Machine learning method for descriptive, predictive and normative facility operations

Abstract

The digital twinning of the facility defines a relationship between different components of the facility and a recording system of the facility. Information from different monitoring systems of the facility is related to events by digital twinning of the facility. Historical operation information of the facility is used to train a machine learning model. The trained machine learning model facilitates operation at the facility by providing descriptive information, predictive information, and / or normative information about the operation at the facility.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims U.S. Provisional Application No. 63 / 338,563, filed on May 5, 2022, entitled “Machine Learning Methods for Descriptive, Predictive, and Prescriptive Facility Operations,” the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure generally relates to the field of using machine learning methods to facilitate facility operations. Background Art

[0004] Different monitoring systems may be used to monitor and troubleshoot operations at a facility. Data collected by the different monitoring systems may be stored in different databases, and using such data to facilitate facility operations may be difficult and time consuming. Summary of the invention

[0005] The present disclosure relates to facilitating facility operations. Historical operational information and / or other information of a facility may be obtained. Historical operational information of a facility may be obtained based on a digital twin of the facility and / or other information. The digital twin of the facility may define relationships between components of the facility. The historical operational information and / or other information of the facility may be used to train a machine learning model. The trained machine learning model may facilitate one or more operations at the facility by outputting descriptive information, predictive information, prescriptive information, and / or other information about the operations at the facility. The trained machine learning model may be stored in a storage medium.

[0006] Facility context information and / or other information may be obtained. The facility context information may define a context for a given operation at the facility. The facility context information may be input into a trained machine learning model. The trained machine learning model may output descriptive information, predictive information, prescriptive information, and / or other information about a given operation at the facility.

[0007] A system for facilitating facility operations may include one or more electronic storage devices, one or more processors, and / or other components. The electronic storage device may store information related to the facility, historical operating information, information related to historical operations at the facility, information related to a digital twin of the facility, information related to components of the facility, information related to relationships between components of the facility, information related to a machine learning model, information related to training of a machine learning model, information related to use of a machine learning model, and / or other information.

[0008] The processor may be configured by machine readable instructions. Execution of the machine readable instructions may cause the processor to facilitate facility operations. The machine readable instructions may include one or more computer program components. The computer program components may include one or more of the following: a history operation component, a training component, a storage component, a scenario component, a facility operation component, and / or other computer program components.

[0009] The historical operation component may be configured to obtain historical operation information and / or other information of the facility. The historical operation information of the facility may be obtained based on the digital twin of the facility and / or other information. The digital twin of the facility may define the relationship between the components of the facility.

[0010] In some embodiments, the digital twin may output historical operating information of the facility based on relationships between components of the facility and / or other information.

[0011] In some embodiments, the historical operation information of the facility may include process control information, alarm information, bypass information, safety information, operator action information, and / or other information. The process control information, alarm information, bypass information, safety information, operator action information, and / or other information related to the event may be associated through the digital twin for use in the machine learning model.

[0012] In some embodiments, process control information, alarm information, bypass information, safety information, operator action information, and / or other information related to the event may be associated based on piping and instrumentation diagrams, cause-and-effect diagrams, and / or other information.

[0013] In some embodiments, associating process control information, alarm information, bypass information, safety information, and operator action information related to an event based on a piping and instrumentation diagram may include: generating a graphical model of a facility based on a piping and instrumentation diagram and / or other information; and associating process control information, alarm information, bypass information, safety information, and operator action information related to an event being executed based on the graphical model of the facility.

[0014] In some embodiments, the graphical model of the facility may include nodes for physical components of the facility and control components of the facility. The graphical model of the facility may include different types of edges between nodes, which are used to represent physical connections and logical connections between corresponding components of the facility. The physical connections between the components of the facility may include process lines between the components of the facility. The logical connections between the components of the facility may include electrical connections and / or input / output connections between the components of the facility.

[0015] The training component may be configured to train one or more machine learning models. The machine learning models may be trained using historical operational information and / or other information of the facility. The trained machine learning models may facilitate one or more operations at the facility. The trained machine learning models may facilitate operations at the facility by outputting descriptive information, predictive information, prescriptive information, and / or other information about the operations at the facility.

[0016] In some implementations, a machine learning model may include one or more sequence models.

[0017] The storage component may be configured to store the trained machine learning model. The trained machine learning model may be stored in one or more storage media.

[0018] In some embodiments, a trained machine learning model may perform one or more classification tasks.

[0019] In some embodiments, the trained machine learning model can perform one or more regression tasks.

[0020] The scene component can be configured to obtain facility scene information and / or other information. The facility scene information can define the scene of one or more operations at the facility.

[0021] The facility operations component may be configured to input facility scenario information and / or other information into a trained machine learning model. The trained machine learning model may output descriptive information, predictive information, and / or prescriptive information about operations at the facility. The trained machine learning model may output descriptive information, predictive information, and / or prescriptive information about the operational scenario at the facility.

[0022] In some implementations, one or more automated operations at a facility may be performed based on normative information and / or other information regarding operations at the facility.

[0023] In some embodiments, the facility operations component may be configured to provide visualizations of descriptive information, predictive information, and / or prescriptive information about operations at the facility. In some embodiments, the facility operations component may be configured to provide visualizations of descriptive information, predictive information, and / or prescriptive information about operating scenarios at the facility.

[0024] These and other objects, features and characteristics of the systems and / or methods disclosed herein, as well as the methods of operation and functions of the related elements and combinations of parts of the structures and the economies of manufacturing will become apparent upon consideration of the following description and the appended claims with reference to the accompanying drawings, all of which form a part of this specification, wherein like reference numerals represent corresponding parts in the various figures. However, it should be clearly understood that the drawings are for illustration and description purposes only and are not intended as definitions of limitations of the invention. As used in this specification and claims, the singular forms "a / an" and "the" include plural referents unless the context clearly dictates otherwise. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 An example system for facilitating facility operations is shown.

[0026] Figure 2A An example method for facilitating facility operations is shown.

[0027] Figure 2B An example method for facilitating facility operations is shown.

[0028] Figure 3 Example processes for facilitating facility operations are shown.

[0029] Figure 4 An example instrument protection layer and operator action overview is shown.

