Information processing system, information processing method, and program

WO2026216369A3PCT designated stage Publication Date: 2026-11-12KURITA WATER INDUSTRIES LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2026/013072
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-04-08
Filing Date
2026-03-30
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing industrial processes face challenges in effectively managing and restoring abnormal states, often requiring significant skill and time, leading to potential process disruptions and prolonged recovery times.

Method used

An information processing system that acquires and accumulates normal and abnormal information, generates correlation information, estimates the likelihood of abnormal states, and specifies effective countermeasures for timely intervention.

Benefits of technology

Enhances the ability to handle abnormal states by providing intuitive countermeasure information and maintaining high estimation accuracy, especially during process changes or increased operating rates.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2026013072_12112026_PF_FP_ABST
    Figure JP2026013072_12112026_PF_FP_ABST
Patent Text Reader

Abstract

[Summary] The present disclosure provides an information processing system, including a processor configured to execute a program so as to: acquire and accumulate normal information generated when a process is in a normal state; acquire and accumulate abnormal information generated when the process is in an abnormal state, the abnormal information including a countermeasure for the abnormal state and a result of the countermeasure; generate correlation information indicating a correlation between the accumulated normal information and the accumulated abnormal information; estimate a possibility of occurrence of the abnormal state based on the generated correlation information and the acquired normal information; specify an effective countermeasure for the abnormal state based on the accumulated abnormal information when the estimated possibility of occurrence is equal to or greater than a first level; and output countermeasure information indicating the specified effective countermeasure.
Need to check novelty before this filing date? Find Prior Art

Description

INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING METHOD, AND PROGRAM

[0001] CROSS REFERENCE TO RELATED APPLICATIONS The present application claims priority to Japanese Patent Application No. 2025-063837, filed April 8, 2025, the contents of which are incorporated herein by reference in their entirety. The present disclosure relates to an information processing system, an information processing method, and a program.

[0002] Various products including paper products are manufactured through industrial processes. In such processes, from the perspective of preventing or reducing the occurrence of trouble in operation and adverse effects on products, troubles are predicted in advance. In connection with this, techniques for estimating product quality and the like in papermaking processes, etc. are known. For example, Patent Literature 1 discloses a technique for estimating an expected result, etc. by using predetermined parameter information and relationship model information.

[0003] JP2022-123880 A

[0004] In various processes, when an abnormal state occurs, it is necessary to resolve the state and restore the process to a normal state; however, depending on the skill and experience of a worker who has dealt with the situation, there may occur a case where the process cannot be restored, or a case where the time required for restoration becomes excessively long.

[0005] In view of the above circumstances, the present disclosure provides an information processing system and the like that can improve the ability to deal with an abnormal state.

[0006] According to one aspect of the present disclosure, there is provided an information processing system including a processor configured to execute a program so as to: acquire and accumulate normal information generated when a process is in a normal state; acquire and accumulate abnormal information generated when the process is in an abnormal state, the abnormal information including a countermeasure for the abnormal state and a result of the countermeasure; generate correlation information indicating a correlation between the accumulated normal information and the accumulated abnormal information; estimate a possibility of occurrence of the abnormal state based on the generated correlation information and the acquired normal information; specify an effective countermeasure for the abnormal state based on the accumulated abnormal information when the estimated possibility of occurrence is equal to or greater than a first level; and output countermeasure information indicating the specified effective countermeasure.

[0007] According to the above aspect, an information processing system or the like that can improve the ability to deal with the abnormal state is provided.

[0008] Furthermore, the present disclosure may be provided in each of the following aspects.

[0009] (1) An information processing system, comprising a processor configured to execute a program so as to: acquire and accumulate normal information generated when a process is in a normal state; acquire and accumulate abnormal information generated when the process is in an abnormal state, the abnormal information including a countermeasure for the abnormal state and a result of the countermeasure; generate correlation information indicating a correlation between the accumulated normal information and the accumulated abnormal information; estimate a possibility of occurrence of the abnormal state based on the generated correlation information and the acquired normal information; specify an effective countermeasure for the abnormal state based on the accumulated abnormal information when the estimated possibility of occurrence is equal to or greater than a first level; and output countermeasure information indicating the specified effective countermeasure.

[0010] According to such an aspect, it is possible to improve the ability to deal with the abnormal state.

[0011] (2) The information processing system according to (1), wherein: the countermeasure information includes a procedure for an operation to resolve the abnormal state, an amount of the operation, a type of substance effective in resolving the abnormal state, or an amount of the substance.

[0012] According to such an aspect, it is possible to easily perform a countermeasure for avoiding the occurrence of the abnormal state.

[0013] (3) The information processing system according to (1) or (2), wherein: the processor is further configured to execute the program so as to: output the countermeasure information in a mode corresponding to effectiveness of the countermeasure indicated by the countermeasure information.

[0014] According to such an aspect, the effectiveness of the countermeasure can be intuitively grasped.

[0015] (4) The information processing system according to any one of (1) to (3), wherein: the processor is further configured to execute the program so as to: repeatedly generate the correlation information at a predetermined generation interval.

[0016] According to such an aspect, it is possible to constantly improve the estimation accuracy.

[0017] (5) The information processing system according to (4), wherein: the processor is further configured to execute the program so as to: shorten the generation interval for a certain period when there is a change in the process.

[0018] According to such an aspect, it is possible to suppress a decrease in estimation accuracy even when there is a change in the process.

[0019] (6) The information processing system according to (4) or (5), wherein: the processor is further configured to execute the program so as to: shorten the generation interval as an operating rate of the process increases.

[0020] According to such an aspect, the estimation accuracy can be more easily maintained at a high level even during a period when the operating rate of the process is high.

[0021] (7) The information processing system according to any one of (4) to (6), wherein: the processor is further configured to execute the program so as to: shorten the generation interval as a scale of the process is larger, a step of the process is more complex, or required accuracy is higher.

[0022] According to such an aspect, it is possible to suppress the occurrence of the abnormal state by performing a countermeasure at a more appropriate timing.

[0023] (8) The information processing system according to any one of (1) to (7), wherein: the processor is further configured to execute the program so as to: repeatedly estimate a possibility of occurrence of the abnormal state at a predetermined estimation interval, and shorten the estimation interval as the possibility of occurrence increases.

[0024] According to such an aspect, it is possible to suppress the occurrence of a situation in which the estimation is delayed and countermeasures cannot be taken in time.

[0025] (9) The information processing system according to any one of (1) to (8), wherein: the processor is further configured to execute the program so as to: acquire the normal information at a predetermined acquisition interval, and shorten the acquisition interval for a certain period when there is a change in the process.

[0026] According to such an aspect, the estimation accuracy of the changed process can be rapidly improved.

[0027] (10) The information processing system according to any one of (1) to (9), wherein: the processor is further configured to execute the program so as to: output preventive information indicating a countermeasure that prevents the abnormal state based on the accumulated abnormal information when the estimated possibility of occurrence becomes equal to or greater than a second level that is lower than the first level.

[0028] According to such an aspect, it is possible to suppress the occurrence of the abnormal state due to a sudden change in the state of the process.

[0029] (11) An information processing method comprising steps of executing operations performed by the information processing system according to any one of (1) to (10).

[0030] (12) A program configured to allow a computer to execute each operation performed by the information processing system according to any one of (1) to (10). Of course, the present disclosure is not limited to the above aspects.

[0031] FIG. 1 is a diagram showing an example of the overall configuration of a process management system 1.FIG. 2 is a diagram showing an example of a process to be managed.FIG. 3 is a diagram showing an example of a hardware configuration of a process management server 2.FIG. 4 is a diagram showing an example of a hardware configuration of a user terminal 3.FIG. 5 is an activity diagram illustrating influence degree specification processing.FIG. 6 is a diagram illustrating an example of a plot of the number A of occurrences of trouble versus index a related to an expected result, for a total of 30 data sets.FIG. 7 is an explanatory diagram showing an example of a method for calculating the degree of influence using a defect index as an index (expected result).FIG. 8 is a graph schematically showing the degree of influence on the event for each parameter.FIG. 9 is a schematic diagram showing the progression of actual measured values (a line graph) of the ORP (an example of a parameter) of a raw material system 1 for an event in a water system and the degree of influence of the ORP of the raw material system 1 on the event.FIG. 10 is an activity diagram illustrating operation management processing.FIG. 11 is a diagram illustrating an example of the specified degrees of influence.FIG. 12 is a diagram illustrating an example of a normal information DB.FIG. 13 is a diagram illustrating an example of a notification screen.FIG. 14 is a diagram illustrating an example of an input screen.FIG. 15 is a diagram illustrating an example of an abnormal information DB.FIG. 16 is a diagram illustrating an example of a countermeasure DB.FIG. 17 is a diagram illustrating an example of an explanatory image.FIG. 18 is a diagram illustrating another example of an explanatory image.FIG. 19 is a diagram illustrating an example of a registration screen for countermeasure information.FIG. 20 is a diagram illustrating an example of a setting screen for factor parameters.FIG. 21 is a diagram illustrating an example of the displayed countermeasure information.FIG. 22 is a diagram illustrating an example of an interval table.FIG. 23 is a diagram illustrating another example of an interval table.FIG. 24 is a diagram illustrating another example of an interval table.FIG. 25 is a diagram illustrating another example of an interval table.FIG. 26 is a diagram illustrating an example of an interval table.FIG. 27 is a diagram showing an example of the displayed prevention information.FIG. 28 is a diagram illustrating another example of a process to be managed.

[0032] <Embodiments> Hereinafter, an embodiment of the present disclosure will be described. Note that various features described in the embodiment below can be combined with each other.

[0033] A program for implementing software described in one embodiment may be provided as a non-transitory computer-readable storage medium, may be made available for download from an external server, or may be provided in such a manner that the program can be activated on an external computer to implement functions thereof on a client terminal (so-called cloud computing).

[0034] In various types of information processing according to one embodiment, an input and an output corresponding to the input can be realized. Here, as long as an output is obtained as a result of the input, the form of information referred to in such information processing (hereinafter referred to as reference information) is not limited. The reference information may be, for example, rule-based information such as a database, a look-up table, a predefined function (including a decision formula such as a regression equation constructed by statistical methods), may be a trained model in which the correlation between an input and an output has been learned in advance, or may be a generative AI such as a large-scale language model or a vision language model that can output a desired result by inputting a prompt.

[0035] In one embodiment, the term "unit" may also include, for example, a combination of hardware resources implemented by a circuit in a broad sense, and information processing performed by software that may be specifically implemented by such hardware resources. Furthermore, various kinds of information are handled in one embodiment, and such information may be represented by, for example, physical values of signal values representing voltage and current, high and low signal values as a set of binary bits consisting of 0 or 1, or quantum superposition (so-called qubits), and communication and computation may be executed on a circuit in a broad sense.

[0036] Furthermore, the term "circuit in a broad sense" refers to a circuit that is implemented by appropriately combining at least one or more of a circuit, circuitry, a processor, memory, and the like. The processor may be a general-purpose processor or a dedicated circuit. In other words, the term "circuit in a broad sense" includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), a field programmable gate array (FPGA)), and the like.

[0037] 1. Hardware Configuration FIG. 1 is a diagram showing an example of the overall configuration of a process management system 1. The process management system 1 is a system that executes information processing relating to industrial processes. Hereinafter, a case where the process management system 1 manages a water system process W will be described. The water system process W is various industrial processes involving water. The type of water system process W is not particularly limited, but as an example, it is a water system in a process of manufacturing a paper product and / or a material related to the manufacture of a paper product. In the following, these may be collectively referred to as "paper products, etc.".

[0038] "A material related to the manufacture of a paper product" does not refer to the paper product itself to be manufactured, but rather to various raw materials, intermediates and the like that are used in the manufacture of the paper product. Raw materials used in the manufacture of the paper product include white water, calcium oxide, etc. generated during the manufacturing process of the paper product. Intermediates used in the manufacture of the paper product include, for example, pulp and the like that is manufactured prior to the papermaking process. Specifically, examples of processes for manufacturing white liquor, calcium oxide, and pulp, which are materials related to the paper product, include a cooking process, a washing process, a black liquor concentration process, and a causticizing process. In addition, as a water system other than a water system of the process of manufacturing paper products, etc., examples of the water system to be a target may include various piping, a heat exchanger, a storage tank, a kiln, and cleaning equipment.

[0039] FIG. 2 is a diagram showing an example of a process to be managed. In FIG. 2, an example of a papermaking process is shown. In the papermaking process shown in FIG. 2, various systems such as a raw material system, a preparation / papermaking system, a recovery system, and a drainage system are shown. The application of the process management system 1 is not limited to the process shown in the figure, and predetermined elements may be added or deleted as appropriate. The process management system 1 includes a process management server 2, a user terminal 3, and a measurement device 4, which are connected via a communication line.

[0040] The communication line includes the Internet, wireless, etc., and serves to mediate exchange of data between apparatuses connected thereto. The measurement device 4 is configured to transmit various parameters measured in the water system process W to the process management server 2. A system exemplified by the process management system 1 comprises one or more devices or components. Thus, even the process management server 2 alone may be an example of the system, and a configuration that also includes the user terminal 3, the water system process W as the application target, and the measurement device 4 may be referred to as the system as well.

[0041] FIG. 3 is a diagram showing an example of a hardware configuration of the process management server 2. The process management server 2 includes a communication unit 21, a storage unit 22, and a controller 23, and is configured such that these units are electrically connected by a communication bus 20. Hereinafter, each unit of the process management server 2 will be described.

[0042] (Communication unit 21) The communication unit 21 is configured to transmit various electrical signals from the process management server 2 to an external component. Furthermore, the communication unit 21 is configured to receive various electrical signals from an external component to the process management server 2. The communication unit 21 includes a network communication function so that various information can be communicated between the process management server 2 and an external device via a communication line.

