Processing system and program

Abstract

We provide a processing system that can properly handle product replacement. [Solution] The processing system 10 includes a CPU 10a as at least one processing unit configured to perform processing related to product replacement. The CPU 10a is configured to acquire the location of the product, acquire environmental data indicating the environment of the location, input the environmental data into an artificial intelligence model and acquire time-trace data of the load the product will receive at the location, which is output by the artificial intelligence model as a future estimation result, and generate a product replacement plan based on the time-trace data.

Description

Technical Field

[0004] ,

[0006] , , ,

[0005] , , , ,

[0001] The disclosure according to this specification relates to a technology for realizing a circular economy.

Background Art

[0002] Patent Document 1 discloses a recycling operation system for a machine as a product. This system calculates the residual value of the current machine and determines to sell it as a used machine if there is a residual value. Also, even if the residual value is low, if it is determined that the parts of the machine are applicable to the regeneration of another machine and the profitability is achievable including the regeneration cost, the system determines to regenerate the other machine.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] As described above, in the technology of Patent Document 1, the current state of the product is judged. However, even if an attempt is made to recover the product such as selling it after obtaining the judgment of the current state, there may be cases where the recovery destination cannot be found immediately or the delivery of the product after replacement cannot be carried out immediately. Thus, since it is difficult to make a planned response, there is a concern that the replacement of the product cannot be appropriately realized.

[0005] One of the purposes according to the disclosure of this specification is to provide a processing system and a program capable of appropriately realizing the replacement of a product.

Means for Solving the Problems

[0006] One aspect disclosed herein is a processing system comprising at least one processing unit (10a) configured to perform processing relating to the replacement of products (90, 90a, 90b, 90c, 491, 492, 493), At least one processing unit, Obtaining the location of the product, Obtaining environmental data that indicates the environment of the location, This involves inputting environmental data into an artificial intelligence model and obtaining time-series data on the load a product receives at its location, which is then output by the AI ​​model as a future prediction. It is configured to generate a product replacement plan based on time-series data and to perform the following actions.

[0007] Another aspect of the disclosed aspect is a program used for processing the exchange of products (90, 90a, 90b, 90c, 491, 492, 493), At least one processor (10a) Obtaining the location of the product, Obtaining environmental data that indicates the environment of the location, This involves inputting environmental data into an artificial intelligence model and obtaining time-series data on the load a product receives at its location, which is then output by the AI ​​model as a future prediction. The system generates a product replacement plan based on time-series data and then executes the process.

[0008] According to these embodiments, the artificial intelligence model, upon inputting environmental data, outputs time-series data of the load a product experiences at its location. This time-series data accurately reflects the product load specific to its location by utilizing environmental data. Based on this time-series data, which represents future estimations, a product replacement plan is generated, enabling planned product replacement. Therefore, product replacement can be appropriately implemented.

[0009] The symbols in parentheses included in the claims, etc., are illustrative examples illustrating the correspondence with the embodiments described later, and are not intended to limit the technical scope. [Brief explanation of the drawing]

[0010] [Figure 1] A diagram showing examples of product locations. [Figure 2] A diagram showing the general configuration of the processing system. [Figure 3] A diagram illustrating the functional architecture of the processing system. [Figure 4] A diagram showing an example of time-series data illustrating the load a product experiences. [Figure 5] A flowchart illustrating an example of a processing method used by a processing system. [Figure 6] A diagram illustrating the functional architecture of the processing system. [Figure 7] A diagram showing the general configuration of the processing system. [Figure 8] A flowchart illustrating an example of a processing method used by a processing system. [Figure 9] A diagram showing the general configuration of the processing system. [Modes for carrying out the invention]

[0011] In this disclosure or claims, the term "processor" means one or more hardware processors configured to execute processing defined by computer program code (i.e., one or more instructions of a computer program) contained in a computer program by reading the code each time. In other words, a "processor" is a hardware device that executes one or more programmed processes. Therefore, computer program code can also be considered software that can define the processing of the processor according to its content. For example, a "processor" may be a general-purpose or specific-purpose processor and may be, but is not limited to, a CPU, microprocessor, GPU, and DFP (Data Flow Processor).