[0030] Figure 5 An example graph model of a facility is shown. DETAILED DESCRIPTION

[0031] The present disclosure relates to facilitating facility operations. A digital twin of a facility defines relationships between different components of the facility and the facility's systems of record. Information from different monitoring systems of the facility is correlated with events through the digital twin of the facility. Historical operational information of the facility is used to train a machine learning model. The trained machine learning model facilitates operations at the facility by providing descriptive information, predictive information, and / or prescriptive information about operations at the facility.

[0032] The methods and systems of the present disclosure may be implemented by a system such as Figure 1The system 10 shown in the figure is implemented and / or implemented in the system. The system 10 may include one or more of the following: a processor 11, an interface 12 (e.g., a bus, a wireless interface), an electronic storage device 13, a display 14, and / or other components. The historical operation information and / or other information of the facility can be obtained by the processor 11. The historical operation information of the facility can be obtained based on the digital twin of the facility and / or other information. The digital twin of the facility can define the relationship between the components of the facility and the recording system of the facility. The historical operation information and / or other information of the facility can be used by the processor 11 to train the machine learning model. The trained machine learning model can facilitate one or more operations at the facility by outputting descriptive information, predictive information, normative information, and / or other information about the operation at the facility. The trained machine learning model can be stored in a storage medium by the processor 11.

[0033] Facility context information and / or other information may be obtained by the processor 11. The facility context information may define a context for a given operation at the facility. The facility context information may be input into a trained machine learning model by the processor 11. The trained machine learning model may output descriptive information, predictive information, prescriptive information, and / or other information about a given operation at the facility.

[0034] The electronic storage device 13 may be configured to include an electronic storage medium that electronically stores information. The electronic storage device 13 may store software algorithms, information determined by the processor 11, information received remotely, and / or other information that enables the system 10 to operate properly. For example, the electronic storage device 13 may store information related to the facility, historical operation information, information related to historical operations at the facility, information related to the digital twin of the facility, information related to components of the facility, information related to the relationship between components of the facility, information related to the record system of the facility, information related to the machine learning model, information related to the training of the machine learning model, information related to the use of the machine learning model, and / or other information.

[0035] Display 14 may refer to an electronic device that provides a visual presentation of information. Display 14 may include a color display and / or a non-color display. Display 14 may be configured to present information in a visual manner. Display 14 may present information using / within one or more graphical user interfaces. For example, display 14 may present information related to the facility, historical operating information, information related to historical operations at the facility, information related to the digital twin of the facility, information related to components of the facility, information related to the relationship between components of the facility, information related to the facility's recording system, information related to the machine learning model, information related to the training of the machine learning model, information related to the use of the machine learning model, and / or other information.

[0036] A facility may refer to a location where one or more specific activities occur. A facility may include equipment that performs one or more activities. A facility may include equipment that performs one or more functions. For example, a facility may include an oil platform, an oil rig, an offshore platform, a refinery, or an oil and / or gas production platform to extract and / or process resources (e.g., hydrocarbons) located in underground rock formations. Other types of facilities are contemplated.

[0037] A process disturbance may refer to a disturbance to the operation at a facility (e.g., an interruption, failure, deviation, damage). A facility may include automatic and / or manual tools to resolve process disturbances at the facility. For example, a facility may include a process control loop to automatically respond to and / or mitigate instabilities in operation caused by process disturbances. If the process disturbance is not properly resolved, a process alarm may prompt the operator of the facility to take action. If the operator's manual actions cannot resolve the process disturbance, other alarms may be triggered, and automatic safeguards may be activated to partially and / or completely shut down the facility (e.g., to prevent catastrophic failures). Such shutdown events may be expensive and disruptive.

[0038] A facility may include multiple monitoring systems to monitor processes, equipment, operator actions, conditions, and / or other aspects of operations at the facility. Different levels of monitoring may be used to monitor and troubleshoot different aspects of operations at the facility. For example, a facility may include separate monitoring for process control, alarms, bypass actions, and instrument protection systems. The information collected by these separate monitoring systems may be maintained separately and used for different purposes.

[0039] The present disclosure provides a tool based on machine learning, which provides descriptive, predictive and / or prescriptive information about facility operations. Tools based on machine learning can achieve more effective prioritization and decision-making by connecting information from separate monitoring systems (isolated systems), thereby improving the reliability of facilities and reducing shutdowns. The digital twin of the facility can be used to contextualize and associate information from separate monitoring systems. The contextualization and association of information enable machine learning models to be used for process automation and operator responses to process disturbances. The historical operating information of the facility can be used to train the machine learning model. The machine learning model can digitize the operator's knowledge / experience from the historical operating information. The machine learning model can be used to describe what is happening at the facility, predict what will happen at the facility (e.g., predict facility response) and / or standardize the actions that the operator should take. The machine learning model can be used to automate the actions at the facility.

[0040] Figure 3An example process 300 for facilitating operations at a facility is shown. In the process 300, historical operational information 302 of the facility may be used to train a machine learning model 312. The historical operational information 302 may include historical operational information from process control 322, alarms 324, instrument protection systems 326, bypasses 328, and / or operator actions 330. Different portions of the historical operational information 302 may be monitored, tracked, and / or stored separately. Different portions of the historical operational information 302 may not be correlated.

[0041] The digital twin 340 of the facility can be used to contextualize and correlate different portions of the historical operational information 302. Information from process controls 322, alarms 324, instrument protection systems 326, bypasses 328, and / or operator actions 330 can be contextualized and correlated through the digital twin 340 of the facility. The digital twin 340 can identify, extract, and package event-related information from controls 322, alarms 324, instrument protection systems 326, bypasses 328, and / or operator actions 330 for use in training the machine learning model 312. The machine learning model 312 can be trained using the historical operational information 302 provided by the digital twin to perform classification tasks and / or regression tasks.

[0042] The digital twin 340 may identify, extract, and package event-related information from controls 322, alarms 324, instrument protection systems 326, bypasses 328, and / or operator actions 330 based on relationships between facility components and / or systems of record for the facility. The digital twin 340 may include piping and instrumentation diagram (P&ID) information digitized in the form of an equipment and instrumentation ontology diagram. The P&ID may include a diagram showing pipelines and process equipment as well as instruments (measurement instruments for indicating, measuring, and recording physical quantities) and control devices. The P&ID may include a diagram showing the interconnection of process equipment and instruments used to control the process. The equipment ontology diagram may define which components are related to a specific component and / or which components are related to a specific event. The equipment ontology may provide information about the strength of connections, interactions, dependencies, and / or hierarchies between components of the facility. The equipment ontology diagram may be used to identify, extract, and package event-related information from controls 322, alarms 324, instrument protection systems 326, bypasses 328, and / or operator actions 330 for training the machine learning model 312.