[0043] (Storage unit 22) The storage unit 22 stores various kinds of information defined by the description above. The storage unit 22 may be implemented, for example, as a storage device such as a solid state drive (SSD) storing various programs, etc. pertaining to the process management server 2 and executed by the controller 23, or as a memory such as a random access memory (RAM) storing temporarily necessary information (arguments, arrays, or the like) pertaining to program operations. The storage unit 22 stores various programs, variables, and the like pertaining to the process management server 2 and executed by the controller 23.

[0044] (Controller 23) The controller 23 includes at least one processor. The at least one processor may be constituted by, for example, an unshown central processing unit (CPU), an MPU (Micro Processing Unit), a GPU (Graphics Processing Unit), one or more Integrated Circuits, one or more Discrete Circuits, or a combination thereof.

[0045] The controller 23 implements various functions pertaining to the process management server 2 by reading a predetermined program stored in the storage unit 22. In other words, information processing by software stored in the storage unit 22 is implemented by the controller 23 that is an example of hardware, so as to function as respective functional units included in the controller 23. These will be described in further detail in the next section. The controller 23 is not limited to being a single unit, and may be implemented with multiple controllers 23 for the individual functions. Moreover, a combination thereof may also be employed.

[0046] FIG. 4 is a diagram showing an example of a hardware configuration of the user terminal 3. The user terminal 3 is typically a terminal used by a user who performs operations relating to the water system process W. The user terminal 3 includes a communication unit 31, a storage unit 32, a controller 33, an output unit 34 and an input unit 35, and these components are electrically connected via a communication bus 30 inside the user terminal 3. The descriptions of the communication unit 31, the storage unit 32, and the controller 33 are omitted since they are substantially the same as those of the communication unit 21, the storage unit 22, and the controller 23 in the process management server 2 described above.

[0047] (Output unit 34) The output unit 34 includes display means such as a display and sound emitting means such as a speaker, and outputs visual information and auditory information. For example, the output unit 34 displays, on a display surface of a display, visual information such as a screen, an image, an icon, text, and the like generated in a form that is visible to the user. Further, the output unit 34 outputs audible sound such as voice or synthetic sound from a speaker.

[0048] The output unit 34 may be included in a housing of the user terminal 3, or may be attached thereto externally. The output unit 34 displays a screen of a graphical user interface (GUI) that can be operated by a user. For example, it is preferable to selectively use, according to the type of the user terminal 3, a display device such as a CRT display, a liquid crystal display, an organic EL display, or a plasma display.

[0049] The input unit 35 may be included in the housing of the user terminal 3 or may be attached thereto externally. For example, the input unit 35 may be implemented as a touchscreen integrated with the output unit 34. If it is a touchscreen, the user can input operations such as tapping, swiping, and the like. Of course, instead of a touchscreen, switch buttons, a mouse, a QWERTY keyboard and the like may be adopted. Further, the input unit 35 may include sound collecting means such as a microphone, and may have a function of collecting voice of a user and receiving input of the collected voice. In other words, the input unit 35 receives an operation input performed by the user. This input is transferred as an instruction signal to the controller 33 via the communication bus 30, and the controller 33 may perform predetermined control or computation as needed.

[0050] The measurement device 4 is configured to measure various parameters related to the water system process W. In other words, this measurement device 4 is a device for measuring a predetermined parameter. In the present embodiment, predetermined calculation processing is performed using two or more parameters selected from a group consisting of a water quality parameter, a control parameter, and a result parameter, and the measurement device 4 acquires a parameter that serves as the basis for this calculation. Note that, although only a single measurement device 4 is shown in FIG. 1, a plurality (or multiple types) of measurement devices 4 may be applied in the process management system 1.

[0051] As the measurement device 4, various sensors and the like can be selected depending on the contents of the parameters to be measured. As the measurement device 4, for example, a pH meter, an electrical conductivity meter, an oxidation-reduction potential meter, a turbidimeter, a thermometer, a level meter for measuring foam height, a COD meter, a UV meter, a particle size distribution analyzer, a flocculation sensor, a digital camera (or a digital video camera), an internal bubble sensor, an absorption photometer, a freeness meter, a dissolved oxygen meter, a zeta potential meter, a residual chlorine meter, a hydrogen sulfide meter, a retention / freeness meter, a color sensor, a hydrogen peroxide meter, and a five-sense sensor can be used. The five-sense sensor here may include an image sensor, a luminous intensity sensor, an acoustic sensor, an ultrasonic sensor, a gas component sensor, an odor sensor, a liquid component sensor, a tactile sensor, a pressure sensor, a temperature sensor, a humidity sensor, a displacement sensor, etc.

[0052] It should be noted that in some cases, the control parameters and the like that are directly input for controlling the device may be used as they are, or such data may be received from the device via communication, or an operator of the device or the like may record the control parameters in a place other than the device.

[0053] 2. Information Processing Hereinafter, information processing according to the embodiment will be described. In the following description, the process management server 2 and the user terminal 3 are described as the entities of respective information processing; however, such information processing is executed by at least one processor included in the process management system 1, that is, by one or more processors included in a controller of each device. Each piece of information processing may also be described as being executed by a functional unit that is specifically implemented through hardware (each processor) executing information processing by software stored in each device. For example, information processing for "acquiring" information may be said to be executed by an "acquisition unit" implemented by a processor. Respective functional units may be appropriately added, omitted, or integrated depending on the application to which the process management server 2 is applied.

[0054] The process management system 1 executes operation management processing for managing the operation of an industrial process. In the present embodiment, the process management system 1 executes the operation management processing for the water system process W. The operation management processing includes influence degree specification processing for specifying the degree of influence of the above-described parameters on an event.

[0055] The "event" is an event in a water system process W or derived from the water system process W, and represents, for example, a concept such as a situation in the manufacturing process of a given product or the quality of the product to be manufactured. In the following, "an event in the water system or derived from the water system" may be simply referred to as an "event". In the embodiment, as an example, the event is described as referring to a case in which there is concern about trouble in the manufacturing process or a deterioration in product quality, but the event is not limited to this and may also be applied to cases in which the event refers to a favorable situation in the manufacturing process or a situation in which product quality can be ensured.

[0056] FIG. 5 is an activity diagram illustrating influence degree specification processing. (Activity A101) Acquisition of Parameter Information First, the process management server 2 acquires parameter information. The parameter information includes two or more parameters selected from the group consisting of the water quality parameter, the control parameter and the result parameter.

[0057] The water quality parameter is a parameter related to water quality of the water system. The control parameter is a parameter related to the water system, equipment related to the water system, or a control condition of a raw material to be added to the water system. The result parameter is a parameter having a different meaning from the expected result, and is a parameter that does not correspond to an index relating to a result having arisen in the water system, equipment related to the water system, or a raw material to be added to the water system, or a result having been derived from the water system, equipment related to the water system, or a raw material to be added to the water system. When acquiring these parameters, the process management server 2 is configured, for instance, to acquire various types of information via the communication unit 21 from the measurement device 4 that can measure at least a part of the parameters.

[0058] The parameter information is related to the water system process W regarding an estimation target and includes two or more parameters selected from the group consisting of the water quality parameter, the control parameter, and the result parameter. Note that the water system process W here is not limited to one present in a single tank or a single flow path or one in which a continuous flow is present. A water system having a plurality of tanks or flow paths, specifically, including a system in which a flow path branches, a water system in which a plurality of flow paths merge, a water system in which material is transferred in batch units from one tank to another, or a water system in which processing is performed during the flow, is also regarded as a single water system. In addition, when the water system related to the water quality parameter, the control parameter, or the result parameter is divided according to tanks or the like, the water quality parameter, the control parameter, or the result parameter with respect to a part of the water system may be used, or the water quality parameter, the control parameter, or the result parameter with respect to the entire water system may also be used.

[0059] The water quality parameter is not particularly limited as long as the parameter is related to a water quality of the water system process W. In addition, the control parameter is not particularly limited as long as the parameter is related to the water system process W, equipment related to the water system process W, or a control condition of a raw material to be added to the water system process W. Furthermore, the result parameter is not particularly limited as long as the parameter has a different meaning from the expected result and is related to a result having arisen in the water system process W, equipment related to the water system process W, or a raw material to be added to the water system process W, or a result having been derived from the water system process W, equipment related to the water system process W, or a raw material to be added to the water system process W.

[0060] Note that while "a parameter that does not correspond to an index" in the result parameter includes a case where assessment indices (for example, physical quantity) differ (for example, when one of the assessment indices is a length and the other assessment index is a mass), a case where the assessment index is identical but objects of assessment differ from each other (for example, a mass of paper and a mass of an additive) and a case where locations of measurement differ from each other (for example, an oxidation-reduction potential of a papermaking raw material system and an oxidation-reduction potential of a papermaking system), whereas, a case where the results have different meanings (for example, in a case where an oxidation-reduction potential is positive, and in a case where an oxidation-reduction potential is negative) is not included.

[0061] Hereinafter, specific examples of the water quality parameter, the control parameter, and the result parameter in a case where the water system process W is a water system in a process of manufacturing paper products, etc. will be described. These parameters may be related to, for example, the papermaking process as shown in FIG. 2, or may relate to, for example, a cooking system, a black liquor treatment system, a green liquor treatment system, a slaking / causticizing system, and a calcination system (lime calcination process) in a pulp manufacturing system, as described in JP 2022-12850A.

[0062] That is, the parameter information may include, as the water quality parameter, one or more selected from the group consisting of pH of the water system, electrical conductivity, an oxidation-reduction potential, a zeta potential, turbidity, temperature, a foam height, biochemical oxygen demand (BOD), chemical oxygen demand (COD), total organic carbon (TOC), inorganic carbon, absorbance, color, appearance, whiteness, transparency, a particle size distribution, a degree of flocculation, an amount of foreign matter, suspended solids (SS), a foaming area on a water surface, an area of underwater contamination, an amount of bubbles, an amount of glucose, an amount of organic acids, an amount of active alkali, total titrated alkali, metal or a metal ion content, a non-metal ion content, an amount of starch, an amount of calcium, an amount of total chlorine, an amount of free chlorine, an amount of dissolved oxygen (DO), a cationic demand, an amount of hydrogen sulfide, a degree of sulfidation, an amount of hydrogen peroxide, ash concentration, respiration rate of microorganisms in the system, a viable bacterial count, a spore-forming bacterial count, and ATP. These water quality parameters are acquired from any point in the water system process W.

[0063] It should be noted that, among the above-mentioned water quality parameters, "appearance" may be acquired from an RGB color sensor or a camera image. The "amount of active alkali" or "total titrated alkali" may refer to the amount of alkali evaluated as sodium, the amount of alkali evaluated as calcium, or the total amount of alkali. Furthermore, the metal in the "metal or metal ion content" may include heavy metals such as iron (Fe), lead (Pb), gold (Au), platinum (Pt), silver (Ag), copper (Cu), chromium (Cr), cadmium (Cd), mercury (Hg), zinc (Zn), arsenic (As), manganese (Mn), cobalt (Co), nickel (Ni), molybdenum (Mo), tungsten (W), tin (Sn), bismuth (Bi), uranium (U), and plutonium (Pu); alkali metals such as sodium (Na) and potassium (K); alkaline earth metals such as calcium (Ca) and barium (Ba); and base metals such as magnesium (Mg) and aluminum (Al). Note that the "metal or metal ion content" may be the content of a specific (one) metal or metal ion, or the content of multiple types of metals. In addition, the "non-metal ion" may include ions containing hetero atoms such as silicon (Si), phosphorus (P), sulfur (S), and nitrogen (N) as constituent components, as well as halogen ions such as chlorine (Cl) and bromine (Br).

[0064] In addition, the parameter information may include, as the control parameter, one or more selected from the group consisting of an operating speed (a papermaking speed) of a paper machine, a rotational speed of a filter cloth of a raw material dehydrator, a rotational speed of a filter cloth of a cleaning machine, an additive amount of chemicals or unit consumption to the water system, an additive amount of chemicals or unit consumption relative to a raw material to be added to the water system, an additive amount of chemicals or unit consumption relative to equipment related to the water system, an amount of vapor or unit consumption for heating, temperature of vapor for heating, pressure of vapor for heating, a flow rate from a seed box, nip pressure of a press part, felt vacuum pressure of a press part, a blending ratio of a papermaking raw material, a blended amount or unit consumption of waste papers of papermaking raw material, a mesh size of a screen for a papermaking raw material, a gap distance between a rotor and a stator of a beater, freeness, a degree of beating, draw, linear pressure of a press part, a gravity dewatering amount of a wire part or equivalent parameters, a forced dewatering amount of a wire part or equivalent parameters, a gravity dewatering amount of a press part, a parameter equivalent to a gravity dewatering amount of a press part, parameters for controlling the gravity dewatering amount of a press part, a forced dewatering amount of a press part, parameters equivalent to the forced dewatering amount of a press part, parameters for controlling the forced dewatering amount of a press part, concentration of interlayer adhesive, an amount or unit consumption of interlayer adhesive, a shower water amount or unit consumption, fresh water unit consumption, water usage unit consumption, timing or frequency of cleaning a wire part, timing or frequency of cleaning a press part, timing or frequency of cleaning a dryer part, timing or frequency of changing tools of a paper machine, a fresh water replenishment amount, an amount of rejects from the screen or parameters for controlling them, differential pressure across the screen or a parameter equivalent thereto, an amount of rejects from the cleaner or a parameter for controlling them, a fuel amount or unit consumption for heating, concentration in a floatater, an amount of air in the floatater or a parameter equivalent thereto, an amount of froth in a floatater or parameters equivalent thereto, the load of a disperser or a parameter equivalent thereto, the number of felt singeing, a flow rate of fluid flowing through the water system, air temperature, an amount of vapor or unit consumption for heating, an amount or unit consumption of calcium carbonate to be added, an amount or unit consumption of calcium bicarbonate to be added, a flow rate or concentration of clarified green liquor, a flow rate or concentration of raw green liquor, a flow rate or concentration of dilute black liquor, a flow rate or concentration of concentrated black liquor, a flow rate or concentration of white liquor, black liquor injection rate in a black liquor treatment system, black liquor injection concentration in a black liquor treatment system, an amount of pulp production, a moisture content of a mud filter, lime input or unit consumption in a slaking / causticizing system, an amount of lime production in a calcination system, fuel usage, a fuel usage unit consumption, causticizing efficiency in a slaking / causticizing system, caustic input or unit consumption in a cooking system, an amount of black liquor generated in a black liquor treatment system, an amount of solids in evaporating black liquor in a black liquor treatment system, an evaporation factor in evaporating black liquor in a black liquor treatment system, temperature of an economizer in a boiler in a black liquor treatment system, water content of sludge in a green liquor treatment system, specific gravity of black liquor, specific gravity of green liquor, specific gravity of white liquor, specific gravity of weak liquor, specific gravity of dregs, specific gravity of the calcium carbonate before calcination, and specific gravity of the calcium oxide after calcination.