[0012] In this disclosure or claims, the term “memory” means one or more hardware memories that are non-transitional tangible recording media configured to record computer program code and / or data in a manner accessible from a processor. “Memory” can be implemented by memory technology such as SRAM, SDRAM, non-volatile flash memory, or other types of memory. The computer program code that constitutes the program is recorded in memory and executed by a processor, thereby enabling the processor to perform the various functions described above.

[0013] In this disclosure or the claims, the term "circuit" refers to a single or multiple logical circuits as hardware, and is configured to perform specific processing defined based on a pre-designed circuit configuration. In other words (and in contrast to a "processor"), the "circuit" in this disclosure or the claims does not refer to something whose processing is defined by software such as the above computer program code, but rather refers to a hardware device that executes specific processing based on a circuit configuration. For example, "circuit" may include custom ICs such as ASICs (Application Specific Integrated Circuits) and FPGAs (Field Programmable Gate Arrays) designed by a hardware description language (HDL: Hardware Description Language). That is, the "circuit" in this disclosure or the claims includes all hardware circuits except the above processor that executes processing by loading computer program code.

[0014] In this disclosure or the claims, the expression "at least one circuit and processor" should be interpreted as disjunctive (logical OR), and should not be interpreted as at least one circuit and at least one processor.

[0015] In this disclosure or the claims, the term "processing unit" means a hardware device that executes processing by a "processor", a "circuit", or a combination thereof. When its function is interpreted as being impossible to be realized by a "circuit" and possible to be realized by a "processor", it may mean the "processor" itself.

[0016] Hereinafter, a plurality of embodiments will be described based on the drawings. In each embodiment, corresponding components may be denoted by the same reference numerals, and redundant descriptions may be omitted. When only a part of the configuration is described in each embodiment, the configuration of other embodiments described previously can be applied to other parts of the said configuration. Also, not only the combinations of configurations explicitly shown in the description of each embodiment, but also the configurations of a plurality of embodiments can be partially combined with each other as long as there is no problem with the combination, even if not explicitly shown.

[0017] (First Embodiment) The processing system 10 of the first embodiment is a system used to realize a circular economy. The processing system 10 is configured to be able to execute processing related to the replacement of the product 90. Here, the product 90 is a product existing at a predetermined location. The product 90 has a lifespan due to deterioration and is a product for which it is assumed that the product itself or a part of it will be replaced.

[0018] As shown in FIG. 1, the product 90 exists at a predetermined location on the earth. The product 90 may be, for example, infrastructure equipment such as power generation equipment, power supply equipment, communication equipment, and signal equipment, or may be parts such as inverters, motors, and antennas provided in these equipment. When the product 90 is fixedly installed, the location may be referred to as the installation location.

[0019] Here, the replacement of the product may refer to the replacement of the entire equipment when the product is equipment, or the replacement of the parts included in the equipment. The replacement of the parts included in the equipment may correspond to so-called equipment repair or maintenance. Also, the replacement of the product may refer to the replacement of the entire part when the product is a part of the equipment, or the replacement of the sub-parts constituting the part (target product). The replacement of the sub-parts constituting the target product may correspond to so-called repair or maintenance of the target product.

[0020] As shown in Figure 2, the processing system 10 is configured primarily around, for example, at least one computer. The computer comprising the processing system 10 has at least one CPU (Central Processing Unit) 10a and at least one memory 10b. The memory 10b may be a non-transitory tangible storage medium that non-temporarily stores computer programs, data, artificial intelligence models, etc., that can be read by the CPU 10a. Furthermore, the computer may have a rewritable volatile storage medium such as RAM (Random Access Memory) 10c. The CPU 10a can execute various processes according to the computer programs stored in the memory 10b.

[0021] Furthermore, the computer includes an interface 10d for exchanging data with the outside world. Interface 10d may include a communication interface for connecting the computer to the internet or external devices. Interface 10d may also include an operation interface such as a keyboard or mouse for accepting human input.

[0022] The computer may also include a display 10e. The display 10e may be a liquid crystal display or an OLED display that displays the results of processing by the computer as an image. The display 10e may also be a projection-type display such as a projector.

[0023] As shown in Figure 3, the processing system 10 includes an information gathering unit P1, a life estimation unit P2, and a planning unit P3 as functional units whose functions are realized when the CPU 10a executes a computer program stored in memory 10b.

[0024] The information gathering unit P1 is configured to acquire various types of information used for planning the replacement of product 90. This information may include environmental data representing the external environment of product 90. The environmental data may include at least one of the following: weather data, space data, and topographic data.

