Information processing apparatus, information processing method and program

Abstract

To provide an information processing apparatus that efficiently acquires an operation rule for controlling an operation of a plant.SOLUTION: An information processing apparatus 100 comprises: an information search unit 105 that searches one or more operation rules of one or more second plant (water purification plant Pn) associated with a status detected for a first plant (water purification plant PA); an extraction unit 103 that extracts one or more second properties being different from first properties being one or more properties of one or more factors affecting an operation of a first plant, of second properties being one or more properties of one or more factors affecting operation of a second plant, in which an operation rule is searched, for each of one or more operation rules; a determination unit 106 that determines whether or not the second property occurs for each of the extracted one or more second properties; and a rule search unit 102 that searches one or more operation rules in which a second property determined to occur is extracted.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] An embodiment of the present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] For example, in plants such as waterworks plants (water purification plants, sewage plants, etc.), a technology has been proposed that uses operation rules to control the behavior of each monitored and controlled object according to the detected state. The operation rules are, for example, rules that are written in advance in a knowledge graph format with description patterns of IF-THEN rules.

[0003] In water supply plants, for example, the effects of global warming can cause raw water to deteriorate. For example, changes in the habitats of migratory birds can cause raw water to be contaminated with organic components from feces and urine that were not previously present in a water supply plant. As a result, predetermined operating rules can no longer be applied, and the plant can no longer be controlled appropriately. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7002978 [Patent Document 2] Japanese Patent Application Publication No. 2023-139389 Summary of the Invention [Problem to be solved by the invention]

[0005] As a countermeasure in the case where the predetermined operation rules cannot be applied, for example, it is possible to acquire and use operation rules applicable to the newly occurring situation from another plant.

[0006] An object of the present invention is to provide an information processing device, an information processing method, and a program that can efficiently acquire operation rules for controlling the operation of a plant. [Means for solving the problem]

[0007] An information processing device according to an embodiment includes a processing unit. The processing unit searches for one or more operation rules for one or more second plants that are associated with a situation detected for the first plant. For each of the one or more operation rules, the processing unit extracts one or more second properties that are different from the first properties, which are the properties of the one or more factors that affect the operation of the first plant, from among second properties, which are the properties of the one or more factors that affect the operation of the second plants for which the operation rules have been searched. The processing unit determines whether the second property has occurred for each of the extracted one or more second properties. The processing unit selects one or more operation rules from which the second property determined to have occurred has been extracted. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram of an information processing device. [Figure 2] FIG. 4 is a diagram showing an example of a data structure of property data. [Figure 3] 1 is a flowchart of information processing. [Figure 4] FIG. 10 is a diagram for explaining a specific example of information processing. [Figure 5] FIG. 10 is a diagram showing an example of searching for an operation rule. [Figure 6] FIG. 4 is a diagram showing an example of a display screen. [Figure 7] FIG. 10 is a diagram showing an example of a graphical display screen. [Figure 8] A diagram showing examples of basic knowledge about water supply and knowledge related to water sources. [Figure 9] FIG. 10 is a diagram showing an example of a new property candidate extraction process. [Figure 10] FIG. 10 is a diagram for explaining details of a process for searching for status information. [Figure 11] FIG. 10 is a diagram for explaining examples of attributes of a water purification plant. [Figure 12] FIG. 10 is a diagram for explaining details of the process of determining whether or not diversion is possible. [Figure 13] FIG. 10 is a diagram for explaining a specific example of a process for determining whether or not diversion is possible. [Figure 14] FIG. 10 is a diagram for explaining a specific example of a process for determining whether or not diversion is possible. [Figure 15] FIG. 1 is a hardware configuration diagram of an information processing device. DETAILED DESCRIPTION OF THE INVENTION

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of an information processing apparatus according to the present invention will be described in detail below with reference to the accompanying drawings.

[0010] The following method is generally used to identify the cause of the inapplicability of the operation rule being used. How to analyze plant monitoring and control logs to detect new trends How to gather relevant information describing situations where it is not applicable

[0011] The operational rules include, for example, operation rules and operation know-how. The operation rules are, for example, rules incorporated into the automatic control of the plant and rules that serve as the basis for operator decisions. The operation know-how is, for example, rules that serve as options for operator decisions.

[0012] With the above method, elements that are not being monitored (elements that are not monitored because they are not of interest and are not recorded in the monitoring and control log) are unlikely to be the subject of analysis and information collection, and may not be identified as possible causes, or it may take a long time to identify the cause. Furthermore, if the conditions for information collection are not properly described, elements that are not being monitored may not be the subject of information collection.

[0013] In the following, we will mainly explain an example where the plant to be monitored and controlled is a water purification plant. A water purification plant is an example of a plant where the input for controlling its operation changes due to changes in the environment, etc. Applicable plants are not limited to water supply plants such as water purification plants, but may be any other plants.

[0014] In the following, the water purification plant (first plant) that is the subject of operation (operation, monitoring) will be referred to as water purification plant PA, and one or more other water purification plants (second plants) different from water purification plant PA will be referred to as water purification plants Pn (n is an integer greater than or equal to 1).

[0015] The information processing device of the embodiment has, for example, the following functions (F1) to (F3). (F1) Information on the nature of factors that may affect the situation to be investigated (survey situation) that occurred at the water purification plant PA is obtained from knowledge data such as the operating rules of other water purification plants Pn (obtaining search candidates). (F2) By comparing the acquired property information with information indicating the occurrence status of the properties of factors that may affect the operation of the water purification plant PA, new property information of factors that may affect the operation of the water purification plant PA is extracted (narrowing down the search candidates). (F3) Obtain (search) behavior rules applicable to the situation under investigation from the behavior rules of other water purification plants Pn that include rules related to the extracted new properties.

[0016] The situation to be investigated corresponds to an unexpected situation (malfunction situation) that has occurred at the water purification plant PA. Factors that can affect the operation of the water purification plant PA include, for example, the geology, ecosystem, and climate of the water source (input source) of the water purification plant PA. Information on the nature of the factor is, for example, information indicating the amount of components removed by water treatment, the presence or absence of events related to the occurrence or increase or decrease of components, and the nature of the events. Information indicating the occurrence status of the nature of the factor is, for example, information indicating the occurrence status of events related to the occurrence or increase or decrease of components.

[0017] The above function makes it possible to search for and acquire behavior rules that can be applied to changes in the environment that have not been experienced or that have not been of interest.