[0065] The control parameters described above are acquired from various pieces of equipment and the like related to the water system process W. In the case where the water system process W is the water system in the process of manufacturing paper products, the equipment here includes, for example, wire and felt or the like of the paper machine to which chemicals are directly added, and a dry part (dryer) or the like of the papermaking system. With respect to the above control parameters, the type of "fuel" can be set appropriately, and may be, for example, heavy oil or petroleum coke. Furthermore, the point where the "temperature" in the "temperature of an economizer in a boiler in a black liquor treatment system" is acquired may be any point within the device. For example, it may be an inlet point of the device, it may be inside the device, or it may be an outlet point of the device.

[0066] In addition, the parameter information may include, as the result parameter, one or more selected from the group consisting of a unit weight (grammage) of a paper product, a yield rate, white water concentration, moisture content of a paper product, an amount of vapor or unit consumption in equipment manufacturing the paper product, temperature of vapor in equipment manufacturing the paper product, pressure of vapor in equipment manufacturing the paper product, thickness of the paper product, concentration of ash in the paper product, a type of defect of the paper product, the number of defects in the paper product, a timing of paper break in a process, freeness, a degree of beating, an amount of aeration, air temperature in a dryer, a joint rate, the number of joints, an amount of paper loss, the number of defects, a defect index, outdoor temperature including indoor temperature, outdoor humidity including indoor humidity, wet paper moisture content, product moisture content, temperature of rolls and cylinders in the dryer, a measurement value obtained by a five-sense sensor, a fuel amount or a fuel usage unit consumption, a calcination rate in a calcination system, a causticizing efficiency in a slaking / causticizing system, caustic input in a cooking system, and lime input in a slaking / causticizing system. Among these result parameters, as the amount of vapor in equipment manufacturing the paper product, for example, an amount of vapor in a dryer of the paper machine, an amount of vapor in a kraft pulp black liquor evaporator, an amount of vapor in a black liquor heater of a kraft pulp digester, and an amount of vapor injected to heat pulp raw material or white water can be used. These result parameters are acquired from various pieces of equipment and the like related to the water system process W.

[0067] Note that there are parameters which originally represent the same matter but are classified according to their purpose into two or more of the water quality parameter, the control parameter, and the result parameter. For example, in a black liquor evaporator, black liquor is heated by indirect heat exchange with vapor generated by a boiler and, as a result of the heating, process vapor is generated from the black liquor. The process vapor is used to heat (thicken) thickened black liquor in a next step. The amount of generated process vapor is a result parameter from the perspective of being generated from the black liquor and is also used as a control parameter from the perspective of being used in heating (thickening) thickened black liquor in a next step. In addition, the vapor generated by the boiler in order to heat the black liquor is used as a control parameter.

[0068] It is ensured that the two or more parameters acquired in A101 are not all substantially identical. For example, in a case where all of the process vapor generated from the black liquor described above is used to heat (thicken) the thickened black liquor in the next step, the process vapor generated from the black liquor as a result parameter and the amount of vapor to be used to heat (thicken) the thickened black liquor as a control parameter are excluded as the two parameters (parameters other than these two are not used). This is because, in such a case, the process vapor generated from the black liquor as a result parameter and the amount of vapor to be used to heat (thicken) the thickened black liquor as a control parameter are substantially identical.

[0069] However, in a case where a part of the process vapor generated from the black liquor is used to heat (thicken) the thickened black liquor in a next step, the process vapor generated from the black liquor as a result parameter and the amount of vapor to be used to heat (thicken) the thickened black liquor as a control parameter can be used as the two parameters. This is because, in such a case, the process vapor generated from the black liquor as a result parameter and the amount of vapor to be used to heat (thicken) the thickened black liquor as a control parameter are not substantially identical.

[0070] Furthermore, in a case where all of the process vapor generated from the black liquor described above is used to heat (thicken) the thickened black liquor in a next step, in addition to using the process vapor generated from the black liquor as a result parameter and the amount of vapor to be used to heat (thicken) the thickened black liquor as a control parameter as the two parameters, a plurality of parameters may be substantially identical if further combining another parameter such as the pH of the water system as a water quality parameter. In addition, for example, freeness and degree of beating may refer to the same parameters, and may be included in both the control parameter and the result parameter.

[0071] These parameters may be quantitative or qualitative. When using a qualitative parameter, a numerical value may be assigned to the parameter and the parameter may be handled as quantitative data.

[0072] Note that the water quality parameter, the control parameter, and the result parameter are respectively concepts encompassing a plurality of parameters. With respect to "acquiring, as parameter information, two or more parameters selected from a group consisting of a water quality parameter, a control parameter, and a result parameter", the two or more parameters included in the parameter information are independent of one another and can be selected from the respective parameters of the water quality parameter, the control parameter, and the result parameter, and two or more parameters (for example, pH and water temperature) may be selected from only one of the water quality parameter, the control parameter, and the result parameter or two or more parameters may be selected from a combination of two or three (for example, pH, press pressure of a press part, and thickness of the paper product) of the water quality parameter, the control parameter, and the result parameter. However, exactly identical parameters (for example, pH of water at point A and pH of water at point A) are not to be selected (however, for example, pH of water at point A and pH of water at point B, which are different measurement points, may be selected).

[0073] The order of performing Activity A101 here and Activity A102 that will be described later is arbitrary. Activity A101 can be performed prior to Activity A102, Activity A102 can be performed prior to Activity A101, or both activities can be performed in parallel (simultaneously).

[0074] (Activity A102) Acquisition of Relationship Model Information In Activity A102, the process management server 2 acquires relationship model information. The relationship model information indicates a relationship between an index created in advance, and the two or more parameters. In a typical example, the process management server 2 acquires relationship model information created by the process management server 2 itself. The process management server 2 is configured to execute a relationship model information creation step. In the relationship model information creation step, the process management server 2 creates or updates relationship model information to be used in the estimation step and the like described later.

[0075] Note that "in advance" means prior to estimating an index and may be either during actual operation of the water system process W or prior to actual operation, as long as it is prior to estimating the index. Furthermore, in one example of the embodiment, an "index" is quantitative and can be regarded, for example, as being for evaluating a situation of the above-described event (an event in the water system process W or an event derived from the water system process W). If the index indicates an unfavorable condition (if it has an unfavorable value), it is evaluated accordingly that the event is not in a favorable state and, for example, that troubles are likely to occur in the manufacturing process or that the quality of the products to be manufactured may deteriorate. In the embodiment, the index can be expressed as an expected result or an index related to the expected result. In the following, "an index related to the expected result" may be referred to as "a related index" to distinguish it from the generic "index".

[0076] The "expected result" related to the relationship model refers to various events related to the water system process W. The process management server 2 estimates the expected result (index), and the expected result related to the relationship model information also corresponds to the result estimated by the process management server 2. The expected result can be set appropriately depending on the estimation target, but in the case where the water system process W is a water system in a process of manufacturing a paper product, the expected result may be as shown below.

[0077] That is, in the present embodiment, the expected result (index) may be related to the quality or the amount of energy of the paper product and / or a material associated with the manufacture of the paper product. The quality may relate to one or more selected from the group consisting of the number of defects, paper strength, a joint rate, a sizing degree, air permeability, smoothness, an ash content, color tone, whiteness, formation, odor, causticizing efficiency, calcination rate, a kappa value, freeness, and a moisture content of a paper product and / or a material related to the manufacture of the paper product. The amount of energy may relate to one or more selected from the group consisting of an amount of vapor, a vapor unit consumption, a fuel amount, and a fuel usage unit consumption, in manufacturing a paper product and / or a material related to the manufacture of the paper product.

[0078] As mentioned above, "a material related to the manufacture of the paper product" does not refer to the paper product itself to be manufactured, but to various raw materials and intermediates used in the manufacture of the paper product. That is, a raw material used in the manufacture of the paper product includes white water, calcium oxide, and the like that are generated during the manufacturing process of the paper product. In other words, the causticizing efficiency and the calcination rate described above may be related to these raw materials. Intermediates used in the manufacture of the paper product include, for example, pulp manufactured prior to the papermaking process. The above-mentioned expected result may be an evaluation item related to such pulp.

[0079] The expected result may be the items listed above themselves, or may be a combination of the items listed above, or related items. For example, examples of the related items related to the calcination rate described above include the amount of fuel used to satisfy the predetermined calcination rate (typically, the amount of fuel used in the kiln when obtaining calcium oxide which is used in the causticizing and slaking processes, from calcium carbonate). The type of fuel for the "fuel amount" in the expected result (index) is not particularly limited, and may be heavy oil, petroleum coke, or the like.

[0080] In addition, the expected result in the present embodiment may correspond to the result parameter described above. In this case, the result parameter is usually not included in the parameter information (two or more parameters).

[0081] Examples of the relationship model information include, but are not particularly limited to, a function or a look-up table indicating a relationship between the expected result or a related index and two or more parameters, or a trained model of a relationship between the expected result or a related index and two or more parameters.

[0082] It is assumed that at least two parameters are common between the parameter information acquired in A101 and the parameter information used in the relationship model information. As described above, the water quality parameter, the control parameter, and the result parameter are each concepts encompassing a plurality of parameters. "Have two or more parameters that are common to each other" may mean two or more of any parameters (for example, two or more of only water quality parameters; pH and water temperature) among the water quality parameter, the control parameter, and the result parameter being common, a combination of the water quality parameter, the control parameter, and the result parameter (for example, one water quality parameter and one control parameter; for example, pH of water and press pressure of the press part) being common, or all of the water quality parameter, the control parameter, and the result parameter (for example, one water quality parameter, one control parameter, and one result parameter) being common.

[0083] In addition, the "related index" in the index may be an index that has a certain correlation with the expected result (e.g., a function of two or more parameters), and may be independently created by the user (operator, etc.) of such an information processing system, rather than a generally known index.

[0084] This relationship model information is created, for example, as follows. That is, prior to estimating the expected result or the related index, a prior measurement result corresponding to the expected result or a prior measurement index related to the prior result is measured. In addition, in the same water system, two or more parameters, each of which is one of the water quality parameter, the control parameter, or the result parameter, are measured. A plurality of sets of data of the prior measurement result or prior measurement index and the parameter are prepared, for example, by changing the day and time on which measurements are taken, so that variations occur in the prior measurement result or prior measurement index and the parameter.

[0085] Next, the prior measurement result or prior measurement index is assumed to be a function of two or more parameters, and the form and coefficients of the function are determined by comparing with the prior measurement result or the prior measurement index, thereby constructing relationship model information. Here, the relationship model information may be a model obtained by performing predetermined processing on a prior measurement result corresponding to the index or an index related to a prior confirmation result, and two or more parameters.

[0086] As the processing here, the following can be used: regression analysis (a linear model, a generalized linear model, a generalized linear mixed model, ridge regression, lasso regression, elastic net, support vector regression, projection pursuit regression, principal component regression, etc.), a time series analysis (a VAR model, a SVAR model, an ARIMAX model, a SARIMAX model, a state space model, a HMM model, etc.), a decision tree (a decision tree, a regression tree, a random forest, XGBoost, LightGBM, etc.), a neural network (a simple perceptron, a multilayer perceptron, a DNN, a CNN, a RNN, a LSTM, a GAN, a VAE, a LLM, etc.), Bayesian methods (naive Bayes, Bayesian optimization, a Bayesian network, etc.), clustering (k-means, k-means++, etc.), classification (a k-nearest neighbor algorithm, a support vector machine, etc.), ensemble learning (Boosting, Adaboost, etc.) or the like.

[0087] In an embodiment, the relationship model information is preferably a model obtained from a regression analysis between a prior confirmation result corresponding to the expected result or an index related to the prior confirmation result and the two or more parameters. Note that the number of sample sets when performing a regression analysis is not particularly limited.

[0088] The creation of the relationship model information is preferably performed in the same water system as the water system of which the expected result is to be estimated. In addition, for example, in a case where the water quality of a water system changes significantly even in the same apparatus (for example, a case where pulp that is a papermaking raw material is changed or the like in a papermaking system at a paper mill), relationship model information with respect to the water system after the water quality changes is preferably created and used.

[0089] From such a viewpoint, the expected result or the related index and the two or more parameters may be measured on a regular or irregular basis during operation of the water system process W, the relationship model information may be created each time, or the relationship model information may be updated by adding data.

[0090] Such relationship model information may be created and updated, for example, by a function of the process management server 2. The process management server 2 can create relationship model information suitable for the information processing of the present embodiment by performing predetermined calculation processing on the acquired expected result or related index and two or more parameters. Note that, the creation of relationship model information may be manually performed by, for example, an operator or the like.

[0091] An example of creating the relationship model information will be described below. Specifically, a case will be described in which an index a related to an expected result is calculated using a water quality parameter x, a water quality parameter y, and a control parameter z, and the index a related to the expected result and the number A of occurrences of trouble (times / day) as the expected result are calculated.

