Researchers Submit Patent Application, “Modeling Of Complex Systems Using A Distributed Simulation Engine”, for Approval (USPTO 20240202834): Patent Application

2024 JUL 05 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- From Washington, D.C., NewsRx journalists report that a patent application by the inventors Crabtree, Jason (Vienna, VA, US); Sellers, Andrew (Monument, CO, US), filed on March 2, 2024, was made available online on June 20, 2024.

No assignee for this patent application has been made.

News editors obtained the following quote from the background information supplied by the inventors: “

“Field of the Invention

“The present invention is in the field of use of computer systems in complex systems operations and planning. Specifically, the present invention is related to the problem of accurately and detailed predictive simulation of large, complex systems.

“Discussion of the State of the Art

“Over the past decade, the amount of operational, infrastructure, risk management and information available to decision makers from such sources as ubiquitous sensors found on a equipment or available from third party sources, detailed cause and effect data, and process monitoring software has expanded to the point where the data has overwhelmed leaders’ abilities to follow all of it and certainly to interpret and make meaningful use of available data in a given environment. In other words, the torrent of information now available to a decision maker or group of decision makers has far outgrown the ability of those most in need of its use to either fully follow it or to reliably use it. Failure to recognize important trends or become aware of information in a timely fashion has led to highly visible customer facing outages at NETFLIX™, FACEBOOK™, and UPS™ over the past few years, just to list a few.

“There have been several developments in recently that have arisen with the purpose of streamlining or automating either data analysis or decision processes. PALANTIR™ offers software to isolate patterns in large volumes of data, DATABRICKS™ offers custom analytics services, ANAPLAN™ offers financial impact calculation services and there are other software sources that mitigate some aspect of data relevancy identification, analysis of that data and decision automation, but none of these solutions handle more than a single aspect of the whole task. This insinuates the technology being used in the decision process as one of the variables as data from one software package often must be significantly and manually transformed to be introduced into the software for the next analysis, if appropriate software exists. This step is both inefficient use of human resources and has potential to introduce error at a critical process point.

“There has also been great progress made in the area of accurate system modeling and simulation. As one will quickly surmise, reliable simulation of prospective systems or ventures, any novel system or venture, in fact, has very high potential to save great amounts of capital, both monetary and human, can often be run to some completion point much more rapidly than the real process and, if unforeseen inefficiencies or risks are uncovered, or even if the original plan fails outright, changes to specific parameters and assumptions can be rapidly and efficiently made until an optimized and successful solution is found or the plan is be shelved, all with a minimal use of resources and loss of capital. As computing processing power has increased, more traditional modeling and simulation methodology such as system dynamic, in which actors of the same type such as a car or a truck will have their descriptive data, such as traveling speed or weight highly aggregated in a simulated system and then each of those aggregated actor types will represent that type during simulation interactions; or discrete event simulation where the simulation is processed by dividing events within it usually thought of as continuous into a series of discrete subevents to show the effects of performing a lengthy process on an object or group of objects, such as a trip into a hospital emergency room being: “open transport vehicle door,” “place patient into vehicle,” “drive vehicle 0.4 miles to corner of patient’s home road,” . . . ; . . . ; “record patient health insurance information in health computing system,” . . . ; . . . ; “examine by doctor,” . . . ; . . . ; “collect payment not covered by insurance,” “patient departs hospital,” where even the steps described may be further subdivided; have found highly useful resurgence. As one might realize, the significant ascent of computing power has given rise to a simulation engine that promises much more accurate modeling by allowing each object, also possibly denoted agent or actor, in a simulation to have its own individual model. This capability allows one to experiment with effects that these differences between individuals of like type have on the progression and outcome of a simulation, but may also require significantly more processing time and data bandwidth to complete and so are often used in smaller scale simulation than a system dynamics simulation.

“Currently there are multiple Open Source simulation engines available. DEUS, a discrete event simulator engine; OM-NET++, another discrete event simulator engine; Siafu, an agent-based simulation engine; and a discrete event-based simulator engine with high scalability. All of these offerings suffer, to varying extents from limited scalability and deployability. As outlined above for other data capture and transformation offerings, these simulation engines act as another, detached service within a complete analysis and prediction system into which data from other services must be often first re-formatted and then manually entered. Further many accurate, yet executable simulations rely on multiple simulation methods, system dynamics, discrete event and agent based, depending on the portion of the simulation being represented and the minimal accuracy needed to produce a reliable outcome.

“What is needed is a fully integrated system that retrieves relevant information from many diverse sources, identifies and analyzes that high volume data, transforming it to a useful format and then uses that data to drive an integrated highly scalable simulation engine which may employ combinations of the system dynamics, discrete event and agent based paradigms within a simulation run such that the most useful and accurate data is obtained and stored for the needs of the analyst.”