[0018] The behavior rules are expressed in the form of a knowledge graph with description patterns of IF-THEN rules, for example. (F1) and (F3) can be realized by scanning the knowledge graph using the description patterns and by determining the similarity of the structure and meaning of the subgraph to be matched. The structure and meaning of the subgraph mean, for example, what meanings the nodes in the subgraph have, what meanings the links have, and in what direction they are connected.

[0019] (F2) can be realized by generating a fill-in-the-blank query (prompt) from the knowledge graph to instruct a large-scale language model (LLM), and then extracting information from the LLM using the generated query. This function utilizes the property that knowledge graphs can be converted to and from natural language sentences.

[0020] Fig. 1 is a block diagram showing an example of the configuration of an information processing device 100 according to an embodiment. As shown in Fig. 1, the information processing device 100 is connected to search target data 200 via a network 300. The information processing device 100 includes an acquisition unit 101, a rule search unit 102, an extraction unit 103, a generation unit 104, an information search unit 105, a determination unit 106, a selection unit 107, an output control unit 108, and a storage unit 120.

[0021] The search target data 200 corresponds to data that is the target of a search by the information search unit 105. The search target data 200 is, for example, public data, which is publicly available data including news. The search target data 200 may include data indicating whether or not a characteristic of a factor that may affect the operation of a water purification plant has occurred, as well as the date and time of the occurrence. Therefore, the search target data 200 can be used as a search target for information to determine whether or not a certain characteristic has occurred.

[0022] The network 300 may be of any type, such as the Internet, and may be a wired network, a wireless network, or a network in which both wired and wireless networks coexist.

[0023] The storage unit 120 stores various types of information used by the information processing device. For example, the storage unit 120 stores the following information for each water purification plant: Driving pattern data Equipment configuration data · Characteristic data Rule data Business record data

[0024] Operation pattern data is data that indicates a pattern of one or more operations (operations) at a water purification plant. For example, operation pattern data is data that indicates the equipment and location where an operation such as chlorine injection is performed. The substance to be injected is not limited to chlorine, and may be any other substance. Furthermore, the operation is not limited to the operation of injecting a substance. For example, at a water purification plant, operations such as injecting activated carbon, injecting a coagulant, injecting a neutralizing agent, and stirring water may be performed.

[0025] The equipment configuration data is data indicating one or more pieces of equipment that make up a water purification plant. Examples of equipment include a water conveyance conduit, a receiving well, an activated carbon contact basin, a mixing basin, a flocculation basin, a sedimentation basin, a sludge tank, a water conveyance conduit, a filtration basin, a backwash tank, a chemical mixing basin, a clear water reservoir, and a distribution reservoir. The equipment configuration data may be data indicating equipment that exists in the corresponding water purification plant from among these equipment configurations (standard equipment configurations).

[0026] The property data is data that indicates the properties of factors that may affect the operation of a water purification plant. Fig. 2 is a diagram showing an example of the data structure of the property data 121. As shown in Fig. 2, the property data 121 includes factors and the properties of the factors at each water purification plant. Note that, although Fig. 2 shows the property data of multiple water purification plants together, the property data may be stored separately for each water purification plant.

[0027] The rule data is data indicating operation rules established for each water purification plant. As described above, the rule data may be expressed in a knowledge graph format. Examples of rule data (operation rules) expressed in a knowledge graph format will be described later.

[0028] The business record data is data that records the past operations of each water purification plant. The business record data may be in any format, for example, electronic data representing a business log. The business record data may record whether or not a characteristic of a factor that may affect the operation of the water purification plant has occurred, as well as the date and time of the occurrence. Therefore, the business record data can be used as a search target for information to determine whether a certain characteristic has occurred.

[0029] The storage unit 120 can be configured with any commonly used storage medium, such as a flash memory, a memory card, a RAM (Random Access Memory), a HDD (Hard Disk Drive), an optical disk, etc. Some or all of the data stored in the storage unit 120 may be stored in physically different storage media, or may be stored in different storage areas of the same physically stored medium.

[0030] Furthermore, some or all of the data in the storage unit 120 may not be stored in the information processing device 100, but may be acquired from a storage unit in a control system that controls each water purification plant, for example.

[0031] Returning to the explanation of Fig. 1, the acquisition unit 101 acquires various types of information used in the information processing device 100. For example, the acquisition unit 101 acquires information indicating the investigation target situation detected for the water purification plant PA. The acquisition unit 101 may acquire information by any method, but for example, a method of receiving information from the control system of each water purification plant connected via the network 300 can be applied.

[0032] The rule search unit 102 searches for one or more operation rules of other water purification plants Pn that are related to the acquired situation under investigation. Any method may be used to search for operation rules related to the situation under investigation. When the operation rules are expressed in a knowledge graph format as described above, the rule search unit 102 searches for operation rules related to the situation under investigation by comparing the situation under investigation converted into the knowledge graph format with the operation rules expressed in the knowledge graph format.

[0033] For example, the rule search unit 102 generates a knowledge graph representing the situation to be investigated from the situation to be investigated. The rule search unit 102 searches the rule data stored in the storage unit 120 for an operation rule including a knowledge graph (subgraph) similar to the generated knowledge graph.

[0034] The extraction unit 103 uses the searched operation rules to extract one or more properties that are different from one or more properties C1 (first properties) of one or more factors that affect the operation of the water purification plant PA. For example, for each of the one or more operation rules, the extraction unit 103 extracts one or more properties C2 that are different from the property C1 from among the properties C2 that are one or more properties of one or more factors that affect the operation of the water purification plant Pn for which the operation rules were searched.

[0035] The generation unit 104 generates a query statement (search statement, search query) for searching for information related to the extracted property C2. Any method for generating a query statement may be used, but for example, the following method may be applied. A method for generating query prompts for other systems that perform searches, such as systems that use generative AI (Artificial Intelligence). -How to generate queries using fill-in-the-blank templates.

[0036] The information search unit 105 searches for information using the generated query sentence. For example, the information search unit 105 searches for situation information indicating the occurrence of property C2 by using at least one of the search target data 200 and the business record data of each water purification plant as the search target.

[0037] The determination unit 106 determines whether or not the property C2 has occurred for each of the one or more properties C2 extracted by the extraction unit 103. For example, the determination unit 106 determines that the property C2 has occurred when the information search unit 105 searches for situation information indicating that the property C2 has occurred.