[0092] When creating the relationship model information, the water quality parameter x, the water quality parameter y, and the control parameter z shown in Table 1 below are measured and, at the same time, the number A of occurrences of trouble is also measured to acquire a total of 30 sets of data. The acquired data are shown in Table 1 below. Of course, the number of sets of data is not limited thereto, and may be set appropriately.

[0093] The index a related to the expected result is represented by the following Equation (1), where the "parameters" are x, y, and z, bn are coefficients of x, y, and z, and a0and b0are constants. (Equation 1)

[0094] A negative binomial regression analysis is performed based on actual measured values of the number A of occurrences of trouble and the index a related to the expected result of Equation (1). The index a related to the expected result calculated by the negative binomial regression analysis is shown in Table 1. FIG. 6 is a diagram illustrating an example of a plot of the number A of occurrences of trouble versus the index a related to the expected result, for a total of 30 data sets. A correlation coefficient between the number A of occurrences of trouble and the index a related to the expected result is r = 0.78 (p < 0.05), and a strong correlation is observed.

[0095]

[0096] Note that a function of the index a related to the expected result used in the regression analysis and the water quality parameter x, the water quality parameter y, and the control parameter z is not limited to Equation (1) described above, and a general Equation (2) can be used. (Equation 2)

[0097] (Activity A103) Estimation of Index After acquiring parameter information in Activity A101 and acquiring relationship model information in Activity A102, the process management server 2 estimates an index (an expected result or a related index) based on the acquired parameter information and the relationship model information. Typically, the process management server 2 can cause an estimated result to be output by inputting the acquired parameter information into the relationship model information.

[0098] Here, it is preferable that the estimated result is presented to the user in association with each parameter included in the parameter information. The estimated index is configured so as to be recognizable by the user or the like. In other words, the process management server 2 creates display information and controls the display information so that it can be visually recognized by the user or the like. Note that the display information may be visual information itself such as a screen, an image, an icon, text, etc. generated in a form that is visible to the user, or the display information may be rendering information for displaying visual information such as a screen, an image, an icon, text and the like on various devices or terminals.

[0099] (Activity A104) Specification of Data Set The process management server 2 specifies a data set in which an index related to an event (an expected result or a related index) is associated with two or more parameters.

[0100] Here, the parameter is a parameter associated with the index, in other words, may be relevant to the quality of the value of the index. The selection of parameters is made, for example, based on the knowledge of a user, etc. or through analysis by an arbitrary information processing apparatus. If this selection is appropriate, the correlation with the index will be appropriate, and thus it is expected that the accuracy of the index and the accuracy of an influence degree described below will also be improved.

[0101] As described above, the data set is data in which an index related to an event and two or more parameters are associated with each other as a set. In an exemplary embodiment, it is preferable that the index and the two or more parameters associated in the data set are more recently obtained than the index and the two or more parameters associated in the relationship model information. In other words, the data (index and two or more parameters) that constitute the data set may be new data that has not necessarily been used to create the relationship model information. That is, the data set in the present embodiment may be a data set based on the most recently obtained parameter information and the corresponding index. The corresponding index here may typically be a value output by inputting the most recently obtained parameter information into the relationship model information. This ensures real-time capability of the calculation of the degree of influence in Activity A105 described later, and also improves calculation accuracy of the degree of influence.

[0102] However, the specified data set is not limited thereto. For example, the process management server 2 may use the specified data set to create the relationship model information, and then use this data set to calculate the degree of influence. In other words, the specified data set may be part of the data set that constitutes the relationship model information, and does not necessarily need to be independent of the data in the relationship model information. Note that there is also an aspect that when the two are independent, the calculation accuracy of the degree of influence can be expected to be improved.

[0103] In addition, in the present embodiment, the data set is described as using an index related to the event which is estimated by the process management server 2, but is not limited thereto. By utilizing a statistical or machine learning method, an index related to the event may be calculated in a simulated manner, and the simulatively calculated index may be specified as a data set. For example, a method called Permutation Importance can be used to generate a simulated index by applying two or more new parameters to an arbitrary range of data sets used to create the relationship model information and pseudo-evaluating them, and this simulated index may be used as the index to be specified.

[0104] (Activity A105) Specification of Degree of Influence The process management server 2 specifies the degree of influence on the event for each parameter included in the data set by comparing the relationship model information acquired in Activity A102 with the data set specified in Activity A104.

[0105] The degree of influence refers to the degree of influence on an event for each parameter included in the data set, for example, when an event (e.g. some kind of trouble in a manufacturing process) occurs, if the degree of influence of a certain parameter in the data set is relatively high, it can be determined that the parameter is the main factor that causes the event (the aforementioned trouble).

[0106] There may be various methods for specifying the degree of influence, but in a typical aspect, the process management server 2 specifies (calculates, estimates, or infers) the degree of influence of each parameter associated in the data set by predetermined calculation processing. The predetermined calculation processing includes at least one of the difference processing, statistical processing, and processing using a trained model.

[0107] Here, a case where difference processing is used as the calculation processing for calculating the degree of influence will be described as an example. FIG. 7 is an explanatory diagram showing an example of a method for calculating the degree of influence using a defect index as an index (expected result). FIG. 8 is a graph schematically showing the degree of influence on the event for each parameter. In the following description, an example is given in which the parameters of the data set include six parameters (parameter P1 to parameter P6) as shown in FIG. 8.

[0108] In the defect index ID as an index (expected result) shown in FIG. 7, t0, t1, t2, t3, t4, t5, and t6 are determined in advance. The equation of the defect index ID shown in FIG. 7 and the values of t0 and the like correspond to the relationship model information. Furthermore, a and b are constants and can be set appropriately. The value of the defect index ID itself corresponds to an index (expected result).

[0109] Each degree of influence V (n) of a parameter Pn (n = 1 to 6) can be expressed as ID (p) - ID (n). In this way, the degree of influence is calculated (specified) by performing the difference processing related to the defect index ID. Note that FIG. 7 shows an example of the calculation of the degree of influence V (1), which is the value of the degree of influence of the parameter P1.

[0110] ID (p) is calculated using parameters obtained more recently than the parameter included in the relationship model information. The parameter obtained more recently is, for example, not a parameter used to create the relationship model information in the relationship model information creation unit described above, but a new parameter obtained separately from those used to create the relationship model information.

[0111] Then, the ID (p) is obtained by substituting the values of the parameters of the data set into the equation of the defect index ID. Specifically, A (p1) in the equation for ID (p) is the more recently obtained value for the parameter P1, B (p2) is the more recently obtained value for the parameter P2, C (p3) is the more recently obtained value for the parameter P3, D (p4) is the more recently obtained value for the parameter P4, E (p5) is the more recently obtained value for the parameter P5, and F (p6) is the more recently obtained value for the parameter P6.

[0112] These values A (p1) to F (p6) are typically values obtained at the same time, but in an exemplary aspect, they may be values obtained at different times. For example, an aspect is exemplified in which the value A (p1) is a value acquired X hours ago, while the value B (p2) is acquired as an average value in a time period from Y days ago to Z hours ago.

[0113] As shown in FIG. 7, ID (1) can be calculated based on the reference value A (s1) for the parameter P1 and the parameters (values B (p2) to F (p6)) obtained more recently than the parameters included in the relationship model information. The reference value A (s1) can be set based on various perspectives, such as a past average value during the relationship model creation period, and the like, a target reference value, a design value, and a numerical value for the period in which operation was good. The same applies to the reference value B (s2) to the reference value F (s6) of the other parameters P2 to P6. In other words, the reference values described here are based on the relationship model information. Note that the ID (1) term is the same as ID (p) except for the reference value A (s1) part, as shown by the arrow Ar in FIG. 7.

[0114] Although not shown in FIG. 7, the degree of influence V (2), which is a value of the degree of influence of the parameter P2, can be calculated by calculating ID (p) - ID (2). The value of ID (p) has been explained above so will be omitted here, but for ID (2), the value of B (p2) becomes the reference value B (s2). All other aspects are the same as ID (p). Note that the degree of influence V (3), which is the value of the degree of influence of the parameter P3, and subsequent degrees of influence can be calculated in the same manner.

[0115] In this way, the degree of influence is specified by comparing the relationship model information acquired in Activity A102 with the data set specified in Activity A104. ID (p) is a value calculated based on the data set, whereas ID (n) is calculated based not only on the above-mentioned data set but also on a reference value, and thus reflects the relationship model information. In one example of an embodiment (difference processing), the degree of influence V (n) of each parameter is given by ID (p) - ID (n), and thus the degree of influence can be said to be a value obtained by comparing the data set with the relationship model.

[0116] In the above-mentioned description, a case is described as an example in which, when calculating ID (p) and ID (n), parameter values (values A (p1) to F (p6)) obtained more recently than the parameters included in the relationship model information were used. In other words, the more recently obtained parameters (values A (p1) to F (p6)) used in calculating ID (p) and ID (n) are the values of parameter P1 to parameter P6 themselves, but are not limited thereto. For example, a moving average may be used instead of the value itself. In addition, composite values of two or more parameters (composite parameter values) may also be used as the parameters (values A (p1) to F (p6)) used in calculating the defect index described here. In this case, the degree of influence may be specified using the composite parameter corresponding to the composite value when calculating the degree of influence.

[0117] Furthermore, in the above description, an example has been described in which difference processing is used as calculation processing, but processing such as statistical processing and a trained model can also be used. For this processing, for example, SHAP, Permutation Importance, Feature Importance, impulse response functions, etc. can be used. Such processing may be combined with the above-mentioned difference processing to calculate the degree of influence with improved reliability and robustness.

[0118] Furthermore, although it has been described above that only one reference value is set for the value of each ID (n), this is not limited thereto. For example, when calculating the value of one ID (n), multiple reference values such as the reference value A (s1) and the reference value B (s2) may be used. In this case, the degree of influence due to the combination of two or more parameters is calculated. In other words, it becomes possible to specify the degree of influence related to the interaction of parameters.

[0119] When the process management server 2 specifies the degree of influence of each parameter as described above, the degree of influence is shown to the user via the output unit 34 of the user terminal 3, for example in a form such as that shown in FIG. 8. In other words, the process management server 2 can output the calculated degree of influence to the user terminal 3 as information that allows the user to understand it (influence degree information), thereby specifying the degree of influence.

[0120] The form of the influence degree information is not particularly limited, and may be, for example, visual information itself generated in a visually recognizable manner such as numerical values or graphs as shown in FIG. 8, or rendering information for displaying visual information. Furthermore, the information may be sound information rather than visual information, or both. In the example of FIG. 8, the user can visually recognize that the parameter having a greatest degree of influence as a factor causing the event is the parameter P2, "ORP (oxidation-reduction potential) of the raw material system 1".

[0121] If users can visually recognize the factor that causes the event as a manufacturing flow diagram in addition to the graph such as the one shown in FIG. 8, the system will be easier for users to use. For example, as shown in FIG. 2, the process management server 2 can specify a parameter having a large degree of influence (in the example of FIG. 2, the parameter P2 "ORP of the raw material system 1") by highlighting it on the manufacturing flow diagram.

[0122] Note that "specification" does not necessarily include content such as output of numerical values or graphs (output of visual information) or output of audio (output of audio information). In other words, "specification" may include, for example, simply calculating the degree of influence through the calculation processing described in FIG. 7. For example, if there is no particular problem with the manufacturing process or product quality, it is not necessary to inform the user, and also, there are cases where the user is not interested in the magnitude of the degree of influence of the parameters on the event, but is interested in countermeasures for the event, as described later.

[0123] The timing at which the process management server 2 specifies the degree of influence may be regular or irregular. When specifying the degree of influence regularly, for example, an interval of several seconds to several tens of minutes can be set. The process management server 2 may also specify the degree of influence based on manual instructions from the user (instructions from the user to specify the degree of influence). Furthermore, in a case where the calculated value of the index exceeds a preset threshold value, the process management server 2 may specify the degree of influence of each parameter.

[0124] FIG. 9 is a schematic diagram showing the progression of actual measured values (a line graph) of the ORP (an example of a parameter) of the raw material system 1 for an event in the water system and the degree of influence of the ORP of the raw material system 1 on the event. The process management server 2 may also output to the user terminal 3 a screen showing the actual measured values and the degree of influence over time, as shown in FIG. 9, for example. This allows the user to understand changes over time in both the actual measured values of the parameter and the degree of influence of the parameter, thereby making the system easier for the user to use.

[0125] As described above, by executing the influence degree specification processing, the degree of influence is specified for the parameters (parameter P1 to parameter P6) related to the data set. This makes it possible to specify, with high accuracy, which factor among all parameters (all factors) is most influential with respect to the event that is currently occurring (e.g., a trouble in the manufacturing process), thereby improving usability for the system user and facilitating the management of events involving the water system.

[0126] Next, the overall operation management processing will be described. FIG. 10 is an activity diagram illustrating the operation management processing.

[0127] (Activity A100) Influence Degree Specification Processing First, the process management server 2 executes the influence degree specification processing shown in FIG. 5. The process management server 2 repeatedly executes the influence degree specification processing, for example, at a predetermined time interval. The predetermined time interval is, for example, an interval of several seconds to several tens of minutes as described above, but is not particularly limited. The process management server 2 specifies the degree of influence for each of the above-described parameters.

[0128] FIG. 11 is a diagram illustrating an example of the specified degrees of influence. FIG. 11 shows a graph in which the vertical axis indicates the type of parameter and the horizontal axis indicates the degree of influence. In the example of FIG. 11, parameters such as a white water_MBI value, operation parameter C, operation parameter A, B tank_fine solids, A tank_ORP, operation parameter B, recovered water_electrical conductivity, recovered water_turbidity, and white water_temperature are shown in order from the top. In the example of FIG. 11, the degree of influence is shown in a range from ?10 to 60. Among the specified degrees of influence, the white water_MBI value is the largest and exceeds 60. Next, the operation parameter C is at approximately 17, and the operation parameter A at approximately 5, and the degrees of influence for the other parameters are less than 0.