As a supplement to the background information on this patent application, NewsRx correspondents also obtained the inventors’ summary information for this patent application: “Accordingly, the inventor has developed a distributed system for accurate and detailed modeling of systems with large and complex datasets using a distributed simulation engine. The system further uses results of information analytics to optimize the making of decisions and allow for alternate action pathways to be simulated using the latest data and machine mediated prediction algorithms. Specifically, portions of the system are applied to the areas reliably predicting the outcomes of differential decision paths and prediction of risk to value for each set of decision choices through simulation of the progression of each decision pathway using the most current sensor data, specific programmed decision defining parameters and environment data available and then presenting that data in a format most useful to the authors of the simulation.

“According to a preferred embodiment of the invention, a system for accurate and detailed modeling of systems with large and complex datasets using a distributed simulation engine comprising: a data retrieval engine stored in a memory of and operating on a processor of a computing device; a data analysis engine stored in a memory of and operating on a processor of a computing device; and an automated planning and value at risk estimation module stored in a memory of and operating on a processor of one of more computing devices. An action outcome simulation module stored in the memory of and operating on a processor of one or more computing devices, wherein, the information retrieval engine: retrieves a plurality of relevant data from a plurality of sources; accepts a plurality of analysis parameters and control commands directly from human interface devices or from one or more command and control storage devices, and stores accumulated retrieved information for processing by data analysis engine or predetermined data timeout. The information analysis engine retrieves a plurality of data types from the information retrieval engine, and performs a plurality of analytical functions and transformations on retrieved data based upon the specific goals and needs set forth in a current campaign by process analysis authors. The automated planning and value at risk estimation module: employs results of data analyses and transformations performed by the information analysis engine, together with available supplemental data from a plurality of sources as well as any current campaign specific machine learning, commands and parameters from process analysis authors to formulate current planning and risk status reports and employs results of data analyses and transformations performed by the information analysis engine, together with available supplemental data from a plurality of sources, any current campaign specific commands and parameters from process analysis authors, as well as input gleaned from machine learned algorithms to deliver decision pathway simulations and value at risk support to a first end user. The action outcome simulation module: retrieves at least a portion of the results of data analyses and transformations performed by the information analysis engine, retrieves at least one piece of raw data from the information retrieval engine, employs a plurality of parameters entered from the automated planning and value at risk estimation module, uses information obtained to execute predictive simulations of venture or decision progress pathway and outcome as originally initialized by simulation author using a simulation method that combines system dynamics method, discrete event method, or agent based method for at least one simulation instance, employs groupings of action profile data and configuration parameters to create computer based models of real-world items to act in the simulation.

“According to another embodiment of the invention, a system for fully integrated collection of relevant data, analysis of that data and generation of both analysis-driven decisions and analysis-driven simulations of alternate candidate decision comprising: a data retrieval engine stored in a memory of and operating on a processor of a computing device, 2. The system of claim 1, wherein the information retrieval engine stored in the memory of and operating on a processor of a computing device, employs a portal for human interface device input at least a portion of which are relevant data and at least another portion of which are commands and parameters related to the conduct of a current venture campaign alternatives. Wherein the automated planning and value at risk estimation module uses at least information theory based statistical analysis to reliably predict future outcome of current decision based analyzed previous data. The automated planning and value at risk estimation module uses at least Monte Carlo heuristic model value at risk principles to reliably estimate future value at risk figures of current decision based analyzed previous data. The automated planning and value at risk estimation module uses a specifically designed graph-based data store service to efficiently store and manipulate the large data structures created during decision outcome analysis. The automated planning and value at risk estimation module has job control function that allows both jobs that run in a single iteration with a single set of parameters and jobs that include multiple iterations and sets of predetermined sets of parameters with termination criteria to stop execution when desired analysis results are obtained. Some jobs are run offline in a batch like mode and other jobs are run online in an interactive mode where users enter parameters for subsequent iterations based upon results of previous iterations until a predesigned analysis result terminates execution. At least one simulation includes models for hazards, vulnerabilities, contractual obligations and financial capital loss. The automated planning and value at risk estimation module acts upon at least one computer-based model to modify it prior to the simulation.

“According to another embodiment of the invention, a system for fully integrated collection of relevant data, analysis of that data and generation of both analysis-driven decisions and analysis-driven simulations of alternate candidate decision comprising: An information retrieval engine stored in the memory of and operating on a processor of a computing device, employs a portal for human interface device input at least a portion of which are relevant data and at least another portion of which are commands and parameters related to the conduct of a current venture campaign alternatives. An automated planning and value at risk estimation module uses a least information theory-based statistical analysis to reliably predict future outcome of current decision based analyzed previous data. An automated planning and value at risk estimation module uses at least Monte Carlo heuristic model value at risk principles to reliably estimate future value at risk figures of current decision based analyzed previous data. An automated planning and value at risk estimation module uses a specifically designed graph-based data store service to efficiently store and manipulate the large data structures created during decision outcome analysis. An automated planning and value at risk estimation module has job control function that allows both jobs that run in a single iteration with a single set of parameters and jobs that include multiple iterations and sets of predetermined sets of parameters with termination criteria to stop execution when desired analysis results are obtained. Some jobs are run offline in a batch like mode and other jobs are run online in an interactive mode where users enter parameters for subsequent iterations based upon results of previous iterations until a predesigned analysis result terminates execution