[0038] The determination unit 106 may determine whether or not the property C2 has occurred at a location identified based on the location (location) of the water purification plant PA. The location identified based on the location may be, for example, the location itself or a location within a certain range including the location. The determination unit 106 may also identify a period during which the occurrence of the property C2 will affect the operation of the water purification plant PA, and determine whether or not the property C1 has occurred during the identified period. In this case, the information search unit 105 may generate a query statement that further includes conditions related to the location and the period. When situation information indicating that the property C2 has occurred at the location and the period specified as the conditions is found using such a query statement, the determination unit 106 determines that the property C2 has occurred at the location and the period.

[0039] The selection unit 107 selects one or more behavior rules from which the property C2 determined by the determination unit 106 to have occurred has been extracted.

[0040] The operation rule selected in this way may be diverted as an operation rule that can be applied to a newly occurring situation. The selection unit 107 may further select a more appropriate operation rule from the selected operation rules to be diverted to the water purification plant PA. In this case, the determination unit 106 and the selection unit 107 may further have the following functions.

[0041] For example, the determination unit 106 determines whether the attributes of the water purification plant PA are similar to the attributes of the water purification plant Pn for which the selected operation rule is defined. The attributes include one or more pieces of equipment included in the water purification plant, the order of operations of multiple pieces of equipment included in the water purification plant, and part or all of the operation pattern of the water purification plant.

[0042] The determining unit 106 may determine similarity by prioritizing attributes with higher priorities than other attributes according to the priority of each attribute. For example, the priority may be determined in the following order: (1) One or more facilities included in a water treatment plant (2) The sequence of operation of multiple pieces of equipment in a water purification plant (3) Driving pattern

[0043] The determination unit 106 may calculate a similarity indicating the degree of similarity between the attribute of the water purification plant PA and the attribute of the water purification plant Pn for which the selected operation rule is defined, and use the similarity to determine whether or not they are similar. The similarity is calculated, for example, based on the number of matching elements among multiple elements included in the attribute. For example, if the attribute is one or more pieces of equipment, the similarity may be calculated based on the number of matching pieces of equipment. The determination unit 106 may determine that the attribute of the water purification plant PA and the attribute of the water purification plant Pn for which the selected operation rule is defined are similar if the similarity is equal to or greater than a predetermined threshold.

[0044] The selection unit 107 selects the operation rules that are determined to be similar. The selection unit 107 may further select (extract) a part of the selected operation rules that can be diverted to the operation of the water purification plant PA, and output the selected part of the operation rules.

[0045] The output control unit 108 controls the output of various information used in the information processing device 100. For example, the output control unit 108 outputs (displays) a display screen for presenting the selected operation rule on a display device such as a display. The method of outputting information by the output control unit 108 is not limited to this, and any other method may be used. For example, a method of transmitting information to an external device connected via the network 300 may be used.

[0046] At least a part of each of the above units (acquisition unit 101, rule search unit 102, extraction unit 103, generation unit 104, information search unit 105, judgment unit 106, selection unit 107, and output control unit 108) may be realized by one or more processing units. Each of the above units is realized, for example, by one or more processors. For example, each of the above units may be realized by having a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) execute a program, that is, by software. Each of the above units may be realized by a processor such as a dedicated IC (Integrated Circuit), that is, by hardware. Each of the above units may be realized by a combination of software and hardware. When multiple processors are used, each processor may realize one of the units, or may realize two or more of the units.

[0047] The information processing device 100 may be physically configured as one device or may be physically configured as multiple devices. For example, the information processing device 100 may be built in a cloud environment. Furthermore, each unit within the information processing device 100 may be distributed across multiple devices.

[0048] Next, a description will be given of information processing by the information processing apparatus 100 according to the embodiment. Fig. 3 is a flowchart showing an example of information processing according to the embodiment.

[0049] The acquisition unit 101 acquires information on the survey target situation detected for the water purification plant PA (step S101). The rule search unit 102 generates a knowledge graph representing the survey target situation (step S102). The rule search unit 102 searches for operation rules of other water purification plants Pn that are similar to the knowledge graph (step S103).

[0050] The extraction unit 103 acquires the properties C2 of the factors related to the searched operation rules (step S104). The extraction unit 103 extracts the properties C2 that are not included in the properties C1 of the water purification plant PA being investigated from the acquired properties C2 (step S105).

[0051] The generation unit 104 generates a query statement for searching for the occurrence status of the extracted property C2 (step S106). The information search unit 105 searches for situation information indicating the occurrence status of the extracted property C2 using the generated query statement (step S107).

[0052] If status information indicating that the property C2 has occurred is found, the determination unit 106 determines that the property C2 has occurred. In this case, the following process is further executed. Although not shown in FIG. 3, if status information indicating that the property C2 has occurred is not found, the following process is not executed and the information processing ends.

[0053] The selection unit 107 selects an operation rule that matches the searched situation (step S108). For example, the selection unit 107 selects one or more operation rules from which the property C2 determined to have occurred has been extracted. The selection unit 107 further selects (extracts) an operation rule that can be applied to the operation of the water purification plant PA from the selected operation rules (step S109).

[0054] The output control unit 108 outputs the selected operation rule (step S110), and ends the information processing.

[0055] Next, a specific example of information processing according to this embodiment will be further described below. Fig. 4 is a diagram for explaining a specific example of information processing according to this embodiment.

[0056] As described above, the rule search unit 102 searches for operation rules of other water purification plants Pn related to the situation (malfunction situation) that has occurred in the water purification plant PA, which is the plant to be investigated (step S201). Hereinafter, the other water purification plants Pn whose operation rules have been searched for may be referred to as conversion source candidates. Also, the searched operation rules may be referred to as conversion candidate rules.

[0057] The conversion candidate rule 122 in FIG. 4 shows an example of a searched operation rule. The conversion candidate rule 122 is stored in, for example, the memory unit 120. The conversion candidate rule includes a rule number, a conversion source candidate, and a conversion candidate. The rule number is identification information that identifies the operation rule. The conversion source candidate is identification information that identifies the water purification plant where the operation rule was searched. The conversion candidate is information that indicates whether or not to adopt the operation rule to be converted. For example, a value (e.g., "x") is set for the conversion candidate if it is adopted, and no value is set if it is not adopted.

[0058] Figure 4 shows an example in which two operation rules identified by rule numbers "#1-1" and "#1-5" are retrieved from water purification plant P1, and one operation rule identified by rule number "#3-4" is retrieved from water purification plant P3.