[0043] The digital twin 340 can facilitate the integration of information from separate monitoring systems to improve the efficiency of facility monitoring. The process 300 can facilitate operations at the facility, such as by reducing the risk of facility shutdown, improving equipment life, and / or accelerating maintenance prioritization. The process 300 can digitize the operator's knowledge / experience by contextualizing operator actions through the use of a digital twin, associating control loops, equipment, alarms, and bypass systems to reflect real-world relationships, enabling identification of the root cause of problems at the facility and / or otherwise facilitating operations at the facility.

[0044] Facility scenario information 314 may be input into machine learning model 312. Facility scenario information 314 may define a scenario for one or more operations at a facility. Machine learning model 312 outputs descriptive information, predictive information, and / or prescriptive information about the operations at the facility. That is, machine learning model 312 outputs descriptive information, predictive information, and / or prescriptive information about the operation scenarios defined by facility scenario information 314. Descriptive information about the operation may include information describing the operation (e.g., the machine learning model 312 identifies the operation and / or event that occurred at the facility). Predictive information about the operation may include information predicting what will happen at the facility (e.g., the machine learning model 312 predicts the result of the operation at the facility - predicts the event that will occur after / due to the operation). Prescriptive information about the operation may include information recommending / requiring what steps / actions should be taken at the facility (e.g., the machine learning model 312 recommends how the operation at the facility should be changed, and the machine learning model 312 is used to automate the change of the operation at the facility). Other uses of machine learning model 312 for facilitating facility operations can be envisioned.

[0045] Return to reference Figure 1 , the processor 11 may be configured to provide information processing capabilities in the system 10. Therefore, the processor 11 may include one or more of the following: a digital processor, an analog processor, a digital circuit designed to process information, a central processing unit, a graphics processing unit, a microcontroller, an analog circuit designed to process information, a state machine, and / or other mechanisms for electronically processing information. The processor 11 may be configured to execute one or more machine-readable instructions 100 to facilitate facility operations. The machine-readable instructions 100 may include one or more computer program components. The machine-readable instructions 100 may include a historical operation component 102, a training component 104, a storage component 106, a scenario component 108, a facility operation component 110, and / or other computer program components.

[0046] The historical operation component 102 may be configured to obtain historical operation information and / or other information of the facility. The historical operation information of the facility may include information about operations at the facility. The historical operation information of the facility may include information about operations that have occurred at the facility. The historical operation information of the facility may include time series data related to the operations at the facility. The historical operation information of the facility may include a collection of measured, sensed, detected and / or recorded characteristics of operations at the facility at different times. For example, the historical operation information of the facility may include timestamp information about event occurrences, sensor readings, operator actions, limits and / or bypass and safety actions.

[0047] The historical operation information of a facility may characterize the operations that have occurred at the facility. The historical operation information may include information about the characteristics of the operations at the facility. The operational characteristics of a facility may refer to the characteristics of the facility during operation (e.g., characteristics in and / or around the facility, characteristics of the equipment at the facility). The operational characteristics of a facility may refer to the properties, qualities, configurations, parameters, and / or other characteristics of materials / equipment within, in, and / or around the facility during operation.

[0048] The historical operational information for a facility may characterize the operations at the facility by including information that defines, describes, depicts, identifies, relates, quantifies, reflects, elaborates, and / or otherwise characterizes values ​​of properties, qualities, configurations, parameters, and / or other characteristics of materials / equipment within, in, and / or around the facility during operations. The historical operational information for a facility may characterize the operations at the facility by including information that may be used to determine values ​​of properties, qualities, configurations, parameters, and / or other characteristics of materials / equipment within, in, and / or around the facility during operations. Other types of historical operational information are contemplated.

[0049] Obtaining historical operation information may include one or more of the following: accessing, acquiring, analyzing, determining, inspecting, generating, identifying, loading, locating, measuring, opening, receiving, retrieving, reviewing, selecting, storing and / or otherwise obtaining historical operation information. The historical operation component 102 may obtain historical operation information from one or more locations. For example, the historical operation component 102 may obtain historical operation information from a storage location, such as an electronic storage device 13, an electronic storage device of a device accessible via a network, and / or other location. The historical operation component 102 may obtain historical operation information from one or more hardware components (e.g., a computing device, a sensor) and / or one or more software components (e.g., software running on a computing device). The historical operation component 102 may obtain historical operation information from multiple databases / storage locations. For example, different types of historical operation information may be stored in different databases / storage locations, and portions of the historical operation information associated with a particular event may be obtained from different databases / storage locations.

[0050] The historical operation information of the facility can be obtained based on the digital twin and / or other information of the facility. The digital twin of the facility may refer to a virtual representation or digital model of the facility. The digital twin may serve as a real-time digital counterpart of the facility and / or the operation / process occurring at the facility. The digital twin of the facility may define the relationship between the components of the facility. The components of the facility may refer to the equipment in the facility, the materials used in the facility, and / or other components of the facility. The relationship between the components of the facility may include connections, interactions, dependencies, hierarchies, and / or other relationships between the components. The relationship between the components of the facility can be used to obtain the historical operation information of the facility. The digital twin of the facility may define one or more record systems of the facility. The record system of the facility may refer to an information storage and retrieval system, which is an authoritative source of data related to the facility. The record system of the facility may refer to the collection and / or connection of relevant and / or contextualized information of the facility, such as design documents, equipment databases, time series data, inspection records, maintenance records, turnaround information, change management, and / or other information of the facility. Information may be stored in different systems / databases, and the information stored in different systems / databases may be connected to each other through digital twins. Information stored in different systems / databases can be accessed through the digital twin.

[0051] In some embodiments, the historical operation information of the facility may include process control information, alarm information, bypass information, safety information, operator action information, and / or other information. Process control information may refer to information from a process control system. Process control information may refer to information that defines and / or characterizes a process (e.g., an operation, a portion of an operation) at a facility and / or process control at a facility. Alarm information may refer to information from an alarm management system. Alarm information may refer to information that defines and / or characterizes an alarm that has been triggered at a facility and / or the operation of an alarm at a facility. Bypass information may refer to information from a bypass management system. Bypass information may refer to information that defines and / or characterizes a bypass at a facility (e.g., the location of the bypass, the equipment affected by the bypass, the process under the bypass). Safety information may refer to information from an instrument protection system (IPS). Safety information may refer to information that defines and / or characterizes a safety condition, the triggering / activation of a safety condition, and / or the operation of a SIS device. Operator action information may refer to information about operator actions at a facility. Operator action information may refer to information that defines and / or characterizes actions taken by one or more operators at a facility.