[0025] Meteorological data includes at least one of the following for the location or the area containing the location: solar radiation, wind direction, wind speed, rainfall, and snowfall. Meteorological data may include at least one of the following: historical weather data, current weather data, and future weather data. Future weather data may be so-called weather forecasts. Meteorological data may be data provided by public meteorological agencies (e.g., meteorological observatories, weather stations) or private meteorological agencies (e.g., weather forecasting companies).

[0026] Space data includes at least one of the following: space weather maps, information on solar activity, and information on asteroids and meteorites approaching Earth. Space data may also include, for example, data on the amount of radiation reaching Earth. Space data may include at least one of the following: past space data, current space data, and future space data. Future space data may be so-called space weather forecasts. Space data may be data provided by research institutions or universities.

[0027] Topographic data is data that includes at least one of the following: map information and survey information, for the surrounding area including the location. Topographic data may include, for example, data on the elevation and elevation differences of the surrounding area, data on seas or rivers in the surrounding area, and data on fields, rice paddies, factories in the surrounding area.

[0028] As shown in Figure 2, the information gathering unit P1 may acquire environmental data from a dedicated data provision center 20 that is connected via communication, or from a cloud server CS on the internet.

[0029] The data provision center 20 is a system primarily composed of at least one computer. The computer comprising the data provision center 20 has at least one CPU 20a and at least one memory 20b. The memory 20b may be a non-transitional tangible storage medium that non-temporarily stores computer programs, data, artificial intelligence models, etc., that can be read by the CPU 20a. Furthermore, the computer may have a rewritable volatile storage medium such as RAM (Random Access Memory) 20c. The CPU 20a can execute various processes according to the computer programs stored in the memory 20b.

[0030] Furthermore, the computer includes an interface 20d for exchanging data with the outside world. Interface 20d may include a communication interface for connecting the computer to the internet or external devices. Interface 20d may also include an operation interface such as a keyboard or mouse for accepting human input.

[0031] The data provision center 20 also includes databases (hereinafter referred to as DBs) 20e, 20f, and 20g for storing environmental data to be provided to the processing system 10. DBs 20e, 20f, and 20g are configured to include at least one type of non-transitional physical storage medium, such as semiconductor memory, magnetic media, and optical media. Multiple DBs 20e, 20f, and 20g may be provided depending on the type of data. For example, the data provision center 20 may include a meteorological DB 20e for storing weather data, a space DB 20f for storing space data, and a topographic DB 20g for storing topographic data. The data provision center 20 sequentially updates the environmental data stored in each DB 20e, 20f, and 20g so that the latest data is included.

[0032] Furthermore, the data provision center 20 can provide a portion of the environmental data stored in each DB 20e, 20f, and 20g to the processing system 10 (for example, the information collection unit P1) or other systems, based on the request.

[0033] The cloud server CS refers to a server on a network realized through cloud computing, and may be realized by multiple processing units located in remote locations that are far apart from each other. The information gathering unit P1 may, for example, use a web search engine to search for information on the internet and obtain environmental data from the cloud server CS, which is the source of the searched information.

[0034] Furthermore, the information gathering unit P1 identifies the location of the product 90. The location may be identified before or after the acquisition of environmental data. If the location is identified before the acquisition of environmental data, the information gathering unit P1 can efficiently use the hardware resources of the processing system 10 by restricting the acquisition of weather and topographic data for locations far from the location. If the location is identified after the acquisition of environmental data, the information gathering unit P1 can quickly handle the replacement of multiple products 90 located in different locations by acquiring weather and topographic data for all locations in advance.

[0035] To locate the product 90, the processing system 10 may be connected to the product 90 via the Internet or other means to enable communication. The information gathering unit P1 may then obtain information about the product 90 from the product 90.

[0036] In this way, the various information collected by the information collection unit P1, including the location and environmental data of product 90, is provided to the artificial intelligence I1. More specifically, the information collection unit P1 converts the data acquired from the data provision center 20, the cloud server CS on the internet, etc., into a format suitable for input to the artificial intelligence I1, and then outputs it to the artificial intelligence I1.