[0059] The extraction unit 103 extracts the difference (new property candidate) between the property C2 of the conversion source candidate having the conversion candidate rule and the property C1 of the water purification plant PA (step S202). For example, assume that property data such as that shown in FIG. 2 is obtained for each water purification plant. In this case, the property C2 of the conversion source candidate water purification plant P1 is "high in manganese" and "summer birds arrive," and the property C2 of the conversion source candidate water purification plant P3 is "high in manganese." Combining these two, the property C2 of the conversion source candidate becomes "high in manganese" and "summer birds arrive." These properties C2 are not included in the property C1 of the water purification plant PA, "high in iron" and "winter snowfall." Therefore, the extraction unit 103 extracts "high in manganese" and "summer birds arrive" as the difference (new property candidate).

[0060] New property candidate 123 in Figure 4 shows an example of a new property candidate that is property C2 extracted as a difference. New property candidate 123 is stored, for example, in storage unit 120. Occurrence / non-occurrence is information indicating whether or not each property that is a new property candidate has occurred. A value for occurrence / non-occurrence is set depending on whether or not situation information has subsequently been searched for. For example, a value (e.g., "x") is set for occurrence / non-occurrence when situation information indicating that the property of the new property candidate has occurred is searched for, and no value is set when it is not searched for.

[0061] The information search unit 105 searches for status information (new substance occurrence information) regarding the occurrence of new property candidates in the water source of the water purification plant PA (step S203). Assume that status information regarding "summer bird arrival" is found among the new property candidates. In this case, for example, the information search unit 105 sets a value to the occurrence / non-occurrence field corresponding to "summer bird arrival." New property candidate 123b in FIG. 4 shows a state in which a value is set to the occurrence / non-occurrence field.

[0062] The selection unit 107 selects a diversion candidate rule of the diversion source candidate that includes the new property candidate whose occurrence has been confirmed as a property (step S204). In the example of Fig. 4, the diversion candidate rule of the water purification plant P1, which is the diversion source candidate, that includes the new property candidate whose occurrence has been confirmed as a property, "summer birds arriving", is selected.

[0063] Specifically, two operation rules identified by rule numbers "#1-1" and "#1-5" are selected. For example, the selection unit 107 sets values ​​to the diversion candidates corresponding to the selected operation rules. The diversion candidate rule 122b in FIG. 4 shows a state in which values ​​are set to the diversion candidates.

[0064] It may be possible to determine whether or not a rule relating to the generated property is included for each operation rule. In such a case, the selection unit 107 may exclude operation rules that do not include a rule relating to the generated property from the selection targets and select operation rules that include a rule relating to the generated property. For example, assume that only the operation rule identified by the rule number "#1-1" includes a rule relating to the generated property. In this case, the selection unit 107 selects the operation rule identified by the rule number "#1-1" and sets a value to the diversion candidate corresponding to the selected operation rule. The diversion candidate rule 122c in FIG. 4 shows a state in which values ​​are set to the diversion candidates in this way.

[0065] The following example will be described below. Although the PA operation at the water purification plant is being controlled as before, "the residual chlorine concentration in the treated water at the outlet of the sedimentation tank has dropped and is no longer meeting the control standard."

[0066] In such a case, an IF-THEN style rule such as "increase the amount of chlorine injected and the residual chlorine concentration of the treated water at the outlet of the sedimentation tank will increase to meet the control standard" is searched for in the rules of other water purification plants Pn. This allows the manager of water purification plant PA to know at what location (equipment) in the water purification plant and by how much the amount of chlorine injected should be increased.

[0067] 5 is a diagram showing an example of searching for an operation rule. First, the rule searching unit 102 converts data (e.g., text data) indicating the situation to be investigated into a knowledge graph. Graph 511 in FIG. 5 shows an example of a knowledge graph representing the situation to be investigated.

[0068] Any method may be used to convert data indicating the situation to be investigated into a knowledge graph, but for example, the following method may be applied. This method uses a generative AI LLM (large-scale language model) and a separation prompt. This makes it possible to eliminate variations in notation. A separation prompt is, for example, a prompt for separating an IF-THEN format knowledge graph into data representing the IF situation and data representing the THEN situation. A separation prompt may also be, for example, a prompt for separating the IF situation data into data representing the target of the situation, data representing the state of the target, and data representing the location of the target. The unit of separation is, for example, a node in the knowledge graph or a set of nodes. The result of the separation is, for example, data represented by triples in the knowledge graph. · A method for converting text data into a knowledge graph according to a grammar and schema that defines the structure of the knowledge graph. A method for generating a knowledge graph using an input means such as an editor that creates data for the knowledge graph. A method for converting an inputted situation to be investigated into a knowledge graph using an input means including a syntax form that can be converted into a knowledge graph.

[0069] Graph 500 in Fig. 5 shows an example of a knowledge graph representing behavior rules that can be searched for. As shown in Fig. 5, graph 500 includes node 501, which represents a state corresponding to an IF condition, and two nodes 502 and 503, which represent states corresponding to the THEN conditions for node 501. Node 502 corresponds to the case where the IF condition is satisfied (OK). Node 503 corresponds to the case where the IF condition is not satisfied (NG).

[0070] The rule search unit 102 can search for an action rule that matches graph 511 by comparing graph 500, which is a knowledge graph having a description pattern of an IF-THEN rule, with graph 511. In the example of Fig. 5, the part having node 503 as the parent node is the same as graph 511, so graph 500 is searched for as a graph that matches graph 511.

[0071] The situation to be investigated may be input by the input means as described above. Fig. 6 shows an example of a display screen that can be used for inputting the situation to be investigated and displaying the processing results based on the situation to be investigated.

[0072] 6, display screen 600 includes, as elements for inputting a question, an input field 601, a content confirmation button 602, a confirmation result field 603, a diagram display button 604, and a search button 605. Display screen 600 also includes, as elements for displaying processing results, an output field 611, a reference display button 612, output fields 621 and 622, a reuse edit button 623, a diversion availability output field 624, and a diagram display button 625.

[0073] The input field 601 is an input field for inputting text data indicating the situation to be investigated. When the content confirmation button 602 is pressed, the output control unit 108 outputs information (OK or NG) indicating the confirmation result as to whether the data entered in the input field 601 can be converted into a knowledge graph to the confirmation result field 603.

[0074] When the diagram display button 604 is pressed, the output control unit 108 outputs a diagram display screen for displaying a knowledge graph into which the survey target situation has been converted.

[0075] 7 is a diagram showing an example of a graphical display screen. The graphical display screen includes text 701 input as the situation to be investigated and a conversion result 702. The conversion result 702 corresponds to a knowledge graph obtained by converting the text 701.