[0129] (Activity A201) Determination of Abnormal State The process management server 2 determines whether the operation of the water system process W is in a normal state or an abnormal state based on the specified degree of influence. For example, when the specified degree of influence is less than a threshold, the process management server 2 determines that the operation of the water system process W is in a normal state, and when the specified degree of influence is equal to or greater than the threshold, the process management server 2 determines that the operation of the water system process W is in an abnormal state. Hereinafter, the threshold used for abnormality determination in this manner is referred to as an "abnormality threshold."

[0130] For example, in the example of FIG. 11, a value of "30" for the degree of influence is set as the abnormality threshold. The abnormality threshold is set to, for example, a value at which it is considered that if the parameter reaches or exceeds that value, the quality of the product manufactured involving the water system process W would fail to satisfy the standards. The abnormality threshold may be common to all parameters, or may be set for each parameter.

[0131] (Activity A202) Accumulation of Normal Information When the process management server 2 determines that the operation of the water system process W is in the normal state, the process management server 2 acquires and accumulates normal information generated when the water system process W is in the normal state. For example, the process management server 2 accumulates, as normal information, parameters acquired in the normal state (water quality parameters, control parameters, and result parameters). The process management server 2 stores and accumulates each parameter in a normal information DB (database).

[0132] FIG. 12 is a diagram illustrating an example of the normal information DB. In the normal information DB, an acquisition date and time, water quality parameters, control parameters, and result parameters are stored in association with one another. The process management server 2 specifies a degree of influence based on the acquired parameters, and when the specified degree of influence is less than the abnormality threshold, stores the parameters in the normal information DB as normal information in association with the acquisition date and time. The normal information stored in this manner may be information such as control histories or measurement results output by machines such as equipment or sensors, or may be information manually input by a worker who operated the equipment or by a worker who checked panels or meters of the equipment.

[0133] (Activity A301) Notification of Abnormal State When the process management server 2 determines that the operation of the water system process W is in the abnormal state, the process management server 2 transmits, to the user terminal 3, notification data for notifying the occurrence of the abnormal state. The user terminal 3 displays notification contents indicated by the transmitted notification data.

[0134] FIG. 13 is a diagram illustrating an example of a notification screen. A notification screen G1 shown in FIG. 13 displays a character string indicating notification contents "An abnormal state has occurred" and display fields C11, C12, and C13. In the display field C11, the parameters determined to be in the abnormal state are displayed as abnormal parameters. The abnormal parameter is one or more of the water quality parameters, the control parameters, and the result parameters. In the display field C12, a date and time at which the abnormal state was determined is displayed as an occurrence time.

[0135] In the display field C13, in the case where the location where the abnormal state has occurred in the water system process W can be specified, the occurrence location is displayed. For example, in the case where the abnormal parameter is a value measured by a sensor, the occurrence location of the abnormal state is the location measured by the sensor in the water system process W. In addition, when the abnormal parameter is a control history of an equipment, the occurrence location of the abnormal state is a location where the equipment is installed. When the occurrence location of the abnormal state cannot be specified (for example, yield rate or the number of defects), the occurrence location is not displayed.

[0136] (Activity A302) Countermeasure Operation for Abnormal State The process management server 2 accepts an operation by the user for taking a countermeasure for the abnormal state depending on the type of abnormal state. Examples of countermeasures for the abnormal state include adjustment of an addition amount of a chemical, backwashing of a filtration device (an operation of washing by flowing water in a reverse direction), switching of a line filter in a manufacturing process, addition of a chemical, and an increase in an amount of blowdown of clear water. When equipment for performing a corresponding countermeasure is a control target of the process management server 2, the user operates the equipment via the user terminal 3 and the process management server 2 to perform the corresponding countermeasure. Depending on the type of the countermeasure, it may be necessary to operate an equipment that cannot be controlled by the process management server 2. In that case, the user directly operates the equipment to perform the countermeasure.

[0137] (Activity A303) Display of Countermeasure Input Screen Next, the process management server 2 causes the user terminal 3 to display an input screen for inputting the contents of the countermeasure for the abnormal state.

[0138] FIG. 14 is a diagram illustrating an example of the input screen. An input screen G2 shown in FIG. 14 displays a character string "Please input countermeasure information." and an input field C21. The countermeasure information is information relating to the countermeasure for the abnormal state, and includes, for example, the contents of the countermeasure and a result obtained when the countermeasure was performed in the past. The contents of the countermeasure include operation details of the equipment used for the countermeasure, a type of a chemical used for the countermeasure, an addition amount thereof, and the like. In the input field C21, a task, a task name, a countermeasure date and time, a person in charge, an occurred event, a fact, an estimated cause, implementation / instruction details, countermeasure content, a countermeasure result, and a comment are input as the countermeasure information.

[0139] (Activity A304) Input of Countermeasure Contents The user terminal 3 accepts input of the countermeasure information by the user on the input screen G2. In the example of FIG. 14, an event in which a vacuum degree rises from ?90 KPa to ?60 KPa is input as the occurred event. Furthermore, the contents of the countermeasure that a vacuum pump is replaced with a new one and a result of the countermeasure that the vacuum degree is normalized are input. As the countermeasure information, not only text but also numerical values, charts, photographs, videos, and the like can be input. The user terminal 3 transmits the input countermeasure information to the process management server 2.

[0140] (Activity A203) Accumulation of Abnormal Information The process management server 2 acquires the transmitted countermeasure information as abnormal information and accumulates the abnormal information. The abnormal information is information generated when the water system process W is in the abnormal state. The countermeasure information is information relating to a countermeasure to be taken when the abnormal state is determined in the water system process W, and is therefore one example of the abnormal information. The process management server 2 stores and accumulates the countermeasure information in an abnormal information DB.

[0141] FIG. 15 is a diagram illustrating an example of the abnormal information DB. In the abnormal information DB, a determination date and time of the abnormal state, a task, a task name, a countermeasure date and time, a person in charge, an occurred event, countermeasure contents, a result of the countermeasure, and the like are stored as the abnormal information. The process management server 2 stores the acquired countermeasure information in the abnormal information DB as abnormal information in association with a date and time at which the abnormal state that triggered the countermeasure indicated by the countermeasure information was determined. The abnormal information stored in this manner may be information generated by a person based on an input operation or the like, or may be information generated by a machine such as an operation history of an equipment used for the countermeasure for the abnormal state.

[0142] (Activity A204) Generation of State Correlation Information The process management server 2 generates state correlation information indicating a correlation between the normal information accumulated in A202 and the abnormal information accumulated in A203. For example, the process management server 2 uses, as training data, past abnormal information and normal information acquired within a predetermined period going back from the determination date and time of the abnormal state indicated by the abnormal information, and causes AI (Artificial Intelligence) to generate a learning model indicating a correlation between the normal information and the abnormal information as the state correlation information.

[0143] For example, when an abnormal state in which the vacuum degree rises as described with reference to FIG. 14 occurs, if normal information in the same period or preceding period includes information having a correlation with the abnormal state, a learning model indicating the correlation is generated by causing the AI to perform the above machine learning. The state correlation information is not limited thereto. For example, the process management server 2 may perform analyses such as the above-described regression analysis, a time-series analysis, a decision tree, a neural network, Bayes, clustering, classification, or ensemble learning using the accumulated normal information and abnormal information, and generate, as the state correlation information, a table or algorithm indicating a correlation between the normal information and the abnormal information. These analyses may be performed autonomously by the process management server 2 or may be performed via a human operation.

[0144] The accumulation processing of the normal information in A201 and A202 described above is executed each time the degree of influence is specified. On the other hand, the accumulation processing of the abnormal information in A201, A301 to A304, and A203 is executed each time the abnormal state occurs. The generation processing of the state correlation information in A204 is executed, for example, at a timing when the abnormal information is accumulated. In parallel with the processing from A201 to A204, when, in the water system process W, a state has not yet reached the abnormal state but a possibility that the abnormal state will occur increases to a certain level or higher if left as it is, the process management server 2 executes countermeasure processing to deal with the state. Hereinafter, such a state in which the abnormal state is predicted to occur is referred to as a predicted abnormal state. Hereinafter, Activities A401 to A403, which are countermeasure processing for the predicted abnormal state, will be described.

[0145] (Activity A401) Determination of Predicted Abnormal State The process management server 2 determines whether the current state of the water system process W is the predicted abnormal state. In detail, the process management server 2 first estimates a possibility of the abnormal state occurring based on the state correlation information generated in A204 and the normal information accumulated in A202. This possibility can be expressed as a probability of the abnormal state occurring, and is hereinafter referred to as an "abnormal occurrence rate." For example, each time the normal information is accumulated in A202, the process management server 2 estimates the abnormal occurrence rate based on the accumulated normal information.

[0146] For example, when a learning model is generated as the state correlation information, the process management server 2 inputs the accumulated normal information into an AI including the generated learning model and instructs the AI to output the abnormal occurrence rate. Then, the process management server 2 estimates the abnormal occurrence rate based on the value output by the AI. The process management server 2 may estimate the abnormal occurrence rate based only on the normal information at the present time, or may estimate the abnormal occurrence rate based on the normal information for a certain period in the past including the present time (for example, from several hours to several days in the past).

[0147] The process management server 2 estimates the abnormal occurrence rate based on a correlation indicated by the generated state correlation information and the currently accumulated normal information, even when other state correlation information is generated. For example, it is assumed that an abnormal state IR1 is correlated with parameters P1, P2, and P3 included in the normal information, and that the occurrence probability of the abnormal state IR1 can be expressed by a function f(P1, P2, P3). In this case, the process management server 2 extracts the parameters P1, P2, and P3 from the accumulated normal information, calculates the function f(P1, P2, P3) into which the extracted parameters are substituted, thereby obtaining the occurrence probability of the abnormal state IR1. Then, the process management server 2 estimates the abnormal occurrence rate based on the calculated value.

[0148] The process management server 2 determines that the current state is a predicted abnormal state when the abnormal occurrence rate estimated as described above is equal to or greater than a predetermined level. The level herein is a value predetermined for determining the predicted abnormal state, and is hereinafter referred to as a determination level. As the determination level, for example, when the abnormal occurrence rate is estimated from 0% to 100%, a value such as 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, or 100% may be set, or a value within a range between any two of the numerical values exemplified here may be set.

[0149] When the process management server 2 determines that the current water system process W is not in the predicted abnormal state (NO), the process management server 2 terminates the countermeasure processing. When the process management server 2 determines that the current water system process W is in the predicted abnormal state (YES), the process management server 2 continues the countermeasure processing and executes the next Activity A402.

[0150] (Activity A402) Specification of Predicted Countermeasure When the process management server 2 determines in A401 that the water system process W is in a predicted abnormal state, the process management server 2 specifies an effective countermeasure for eliminating the predicted abnormal state. Hereinafter, this countermeasure is referred to as a "predicted countermeasure." For example, the process management server 2 specifies the predicted countermeasure based on the parameter determined to be in the abnormal state in A201, the abnormal information accumulated in A203, and the countermeasure information indicating a pre-registered countermeasure method. The countermeasure information is stored in the countermeasure DB stored in the process management server 2.

[0151] FIG. 16 is a diagram illustrating an example of the countermeasure DB. In the countermeasure DB shown in FIG. 16, explanatory images, countermeasure names, detailed explanations, and factor parameters are stored in association with one another. For example, the countermeasure "Adjustment of Chemical A" is associated with a detailed explanation "Adjust the addition amount of Chemical A so that MBI decreases." MBI (Microbial Biomass Index) is an index for evaluating the efficiency of water purification. A countermeasure "Backwashing of Filtration Device" is associated with a detailed explanation "Backwash the filtration device in the recovered water treatment system to reduce xx leakage." The detailed explanation may include more detailed operation procedures.

[0152] FIG. 17 is a diagram illustrating an example of an explanatory image. In an explanatory image D1 of "Adjustment of Chemical A" shown in FIG. 17, a table associating a value of "White Water MBI" with a value of "Recommended Addition Amount of Chemical A" (mg / l) and a graph showing a relationship between those values are shown. When "White Water MBI" is included in the water quality parameters and it is determined that there is a possibility of the abnormal state occurring in "White Water MBI," the explanatory image D1 can be used to determine the "Recommended Addition Amount of Chemical A" corresponding to the value of "White Water MBI," thereby enabling appropriate countermeasure.

[0153] FIG. 18 is a diagram illustrating another example of an explanatory image. As an explanatory image D2 of "Backwashing of Filtration device" shown in FIG. 18, an operation panel of a filtration device in the recovered water treatment system and a position of a backwash start switch on the operation panel are shown. When the countermeasure indicated by the abnormal information includes "Backwashing of Filtration device" and it is determined that there is a possibility of occurrence of the abnormal state identical to the abnormal state indicated by the abnormal information, the explanatory image D2 enables the operation of the operation panel of the filtration device to perform "Backwashing of Filtration device," thereby enabling an appropriate countermeasure.

[0154] A factor parameter is a parameter used when determining a predicted abnormal state, and is associated with a predicted countermeasure for an abnormal state predicted in the predicted abnormal state that is determined based on the parameter. For example, the countermeasure "Adjustment of Chemical A" is associated with the factor parameter "Operation Parameter A." In this case, when it is determined that the state is a predicted abnormal state based on "Operation Parameter A," it is indicated that the countermeasure "Adjustment of Chemical A" is effective to prevent the predicted abnormal state from occurring. The association between the factor parameter and the predicted countermeasure is performed, for example, by a user. Details thereof will be described later.