[0052] Process control information, alarm information, bypass information, safety information, operator action information, and / or other information associated with an event can be associated through a digital twin for training one or more machine learning models. The association of different information may include establishing / determining the relationship between different information. An event may refer to the occurrence of one or more things. An event may refer to one or more changes at a facility. For example, an event may include a change in process conditions, such as a change in pressure, temperature, and / or flow, the start / shutdown of a piece of equipment, the triggering of an alarm, an operator action, a change in operation, and / or other changes at a facility. Information associated with an event can be used to train a machine learning model. A digital twin can be used to determine what information is associated with an event and what information is not associated with an event.

[0053] For example, a digital twin can be used to identify, extract, and / or package portions of the historical operational information of an event based on relationships between components of the facility and / or the facility's systems of record. The digital twin can include an equipment ontology graph that defines which components are related to a particular component and / or which components are related to a particular event. The equipment ontology graph can be used to determine what information is related to the event, and the relevant information about the event can be used as the historical operational information of the facility to train a machine learning model.

[0054] In some embodiments, process control information, alarm information, bypass information, safety information, operator action information, and / or other information associated with an event may be associated based on a piping and instrumentation diagram, a cause-and-effect diagram, and / or other information. Piping and instrumentation diagrams and cause-and-effect diagrams may be used to construct relationships between different types of information. For example, the interconnections of process equipment and instruments used to control a process in a piping and instrumentation diagram may be used to associate different information with an event. A cause-and-effect diagram may define specific sequential relationships between components, such as how equipment is affected / changed based on the occurrence of a particular event (e.g., what components may be activated to shut down equipment / facility in response to an alarm), and sequential relationships between components may be used to associate different information with an event.

[0055] Figure 4An example instrument protection layer and operator action overview 400 is shown. As shown in overview 400, the relationship between information from different monitoring systems can be established using piping and instrumentation diagrams (P&IDs) and cause-and-effect diagrams (C&Es). Between process disturbances (e.g., unwanted operating conditions such as pressure spikes, equipment failures) and process safety events, there may be different monitoring applications: (1) automated process control to help regulate and stabilize the process, (2) alarms to alert operators; (3) operator actions to manage process disturbances and restore the facility to normal operation, and (4) instrument protection system activation to shut down the facility if a safety condition is violated. The relationship between information from process control and alarms can be established using piping and instrumentation diagrams (e.g., correlating alarms to process control loop performance), while the relationship between information from alarms and instrument protection systems can be established using cause-and-effect diagrams. Other information (e.g., inspection records, design documents) can be used to establish the relationship between information from different monitoring systems.

[0056] In some embodiments, the digital twin may output historical operating information of the facility based on relationships between components of the facility, the facility's systems of record, and / or other information. The digital twin itself may determine what portion of the historical operating information is relevant to the event, and those relevant portions of the historical operating information may be output by the digital twin for use in training a machine learning model.

[0057] In some embodiments, the association of different portions of the historical operation information of the facility related to a specific event (e.g., process control information, alarm information, bypass information, safety information, operator action information, and / or other information) performed based on the piping and instrumentation diagram may include (1) generating a graphical model of the facility based on the piping and instrumentation diagram and / or other information, and (2) performing the association of different portions of the historical operation information of the facility related to the specific event based on the graphical model of the facility and / or other models. That is, in order to determine whether different portions of the historical operation information are related to a specific event (e.g., in order to train a machine learning model), a graphical model of the facility may be generated, and the graphical model may be used to determine different portions of the historical operation information related to the specific event.

[0058] A graph model may refer to a model of components of a facility that uses edges between nodes to represent nodes and connections between components. A graph model of a facility may be generated based on a piping and instrumentation diagram of the facility and / or other information. For example, a piping and instrumentation diagram of a facility may be converted to a graph model of the facility. For example, a piping and instrumentation diagram of a facility may be scanned, and component boxes of the piping and instrumentation diagram may be converted to nodes, and lines between component boxes may be converted to edges. Other generations of graph models are contemplated.

[0059] The graphical model of a facility may include nodes for physical components of the facility, control components of the facility, and / or other components of the facility. The graphical model may include different types of nodes for different types of components of the facility. The physical components of a facility may refer to parts, equipment, and / or other components of a facility that operate and / or are operated to process, contain, move materials at the facility, and / or otherwise interact with them. The physical components of a facility may refer to parts, equipment, and / or other components of a facility that receive signals from control components of the facility and operate according to the received signals (e.g., commands transmitted by the received signals). For example, the physical components of a fluid facility may include pumps, actuators, motors, valves, doors, and / or other physical components that can control the flow of fluids.

[0060] A control component of a facility may refer to a part, device, and / or other component of a facility that operates and / or is operated to control the operation of the physical components of the facility. A control component of a facility may refer to a part, device, and / or other component of a facility that transmits signals to and controls the operation of the physical components of the facility. A control component of a facility may refer to a part, device, and / or other component of a facility that receives signals from sensors of the facility to monitor the operation of the physical components of the facility. For example, a control component of a fluid facility may include a logic box that performs sensor measurements on the facility and, in response to deviations of fluid flow in the facility from normal operating conditions, sends signals to pumps, actuators, and / or motors to control fluid flow in the facility.

[0061] A facility may include separate and independent control components to maintain safe operations at the facility. For example, a facility may include a distributed control system and an instrumentation protection system that operate independently of each other to recover from deviations in facility operations and shut down operations when necessary. Different types of control components may be represented by the same or different types of nodes within the graph model.

[0062] The graphical model of a facility may include different types of edges between nodes to represent different connections between corresponding components of the facility. The connections between components of the facility may include physical connections, logical connections, and / or other connections. Physical connections may be represented by one type of edge, while logical connections may be represented by another type of edge.

