US-12626801-B2 - Determining a regimen to improve health parameter of an object

US12626801B2US 12626801 B2US12626801 B2US 12626801B2US-12626801-B2

Abstract

A process includes receiving a set of static characteristics, a set of dynamic characteristics, and a health objective of a target object. A set of reference object data is retrieved and includes multiple reference objects of the same type as the target object. The reference object data includes the same data as the target object data. The dynamic data of the target object and the reference objects is parameterized over time. Each a set of dynamic characteristics for each reference object is split into a plurality of reference profiles. A set of similar reference profiles having a higher similarity metric than a remainder of the reference profiles is identified. A goal difference between the health objective of the target object and an end state of the similar reference profiles is determined. The reference profile having the smallest goal difference is identified and a corresponding health regimen is implemented.

Inventors

Xiao Ling Yang
Lei Tian
Jing James Xu
Si Er Han
Xue Ying ZHANG
XIAO MING MA

Assignees

INTERNATIONAL BUSINESS MACHINES CORPORATION

Dates

Publication Date: 20260512
Application Date: 20240126

Claims (12)

1 . A computer-implemented method comprising: receiving a set of static characteristics of a target object, a set of dynamic characteristics of the target object at a processor, and a health objective of the target object, wherein the set of dynamic characteristics vary over time, wherein the target object is a computer system, wherein the health objective is a computer system lifecycle, wherein the static characteristics include a memory capacity, a storage capacity, and one of a number of cores and a a number of threads of the computer system, and wherein the set of dynamic characteristics include central processing unit (CPU) usage as a percentage of CPU utilization over time, memory usage as an amount of RAM in use over time, network traffic as incoming and outgoing data transfer rates over time, disk input/output (I/O) as read and write speeds on storage devices over time, and system load as average system load over time and/or the number of jobs running over time; and defining a goal of the target object and identifying quantifiable metrics reflective of the goal; retrieving a set of reference object data, the reference object data including a plurality of reference objects of the same object type as the target object, the reference object data having a set of static characteristics for each reference object in the plurality of reference objects and a set of dynamic characteristics for each reference object in the plurality of reference objects, the set of dynamic characteristics for each reference object varying over time; parameterizing the set of dynamic characteristics of the target object and the set of dynamic characteristics for each reference object in the plurality of reference objects using a time series model; splitting each set of dynamic characteristics for each reference object into a plurality of reference profiles using the parameterized dynamic characteristics, wherein each reference profile in the plurality of reference profiles includes a current health status of the reference object at a time step and time series information of the reference object proceeding onwards from the time step; identifying a set of similar reference profiles, the set of similar reference profiles being a subset of the plurality of reference profiles with each reference profile in the set of similar reference profiles having a higher similarity metric than each reference profile excluded from the set of similar reference profiles; determining a goal difference between the goal of the target object and an end state of each reference profile in the set of similar reference profiles, wherein determining the goal difference includes determining a difference between the quantifiable metrics of the goal and corresponding quantifiable metrics of each reference profile in the set of similar reference profiles; and identifying the reference profile in the set of similar reference profiles having the smallest goal difference and implementing a health regimen of the reference object corresponding to the reference profile having the smallest goal difference in the target object, wherein implementing the health regimen of the reference object corresponding to the reference profile having the smallest goal difference in the target object includes implementing a maintenance and update schedule of the reference object corresponding to the reference profile having the smallest goal difference in the target object.
2 . The computer-implemented method of claim 1 , wherein parameterizing the set of dynamic characteristics of the target object and the set of dynamic characteristics for each reference object in the plurality of reference objects using a time series model comprises applying a Holt-Winter exponential smoothing model including three parameters to the set of dynamic characteristics of the target object and the set of dynamic characteristics for each reference object in the plurality of reference objects.
3 . The computer-implemented method of claim 1 , wherein within the plurality of reference profiles each reference profile represents a current health status of the corresponding reference object at a given time step, and includes time series information of the corresponding reference object proceeding onwards from the given time step.
4 . The computer-implemented method of claim 3 , wherein a number of reference profiles is equal to a number of time steps in the set of dynamic characteristics for each reference object in the plurality of reference objects.
5 . A system comprising: a processor and a memory, wherein the memory stores instructions for causing the processor to perform the steps of: receiving a set of static characteristics of a target object, a set of dynamic characteristics of the target object at a processor, and a health objective of the target object, wherein the set of dynamic characteristics vary over time, wherein the target object is a computer system, wherein the health objective is a computer system lifecycle, wherein the static characteristics include a memory capacity, a storage capacity, and one of a number of cores and a number of threads of the computer system, and wherein the set of dynamic characteristics include central processing unit (CPU) usage as a percentage of CPU utilization over time, memory usage as an amount of RAM in use over time, network traffic as incoming and outgoing data transfer rates over time, disk input/output (I/O) as read and write speeds on storage devices over time, and system load as average system load over time and/or the number of jobs running over time; and; retrieving a set of reference object data, the reference object data including a plurality of reference objects of the same object type as the target object, the reference object data having a set of static characteristics for each reference object in the plurality of reference objects and a set of dynamic characteristics for each reference object in the plurality of reference objects, the set of dynamic characteristics for each reference object varying over