[0076] The user can use the diagram display screen to check whether the converted knowledge graph is correct (whether it can be used for searching). Whether the knowledge graph is correct can be confirmed, for example, by checking whether it contains basic nodes (nodes that represent the target of the rule, the state of the target, the location of the target, etc.) that represent the meaning of the IF-THEN rule and edges (relationships between nodes) that also represent the meaning of the IF-THEN rule.

[0077] A function for comparing text reconverted from the conversion result 702 with the input text 701 may be provided. The user can check whether the knowledge graph is correct by checking whether the reconverted text and the text 701 have the same meaning. The function for comparing texts may be realized by using the LLM (Large Scale Language Model) of the generation AI.

[0078] A facility may be provided for editing the knowledge graph, where, for example, the edited knowledge graph is used to search for relevant action rules.

[0079] Returning to the explanation of Fig. 6, when the search button 605 is pressed, the rule search unit 102 uses the converted knowledge graph to search for behavior rules related to the knowledge graph as diversion candidate rules. After that, new property candidates and diversion source candidates are obtained by the processing of steps S202 to S204 in Fig. 4, for example. These processing results are displayed on the display screen 600.

[0080] The output field 611 is a field for displaying the searched new property candidates. When the reference display button 612 is pressed, the output control unit 108 may output a screen showing details of the new property candidates.

[0081] Output columns 621 and 622 are columns for displaying the name (station name) of the water purification plant that is a candidate for conversion source and an overview of the water purification plant.

[0082] When the diversion edit button 623 is pressed, the output control unit 108 may display an edit screen for editing the data of the diversion candidate rule to generate a new operation rule.

[0083] The conversion feasibility output column 624 is a column for outputting the determination result of whether the corresponding conversion candidate rule can be converted. For example, in the above step S109 (described in detail later), the conversion candidate rule of the conversion source candidate water purification plant Pn, which has attributes similar to those of the water purification plant PA, is determined to be a convertible operation rule. The attributes include, for example, operation patterns and equipment configuration. "Operation" and "Equipment" in the conversion feasibility output column 624 indicate the results of determining whether conversion is possible based on the similarity (commonalities) of the operation patterns and equipment configuration, respectively. For example, a check mark is displayed in the conversion feasibility output column 624, indicating that the operation rules are similar and therefore it is determined that conversion is possible.

[0084] When the diagram display button 625 is pressed, the output control unit 108 may further display a diagram display screen for displaying the result of converting the diversion candidate rule into a knowledge graph and an interpretation of the knowledge graph.

[0085] If there is no check mark in the conversion feasibility output field 624, the diagram display screen may display a note, such as the following, indicating caution regarding how to handle the operating rules due to the fact that the attributes of the water purification plant PA being investigated and the water purification plant Pn that is a candidate for conversion are not similar. "Because these rules are for plants that are not highly similar to the plant under investigation, if you are going to use them for other purposes, we recommend that you only refer to the control operation direction (i.e. increase, decrease, stop, etc.)."

[0086] The annotation to be displayed may be changed according to the check display pattern of the diversion permission output column 624 as follows. Neither "Operation" nor "Facility" is checked Only "Driving" is checked Only "Facilities" is checked

[0087] In the process of extracting new property candidates, knowledge data separate from the operation rules may be referenced to extract properties that could be the cause of a malfunction (such as a decrease in residual chlorine). The knowledge data may include, for example, basic knowledge about waterworks and knowledge related to water sources. The basic knowledge about waterworks represents basic knowledge previously acquired about waterworks. The knowledge related to water sources represents knowledge previously acquired about water sources.

[0088] 8 is a diagram showing examples of basic water supply knowledge and water source-related knowledge. As shown in Fig. 8, basic water supply knowledge 125 is information that associates basic knowledge numbers with basic knowledge. The basic knowledge numbers are identification information that identify basic knowledge.

[0089] The water source related knowledge 126 includes a substitution knowledge number, a raw water component to be managed, a substitution target, an impact direction, an impact onset time, and an impact duration. The substitution knowledge number is identification information for identifying the water source related knowledge.

[0090] The managed raw water component represents the component that is managed at the water source. The substitute object represents an object that can substitute for the properties of factors that can affect the operation of the water treatment plant. The impact direction represents the direction in which the impact of the substitute object affects the managed raw water component. The impact direction indicates, for example, whether the component increases (increase) or decreases (decrease). The impact onset time represents the time from when the properties corresponding to the substitute object occur until the impact on the managed raw water component occurs. The impact duration represents the time from when the impact occurs until the impact ends.

[0091] FIG. 9 is a diagram showing an example of a process for extracting new property candidates using knowledge data (basic knowledge of waterworks, knowledge related to water sources).

[0092] The extraction unit 103 inputs the status to be investigated acquired by the acquisition unit 101 (step S301). As in the above, an example will be described in which the status to be investigated indicating that "the residual chlorine concentration of the treated water at the outlet of the settling tank has decreased and the control standard is not met" is input.

[0093] The extraction unit 103 extracts a new property candidate, which is the difference between the property C2 of the diversion source candidate and the property C1 of the water purification plant PA, from the property data stored in the storage unit 120 (step S302). This process corresponds to step S202 in FIG. 4.

[0094] The extraction unit 103 searches the basic water supply knowledge 125 for knowledge that has the same situation as the input situation to be investigated (step S303). For example, the following method can be applied for the search.

[0095] First, the extraction unit 103 converts the basic knowledge of waterworks into a knowledge graph in an IF-THEN format, and separates the knowledge graph into data representing the IF situation and data representing the THEN situation. This process can be performed using a method that uses a large-scale language model (LLM) of the generation AI and a prompt for separation. Note that the basic knowledge of waterworks may be stored in advance, separated into data representing the IF situation and data representing the THEN situation. In the basic knowledge of waterworks 125 in FIG. 8, for example, data 801 corresponds to the data representing the IF situation, and data 802 corresponds to the data representing the THEN situation.

[0096] The extraction unit 103 identifies basic water supply knowledge represented in an IF-THEN format knowledge graph that includes the input target situation in data representing the THEN situation. The extraction unit 103 acquires data representing the IF situation of the identified basic water supply knowledge.