[0063] The physical connection between the components of a facility may refer to the connection that transmits physical materials between the different components of a facility. For example, the physical connection between the components of a fluid facility may include one or more process lines between the components of the fluid facility. The process line may include interconnected pipeline components, such as tubing, pipes, pressure hoses, valves, separators, water dividers, flanges, fittings, washers, filters and / or other components. The physical connection between the facility components may refer to the connection that physically links the facility components. For example, the physical connection between the components of a fluid facility may include the physical link between the actuator and the valve in the facility.

[0064] A physical connection between components of a facility may refer to a connection that logically links the components of the facility. A logical connection between components of a facility may refer to a connection that transmits information between different components of the facility. For example, a local connection between components of a fluid facility may include one or more electrical connections (e.g., transmitting sensor signals, transmitting command signals, transmitting information signals) and / or one or more input / output connections (e.g., facilitating communication) between components of the facility.

[0065] Figure 5 An example diagram model 500 of a facility is shown. The diagram model 500 may display a portion of a larger diagram model of a facility. The diagram model 500 may include nodes 502, 512, 514, 522, 524, 526 for physical components of the facility and nodes 532, 534, 536, 542 for control components of the facility. The solid edges between the nodes of the diagram model 500 may represent physical connections between corresponding components. The dashed edges between the nodes of the diagram model 500 may represent logical connections between corresponding components. For example, node 502 may represent a heat exchanger, node 514 may represent a valve of the heat exchanger, and node 526 may represent an actuator of the valve. Nodes 532, 534, 536 may represent a control system / box that can control the actuator (by sending a signal to the actuator) to operate the valve. The control systems / boxes represented by nodes 532, 534 may communicate with each other. The control system / block represented by node 536 may send and / or receive signals to and from the heat exchanger represented by node 502. The control systems / blocks represented by nodes 532, 534, 536 may be part of a distributed control system of a facility to keep the heat exchanger and other components of the facility in operation.

[0066] The facility may include an instrument protection system represented by node 542. The instrument protection system may operate independently of the distributed control system to prevent the facility from operating in hazardous conditions. For example, if the distributed control system is unable to restore the operating deviation to normal operating conditions (e.g., operation approaches or exceeds safety limits), the instrument protection system may shut down operations at the facility. The instrument protection system may provide an automated shutdown response to a safety violation at the facility.

[0067] In some embodiments, different types of physical connections and / or different types of logical connections may be represented by different types of edges. For example, a logical connection to an instrument protection system may be represented by one type of edge, while a logical connection to a distributed control system may be represented by another type of edge.

[0068] The historical operation information of a facility may include a large amount of data about the operations performed at the facility. For example, the historical operation information may include information about actions taken by multiple operators at the facility. When an event occurs at the facility, one or more operators may take actions to resolve the event, and one or more operators may take actions unrelated to the event. For example, when an alarm is triggered, one or more operators may take actions to resolve the alarm, while one or more other operators may take actions unrelated to the alarm. In order to use historical operation information for machine learning training, it may be necessary to associate information related to the event. That is, different parts of historical information that are related to each other may need to be associated for training a machine learning model.

[0069] The graph model can be used to correlate information related to an event (e.g., a change in an operating parameter, a triggered alarm). The historical operating information of a facility can be filtered using time to identify information that may be related to an event. For example, the historical operating information can be filtered to identify information about actions taken after an alarm was triggered and before the alarm was resolved (e.g., a drift / deviation in operation). Such filtering can identify historical operating information that may be temporally related to an event.

[0070] After time filtering, the graph model can be used to identify historical operation information related to the event. The event may be associated with a specific node, and historical operation information related to components of the facility that are a certain distance (e.g., a jump) from the node associated with the event can be identified as being related to the event. For example, referring to Figure 5 , a component (e.g., a control system / box, a sensor) represented by node 536 may trigger an alarm. Nodes within a threshold distance from node 536 may be identified, and historical operational information of actions taken at the corresponding components may be identified as being associated with the alarm. This identification of historical operational information may be used to generate a repository of actions taken for different events at the facility. Actions taken by one or more operators may be associated with specific events at the facility.

[0071] In some embodiments, different types of edges between nodes of a graph model may be treated in the same manner in distance (e.g., jump) calculations. For example, when determining the distance of a traversal, it may not be important whether the traversed edge represents a physical connection or a logical connection. In some embodiments, different types of edges between nodes of a graph model may be treated in different manners in distance calculations. For example, a traversal of an edge representing a physical connection may have a greater or lesser weight than a traversal of an edge representing a logical connection. As another example, a traversal of an edge representing a certain type of physical or logical connection may have a greater or lesser weight than a traversal of an edge representing another type of physical or logical connection. For example, a traversal of an edge representing a logical connection to an instrument protection system may have a greater or lesser weight than a traversal of an edge representing a logical connection to a distributed control system. In some embodiments, the weights of the edges may be customized. For example, an edge representing a particular connection may be assigned a different weight than other edges.

[0072] The results of actions taken at the facility can be analyzed and associated with specific actions taken at the facility. For example, for individual actions and / or combinations of actions taken for specific types of events, the results of the actions can be associated with the actions. The results between different actions / action combinations can be compared to determine which actions should be suggested / recommended to the operator and / or which actions should be performed automatically. For example, the time taken by the action to resolve the offset / deviation in the operation and the magnitude of the change in the operating parameters caused by the action can be used to rank the actions. For example, the speed at which the action resolves the offset / deviation in the operation and the impact of the action on the operation of the facility (e.g., the amount by which the action changes the temperature, flow, pressure, etc.) can be used to rank the actions based on the results. The association between the actions taken at the facility and the results of the actions can be used to train one or more machine learning models, and the machine learning models can be used to suggest / recommend or automate operator actions based on events that occur at the facility.

[0073] The training component 104 may be configured to train one or more machine learning models. The machine learning model may be trained using historical operational information and / or other information of the facility. The machine learning model may be trained to perform a classification task and / or a regression task. That is, the trained machine learning model may perform one or more classification tasks, or one or more regression tasks to generate an output. The trained machine learning model may facilitate one or more operations at the facility. The trained machine learning model may facilitate the operation at the facility by outputting descriptive information, predictive information, normative information, and / or other information about the operation at the facility. Descriptive information may include information about what is happening at the facility (e.g., identifying operations / events occurring at the facility). Predictive information may include information about what will happen at the facility (e.g., predicting events that will occur at the facility). Normative information may include information about what steps / actions should be taken at the facility (e.g., recommendations about how an operator should change operations, automatically change operations at the facility to prevent the facility from being shut down).