time; parameterizing the set of dynamic characteristics of the target object and the set of dynamic characteristics for each reference object in the plurality of reference objects using a time series model; splitting each set of dynamic characteristics for each reference object in the plurality of reference profiles into a plurality of new reference profiles using the parameterized dynamic characteristics, and creating a new set of reference profiles including each reference profile in the plurality of new reference profiles, wherein each reference profile in the plurality of new reference profiles includes a current health status of the reference object at a time step and time series information of the reference object proceeding onwards from the time step; identifying a set of similar reference profiles, the set of similar reference profiles being a subset of the plurality of reference profiles with each reference profile in the set of similar reference profiles having a higher similarity metric than each reference profile excluded from the set of similar reference profiles; determining a goal difference between the health objective of the target object and an end state of each reference profile in the set of similar reference profiles; and identifying the reference profile in the set of similar reference profiles having the smallest goal difference and implementing a health regimen of the reference object corresponding to the reference profile having the smallest goal difference in the target object, wherein implementing the health regimen of the reference object corresponding to the reference profile having the smallest goal difference in the target object includes implementing a maintenance and update schedule of the reference object corresponding to the reference profile having the smallest goal difference in the target object.
6 . The system of claim 5 , wherein parameterizing the set of dynamic characteristics of the target object and the set of dynamic characteristics for each reference object in the plurality of reference objects using a time series model comprises applying a Holt-Wagner exponential smoothing model including three parameters to the set of dynamic characteristics of the target object and the set of dynamic characteristics for each reference object in the plurality of reference objects.
7 . The system of claim 5 , wherein within the plurality of reference profiles each reference profile represents a current health status of the corresponding reference object at a given time step, and includes time series information of the corresponding reference object proceeding onwards from the given time step.
8 . The system of claim 7 , wherein a number of reference profiles is equal to a number of time steps in the set of dynamic characteristics for each reference object in the plurality of reference objects.
9 . A computer program product comprising a non-transitory memory storing instructions configured to cause a computer system to perform a method including the steps of: receiving a set of static characteristics of a target object, a set of dynamic characteristics of the target object at a processor, and a health objective of the target object, wherein the set of dynamic characteristics vary over time; retrieving a set of reference object data, the reference object data including a plurality of reference objects of the same object type as the target object, the reference object data having a set of static characteristics for each reference object in the plurality of reference objects and a set of dynamic characteristics for each reference object in the plurality of reference objects, the set of dynamic characteristics for each reference object varying over time, wherein the target object is a computer system, wherein the health objective is a computer system lifecycle, wherein the static characteristics include a memory capacity, a storage capacity, and one of a number of cores and a number of threads of the computer system, and wherein the set of dynamic characteristics include central processing unit (CPU) usage as a percentage of CPU utilization over time, memory usage as an amount of RAM in use over time, network traffic as incoming and outgoing data transfer rates over time, disk input/output (I/O) as read and write speeds on storage devices over time, and system load as average system load over time and/or the number of jobs running over time; and; parameterizing the set of dynamic characteristics of the target object and the set of dynamic characteristics for each reference object in the plurality of reference objects using a time series model; splitting each set of dynamic characteristics for each reference object in the plurality of reference profiles into a plurality of new reference profiles using the parameterized dynamic characteristics and creating a new set of reference profiles including each reference profile in the plurality of new reference profiles, wherein each reference profile in the plurality of new reference profiles includes a current health status of the reference object at a time step and time series information of the reference object proceeding onwards from the time step; identifying a set of similar reference profiles, the set of similar reference profiles being a subset of the plurality of new reference profiles with each reference profile in the set of similar reference profiles having a higher similarity metric than each reference profile excluded from the set of similar reference profiles; determining a goal difference between the health objective of the target object and an end state of each reference profile in the set of similar reference profiles; and identifying the reference profile in the set of similar reference profiles having the smallest goal difference and implementing a health regimen of the reference object corresponding to the reference profile having the smallest goal difference in the target object, wherein implementing the health regimen of the reference object corresponding to the reference profile having the smallest goal difference in the target object includes implementing a maintenance and update schedule of the reference object corresponding to the reference profile having the smallest goal difference in the target object.
10 . The computer program product of claim 9 , wherein parameterizing the set of dynamic characteristics of the target object and the set of dynamic characteristics for each reference object in the plurality of reference objects using a time series model comprises applying a Holt-Winter exponential smoothing model including three parameters to the set of dynamic characteristics of the target object and the set of dynamic characteristics for each reference object in the plurality of reference objects.
11 . The computer program product of claim 9 , wherein within the plurality of reference profiles each reference profile represents a current health status of the corresponding reference object at a given time step, and includes time series information of the corresponding reference object proceeding onwards from the given time step.
12 . The computer program product of claim 10 , wherein a number of reference profiles is equal to a number of time steps in the set of dynamic characteristics for each reference object in the plurality of reference objects.