[0037] Artificial intelligence I1 has, for example, multiple factor-specific models M1, M2, M3 and an integrated model M4. In Figure 3, three factor-specific models M1, M2, and M3 are shown as an example, but there may be two or more. The number of factor-specific models provided corresponds to the number of degradation factors (parameters) that degrade product 90 and are used to estimate the remaining lifespan of product 90. That is, each factor-specific model M1, M2, and M3 corresponds individually to different degradation factors, and each model handles the degradation factor it corresponds to individually.

[0038] Factor-specific models M1, M2, and M3 are artificial intelligence models that, when various information including environmental data is input, output time-series data showing the load a product experiences at its location. In other words, factor-specific models M1, M2, and M3 are estimated using artificial intelligence learning. Factor-specific models M1, M2, and M3 are mainly constructed using neural networks. A neural network is a mathematical model that mimics the nerve cells of the human brain. The neural network may be a deep neural network used in deep learning, a convolutional neural network, or a recurrent neural network.

[0039] Each factor-specific model M1, M2, and M3 outputs time-course data showing the load on the product due to the corresponding factor, when environmental data or other information is input. As shown in Figure 4, the time-course data may be, for example, a load curve that expresses the load as a function of time. In this case, the load curve may be represented by a polynomial, and factor-specific models M1, M2, and M3 may be configured to output the coefficients of the polynomial. Alternatively, factor-specific models M1, M2, and M3 may output the values ​​at each point on the load curve as a pair of (time value, load value), and the output data may be configured so that the load curve can be reproduced using spline interpolation or the like. Furthermore, factor-specific models M1, M2, and M3 may be configured to output a graph with the load curve drawn on it as image data, as shown in Figure 4.

[0040] The time progression data shows a period that includes the future replacement timing of product 90. In subsequent processing, it is necessary to estimate the load accumulated up to the replacement timing in order to formulate a replacement plan. In other words, the time progression period includes a period corresponding to the future at the time of output. If the current state of product 90 is unknown, it is also necessary to estimate the load that product 90 has received up to the present, so the time progression period also includes a period corresponding to the past. Conversely, if the current deterioration state of product 90 is clear, for example, by measuring the state of product 90, the time progression period may include only a period corresponding to the future.

[0041] For example, if the average lifespan of product 90 is about one year, and product 90 has just been replaced, it is preferable that the period showing the time progression be the period corresponding to the average lifespan plus a buffer period (for example, one and a half years). If the period showing the time progression is long, the output time progression data may be represented by multiple load curves for each divided period obtained by dividing the data into intervals.

[0042] The input environmental data preferably includes meteorological data. However, the accuracy of meteorological data provided by meteorological agencies (e.g., the accuracy of weather forecasts) is evaluated based on the average value. Therefore, if there are other factors that cause deviations from the average value, it may be difficult to improve the estimation accuracy when estimating the load on product 90 using only the meteorological data. For this reason, it is even more preferable that the input environmental data includes space data and topographic data in addition to meteorological data. By comprehensively reflecting other factors that cause deviations from the average value, the accuracy of load estimation can be improved.

[0043] For example, as shown in Figure 1, if there is a sea area 91 around product 90, salt carried by the wind from the sea area 91 may reach product 90 and degrade it. In this example, the factor corresponding to factor-specific model M1 is "salt damage". Based on the relationship between the environmental data, including topographic data with the location of the sea area 91 and meteorological data including wind direction and wind speed around the installation site, and the factor "salt damage", factor-specific model M1 can output the time progression of the load on product 90 due to the factor "salt damage" as a load curve.

[0044] Furthermore, as shown in Figure 1, for example, if a field 92 exists near product 90, pesticides used in the field 92 may be carried by the wind to product 90, potentially degrading it. In this example, the factor corresponding to factor-specific model M2 is "pesticide damage." Based on the relationship between the environmental data, including topographic data such as the location of field 92 and meteorological data including wind direction and wind speed around the installation site, and the factor "pesticide damage," factor-specific model M2 can output a load curve showing the time progression of the load on product 90 due to the factor "pesticide damage."

[0045] Furthermore, radiation exposure to product 90 can also degrade it. In the example corresponding to this, the factor corresponding to factor-specific model M3 is "radiation." By examining the relationship between cosmic data, including radiation levels from solar activity, etc., and meteorological data, the time course of the load on product 90 due to the factor "radiation" can be output as a load curve.