[0097] The extraction unit 103 searches for water source-related knowledge that has the same situation as the situation of the IF in the acquired basic water supply knowledge (step S304). For example, the extraction unit 103 acquires water source-related knowledge that includes data on the situation of the IF acquired in step S303. In the water source-related knowledge 126 in FIG. 8, for example, data 811 corresponds to the water source-related knowledge that includes data on the situation of the IF. Note that the extraction unit 103 may further convert the data on the situation of the IF into a knowledge graph that includes data on item strings of subject, predicate, and object by syntactic analysis or the like, and compare it with each element of the water source-related knowledge.

[0098] The extraction unit 103 adopts, from among the new property candidates extracted in step S302, a new property candidate that is identical to a substitute object included in the acquired water source-related knowledge (step S305).

[0099] The extraction process of Figure 9 makes it possible to extract properties that exist in the water purification plant Pn from which the conversion candidate rule was extracted but do not exist in the water purification plant PA, and that may be the cause of the input malfunction situation.

[0100] 4, the situation information may be searched for using the location and period in which the property C2 (new property candidate) occurred as conditions. The period can be specified by referring to knowledge data (water source-related knowledge), for example.

[0101] FIG. 10 is a diagram for explaining the details of the process of searching for situation information (step S203).

[0102] The generation unit 104 generates a query statement for checking whether a new property candidate has occurred (step S401). First, the generation unit 104 prepares the following data: Location of the water source of the water purification plant PA Name of new property candidate - Confirmation period for the occurrence of new property candidates

[0103] The location of the water source for each water purification plant can be identified, for example, from the information stored for each water purification plant in the memory unit 120. The location is an example of a location identified based on the location of the water purification plant. The name of the new property candidate can be obtained, for example, by the extraction process of Figure 9. The confirmation period for the occurrence of the new property candidate can be determined from the impact onset time and impact duration included in the water source-related knowledge. For example, when checking from the present, the shortest time in the retroactive period is set as the impact onset time, and the longest time in the retroactive period is set as the impact onset time + impact duration, with the period between the shortest and longest times being the confirmation period. The confirmation period is an example of a period identified as a period during which the occurrence of the new property candidate (property C2) will affect the operation of the water purification plant.

[0104] The generation unit 104 uses the prepared data to generate a query statement that includes the location as a location condition and the confirmation period as a period condition. In the case of a method that generates a query prompt for the generation AI as a query statement, for example, the following query statement is generated: Please list the sources from which you found the following information: Location: Lake Okutama and surrounding areas Event: Summer birds arrive Period: May 2023 to November 2023

[0105] The information search unit 105 uses the generated query sentence to search for situation information indicating whether or not a new property candidate has occurred (step S402).

[0106] The determination unit 106 determines whether a new property candidate has occurred using the search result of the situation information (step S403). As described above, for example, when situation information indicating that a new property candidate (property C2) has occurred is found, the determination unit 106 determines that property C2 has occurred.

[0107] In a case where inaccurate information may be retrieved, the determination unit 106 may determine whether or not characteristic C2 has occurred by using the reliability (score) of the retrieved situation information. For example, the determination unit 106 may determine that characteristic C2 has occurred when the number of retrieved situation information pieces is equal to or greater than a preset threshold.

[0108] Next, the details of the process of extracting operation rules that can be applied to the operation of the water purification plant PA (step S109) will be described. In this process, whether or not the rules can be applied is determined based on whether or not the attributes, including the equipment included in the water purification plant, the order of the equipment's operations, and the operation patterns, are similar.

[0109] FIG. 11 is a diagram illustrating examples of attributes of a water purification plant. The upper part of FIG. 11 shows the attributes of water purification plant PA. The upper part of FIG. 11 shows the attributes of water purification plant P1, which is a candidate for conversion. Data 1101 shows an example of a standard equipment configuration. The standard equipment configuration represents the standard equipment configuration of a water purification plant. Data 1102 shows an example of an equipment configuration adopted at water purification plant PA. As such, each water purification plant may adopt some of the standard equipment configurations. The upper part of data 1102 represents the configuration, and the lower part represents the operating method of the corresponding configuration (normal, slow, rapid, etc.). Data 1103 shows an example of an operation pattern adopted at water purification plant PA. FIG. 11 shows an example of an operation pattern that indicates the equipment (location) where operation operations such as chlorine injection are performed. Attributes (equipment configuration, operation pattern) are also specified for water purification plant P1 in a similar format.

[0110] The determination unit 106 determines whether the operation rules are applicable depending on whether the attributes of the equipment configurations, operation patterns, etc. are similar to each other. Similarity includes, for example, the following situations. Equipment configuration Example 1: The configuration and method of the original water purification plant Pn and the water purification plant PA under investigation are the same Example 2: The configuration of the water treatment plant PA under investigation includes the configuration of the water treatment plant Pn from which the water is to be converted. About driving patterns Example 1: The operation patterns of the original water purification plant Pn and the water purification plant PA under investigation are the same Example 2: The operation pattern of the water purification plant PA under investigation includes the operation pattern of the water purification plant Pn from which the water purification plant is converted.

[0111] FIG. 12 is a diagram for explaining details of the process for determining whether or not diversion is permitted.

[0112] The determination unit 106 acquires the equipment configuration and operation pattern for each of the survey target plant (water purification plant PA) and the conversion source candidate water purification plant Pn from, for example, the operation pattern data and equipment configuration data stored in the storage unit 120 (step S501).

[0113] The determination unit 106 acquires an action from the conversion candidate rule (step S502). As shown in FIG. 5, the action rule of the knowledge graph includes "Action" as an element. "Action" corresponds to the action of the water purification plant. The determination unit 106 acquires "Action" included in the conversion candidate rule as the action.

[0114] The selection unit 107 selects a conversion candidate rule in which the acquired action is included in the operation pattern of the water purification plant PA (step S503).

[0115] The determination unit 106 determines commonalities (similarities) in the equipment configuration between the water purification plant PA and the diversion candidate (step S504). The determination unit 106 also determines commonalities in the operation patterns between the water purification plant PA and the diversion candidate (step S505). The determination unit 106 may, for example, assign a diversion feasibility flag to set the diversion feasibility output field 624 on the display screen 600 in FIG. 6 .

[0116] The selection unit 107 selects a diversion candidate rule corresponding to the diversion source candidate rule that is determined to be similar (step S506).

[0117] Next, a specific example of the process of determining whether or not diversion is possible will be described. Figures 13 and 14 are diagrams for explaining a specific example of the process of determining whether or not diversion is possible.