[0155] The process management server 2, for example, uses, as training data, an abnormal state indicated by the past abnormal information, a countermeasure taken for the abnormal state, a result of the countermeasure, and countermeasure information (explanatory images and detailed explanations) and factor parameters stored in the countermeasure DB for the identical abnormal state, and causes AI to generate, as the countermeasure correlation information, a learning model indicating a correlation between the abnormal state and a predicted countermeasure for the abnormal state. The countermeasure correlation information is not limited thereto, and a table or algorithm indicating a correlation between the abnormal state and a predicted countermeasure for the abnormal state may be generated as the countermeasure correlation information by performing analyses such as the above-described regression analysis, a time-series analysis, a decision tree, a neural network, Bayes, clustering, classification, or ensemble learning.

[0156] The process management server 2 specifies, as the predicted countermeasure for the predicted abnormal state, a countermeasure that is correlated with the abnormal state predicted when it is determined in A401 that the current state is a predicted abnormal state, based on the countermeasure correlation information. Specifically, when a learning model is generated as the countermeasure correlation information, the process management server 2 inputs the predicted abnormal state to the AI including the generated learning model and instructs the AI to output a predicted countermeasure for the abnormal state. The process management server 2 specifies the countermeasure output by the AI as the predicted countermeasure for the predicted abnormal state.

[0157] (Activity A305) Registration of Countermeasure Information The countermeasure information stored in the countermeasure DB shown in FIG. 16 is registered, for example, by a user who performed the countermeasure in A302 after completion of the countermeasure for the abnormal state, so as to be useful when a similar abnormal state occurs in the future. The process management server 2 causes the user terminal 3 to display a registration screen for registering the countermeasure information.

[0158] FIG. 19 is a diagram illustrating an example of the registration screen for the countermeasure information. On a registration screen G3 shown in FIG. 19, a character string "Please register countermeasure information.", input fields C31, C32 and C33 and a save button B31 are displayed. A name of the countermeasure is input in the input field C31. A detailed explanation of the countermeasure is input in the input field C32. An explanatory image of the countermeasure is input in the input field C33. There is no particular format for the detailed explanation and the explanatory image of the countermeasure, and any information may be used as long as it assists the user in performing the countermeasure.

[0159] For example, a type of an operation performed for the countermeasure, an operation amount, timing of the operation, equipment that accepts the operation, a location of the equipment, a usage method of the equipment, behavior after the countermeasure, and the like may be input as the countermeasure information. When the save button B31 is operated, the user terminal 3 stores the countermeasure information input in each input field in the countermeasure DB shown in FIG. 16. In this manner, the countermeasure information indicating countermeasures for the abnormal states that have occurred in the water system process W is accumulated.

[0160] (Activity A306) Setting of Factor Parameter When the countermeasure information is registered as described above, the parameters used for determining an abnormal state that requires the execution of the countermeasure indicated by the registered countermeasure information, that is, the factor parameters described with reference to FIG. 16, are set. Furthermore, the setting of the factor parameters is performed not only at the time of registering the countermeasure information, but also when an abnormal state newly occurs that requires the execution of the countermeasure indicated by the registered countermeasure information, or when there is a change in the required product quality. The process management server 2 causes the user terminal 3 to display a setting screen for setting the factor parameters.

[0161] FIG. 20 is a diagram illustrating an example of the setting screen for the factor parameters. On a setting screen G4 shown in FIG. 20, a character string "Please set the factor parameters.", input fields C41 and C42, an add button B41, input fields C43 and C44, and a delete button B42 are displayed. A name of the predicted countermeasure is input in the input field C41. A factor parameter corresponding to the predicted countermeasure is input in the input field C42. In the example of FIG. 20, the countermeasure name "Adjustment of Chemical A" is input in the input field C41, and the factor parameter "Operation Parameter A" is input in the input field C42.

[0162] When the add button B41 is operated, the process management server 2 displays new input fields for factor parameters. In the example of FIG. 20, the input fields C43 and C44 are additionally displayed. The factor parameter "White Water Tank_MBI Value" is input in the input field C43, and the factor parameter "A Tank_Operation Parameter A" is input in the input field C44. When any one of the input fields is selected and the delete button B42 is operated, the process management server 2 deletes the selected factor parameter.

[0163] As described above, when causing the AI to generate the countermeasure correlation information described in A401, the process management server 2 uses the countermeasure information and the factor parameters stored in the countermeasure DB as training data. The information included in the training data is not limited thereto. The training data may include, for example, factor parameters and countermeasure information in similar processes, or may include information found in books, on the Internet and the like (which may be publicly available or non-public documents). By setting the factor parameters used as the training data on the setting screen G4, it becomes easier to specify the countermeasure associated with those factor parameters as a predicted countermeasure for an abnormal state determined based on the set factor parameters.

[0164] (Activity A403) Output of Countermeasure Information The process management server 2 generates countermeasure information indicating the predicted countermeasure specified as described above and outputs the generated countermeasure information. For example, the process management server 2 outputs the countermeasure information to the user terminal 3 and causes the output countermeasure information to be displayed.

[0165] FIG. 21 is a diagram illustrating an example of the displayed countermeasure information. On a notification screen G5 shown in FIG. 21, a character string "An abnormal state is predicted. Please consider taking a countermeasure.", display fields C51, C52, C53, C54 and C55, an adjustment button B51, and an execution button B52 are displayed. The degree of influence of the factor parameter for which an abnormal state is predicted is displayed in the display field C51. In the example of FIG. 21, the parameter "White Water_MBI Value" having the degree of influence equal to or greater than a determination level (30 in the example of FIG. 21) is displayed.

[0166] A name of a predicted countermeasure for the predicted abnormal state is displayed in the display field C52, and a detailed explanation and an explanatory image of the predicted countermeasure are displayed in the display fields C53 and C54, respectively. A planned operation amount to be executed in the predicted countermeasure is displayed in the display field C55. In the example of FIG. 21, an amount of Chemical A to be added in the predicted countermeasure "Adjustment of Chemical A" is displayed as the operation amount. When the user determines that the operation amount is not appropriate, the user can operate the adjustment button B51 to adjust the operation amount.

[0167] Note that the adjustment of the operation amount is not limited to being performed from the user terminal 3 as in the above example. For example, the adjustment may be performed by operating an existing system that controls the operation of equipment, or by directly operating the equipment. Regardless of whether the process management system 1 or another system or device other than the process management system 1 is operated, once the displayed predicted countermeasure is executed, the parameters changed by the executed countermeasure are acquired and accumulated as normal information, and thus the same result can be obtained.

[0168] The amount of chemical to be added in the countermeasure processing and an operation amount for adjustment (hereinafter referred to as a "controlled amount") may be controlled by the process management server 2. The process management server 2 may determine the controlled amount by using, for example, PID control (P control, I control, D control, PI control, PID control, Ziegler-Nichols method, CHR method, ultimate sensitivity method, or optimal control method), advanced control (robust control, fuzzy control, neural network control, or adaptive control), statistical process control (control charts, process capability analysis, or sampling inspection), or other control methods (on-off control, expert system control, autonomous control). The influence degree specification processing shown in FIG. 5 may be performed in combination with an existing control method such as feedback control (a distributed control system (DCS), SCADA, and the like).

[0169] When the user determines that there is no problem with the content of the predicted countermeasure, the user operates an execution button B52 to execute the predicted countermeasure. In the example of FIG. 21, when the execution button B52 is operated, an amount of Chemical A displayed in the display field C55 is added. By executing this predicted countermeasure, the addition amount of Chemical A is adjusted, and the degree of influence of "White Water_MBI Value" decreases. Since the determination of the predicted abnormal state in A401 is repeatedly executed, the countermeasure information is repeatedly output as long as there are factor parameters whose degree of influence is equal to or greater than the determination level, and when there are no more factor parameters whose degree of influence is equal to or greater than the determination level, the output of the countermeasure information terminates.

[0170] The information displayed as the countermeasure information may be visual information itself generated in a form visually recognizable by the user, such as characters, numerical values, charts, photographs, videos, screens, images, icons, or text, or may be rendering information for displaying such visual information on various devices or terminals, for example.

[0171] Furthermore, the output destination of the countermeasure information is not limited to the above. For example, the process management server 2 may directly output the countermeasure information to the equipment that operates when performing the predicted countermeasure and cause the equipment to execute the predicted countermeasure indicated by the countermeasure information. In this case, the process management server 2 may output, for example, only the operation amount displayed on the planned operation amount shown in FIG. 21 as the countermeasure information to the target equipment. The equipment performs an operation of the predicted countermeasure based on the output countermeasure information. The process management server 2 may concurrently output the countermeasure information indicating what predicted countermeasure has been performed to the user terminal 3. This allows the countermeasure to be performed automatically while enabling the user to know what predicted countermeasure has been performed.

[0172] In the above example, the countermeasure processing from A401 to A403 is executed each time the normal information is accumulated in A202, and thus is performed, for example, at the above-described predetermined time interval (such as an interval of several seconds to several tens of minutes); however, the execution timing of the countermeasure processing is not limited thereto. The process management server 2 may execute the countermeasure processing when an instruction is received from a user, or may execute the countermeasure processing when the abnormal information is accumulated or when the state correlation information is updated. The process management server 2 may execute the countermeasure processing not based on the degree of influence of the parameter, but when the value of the parameter itself falls within a predetermined range.

[0173] As described above, the process management system 1 acquires and accumulates the normal information generated when the process is in the normal state. In addition, the process management system 1 also acquires and accumulates the abnormal information generated when the process is in the abnormal state. The water system process W shown in FIG. 1 is an example of the process. The information stored in the normal information DB shown in FIG. 12 is an example of the accumulated normal information. The information stored in the abnormal information DB shown in FIG. 15 is an example of the accumulated abnormal information.

[0174] The abnormal information includes a countermeasure for the abnormal state and a result of the countermeasure. The "Countermeasure Content" and "Countermeasure Result" shown in FIG. 15 are examples of the countermeasure and result. The "countermeasure" referred to herein is a countermeasure performed when an abnormal state has actually occurred, and does not include the above-described predicted countermeasure performed before the abnormal state occurs in anticipation of the occurrence of the abnormal state.

[0175] The process management system 1 generates correlation information indicating a correlation between the accumulated normal information and the accumulated abnormal information. The process management system 1 generates state correlation information in Activity A204 shown in FIG. 10. The state correlation information is an example of the correlation information. Next, the process management system 1 estimates a possibility of occurrence of an abnormal state based on the generated correlation information (state correlation information) and the acquired normal information. The process management system 1 estimates the possibility of occurrence of the abnormal state in Activity A401 shown in FIG. 10.

[0176] Subsequently, when the estimated possibility of occurrence becomes equal to or greater than a first level, the process management system 1 specifies an effective countermeasure for the abnormal state based on the accumulated abnormal information. The process management system 1 specifies an effective countermeasure for the abnormal state in Activity A402 shown in FIG. 10. The determination level described in the explanation of A402 is an example of the "first level." Then, the process management system 1 outputs countermeasure information indicating the specified effective countermeasure.

[0177] The process management system 1 outputs the countermeasure information in Activity A403 shown in FIG. 10. The information displayed on the notification screen G5 shown in FIG. 21 is an example of the output countermeasure information. When the output destination of the countermeasure information is equipment or the like, the predicted countermeasure is always executed before the predicted abnormal state occurs. On the other hand, when the output destination of the countermeasure information is the user terminal 3, the user who views the countermeasure information executes the predicted countermeasure, and thus the predicted countermeasure may be executed before the predicted abnormal state occurs, or may be executed after the abnormal state has occurred.

[0178] However, regardless of whether the output destination is equipment or the user terminal 3, in the case where the predicted countermeasure is executed before the abnormal state occurs, the abnormal state can be more easily avoided as compared with the case where the predicted countermeasure is not executed. Even in the case where the predicted countermeasure is executed after the abnormal state has occurred, the countermeasure information can be referred to, making it easier to resolve the abnormal state and restore the normal state. In this manner, according to the process management system 1, as compared with the case where no countermeasure information is output, it is possible to improve the ability to deal with an abnormal state that is predicted to occur in the process.

[0179] The countermeasure information includes a procedure for an operation to resolve the abnormal state or the amount of the operation. In the example of FIG. 21, the procedure for the operations displayed in the display field C53 is included in the countermeasure information, and an operation amount displayed in the display field C55 is included in the countermeasure information. The countermeasure information also includes a type of substance effective in resolving the abnormal state or an amount of the substance. The "Chemical A" displayed in the display field C53 of FIG. 21 is an example of the substance effective in resolving the abnormal state, and the operation amount (an amount of Chemical A to be added) displayed in the display field C55 is an example of an amount of the substance.

[0180] The operation procedure, the operation amount, the type of substance, and the amount of the substance are not limited to the above. For example, the detailed explanation shown in FIG. 16 provides an example of other operation procedures. Further, in the operation procedure of increasing an "amount of blowdown of clear water" described in the detailed explanation, for example, when using the equipment that adjusts an increase in the "amount of blowdown" through an operation amount such as a dial, the amount of blowdown to be increased serves as an example of the operation amount. Furthermore, examples of the substance effective in resolving the abnormal state include a replacement part such as a filter and a consumable item such as lubricating oil. In any case, according to the above aspect, in a situation where the occurrence of the abnormal state is predicted, it is possible to easily perform a countermeasure for avoiding the occurrence of the abnormal state.

[0181] <Modification: Output Mode of Countermeasure Information> In the example of FIG. 21, the process management server 2 outputs the countermeasure information indicating one predicted countermeasure; however, the process management server 2 may output countermeasure information indicating two or more predicted countermeasures. In this case, the process management server 2 may calculate a recommendation level of each predicted countermeasure and make a display mode of the countermeasure information of a predicted countermeasure having a high recommendation level different from a display mode of the countermeasure information of a predicted countermeasure having a low recommendation level.