“According to a preferred embodiment of the invention, a method for fully integrated collection of relevant data, analysis of that data and generation of both analysis-driven decisions and analysis-driven simulations of alternate candidate decision comprising the steps of: a) receiving decision parameters and objectives using a client access interface stored in a memory of and operating on a processor of a computing device; b) retrieving a plurality of data from a plurality of sources using a data retrieval engine stored in a memory of and operating on a processor of a computing device; c) creating simulation models of real-world objects from available data using an action outcome simulation module stored in a memory of and operating on a processor of one of more computing devices; d) predicting the outcome of predetermined decision or venture candidates and estimating the value at risk attached to each candidate by simulation of the outplay of the decision or venture using the action outcome simulation module.”

The claims supplied by the inventors are:

“1. A system for modeling of complex systems with large and complex datasets using a distributed simulation engine comprising: a plurality of computing devices each comprising at least a processor, a memory, and a network interface; wherein a plurality of programming instructions stored in one or more of the memories and operating on one or more of the processors of the plurality of computing devices causes the plurality of computing devices to: automatically retrieve a plurality of data from a plurality of sources over a network; create a world model for a simulation using parameters and the retrieved data, the world model configured to represent an aspect of the real world and comprising: a plurality of constraints establishing boundary conditions matching those of the real world under expected conditions of the simulation; a plurality of actor models, each configured to act independently of all other actors within the boundary conditions; and an environment monitoring layer to monitor the actions of each actor model, collect and report results of actor model actions and interactions, and pass control directives to one or more of the actor models; operate the world model until a simulation result is achieved; and perform a plurality of statistical data analyses on the data based on the simulation parameters and the simulation result to obtain an analysis result; and deliver decision pathway simulations and value at risk calculations to a user based on the analysis result.

“2. The system of claim 1, wherein the statistical data analyses comprise information theory based statistical analysis.

“3. The system of claim 1, wherein the statistical data analyses comprise Monte Carlo heuristic model value at risk.

“4. The system of claim 1, wherein the system uses a graph-based data store service to efficiently store and manipulate large data structures created during statistical data analyses.

“5. The system of claim 1, wherein the system allows both jobs that run in a single iteration with a single set of parameters and jobs that include multiple iterations and sets of predetermined sets of parameters with termination criteria to stop execution when desired analysis results are obtained.

“6. The system of claim 5, wherein some jobs are run offline in a batch mode and other jobs are run online in an interactive mode.

“7. The system of claim 1, wherein at least one simulation includes models for hazards, vulnerabilities, contractual obligations and financial capital loss.

“8. The system of claim 1, wherein the analysis result from the automated planning and value at risk estimation is used as feedback to a subsequently run simulation.

“9. A method for accurate and detailed modeling of systems with large and complex datasets using a distributed simulation engine comprising the steps of: automatically retrieving a plurality of data from a plurality of sources over a network; creating a world model for a simulation using parameters and the retrieved data, the world model configured to represent an aspect of the real world and comprising: a plurality of constraints establishing boundary conditions matching those of the real world under expected conditions of the simulation; a plurality of actor models, each configured to act independently of all other actors within the boundary conditions; and an environment monitoring layer to monitor the actions of each actor model, collect and report results of actor model actions and interactions, and pass control directives to one or more of the actor models; operating the world model until a simulation result is achieved; and performing a plurality of statistical data analyses on the data based on the simulation parameters and the simulation result to obtain an analysis result; and delivering decision pathway simulations and value at risk calculations to a user based on the analysis result.

“10. The method of claim 9, wherein the statistical data analyses comprise information theory based statistical analysis.

“11. The method of claim 9, wherein the statistical data analyses comprise Monte Carlo heuristic model value at risk principles.

“12. The method of claim 9, further comprising the step of using a graph-based data store service to efficiently store and manipulate large data structures created during statistical data analyses.

“13. The method of claim 9, further comprising the step of allowing both jobs that run in a single iteration with a single set of parameters and jobs that include multiple iterations and sets of predetermined sets of parameters with termination criteria to stop execution when desired analysis results are obtained.

“14. The method of claim 13, wherein some jobs are run offline in a batch mode and other jobs are run online in an interactive mode.

“15. The method of claim 9, wherein at least one simulation includes models for hazards, vulnerabilities, contractual obligations and financial capital loss.

“16. The method of claim 9, further comprising the step of feeding back the analysis result as input to a subsequently run simulation.

“17. A computer-readable, non-transitory medium comprising a plurality of programming instructions that, when operating on a plurality of computing devices each comprising at least a processor, a memory, and a network interface, cause the plurality of computing devices to carry out the method of claim 9.”

For additional information on this patent application, see: Crabtree, Jason; Sellers, Andrew. Modeling Of Complex Systems Using A Distributed Simulation Engine. U.S. Patent Application Number 20240202834, filed March 2, 2024 and posted June 20, 2024. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(20240202834)&db=US-PGPUB&type=ids

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