[0118] Fig. 13 shows an example of the following behavior rule 1300. Note that an arrow 1301 in Fig. 13 indicates an action corresponding to the Action A2 below. IF: Odor-producing algae dominated by viable bacteria occur more frequently in the raw water than in steady state. Action: Reduce chlorine injection in receiving wells (A1) Increase chlorine injection before the filter (A2) THEN: The residual chlorine concentration of the treated water at the outlet of the sedimentation tank falls below the control standard The residual chlorine concentration of purified water in the purified water reservoir is within the water quality standards. No moldy smell occurs in the purified water from the purified water reservoir

[0119] Fig. 14 shows an example of the following behavior rule 1400. Note that an arrow 1401 in Fig. 14 indicates an action corresponding to the Action A3 below. IF: Odor-producing algae dominated by viable bacteria occur more frequently in the raw water than in steady state. Action: Reduce chlorine injection in receiving wells (A1) Increase chlorine injection in the chemical mixing pond (A3) THEN: The residual chlorine concentration of the treated water at the outlet of the sedimentation tank falls below the control standard The residual chlorine concentration of purified water in the purified water reservoir is within the water quality standards. No moldy smell occurs in the purified water from the purified water reservoir

[0120] Between the behavior rule 1300 and the behavior rule 1400, the actions A2 and A3 are different.

[0121] Assume that the behavior rules 1300 and 1400 are found as diversion candidate rules. In this case, the determination unit 106 acquires the above A1 and A2 as Actions from the behavior rule 1300, and acquires the above A1 and A3 as Actions from the behavior rule 1400 (step S502 in FIG. 12).

[0122] As shown in Figure 11, the operation pattern of the water purification plant PA includes the operation "Chlorine (before)" corresponding to the injection of chlorine into the receiving well (A1) and the operation "Chlorine (after)" corresponding to the injection of chlorine into the chemical mixing basin (A3). On the other hand, the operation pattern of the water purification plant PA does not include the operation corresponding to the injection of chlorine before the filtration basin (A2).

[0123] Therefore, the selection unit 107 selects the operation rule 1400 including A1 and A3 as a conversion candidate rule including the operation included in the operation pattern of the water purification plant PA (step S503 in FIG. 12).

[0124] In steps S504 and S505 of Fig. 12, commonalities in equipment configuration and operation patterns are determined between the water purification plant PA and the diversion candidate. In the example of the attributes of the water purification plant PA and the attributes of the diversion candidate (water purification plant P1) shown in Fig. 11, the determination unit 106 determines commonalities (similarity) between the two, for example, as follows: Equipment configuration Example 1: The configuration and method of water treatment plant P1 and the water treatment plant PA under investigation are not the same. Example 2: The configuration of the investigated water treatment plant PA does not encompass the configuration of water treatment plant P1. About driving patterns Example 1: The operation patterns of water purification plant P1 and the water purification plant PA under investigation are not the same. Example 2: The operation pattern of the water purification plant PA under investigation does not include the operation pattern of the water purification plant P1.

[0125] In this example, the determining unit 106 determines that the attributes of the water purification plant PA and the diversion source candidate (water purification plant P1) are not similar. In this case, the selecting unit 107 does not select the diversion candidate rule corresponding to the diversion source candidate (water purification plant P1).

[0126] In this way, in this embodiment, even if an event is something that a certain plant (water purification plant PA) has never experienced and is not of interest to it, it is possible to acquire knowledge (operation rules) acquired in other plants (water purification plants Pn) that have experienced a similar event. The operation rules are acquired based on information related to the cause of the event (situation information related to the water purification plant PA). Therefore, measures on how to respond to events that have not been addressed until now can be acquired without relying on, for example, personal knowledge. As a result, more appropriate control of the water purification plant PA becomes possible. It is also possible to obtain measures according to the degree of similarity of the equipment configuration with the original plant (water purification plant Pn).

[0127] As described above, according to the embodiment, it is possible to efficiently acquire operation rules for controlling the operation of a plant (such as a water purification plant).

[0128] Next, the hardware configuration of the information processing apparatus according to the embodiment will be described with reference to Fig. 15. Fig. 15 is an explanatory diagram illustrating an example of the hardware configuration of the information processing apparatus according to the embodiment.

[0129] The information processing device of the embodiment includes a control device such as a CPU (Central Processing Unit) 51, a storage device such as a ROM (Read Only Memory) 52 and a RAM (Random Access Memory) 53, a communication I / F 54 that connects to a network and communicates, and a bus 61 that connects each part.

[0130] The programs executed by the information processing apparatus according to the embodiment are provided in advance in the ROM 52 or the like.

[0131] The program executed by the information processing device of the embodiment may be configured to be provided as a computer program product by being recorded in an installable or executable format on a computer-readable recording medium such as a CD-ROM (Compact Disk Read Only Memory), a flexible disk (FD), a CD-R (Compact Disk Recordable), or a DVD (Digital Versatile Disk).

[0132] Furthermore, the program executed by the information processing apparatus of the embodiment may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Also, the program executed by the information processing apparatus of the embodiment may be provided or distributed via a network such as the Internet.

[0133] The programs executed by the information processing device of the embodiment can cause a computer to function as each of the above-mentioned parts of the information processing device. In this computer, the CPU 51 can read the programs from a computer-readable storage medium onto a main storage device and execute them.