[0074] Training the machine learning model may include facilitating learning of the machine learning model by processing examples with the machine learning model. Pairs of historical operation information and information about expected output types may be provided to the machine learning model as examples of input and expected results, respectively. The historical operation information may be used as an input type to be received by the trained machine learning model, and the information about the expected output type paired with the historical operation information may be used as an output type to be generated by the machine learning model.

[0075] For example, to train a machine learning model to output descriptive information, historical operational information related to an operation / event may be paired with an identification / description of the operation / event to train the machine learning model. Historical operational information related to an operation / event may be paired with an identification / description of what is happening at the facility to train the machine learning model. Processing this pairing of information through a machine learning model may enable the machine learning model to learn patterns in the historical operational information associated with specific operations / events at the facility.

[0076] To train the machine learning model to output predictive information, historical operational information related to an operation / event may be paired with an identification / description of an event that occurred later at the facility. Historical operational information related to an operation / event may be paired with an identification / description of what will happen at the facility to train the machine learning model. Processing this pairing of information through the machine learning model may enable the machine learning model to learn patterns in the historical operational information that are associated with specific future events at the facility.

[0077] To train the machine learning model to output normative information, historical operational information associated with an operation / event may be paired with an identification / description of steps / actions to be taken at the facility (e.g., automatically performed by an operator). Historical operational information associated with an operation / event may be paired with an identification / description of steps / actions to be taken at the facility to train the machine learning model. Processing this pairing of information through the machine learning model may enable the machine learning model to learn patterns in the historical operational information associated with specific steps / actions to be taken at the facility.

[0078] In some embodiments, pairing historical operational information with specific steps / actions to be taken at the facility can digitize the operator's historical actions. For example, the operator's actions on different operations / events and the results of the actions can be recorded. Historical operational information can be paired with specific steps / actions based on the expected results. That is, the results of previous operator actions can be used to guide how to train the machine learning model. Historical operational information can be paired with the results of previous operator actions so that the machine learning model can provide possible results of the steps / actions output by the machine learning model. For example, in addition to recommending specific steps / actions taken by the operator of the facility, the machine learning model can also output the probability of the results of the operator taking the recommended steps / actions. The machine learning model can output multiple steps / actions that the operator can take, as well as the possible results of different steps / actions.

[0079] In some embodiments, a machine learning model may include one or more sequence models. A sequence model may refer to a machine learning model that receives a sequence of data as input and / or output. For example, a sequence model may refer to a machine learning model that receives time series data as input and / or output. For example, a machine learning model may include a Markov model. The use of other types of machine learning models is contemplated.

[0080] The storage component 106 may be configured to store the trained machine learning model. The trained machine learning model may be stored in one or more non-transitory storage media and / or other storage media. For example, the storage component 106 may store the trained machine learning model / information defining the trained machine learning model in a storage location, such as the electronic storage device 13, an electronic storage device of a device accessible via a network, and / or other location. The trained machine learning model may be stored for use in facilitating operations at a facility. The trained machine learning model may be stored for use in (1) identifying what is happening at the facility (e.g., outputting descriptive information identifying operations / events occurring at the facility), (2) predicting what will happen at the facility (e.g., outputting predictive information of events that will occur at the facility), and / or (3) recommending / guiding what actions should be taken at the facility (e.g., outputting normative information recommending specific actions to be taken by an operator; outputting normative information controlling how automated operations are controlled at the facility). The trained machine learning model may be stored for retrieval / execution in facilitating operations at the facility.

[0081] The scenario component 108 may be configured to obtain facility scenario information and / or other information. The facility scenario information may define a scenario for one or more operations at the facility. A scenario for operations at the facility may refer to a situation in which one or more operations occur at the facility. A scenario for operations may cover a certain moment or a period of time. A scenario for operations may include the occurrence of one or more events during an operation. The facility scenario information may define a scenario for operations at the facility by including information that defines one or more content, qualities, attributes, features, and / or other aspects of an operational scenario at the facility.

[0082] For example, facility scenario information may define the scenario of the operation at the facility by including information characterizing the operation performed at the facility. Facility scenario information may include information about the operational characteristics of the facility at a specific time (a certain moment, a period of time). Facility scenario information may include real-time facility scenario information. Real-time facility scenario information may refer to facility scenario information that defines the current operation at the facility. For example, real-time facility scenario information may characterize the operational characteristics of the facility currently reported by one or more sensors and / or the operational characteristics of the facility that have been measured within a threshold amount of time (e.g., operational characteristics measured within the past minute / hour / day). The facility scenario information of the facility may include the same type of information as the historical operational information of the facility. Other types of facility scenario information are conceivable.

[0083] Obtaining facility scene information may include one or more of the following: access, acquire, analyze, determine, inspect, generate, identify, load, locate, measure, open, receive, retrieve, review, select, store and / or otherwise obtain facility scene information. Scene component 108 may obtain facility scene information from one or more locations. For example, scene component 108 may obtain facility scene information from a storage location, such as electronic storage device 13, an electronic storage device of a device accessible via a network, and / or other locations. Scene component 108 may obtain facility scene information from one or more hardware components (e.g., computing device, sensor) and / or one or more software components (e.g., software running on a computing device). Scene component 108 may obtain facility scene information from multiple databases / storage locations. For example, different types of facility scene information may be stored in different databases / storage locations, and parts of facility scene information related to a specific event / scene may be obtained from different databases / storage locations. The digital twin of a facility may be used to obtain relevant parts of information from different monitoring systems as facility scene information.

[0084] The facility operation component 110 may be configured to input facility context information and / or other information into a trained machine learning model. The trained machine learning model may use the facility context information to generate an output. The trained machine learning model may output descriptive information, predictive information, and / or prescriptive information and / or other information about the operation at the facility (about the operation context).

[0085] For example, the trained machine learning model may output descriptive information about operations at the facility by outputting information describing operations, events, and / or other conditions at the facility. For example, based on the facility context information input into the trained machine learning model, the trained machine learning model may output that a column overflow is occurring / has occurred at the facility. The trained machine learning model may output details about the column overflow, such as the location and / or time of the column overflow, the facility components / equipment affected by the column overflow, and / or the source / root cause of the column overflow.

