode to the analytical mode when the scheduler engine determines, based on the probabilistic determination of delay and the associated schedule risk, that at least one of the one or more binding constraints is not satisfied and to switch from the analytical mode to the operating mode when the scheduler engine has recalculated the one or more binding constraints, and wherein the scheduler engine transmits a reconfigured schedule to the one or more clinical systems to provide load balancing and dynamic schedule adjustment to the one or more clinical systems.

“2. The apparatus of claim 1, wherein the schedule risk is associated with a probability density function of time for a duration estimation of the one or more tasks in the schedule.

“3. The apparatus of claim 2, wherein the schedule risk is to be determined based on a simulation of task durations and logic for interdependencies between resources.

“4. The apparatus of claim 2, wherein the schedule risk is to be determined based on measures of duration, availability, and reliability for the schedule.

“5. The apparatus of claim 1, wherein the scheduler engine is to dynamically switch between the analytical mode and the operating mode based at least in part on a change in location of resources for adjacent tasks such that an estimation of transport time relative to task duration decreases a probability of task completion below an adjustable level.

“6. The apparatus of claim 1, wherein, when in the analytical mode, the scheduler engine is to calculate slack values for relaxing the binding constraints.

“7. The apparatus of claim 1, wherein, when in the analytical mode, the scheduler engine is to perform at least one of a what-was, what-is, what-if, or forecast analysis with respect to the one or more clinical systems for throughput and resource utilization.

“8. The apparatus of claim 1, wherein the scheduler engine is to determine the schedule based on at least one of a constraint satisfaction problem or heuristics to determine resource availability for the schedule.

“9. A non-transitory computer-readable storage medium comprising a set of instructions, which, in response to execution by a processor, cause the processor to implement a scheduler engine operating in conjunction with one or more clinical systems, the scheduler engine configured to: enable the one or more clinical systems to operate with the scheduler engine in an analytical mode and an operating mode, wherein, when in the analytical mode; the scheduler engine dynamically calculates one or more binding constraints on the one or more clinical systems for scheduling, wherein the calculation of the one or more binding constraints is based on one or more event inputs received from the one or more clinical systems, the event inputs comprising one or more tasks and sub-tasks, and associated resource locations for the one or more tasks and sub-tasks; and when in the operating mode: the scheduler engine configures and transmits a schedule to one or more scheduler interfaces of the one or more clinical systems, wherein the schedule is based on the one or more binding constraints calculated in the analytical mode, and wherein the schedule engine monitors, via a computer network, execution of the schedule by the one or more clinical systems, wherein the monitoring comprises tracking the resource locations during execution of the one or more tasks and sub-tasks; wherein the scheduler engine dynamically switches between the analytical mode and the operating mode based at least in part on a probabilistic determination of delay associated with the schedule and the scheduler engine continually calculates the probabilistic determination of delay based on the monitoring, the scheduler engine triggered to switch from the operating mode to the analytical mode when the scheduler engine determines, based on the probabilistic determination of delay, that at least one of the one or more binding constraints is not satisfied and to switch from the analytical mode to the operating mode when the scheduler engine has recalculated the one or more binding constraints; and wherein the scheduler engine transmits a reconfigured schedule to the one or more clinical systems to provide load balancing and dynamic schedule adjustment to the one or more clinical systems.

“10. The computer-readable storage medium of claim 9, wherein the probabilistic determination of delay associated with the schedule includes a schedule risk associated with a probability density function of time for a duration estimation of a task in the schedule.

“11. The computer-readable storage medium of claim 10, wherein the schedule risk is to be determined based on a simulation of task durations and logic for interdependencies between resources.

“12. The computer-readable storage medium of claim 10, wherein the schedule risk is to be determined based on measures of duration, availability, and reliability for the schedule.

“13. The computer-readable storage medium of claim 9, wherein the scheduler engine is to dynamically switch between the analytical mode and the operating mode based at least in part on a change in location of resources for adjacent tasks such that an estimation of transport time relative to task duration decreases a probability of task completion below an adjustable level.

“14. The computer-readable storage medium of claim 9, wherein, when in the analytical mode, the scheduler engine is to calculate slack values for relaxing the binding constraints.

“15. The computer-readable storage medium of claim 9, wherein, when in the analytical mode, the scheduler engine is to perform at least one of a what-was, what-is, what-if, or forecast analysis with respect to the one or more clinical systems for throughput and resource utilization.

“16. The computer-readable storage medium of claim 9, wherein the scheduler engine is to determine the schedule based on at least one of a constraint satisfaction problem or heuristics to determine resource availability for the schedule.

“17. A computer-implemented method for multi-resource scheduling, the method comprising: configuring, by a scheduler engine including a processor, the scheduler engine to operate with one or more clinical systems in an analytical mode and an operating mode; when in the analytical mode, dynamically calculating, by the scheduler engine, one or more binding constraints on the one or more clinical systems for scheduling, wherein the calculation of the one or more binding constraints is based on one or more event inputs received from the one or more clinical systems, the event inputs comprising at least one or more tasks and sub-tasks, and associated resource locations for the one or more tasks and sub-tasks; when in the operating mode, configuring and transmitting a schedule, by the scheduler engine to one or more scheduler interfaces of the one or more clinical systems, wherein the schedule is based on the one or more binding constraints calculated in the analytical mode, and monitoring, via a computer network, execution of the schedule by the one or more clinical systems, wherein the monitoring comprises tracking the resource locations during execution of the one or more tasks and sub-tasks, the schedule to be output to the one or more clinical systems to provide load balancing and dynamic schedule adjustment to the one or more clinical systems; and dynamically switching, by the scheduler engine, between the analytical mode and the operating mode based at least in part on a probabilistic determination of delay associated with the schedule, wherein the probabilistic determination of delay is associated with a schedule risk and the scheduler engine calculates the probabilistic determination of delay based on the monitoring, the scheduler engine triggered to switch from the operating mode to the analytical mode when the scheduler engine determines, based on the probabilistic determination of delay and the associated schedule risk, that at least one of the one or more binding constraints is not satisfied and to switch from the analytical mode to the operating mode when the scheduler engine has recalculated the one or more binding constraints; and transmitting a reconfigured schedule to the one or more clinical systems to provide load balancing and dynamic schedule adjustment to the one or more clinical systems.

“18. The method of claim 17, wherein the schedule risk is associated with a probability density function of time for a duration estimation of the one or more tasks in the schedule.

“19. The method of claim 17, wherein the scheduler engine is to dynamically switch between the analytical mode and the operating mode based at least in part on a change in location of resources for adjacent tasks such that an estimation of transport time relative to task duration decreases a probability of task completion below an adjustable level.

“20. The method of claim 17, wherein, when in the analytical mode, the scheduler engine is to calculate slack values for relaxing the binding constraints.”

For additional information on this patent, see: Bollapragada, Srinivas; Johnson, Christopher Donald; Thomas, Bex; Chace, Dianne. Systems And Methods For Multi-Resource Scheduling. U.S. Patent Number 10,997,530, filed June 29, 2017, and published online on May 17, 2021. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=10,997,530.PN.&OS=PN/10,997,530RS=PN/10,997,530

(Our reports deliver fact-based news of research and discoveries from around the world.)