Description

BACKGROUND The present invention generally relates to monitoring and tracking an object's health, and more specifically, to a process for determining a regimen to optimize a health parameter of the object. Tracking the health of an object typically involves monitoring one or more specific health parameters over time. Objects that are tracked in this way can include individual adult people with a health parameter focused on a fitness or sports goal, computer servers or systems within a cloud environment with a health parameter focused on a lifecycle goal, manufacturing systems with a health parameter directed toward minimization of downtime, or any similar systems. Consequently, attempts to improve the overall health of the object often focus on activities that are specifically targeted to improving the monitored parameters instead of toward providing a general health benefit. SUMMARY Embodiments of the present invention are directed to a computer-implemented method for monitoring the health of an object. A non-limiting example of the computer-implemented method includes receiving a set of static characteristics, a set of dynamic characteristics, and a health objective of a target object. A set of reference object data is retrieved and includes multiple reference objects of the same type as the target object. The reference object data includes the same data as the target object data. The dynamic data of the target object and the reference objects is parameterized over time. Each a set of dynamic characteristics for each reference object is split into a plurality of reference profiles. A set of similar reference profiles having a higher similarity metric than a remainder of the reference profiles is identified. A goal difference between the health objective of the target object and an end state of the similar reference profiles is determined. The reference profile having the smallest goal difference is identified and a corresponding health regimen is implemented. Embodiments of the present invention likewise include a system and a computer program product for implementing the same. Additional technical features and benefits are realized through the techniques of the present invention. Embodiments and aspects of the invention are described in detail herein and are considered a part of the claimed subject matter. For a better understanding, refer to the detailed description and to the drawings. BRIEF DESCRIPTION OF THE DRAWINGS The specifics of the exclusive rights described herein are particularly pointed out and distinctly claimed in the claims at the conclusion of the specification. The foregoing and other features and advantages of the embodiments of the invention are apparent from the following detailed description taken in conjunction with the accompanying drawings in which: FIG. 1 depicts one exemplary cloud computing system configured to implement the system and method according to one embodiment; FIG. 2 depicts a flow chart of a method for identifying a best course of action to achieve a healthy object according to one embodiment; FIG. 3 depicts an example data set for target and reference objects in the process of FIG. 1; FIG. 4. depicts a single object's data split into multiple profiles in the process of FIG. 1; FIG. 5 depicts a similarity score for a reference profile; and FIG. 6 depicts a difference between a target goal and a most similar reference profile. The diagrams depicted herein are illustrative. There can be many variations to the diagram or the operations described therein without departing from the spirit of the invention. For instance, the actions can be performed in a differing order or actions can be added, deleted or modified. Also, the term “coupled” and variations thereof describes having a communications path between two elements and does not imply a direct connection between the elements with no intervening elements/connections between them. All of these variations are considered a part of the specification. In the accompanying figures and following detailed description of the disclosed embodiments, the various elements illustrated in the figures are provided with two- or three-digit reference numbers. With minor exceptions, the leftmost digit(s) of each reference number correspond to the figure in which its element is first illustrated. DETAILED DESCRIPTION In one embodiment, a computer-implemented method includes receiving a set of static characteristics of a target object and a set of dynamic characteristics of the target object, and a health objective of the target object. The set of dynamic characteristics vary over time. A set of reference object data including a plurality of reference objects of the same object type as the target object is retrieved. The reference object data has a set of static characteristics for each reference object in the plurality of reference objects and a set of dynamic characteristics for each reference object in the plurality of reference object