[0086] The trained machine learning model may output predictive information about the operation at the facility by outputting information predicting what will happen at the facility. For example, based on the facility scenario information input into the trained machine learning model, the trained machine learning model may output its prediction of the time of failure of a facility component / equipment due to a column overflow. The trained machine learning model may output details about the prediction, such as which component / equipment will fail, the predicted time of failure to occur, and / or the predicted extent of the failure.

[0087] The trained machine learning model may output normative information about operations at the facility by outputting information detailing what steps / actions should be taken at the facility. For example, based on facility scenario information (e.g., a pump trip) input into the trained machine learning model, the trained machine learning model may output optimal steps / actions that should be taken by one or more operators to prevent further disruptions at the facility and restore the facility to normal operating conditions. In some embodiments, one or more automated operations at the facility may be performed based on normative information and / or other information about operations at the facility. That is, rather than outputting normative information to operators to guide their actions to restore normal operations at the facility, operations at the facility may be automatically changed based on the normative information.

[0088] In some embodiments, the inputs and outputs of multiple trained machine learning models can be linked together. For example, facility context information can be input into a machine learning model that is trained to output descriptive information and / or predictive information. The descriptive information and / or predictive information output by the trained machine learning model can be used as facility context information, which is input into a machine learning model that is trained to output prescriptive information. A machine learning model that is trained to output prescriptive information can utilize models output by other machine learning models to prescribe what steps / actions should be taken at the facility.

[0089] In some embodiments, the facility operations component 110 may be configured to provide visualizations of descriptive information, predictive information, and / or normative information about operations at the facility (about an operations scenario). The visualization of the descriptive information, predictive information, and / or normative information may include a visual / graphical representation of the descriptive information, predictive information, and / or normative information. The visualization of the descriptive information, predictive information, and / or normative information may be provided on the display 14 (e.g., provided to one or more operators within one or more graphical user interfaces). For example, the descriptive information, predictive information, and / or normative information may be modeled in three-dimensional space along with three-dimensional modeling of the facility to visualize what is happening at the facility, what is predicted to happen at the facility, and / or what steps / actions should be performed at the facility. Other uses of visualizations are contemplated.

[0090] Embodiments of the present disclosure may be carried out in hardware, firmware, software, or any suitable combination thereof. Aspects of the present disclosure may be implemented as instructions stored on a machine-readable medium, which may be read and executed by one or more processors. A machine-readable medium may include any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computing device). For example, a tangible computer-readable storage medium may include a read-only memory, a random access memory, a disk storage medium, an optical storage medium, a flash memory device, etc., and a machine-readable transmission medium may include a form of a propagation signal, such as a carrier wave, an infrared signal, a digital signal, etc. With respect to specific exemplary aspects and embodiments of the present disclosure and performing certain actions, firmware, software, routines, or instructions may be described herein.

[0091] In some embodiments, some or all of the functionality attributed herein to system 10 may be provided by external resources not included in system 10. External resources may include hosts / sources of information, computations, and / or processing external to system 10 and / or other providers of information, computations, and / or processing.

[0092] Although processor 11, electronic storage device 13 and display 14 are Figure 1 10 is shown as being connected to interface 12, but any communication medium may be used to facilitate interaction between any components of system 10. One or more components of system 10 may communicate with each other via hardwired communication, wireless communication, or both. For example, one or more components of system 10 may communicate with each other via a network. For example, processor 11 may communicate with electronic storage device 13 wirelessly. As non-limiting examples, wireless communication may include one or more of the following: radio communication, Bluetooth communication, Wi-Fi communication, cellular communication, infrared communication, or other wireless communication. The present disclosure contemplates other types of communication.

[0093] Although processor 11, electronic storage device 13 and display 14 are Figure 1 10 is shown as a single entity, but this is for illustrative purposes only. One or more of the components of system 10 may be included within a single device or across multiple devices. For example, processor 11 may include multiple processing units. These processing units may be physically located within the same device, or processor 11 may represent processing functions of multiple devices operating in coordination. Processor 11 may be separate from one or more components of system 10 and / or be part of the one or more components. Processor 11 may be configured to execute one or more components through the following: software, hardware, firmware; some combination of software, hardware and / or firmware; and / or other mechanisms for configuring processing capabilities on processor 11.

[0094] It should be appreciated that although the computer program component Figure 1Although shown as being co-located within a single processing unit, one or more of the computer program components may be located remotely from the other computer program components. Although the computer program components are described as performing or being configured to perform operations, the computer program components may include instructions that may program the processor 11 and / or the system 10 to perform the operations.

[0095] Although the computer program components are described herein as being implemented via the processor 11 through the machine-readable instructions 100, this is for ease of reference only and is not meant to be limiting. In some embodiments, one or more functions of the computer program components described herein may be implemented via hardware ( For example , dedicated chips, field programmable gate arrays) rather than software. One or more functions of the computer program components described herein may be implemented by software, hardware, or both.

[0096] The description of functionality provided by the different computer program components described herein is for illustrative purposes and is not intended to be limiting, as any computer program component may provide more or less functionality than described. For example, one or more of the computer program components may be eliminated, and some or all of their functionality may be provided by other computer program components. As another example, processor 11 may be configured to execute one or more additional computer program components that may perform some or all of the functionality attributed to one or more of the computer program components described herein.

[0097] The electronic storage media of electronic storage device 13 may be provided integrally with one or more components of system 10 ( Right now , essentially non-removable) and / or provided as removable via, for example, a port ( For example , USB port, FireWire port wait ) or drive ( example like , disk drive wait ) is a removable storage device connected to one or more components of system 10. Electronic storage device 13 may include one or more of the following: an optically readable storage medium ( For example ,CD wait ), magnetically readable storage media ( example like , magnetic tapes, magnetic hard drives, floppy drives wait ), charge-based storage media ( For example , EPROM, EEPROM, RAM wait )、Solid State Storage Media( For example , Flash Drive wait) and / or other electronically readable storage media. Electronic storage device 13 may be a separate component within system 10, or electronic storage device 13 may be associated with one or more other components of system 10 ( For example , the processor 11) is provided integrally. Although the electronic storage device 13 is Figure 1 1 is shown as a single entity, but this is for illustrative purposes only. In some embodiments, electronic storage device 13 may include multiple storage units. These storage units may be physically located in the same device, or electronic storage device 13 may represent the storage functions of multiple devices operating in coordination.