[0182] The display mode of the countermeasure information can be differentiated by, for example, highlighting, text size, bolding, underlining, different fonts, or a display position. The display mode may also be differentiated by displaying a score indicating the recommendation level. In any case, the process management server 2 displays the countermeasure information of the predicted countermeasure with a higher recommendation level in a more conspicuous mode, so that the effectiveness of the predicted countermeasure indicated by each piece of countermeasure information is expressed by the display mode.

[0183] The recommendation level of the countermeasure can be calculated, for example, by having a person evaluate the countermeasure in advance based on the results of similar countermeasures in the past and then referring to that evaluation value. Furthermore, the process management server 2 may instruct the AI to calculate a higher recommendation level for countermeasures that are expected to result in a higher degree of decrease in the degree of influence when the countermeasure is performed, and may have the AI calculate the recommendation level. The AI can calculate the recommendation level by learning, as training data, the contents of the past countermeasures and results thereof. Note that the information included in the training data is not limited thereto. The training data may include, for example, the contents of the past countermeasures and the results thereof in similar processes, and may include information found in books, on the Internet, and the like (which may be publicly available or non-public documents).

[0184] As described above, the process management system 1 may output the countermeasure information in a mode corresponding to effectiveness of the countermeasure indicated by the countermeasure information. According to such an aspect, the user can intuitively grasp the effectiveness of the predicted countermeasure indicated by the output countermeasure information.

[0185] <Modification: Timing of generating state correlation information> In the above example, the generation processing of the state correlation information shown in A204 of FIG. 10 is executed at the timing when abnormal information is accumulated; however, the present disclosure is not limited thereto. The generation processing of the state correlation information may be executed, for example, at the timing when the user performs an operation instructing generation. The process management system 1 may repeatedly generate the state correlation information at a predetermined generation interval.

[0186] The generation interval is, for example, an interval at which the normal information is accumulated. In the above example, as long as no abnormal state occurs, the normal information is accumulated each time the influence degree specification processing is executed, and thus, for example, the above-described interval of several seconds to several tens of minutes becomes the generation interval of the state correlation information. Although the correlation between the normal information and the abnormal information does not change significantly unless a new abnormal state occurs, normal information indicating various normal states continues to be accumulated even when no abnormal state occurs, and thus the data for estimating the change trend of parameters continue to increase. Therefore, by repeatedly generating the state correlation information regardless of whether an abnormal state occurs or not as described above, it is possible to constantly improve the accuracy of estimating the possibility of occurrence of the abnormal state.

[0187] The process management system 1 may vary the generation interval according to the situation. For example, when there is a change in the process, the process management system 1 shortens the generation interval of the state correlation information for a certain period as compared with a period during which there is no change in the process. When there is a change in the process, the correlation between the normal information and the abnormal information may change. In such a case, when the predicted abnormal state is estimated using the state correlation information generated before the change in the process, the estimation accuracy decreases, making it more likely that an abnormal state occurs without performing any predicted countermeasures.

[0188] Therefore, by shortening the generation interval for a certain period after the process change, as compared with a case where the generation interval is always constant, it is possible to rapidly generate state correlation information that reflects the correlation between normal information and abnormal information accumulated after the process change, thereby suppressing a decrease in estimation accuracy even when there is a change in the process.

[0189] The process management system 1 may shorten the generation interval of the state correlation information as the operating rate of the process increases. In this case, for example, the process management server 2 determines a generation timing of the state correlation information by using an interval table that associates the operating rate of the process and the generation interval.

[0190] FIG. 22 is a diagram illustrating an example of an interval table. In an interval table TB1 shown in FIG. 22, generation intervals "T1," "T2," and "T3" (where T1 > T2 > T3) are associated with operating rates "less than Th1," "Th1 or more and less than Th2," and "Th2 or more." For example, the process management server 2 acquires the operating rate of the process as one of the parameters and generates the state correlation information at a generation interval associated with the acquired operating rate in the interval table TB1.

[0191] In industrial processes, generally, the higher the operating rate, the greater the burden on the equipment and workers, making an abnormal state more likely to occur. Furthermore, when an abnormal state occurs in a state where the operating rate is high, the process may stop or productivity may decrease, resulting in a greater influence on product production volume. Therefore, in the process management system 1, by using the interval table TB1, the process management server 2 shortens the generation interval as the operating rate of the process increases and generates the state correlation information. Accordingly, as compared with a case where the generation interval is always constant, the estimation accuracy is more easily maintained at a high level even during a period when the operating rate of the process is high, and a predicted countermeasure can be performed at a more appropriate timing to suppress the occurrence of the abnormal state.

[0192] The process management system 1 may shorten the generation interval of the state correlation information as the throughput per unit time of the process becomes larger. A larger throughput per unit time tends to increase the operating rate. Thus, similar to the example described with reference to FIG. 22, the estimation accuracy is easily maintained at a high level even during periods of high process throughput, and a predicted countermeasure can be performed at a more appropriate timing to suppress the occurrence of an abnormal state.

[0193] The process management system 1 may vary the generation interval of the state correlation information depending on the current period. For example, the process management system 1 may shorten the generation interval of the state correlation information during peak periods when demand increases compared to normal periods, and may lengthen the generation interval of the state correlation information during off-peak periods when demand decreases compared to normal periods. This makes it possible to maintain high estimation accuracy during periods of high operating rates, while reducing the generation cost of the state correlation information during periods of low operating rates.

[0194] The process management system 1 may shorten the generation interval of the state correlation information as the scale of the process is larger. In this case, for example, the process management server 2 determines the generation timing of the state correlation information by using an interval table that associates the scale of the process with the generation interval. The scale of the process is represented by, for example, a production volume of products, an area of a factory or equipment, the number of equipment units, an investment amount, an energy consumption (electric power, fuel, etc.), a resource input amount (raw materials, water, and consumption amounts of other resources), the number of employees, or an emission amount (emission amounts of air pollutants, water pollutants, waste, and the like).

[0195] FIG. 23 is a diagram illustrating another example of an interval table. In an interval table TB2 shown in FIG. 23, generation intervals "T11," "T12," and "T13" (where T11 > T12 > T13) are associated with the scales of the process "small," "medium," and "large." For example, the process management server 2 acquires scale information representing the scale of the process as one of the parameters and generates the state correlation information at a generation interval associated with the scale of the process indicated by the acquired scale information in the interval table TB2.

[0196] As the scale of the process is larger, the factors that cause occurrence of an abnormal state increase, and therefore the abnormal state is more likely to occur. Accordingly, by using the interval table TB2, the process management system 1 shortens the generation interval as the scale of the process is larger and generates the state correlation information. Accordingly, as compared with a case where the generation interval is always constant, the estimation accuracy is more easily maintained at a high level even when the scale of the process is large, and a countermeasure can be performed at a more appropriate timing to suppress the occurrence of the abnormal state.

[0197] The process management system 1 may shorten the generation interval of the state correlation information as the steps included in the process are more complex. In this case, for example, the process management server 2 determines a generation timing of the state correlation information by using an interval table that associates complexity of the steps included in the process with the generation interval. The complexity of steps is represented by, for example, the number of steps, the number of combinations of mutually dependent steps, the number of types of steps, a level of specialization of the steps, the number of fluctuation factors affecting the steps (quality of raw materials, failure rate of equipment, worker turnover, etc.), or an amount of information handled in the steps.

[0198] FIG. 24 is a diagram illustrating another example of an interval table. In the interval table TB3 shown in FIG. 24, generation intervals "T21," "T22," and "T23" (where T21 > T22 > T23) are associated with step complexity of "low," "medium," and "high." For example, the process management server 2 acquires the complexity of steps included in the process as one of the parameters and generates the state correlation information at a generation interval associated with the acquired complexity in the interval table TB3.

[0199] As the complexity of steps included in the process is higher, mistakes are more likely to occur, and an abnormal state caused by such mistakes is more likely to occur. Accordingly, by using the interval table TB3, the process management system 1 shortens the generation interval as the complexity of the steps is higher and generates the state correlation information. As a result, as compared with a case where the generation interval is always constant, the estimation accuracy is more easily maintained at a high level even when the complexity of the steps included in the process is high, and a predicted countermeasure can be performed at a more appropriate timing to suppress the occurrence of the abnormal state.

[0200] The process management system 1 may shorten the generation interval of the state correlation information as the accuracy required for the process is higher. In this case, for example, the process management server 2 determines the generation timing of the state correlation information by using an interval table that associates the level of accuracy required for the process with the generation interval. The level of required accuracy is represented by, for example, dimensional accuracy allowable for a product involved in the process, accuracy required for measurements performed during the process, accuracy of attachment positions of components, accuracy required for control of temperature and the like, or a yield rate of the product.

[0201] FIG. 25 is a diagram illustrating another example of an interval table. In an interval table TB4 shown in FIG. 25, generation intervals "T31," "T32," and "T33" (where T31 > T32 > T33) are associated with accuracies required for the process, "low," "medium," and "high." For example, the process management server 2 acquires the accuracy required for the process as one of the parameters and generates the state correlation information at a generation interval associated with the acquired required accuracy in the interval table TB4.

[0202] As the accuracy required for the process is higher, stricter operation is required in which even a slight fluctuation of a parameter is determined to be an abnormal state. Accordingly, by using the interval table TB4, the process management system 1 shortens the generation interval and generates the state correlation information as the accuracy required for the process is higher. As a result, as compared with a case where the generation interval is always constant, the estimation accuracy is more easily maintained at a high level even when the accuracy required for the process is high, and a countermeasure can be performed at a more appropriate timing to suppress the occurrence of the abnormal state.

[0203] <Modification: Timing of Estimation> In A401 of FIG. 10, the process management system 1 repeatedly estimates the possibility of occurrence of an abnormal state at a predetermined estimation interval (for example, an interval of several seconds to several tens of minutes), but the estimation interval may be varied. For example, the process management system 1 shortens the estimation interval as the possibility of occurrence of the abnormal state increases. In this case, for example, the process management server 2 determines a timing of estimating the abnormal occurrence rate by using an interval table that associates the level of the abnormal occurrence rate (possibility of occurrence of the abnormal state) with the estimation interval.

[0204] FIG. 26 is a diagram illustrating an example of an interval table. In an interval table TB5 shown in FIG. 26, estimation intervals "T41," "T42," and "T43" (where T41 > T42 > T43) are associated with abnormal occurrence rates "low," "medium," and "high." When determining the predicted abnormal state in A401, the process management server 2 calculates the abnormal occurrence rate and estimates the predicted abnormal state at an estimation interval associated with the calculated abnormal occurrence rate in the interval table TB5.

[0205] Even when the abnormal occurrence rate is lower than the determination level, as the abnormal occurrence rate increases and approaches the determination level, the abnormal state naturally becomes more likely to occur. Accordingly, by using the interval table TB5, the process management system 1 shortens the estimation interval as the abnormal occurrence rate increases so that the predicted abnormal state is estimated more frequently. As a result, as compared with a case where the estimation interval is always constant, it is possible to suppress the occurrence of a situation in which the estimation is delayed and countermeasures cannot be taken in time.

[0206] The process management system 1 may also shorten the estimation interval of the possibility of occurrence of the abnormal state, for example, as the operating rate increases, during periods of higher operating rates, or during periods with a larger process throughput. In any case, by shortening the estimation interval in a situation where the abnormal state is more likely to occur or where the influence of occurrence of the abnormal state tends to be greater, as compared with a case where the estimation interval is always constant, it is possible to suppress occurrence of a situation in which the estimation is delayed and countermeasures cannot be taken in time.

[0207] The process management system 1 may also shorten the estimation interval of the possibility of occurrence of the abnormal state as, for example, the scale of the process is larger, the steps included in the process are more complex, or the accuracy required for the process is higher. In these cases as well, by shortening the estimation interval in an environment where the abnormal state is more likely to occur or where the influence of occurrence of the abnormal state tends to be greater, as compared with a case where the estimation interval is always constant, it is possible to suppress occurrence of a situation in which estimation is delayed and countermeasures cannot be taken in time.

[0208] <Modification: Timing of acquiring normal information> In A202 of FIG. 10, the process management system 1 acquires normal information at a predetermined acquisition interval. The acquisition interval of the normal information substantially coincides with an interval at which the influence degree specification processing is performed (since the normal information is acquired as long as an abnormal state does not occur), and is, for example, an interval of several seconds to several tens of minutes. The process management system 1 may vary the acquisition interval of the normal information.

[0209] The process management system 1 shortens the acquisition interval for a certain period when, for example, there is a change in the process. When there is a change in the process, as described above, the correlation between the normal information and the abnormal information may change, and the estimation accuracy of the predicted abnormal state is likely to decrease. Accordingly, by shortening the acquisition interval of the normal information for a certain period after the process change, the amount of the normal information is increased, and as compared with a case where the acquisition interval is always constant, the estimation accuracy of the predicted abnormal state in the changed process can be rapidly improved.

[0210] Additionally, the process management system 1 may shorten an acquisition interval of the normal information, for example, as the operating rate increases, during a period in which the operating rate is higher, or during a period with a larger process throughput. In any case, by shortening the acquisition interval of the normal information in a situation where an abnormal state is more likely to occur or where the influence of occurrence of the abnormal state tends to be greater, the amount of the normal information is increased, and as compared with a case where the acquisition interval is always constant, the estimation accuracy of a predicted abnormal state can be improved and the occurrence of the abnormal state can be suppressed.

[0211] The process management system 1 may shorten the acquisition interval of the normal information as, for example, the scale of the process is larger, the steps included in the process are more complex, or the accuracy required for the process is higher. In these cases as well, by shortening the acquisition interval of the normal information in an environment where the abnormal state is more likely to occur or where the influence of occurrence of the abnormal state tends to be greater, the amount of the normal information is increased, and as compared with a case where the acquisition interval is always constant, the estimation accuracy of the predicted abnormal state can be improved and the occurrence of the abnormal state can be suppressed.