[0046] The integrated model M4, upon receiving multiple time-transition data for each factor output from individual factor models M1, M2, and M3, integrates this time-transition data and outputs the integrated time-transition data. The integrated model M4 may be implemented by a computer program or by a neural network.

[0047] The integrated model M4 may calculate the integrated load curve by simply adding up each load curve. If there is a nonlinear correlation between the load of one factor and the load of another factor, or if such a correlation is possible, the integrated model M4 may estimate the integrated load curve by learning, such as through deep learning.

[0048] The life estimation unit P2 estimates the remaining life of product 90 based on the integrated load curve. Specifically, if the relationship between the load and remaining life of product 90 is already formulated or held as data, the life estimation unit P2 estimates the remaining life based on that formula or data. For example, the life estimation unit P2 compares the cumulative load calculated based on time-evolution data with the upper limit of the cumulative load that is permissible to maintain product 90 in a state where it exhibits the required functions and performance. The life estimation unit P2 then acquires the timing at which the cumulative load based on time-evolution data reaches the permissible upper limit. This timing becomes the timing at which the product reaches the end of its life (hereinafter referred to as the life timing), and the remaining life is from the present until the timing at which the product reaches the end of its life.

[0049] On the other hand, if such a formula or data is not available, the life estimation unit P2 may estimate the remaining life by comparing information on the specifications and average life of the product 90 with time-evolved data.

[0050] The planning unit P3 formulates a replacement plan for product 90 based on the remaining life estimated by the life estimation unit P2. Formulating the replacement plan may include determining the timing of product 90 replacement. The replacement timing may include a recommended replacement timing. The replacement timing may be a predetermined period before the life timing. The predetermined period may be a buffer time that takes into account, for example, the possibility of replacement being delayed due to trouble.

[0051] Planning Department P3 may determine the replacement timing based on weather data, choosing a day with a low probability of precipitation (a sunny day). For example, if product 90 is a large piece of equipment installed outdoors, replacing it on a rainy day would require rain countermeasures, potentially increasing the cost and time required for the replacement. By scheduling the replacement on a day with a low probability of precipitation, such increases in cost and time can be mitigated.

[0052] The planning of replacements may include proposing an extension of the lifespan of product 90. For example, the planning unit P3 may acquire environmental data or time-dependent data of load by factor, identify periods when product 90 is subjected to high temperatures due to solar radiation, etc., and propose cooling product 90 during those periods. If cooling equipment (e.g., a water-cooled radiator, an air conditioner) is available for use with product 90 and is communicatively connected to the cooling equipment, the processing system 10 may output a request signal to the cooling equipment to enhance its cooling function during the periods when it is subjected to the load.

[0053] The development of a replacement plan may include selecting a destination (or recovery destination) for product 90. For example, if product 90 is used in critical equipment used as infrastructure, it may be operated under high safety standards, but it may be possible to repurpose it for equipment that can be operated under lower safety standards than those required for infrastructure (e.g., factory equipment). For this reason, the planning department P3 obtains information via the internet or the data provision center 20 to obtain information on potential repurposing destinations. The planning department P3 then selects a candidate repurposing destination.

[0054] The planning unit P3 then generates display content to show the planned replacement plan. The planning unit P3 displays the generated display content on the display 10e of the processing system 10. By confirming the replacement plan displayed on the display 10e, the user can schedule the replacement and begin arranging for the replacement product and replacement company, negotiating with the recipient, etc.

[0055] Next, an example of how the processing system 10 plans product replacement will be explained using the flowchart in Figure 5. The series of processes in steps S1 to S6 of this flowchart may be realized by the CPU 10a executing a computer program stored in memory 10c.

[0056] In S1, the information gathering unit P1 acquires information regarding the installation location of product 90. In S2, following the processing in S1, the information gathering unit P1 acquires weather data, space data, and topographic data.

[0057] In S3, following the processing in S2, the processing system 10 uses artificial intelligence I1 to learn and estimate the load curve. In S4, following the processing in S3, the life estimation unit P2 estimates the remaining life of product 90 from the load curve.

[0058] In S5, following the processing in S4, the planning unit P3 formulates a replacement plan for product 90. Specifically, the planning unit P3 selects a repurposing destination and calculates the replacement timing. In S6, following the processing in S5, the planning unit P3 displays the plan on the display 10e. The series of processes ends with S6.