[0134] A configuration example of the embodiment will be described below. (Configuration example 1) retrieving one or more behavior rules for one or more second plants associated with the detected situation for the first plant; For each of the one or more operation rules, extract one or more second properties that are one or more properties of one or more factors that affect the operation of the second plant for which the operation rule is searched, the second properties being different from the first properties being one or more properties of the one or more factors that affect the operation of the first plant; For each of the one or more extracted second properties, determining whether or not the second property occurs; selecting one or more of the behavior rules from which the second property determined to have occurred has been extracted; Processing section An information processing device comprising: (Configuration example 2) The processing unit extracting, from the second properties, properties that may be the cause of the situation based on knowledge data; extracting, from among the second properties for each of the one or more behavior rules, one or more second properties that are different from the first property and that may be the cause of the situation; The information processing device according to configuration example 1. (Configuration example 3) The processing unit determining whether an attribute of the first plant is similar to an attribute of the second plant for which the selected operation rule is defined; selecting the action rule determined to be similar; The information processing device according to configuration example 1 or 2. (Configuration Example 4) The attributes are one or more pieces of equipment included in the plant, an operation sequence of a plurality of pieces of equipment included in the plant, and part or all of an operation pattern. The information processing device according to configuration example 3. (Configuration Example 5) The attributes include one or more pieces of equipment included in the plant, an operation sequence of a plurality of pieces of equipment included in the plant, and an operation pattern; Higher priority is given to one or more pieces of equipment included in the plant, the order of operation of multiple pieces of equipment included in the plant, and the operation pattern, The processing unit determining whether or not the attribute of the first plant is similar to the attribute of the second plant for which the selected operation rule is defined, giving priority to an attribute having a higher priority than other attributes; The information processing device according to configuration example 3 or 4. (Configuration Example 6) The processing unit Searching for situation information indicating the occurrence of the second property; determining that the second property has occurred when the situation information indicating that the second property has occurred is found; 6. The information processing device according to any one of configuration examples 1 to 5. (Configuration Example 7) The processing unit searching for the status information from at least one of record data that records past operations of a plurality of plants including the first plant and the second plant and public data that is publicly available data including news; The information processing device according to configuration example 6. (Configuration Example 8) The processing unit For each of the one or more extracted second properties, determining whether or not the second property has occurred at a location identified based on the location of the first plant. The information processing device according to any one of configuration examples 1 to 7. (Configuration Example 9) The processing unit using knowledge data on one or more of the factors, for each of the extracted one or more second properties, identifying a period in which occurrence of the second property affects the operation of the first plant; For each of the one or more extracted second characteristics, determine whether or not the second characteristic occurred during the specified period. The information processing device according to any one of configuration examples 1 to 8. (Configuration Example 10) The processing unit searching for one or more action rules associated with the situation by matching the situation represented in a knowledge graph format with the action rules represented in the knowledge graph format; The information processing device according to any one of configuration examples 1 to 9. (Configuration Example 11) An information processing method executed by an information processing device, retrieving one or more operating rules for one or more second plants associated with the situation detected for the first plant; extracting, for each of the one or more operation rules, one or more second properties that are one or more properties of one or more factors that affect the operation of the second plant for which the operation rule is searched, the one or more second properties that are different from the first properties that are one or more properties of the one or more factors that affect the operation of the first plant; determining whether or not the second property occurs for each of the one or more extracted second properties; selecting one or more of the behavior rules from which the second property determined to have occurred has been extracted; An information processing method including: (Configuration Example 12) On the computer, retrieving one or more operating rules for one or more second plants associated with the situation detected for the first plant; extracting, for each of the one or more operation rules, one or more second properties that are one or more properties of one or more factors that affect the operation of the second plant for which the operation rule is searched, the one or more second properties that are different from the first properties that are one or more properties of the one or more factors that affect the operation of the first plant; determining whether or not the second property occurs for each of the one or more extracted second properties; selecting one or more of the behavior rules from which the second property determined to have occurred has been extracted; A program to execute.

[0135] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0136] 100 Information processing device 101 Acquisition Department 102 Rule Search Unit 103 Extraction part 104 Generation part 105 Information Search Department 106 Judgment section 107 Selection section 108 Output control section 120 Storage section 200 Search target data 300 Network

Claims

1. retrieving one or more operational rules for one or more second plants associated with the detected situation for the first plant; For each of the one or more operation rules, extract one or more second properties that are different from first properties that are one or more properties of the one or more factors that affect the operation of the first plant from second properties that are one or more properties of the one or more factors that affect the operation of the second plant for which the operation rule is searched; For each of the one or more extracted second properties, determining whether or not the second property has occurred; selecting one or more of the behavior rules from which the second property determined to have occurred has been extracted; Processing section An information processing device comprising:

2. The processing unit extracting, from the second properties, properties that may be the cause of the situation based on knowledge data; extracting, from the second properties for each of the one or more behavior rules, one or more second properties that are different from the first properties and that may be the cause of the situation; The information processing device according to claim 1 .

3. The processing unit determining whether an attribute of the first plant is similar to an attribute of the second plant for which the selected operation rule is defined; selecting the action rule determined to be similar; The information processing device according to claim 1 .

4. The attributes are one or more pieces of equipment included in the plant, an operation sequence of a plurality of pieces of equipment included in the plant, and part or all of an operation pattern. The information processing device according to claim 3 .

5. The attributes include one or more pieces of equipment included in the plant, an operation sequence of a plurality of pieces of equipment included in the plant, and an operation pattern; A higher priority is given to one or more pieces of equipment included in the plant, the order of operations of a plurality of pieces of equipment included in the plant, and the operation pattern, in that order. The processing unit determining whether or not an attribute of the first plant is similar to an attribute of the second plant for which the selected operation rule is defined, giving priority to an attribute having a higher priority than other attributes; The information processing device according to claim 3 .

6. The processing unit Searching for situation information indicating the occurrence of the second property; determining that the second property has occurred when the status information indicating that the second property has occurred is found; The information processing device according to claim 1 .

7. The processing unit searching for the status information from at least one of record data that records past operations of a plurality of plants including the first plant and the second plant and public data that is publicly available data including news; The information processing device according to claim 6 .

8. The processing unit For each of the one or more extracted second properties, determining whether or not the second property has occurred at a location identified based on the location of the first plant. The information processing device according to claim 1 .

9. The processing unit using knowledge data on one or more of the factors, for each of the extracted one or more second characteristics, identifying a period in which occurrence of the second characteristic affects the operation of the first plant; determining whether or not the second property has occurred during a specified period for each of the extracted one or more second properties; The information processing device according to claim 1 .

10. The processing unit searching for one or more action rules associated with the situation by matching the situation represented in a knowledge graph format with the action rules represented in the knowledge graph format; The information processing device according to claim 1 .

11. An information processing method executed by an information processing device, retrieving one or more operation rules for one or more second plants associated with the situation detected for the first plant; extracting, for each of the one or more operation rules, one or more second properties that are one or more properties of one or more factors that affect the operation of the second plant for which the operation rule is searched, the one or more second properties that are different from the first properties that are one or more properties of the one or more factors that affect the operation of the first plant; determining whether or not the second property occurs for each of the one or more extracted second properties; selecting one or more of the behavior rules from which the second property determined to have occurred was extracted; An information processing method including:

12. On the computer, retrieving one or more operation rules for one or more second plants associated with the situation detected for the first plant; extracting, for each of the one or more operation rules, one or more second properties that are one or more properties of one or more factors that affect the operation of the second plant for which the operation rule is searched, the one or more second properties that are different from the first properties that are one or more properties of the one or more factors that affect the operation of the first plant; determining whether or not the second property occurs for each of the one or more extracted second properties; selecting one or more of the behavior rules from which the second property determined to have occurred was extracted; A program to execute.

Images