[0098] Figure 2A and Figure 2B Methods 200, 250 for facilitating facility operations are shown. The operations of the methods 200, 250 presented below are intended to be illustrative. In some embodiments, the methods 200, 250 may be accomplished via one or more additional operations not described and / or not via one or more of the operations discussed. In some embodiments, two or more of the operations may occur substantially simultaneously.

[0099] In some embodiments, methods 200, 250 may be performed in one or more processing devices ( For example , digital processors, analog processors, digital circuits designed to process information, central processing units, graphics processing units, microcontrollers, analog circuits designed to process information, state machines, and / or other mechanisms for electronically processing information). The one or more processing devices may include one or more devices that perform some or all of the operations of method 200 in response to instructions stored electronically on one or more electronic storage media. The one or more processing devices may include one or more devices configured by hardware, firmware, and / or software specifically designed to perform one or more of the operations of method 200.

[0100] refer to Figure 2A In the method 200, at operation 202, historical operation information and / or other information of the facility may be obtained. The historical operation information of the facility may be obtained based on the digital twin of the facility and / or other information. The digital twin of the facility may define relationships between components of the facility and a system of record for the facility. In some embodiments, operation 202 may be performed by communicating with the historical operation component 102 (in Figure 1 The invention may be executed by the same or similar processor components as shown in and described herein.

[0101] At operation 204, the machine learning model may be trained using historical operational information and / or other information of the facility. The trained machine learning model may facilitate one or more operations at the facility by outputting descriptive information, predictive information, prescriptive information, and / or other information about the operations at the facility. In some embodiments, operation 204 may be performed by a training component 104 (in Figure 1 The invention may be executed by the same or similar processor components as shown in and described herein.

[0102] At operation 206, the trained machine learning model may be stored in a storage medium. In some embodiments, operation 206 may be performed by the storage component 106 (in Figure 1 The invention may be executed by the same or similar processor components as shown in and described herein.

[0103] refer to Figure 2B In the method 250, at operation 252, facility context information and / or other information may be obtained. The facility context information may define a context for a given operation at the facility. In some embodiments, operation 252 may be performed by a scene component 108 (in Figure 1 The invention may be executed by the same or similar processor components as shown in and described herein.

[0104] At operation 254, the facility context information may be input into a trained machine learning model. The trained machine learning model may output descriptive information, predictive information, prescriptive information, and / or other information about a given operation at the facility. In some embodiments, operation 254 may be performed by a facility operations component 110 (in Figure 1 The invention may be executed by the same or similar processor components as shown in and described herein.

[0105] Although the systems and / or methods of the present disclosure have been described in detail for illustrative purposes based on what are presently considered to be the most practical and preferred embodiments, it is to be understood that such details are provided for that purpose only and that the present disclosure is not limited to the disclosed embodiments, but rather is intended to cover modifications and equivalent arrangements within the spirit and scope of the appended claims. For example, it is to be understood that the present disclosure contemplates that, to the extent possible, one or more features of any embodiment may be combined with one or more features of any other embodiment.

Claims

1. A system for facilitating facility operations, the system comprising: One or more physical processors configured by machine-readable instructions to: obtaining historical operating information of the facility based on a digital twin of the facility, the digital twin of the facility defining relationships between components of the facility and a system of record for the facility; training a machine learning model using the historical operational information of the facility, wherein the trained machine learning model facilitates the one or more operations at the facility by outputting descriptive information, predictive information, and / or prescriptive information about the one or more operations at the facility; as well as The trained machine learning model is stored in a storage medium.

2. The system of claim 1, wherein the one or more physical processors are further configured by the machine-readable instructions to: obtaining facility context information, the facility context information defining a context for a given operation at the facility; and The facility scenario information is input into the trained machine learning model, wherein the trained machine learning model outputs the descriptive information, the predictive information and / or the prescriptive information about the given operation at the facility.

3. The system of claim 1, wherein the trained machine learning model performs a classification task.

4. The system of claim 1, wherein the trained machine learning model performs a regression task.

5. The system of claim 1, wherein the one or more physical processors are further configured via the machine-readable instructions to provide visualization of the descriptive information, the predictive information, and / or the prescriptive information regarding the one or more operations at the facility.

6. The system of claim 1, wherein one or more automated operations at the facility are performed based on the normative information regarding the one or more operations at the facility.

7. The system of claim 1, wherein the digital twin outputs the historical operating information of the facility based on the relationship between the components of the facility and the recording system of the facility.

8. The system of claim 1, wherein the machine learning model comprises a sequence model.

9. The system of claim 1, wherein: The historical operation information of the facility includes process control information, alarm information, bypass information, safety information, and operator action information; and The process control information, the alarm information, the bypass information, the safety information, and the operator action information related to the event are associated through the digital twin for use in the machine learning model.

10. The system of claim 9, wherein the process control information, the alarm information, the bypass information, the safety information, and the operator action information related to the event are associated based on a piping and instrumentation diagram and a cause-and-effect diagram.

11. The system of claim 10, wherein correlating the process control information, the alarm information, the bypass information, the safety information, and the operator action information related to the event based on the piping and instrumentation diagram comprises: generating a graphical model of the facility based on the piping and instrumentation diagram; as well as The associating of the process control information, the alarm information, the bypass information, the safety information, and the operator action information related to the event is performed based on the graphical model of the facility.

12. The system of claim 11, wherein the graph model of the facility includes nodes for physical components of the facility and control components of the facility.

13. The system of claim 12, wherein the graph model of the facility includes different types of edges between nodes for representing physical connections and logical connections between corresponding components of the facility.

14. The system of claim 13, wherein: The physical connections between components of the facility include process lines between components of the facility; and The logical connections between components of the facility include electrical connections and / or input / output connections between components of the facility.

15. A method for facilitating facility operations, the method comprising: obtaining historical operating information of the facility based on a digital twin of the facility, the digital twin of the facility defining relationships between components of the facility and a system of record for the facility; training a machine learning model using the historical operational information of the facility, wherein the trained machine learning model facilitates the one or more operations at the facility by outputting descriptive information, predictive information, and / or prescriptive information about the one or more operations at the facility; Storing the trained machine learning model in a storage medium; obtaining facility context information, the facility context information defining a context for a given operation at the facility; as well as The facility scenario information is input into the trained machine learning model, wherein the trained machine learning model outputs the descriptive information, the predictive information and / or the prescriptive information about the given operation at the facility.

Images