[0212] <Modification: Output of Preventive Information> In the above example, the predicted countermeasure is performed when a predicted abnormal state is determined; however, a preventive countermeasure may be performed at an earlier timing. Hereinafter, this countermeasure is referred to as an "abnormality preventive countermeasure." In this case, for example, when the estimated possibility of occurrence becomes equal to or greater than a second level that is lower than the first level, the process management system 1 outputs preventive information indicating an abnormality preventive countermeasure (a countermeasure that prevents an abnormal state) based on the accumulated abnormal information.

[0213] The second level is a value predetermined for preventing an abnormal state, and is hereinafter referred to as a prevention level. As the prevention level, for example, when the abnormal occurrence rate is estimated between 0% and 100%, a value smaller than the determination level may be set from among values, specifically such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, and 100%, or may be set within a range between any two of the numerical values exemplified herein. The process management server 2 outputs the prevention information to the user terminal 3, for example, and causes the user terminal 3 to display the output prevention information.

[0214] FIG. 27 is a diagram showing an example of the displayed prevention information. In a notification screen G6 shown in FIG. 27, a character string "Please consider taking countermeasures to prevent an abnormal state.", display fields C61, C62, C63, C64, and C65, an adjustment button B61, and an execution button B62 are displayed. In the display field C61, the degree of influence of the factor parameters is displayed. In the example of FIG. 27, a parameter P4 is displayed whose degree of influence is less than the determination level (30 in the example of FIG. 27) and equal to or greater than the prevention level (15 in the example of FIG. 27).

[0215] In the display field C62, "Addition of Chemical B" is displayed as the name of the abnormality prevention countermeasure. In the display fields C63 and C64, a detailed explanation and an explanatory image of the abnormality preventive countermeasure are displayed, respectively. In the display field C65, a planned operation amount to be executed in the abnormality preventive countermeasure is displayed, and the operation amount is adjustable by operating the adjustment button B61. When the user determines that there is no problem with the content of the countermeasure, the user operates the execution button B62 to execute the abnormality preventive countermeasure. By performing the abnormality preventive countermeasure in this manner, it is possible to suppress occurrence of an abnormal state due to a sudden change in the state of the process or the like, as compared with a case where the prevention information is not output.

[0216] <Modification: Countermeasure Intensity> Basically, the predicted countermeasure or the preventive countermeasure is performed with a set constant intensity. The intensity of the predicted countermeasure or the preventive countermeasure (hereinafter referred to as "countermeasure intensity") is represented by, for example, an amount of a chemical to be added or an operation amount of equipment to be operated. The process management server 2 may vary the countermeasure intensity according to the status of the process. The status of the process includes, for example, an operating rate of the process, throughput of the process, a current period, quality required for a product produced in the process, or the change trend of the normal information over a predetermined period.

[0217] For example, the process management server 2 increases the countermeasure intensity, as the operating rate of the process increases or as the process throughput is larger. The process management server 2 also increases the countermeasure intensity during a period in which the operating rate of the process is likely to be higher or in which the process throughput is likely to be large (for example, during a peak period). Since an abnormal state is more likely to occur as the operating rate is higher or the throughput is larger, and an influence when the process stops becomes greater, by strengthening the countermeasure intensity as described above, as compared with a case where the countermeasure intensity is uniform, it is possible to promptly resolve a state in which an abnormality is likely to occur and to reduce a risk of occurrence of the abnormal state.

[0218] Additionally, as the quality required for a product is higher, an influence when the process stops becomes greater. Accordingly, by strengthening the countermeasure intensity as the quality required for the product is higher, as compared with a case where the countermeasure intensity is uniform, the process management server 2 can reduce a risk of occurrence of the abnormal state and suppress occurrence of a situation in which the process stops. Furthermore, as the change trend of the normal information becomes greater, parameters indicated by the normal information are more likely to change, and it can be said that the process is in a state where the parameters are likely to change abruptly and an abnormal state such as an unexpected stop is likely to occur. Accordingly, by strengthening the countermeasure intensity as the change trend of the normal information becomes greater, as compared with a case where the countermeasure intensity is uniform, the process management server 2 can reduce a risk of occurrence of the abnormal state and suppress occurrence of a situation in which the process stops.

[0219] <Modification: Process to be managed> The process to be managed by the process management system 1 is not limited to the process shown in FIG. 2. FIG. 28 is a diagram illustrating another example of a process to be managed. FIG. 28 shows an example of a standardized pure water supply process. In the standardized pure water supply process shown in FIG. 28, raw water is supplied from a tank, pretreated by a carboner and a security filter, and then passed through an RO membrane, a degassing membrane, and CEDI (Continuous Electrodeionization) to supply pure water. In the treatment using the RO membrane, the water is separated into treated water and brine water due to osmotic pressure. The brine water is water in which impurities are concentrated, and is discharged or recovered for use in other applications.

[0220] In the standardized pure water supply process, the process management system 1 is applied, for example, as follows. First, as an event indicating an abnormal state, for example, a decrease in the brine flow rate of the first-stage RO is predicted. In this case, the following are displayed on the user terminal 3 as countermeasure information: (1) a search of an event history reveals that there have been cases in which the reason for a decrease in the brine flow rate in the past was due to changes in an inverter frequency, RO feed water pressure, and brine pressure; (2) an examination of relevant current sensor values reveals that the inverter frequency has significantly decreased, and (3) a priority assigned by summarizing countermeasures to be taken from the event history and a prior knowledge unit.

[0221] The parameter information used in the standardized pure water supply process may include, as water quality parameters, at least one selected from the group consisting of, for example, acid consumption (4.8) (CaCO3), total hardness (CaCO3), calcium hardness (CaCO3), magnesium hardness (CaCO3), chloride ion, sulfate ion, silica (SiO2), trace phosphate ion, trace iron, trace manganese, ammonium ion, nitrate ion, turbidity (degree), CODMn, TOC, sodium, trace aluminum, potassium, fluoride ion, total evaporation residue, vanadium, UV260, free available chlorine, combined available chlorine, ORP, pressure, flow rate, temperature, carbon dioxide, vacuum degree, water level, pressure, differential pressure, flow rate, temperature, frequency, number of operations, number of starts and stops, operation time, water flow time, air injection time, number of valve openings and closings, operation content, resistivity, drainage amount, current value, and voltage value.

[0222] Furthermore, the parameter information may include, as control parameters, at least one selected from the group consisting of the number of membranes, the number of elements, the number of cells, apparatus, the number of devices, membrane filtration area, activated carbon capacity, standard water flow rate of activated carbon, piping connection information, effective pressure, membrane recovery rate, operation instruction signal, stop signal, operation determination signal, correction coefficient, water flow signal, start / stop signal, opening / closing signal, PLC signal, and inverter signal.

[0223] In addition, the parameter information may include, as result parameters, at least one selected from the group consisting of water level, pressure, differential pressure, flow rate, temperature, chlorine concentration, ORP, turbidity, pH, frequency, number of starts and stops, operation time, number of valve openings and closings, silica concentration, resistivity, drainage amount, silica removal rate, scale risk rate, drainage amount, replacement timing of equipment, replacement timing of sensors, membrane fouling timing, carboner backwash timing, correction coefficient, recovery rate, current efficiency, water saving amount, power saving amount, gas saving amount, impurity removal rate, current value, and voltage value.

[0224] <Modification: Controlled amount> In the above example, the controlled amount of the equipment and the like executed in the state improvement countermeasure is determined by PID control or the like, but the present disclosure is not limited thereto. For example, when two or more related indices (for example, a defect index and a paper-break index ) are operated in the same water system process W, a trade-off relationship may arise, and the controlled amount may not be simply determined. For example, there may be a case where there is a product quality prediction index and an energy consumption prediction index, and when pH, which is one of their common explanatory variables, is increased, defect prediction improves but energy prediction deteriorates. In such a case, the process management server 2 uses an optimization method to search for and control the optimal controlled amount. As specific optimization methods, for example, Pareto optimization, a -constraint method, a penalty method, a barrier method, a Lagrange multiplier method, a genetic algorithm, or differential evolution may be used.

[0225] <Variation Example: Configuration Variations> The configuration shown in FIG. 1 and the like (overall configuration, hardware configuration, functional configuration, and the like) is an example, and other configurations may be adopted as long as no inconvenience arises in implementation. For example, the process management server 2 may be distributed across two or more devices, and may be provided in any form such as on-premises, SaaS (Software as a Service), or a cloud computing system.

[0226] Furthermore, the process management server 2 may have, for example, a hybrid configuration including a cloud server and an edge server installed in a facility of the water system process W. In this case, the cloud server performs the acquisition of relationship information, the estimation of expected results, and the like, while the edge server executes the acquisition of parameters, the countermeasure processing, and the like. Note that the relationship information may also be acquired by the edge server so that the countermeasure processing can be continued even when communication between the cloud server and the edge server is interrupted.

[0227] Furthermore, part or all of the processing executed by the process management server 2 may be executed by the user terminal 3, or part or all of the processing executed by the user terminal 3 may be executed by the process management server 2. In short, as long as the necessary information processing is executed in the process management system 1 as a whole, the devices that execute such information processing are not limited.

[0228] The output destination of information or data (hereinafter referred to as "information and the like") may be another device, a display, a storage unit (including an internal storage unit and an external storage unit), an email address, an account of another system, or the like. The acquisition of the information and the like includes, in addition to an aspect of acquiring information and the like transmitted from another device, an aspect of acquiring information and the like generated by the device itself or information and the like stored in the device itself. The database may be stored in an information processing device such as the process management server 2, or may be stored in an external device. Furthermore, the information stored in the database may be used in an aspect in which the information is temporarily stored in a memory.

[0229] A table or the like (a table, a database, or the like) that associates input information with output information is not limited to the illustrated table or the like, and the number of pieces of input information and output information may be reduced or increased. The relationship between the input information and the output information may be one-to-one, one-to-N, M-to-one, or M-to-N (where M and N are natural numbers). Furthermore, instead of using a table or the like, output information corresponding to input information may be obtained by a mathematical formula, a conditional expression, or the like. A trained model generated by performing machine learning using the input information and correct output information corresponding to the input information as training data may be used. In this manner, any method may be used as long as output information corresponding to input information is obtained based on a correlation between the input information and the output information.

[0230] Although the above-mentioned embodiment is described as a configuration of the process management system 1, an information processing method executed by an information processing system may be provided in which the method includes steps of executing processing of each unit of the information processing system. In addition, a program causing a computer to execute processing of each unit of the process management system 1 may also be provided.

[0231] The embodiment described above shows an information processing method using the relationship model information, but information associated when creating the relationship model information is not limited thereto. That is, the relationship model information used in the present embodiment may be associated with various other conditions such as a weather condition, a regional condition, and a condition related to the age of the facility.

[0232] The above-described embodiments of the present disclosure are examples of the present disclosure, and various configurations other than the above-described configurations may be adopted. The present disclosure is not limited to the embodiments described above, and the present disclosure also encompasses modifications, improvements, etc. within the scope of the present disclosure.

[0233] 1: Process management system 2: Process management server 3: User terminal 4: Measurement device 20: Communication bus 21: Communication unit 22: Storage unit 23: Controller 30: Communication bus 31: Communication unit 32: Storage unit 33: Controller 34: Output unit 35: Input unit W: Water system process

Claims

1. An information processing system, comprising a processor configured to execute a program so as to: acquire and accumulate normal information generated when a process is in a normal state; acquire and accumulate abnormal information generated when the process is in an abnormal state, the abnormal information including a countermeasure for the abnormal state and a result of the countermeasure; generate correlation information indicating a correlation between the accumulated normal information and the accumulated abnormal information; estimate a possibility of occurrence of the abnormal state based on the generated correlation information and the acquired normal information; specify an effective countermeasure for the abnormal state based on the accumulated abnormal information when the estimated possibility of occurrence is equal to or greater than a first level; and output countermeasure information indicating the specified effective countermeasure.

2. The information processing system according to claim 1, wherein: the countermeasure information includes a procedure for an operation to resolve the abnormal state, an amount of the operation, a type of substance effective in resolving the abnormal state, or an amount of the substance.

3. The information processing system according to claim 1 or 2, wherein: the processor is further configured to execute the program so as to: output the countermeasure information in a mode corresponding to effectiveness of the countermeasure indicated by the countermeasure information.

4. The information processing system according to any one of claims 1 to 3, wherein: the processor is further configured to execute the program so as to: repeatedly generate the correlation information at a predetermined generation interval.

5. The information processing system according to claim 4, wherein: the processor is further configured to execute the program so as to: shorten the generation interval for a certain period when there is a change in the process.

6. The information processing system according to claim 4 or 5, wherein: the processor is further configured to execute the program so as to: shorten the generation interval as an operating rate of the process increases.

7. The information processing system according to any one of claims 4 to 6, wherein: the processor is further configured to execute the program so as to: shorten the generation interval as a scale of the process is larger, a step of the process is more complex, or required accuracy is higher.

8. The information processing system according to any one of claims 1 to 7, wherein: the processor is further configured to execute the program so as to: repeatedly estimate a possibility of occurrence of the abnormal state at a predetermined estimation interval, and shorten the estimation interval as the possibility of occurrence increases.

9. The information processing system according to any one of claims 1 to 8, wherein: the processor is further configured to execute the program so as to: acquire the normal information at a predetermined acquisition interval, and shorten the acquisition interval for a certain period when there is a change in the process.

10. The information processing system according to any one of claims 1 to 9, wherein: the processor is further configured to execute the program so as to: output preventive information indicating a countermeasure that prevents the abnormal state based on the accumulated abnormal information when the estimated possibility of occurrence becomes equal to or greater than a second level that is lower than the first level.

11. An information processing method comprising: steps of executing operations performed by the information processing system according to any one of claims 1 to

10. 12. A program configured to allow a computer to execute each operation performed by the information processing system according to any one of claims 1 to

10.