[0059] Furthermore, this series of processes may be performed periodically at predetermined intervals (for example, every 30 minutes) to update the remaining life estimate and replacement plan. This would allow for a flexible response to the latest situation.

[0060] According to the first embodiment described above, when environmental data is input, the artificial intelligence model outputs time-course data of the load that product 90 receives at its location. In other words, this time-course data accurately reflects the product load specific to the location of product 90 by using environmental data. Based on this time-course data, which is an estimated result for the future, a replacement plan for product 90 is generated, making it possible to replace the product in a planned manner. Therefore, the replacement of product 90 can be appropriately implemented.

[0061] Furthermore, according to the first embodiment, the environmental data includes meteorological data of the location. By using meteorological data, the product load specific to the location can be reflected in future estimations with high accuracy. Furthermore, according to the first embodiment, the environmental data includes space data indicating the space environment. By using space data, the accuracy of product load estimation can be further improved. Furthermore, according to the first embodiment, the environmental data includes topographic data of the location. By using topographic data, the product load specific to the location can be reflected in future estimations with high accuracy.

[0062] Furthermore, according to the first embodiment, multiple factor-specific models M1, M2, and M3 are provided for each factor of the load, serving as artificial intelligence models. Environmental data is input to each factor-specific model M1, M2, and M3, and time-course data for the corresponding factor is output individually from each factor-specific model M1, M2, and M3. In addition, the time-course data for each factor is integrated to obtain time-course data that integrates the effects of multiple factors. By configuring the system to output factor-specific time-course data using models specialized for predetermined factors, accuracy can be improved compared to estimating the load without distinguishing between factors.

[0063] (Second Embodiment) As shown in Figure 6, the second embodiment is a modified version of the first embodiment. The second embodiment will be described focusing on the differences from the first embodiment.

[0064] In the processing system 210 of the second embodiment, the artificial intelligence I2 has a plurality of load estimation models M21, M22, and M23 that output load curves. These load estimation models M21, M22, and M23 are models that assume that the load modes are different from each other. The load mode referred to here is a mode that indicates the tendency of the load that the product is subjected to. The load estimation models M21, M22, and M23 may be factor-based models similar to those of the first embodiment. On the other hand, unlike the first embodiment, the load estimation models M21, M22, and M23 may output load curves that take into account a combination of various degradation factors, and are equivalent to the integrated load curve of the first embodiment.

[0065] In the second embodiment, the artificial intelligence I2 has a mode selection model M24 instead of the integrated model M4 of the first embodiment. The mode selection model M24 can select the model to be adopted from among a plurality of load estimation models M21, M22, and M23 using a linear regression model.

[0066] For example, the mode selection model M24 acquires time-lapse data (hereinafter referred to as load measurement data) of loads previously received by products similar to product 90 or products located in the same place, in order to select a model. This load measurement data can be acquired by the information collection unit P1 from the data provision center 20 or other databases, etc., and provided to the mode selection model M24.

[0067] The mode selection model M24 uses the measured load data as the explanatory variable and the load curve output from one of the multiple load estimation models M21, M22, and M23 as the dependent variable, and calculates the sum of squared errors. This process is performed for all load estimation models M21, M22, and M23. Then, the mode selection model M24 decides to adopt the load estimation model that outputs the load curve with the smallest sum of squared errors among the load curves output from each load estimation model M21, M22, and M23.

[0068] Furthermore, when re-running the load estimation for product 90 after mode selection, only the selected load estimation model from among the multiple load estimation models M21, M22, and M23 is used; the other load estimation models do not need to be used. In other words, the time-evolution data (load curve) output by the selected load estimation model is used directly for estimating the remaining life.

[0069] Here, the period for which the load estimation model determined by the mode selection model M24 is adopted may be predetermined. When the adoption period ends, the model selection process using the mode selection model M24 is executed again. In this way, the load mode is reviewed periodically, and it is possible to respond when changes in the load mode (changes in load trends) occur.

[0070] According to the second embodiment described above, multiple load estimation models M21, M22, and M23 are provided as artificial intelligence models, and each of them assumes a different load mode, which is a mode that indicates the tendency of load occurrences on product 90. Environmental data is input to each load estimation model M21, M22, and M23, and time-transition data for the corresponding load mode is output individually from each load estimation model M21, M22, and M23. Furthermore, multiple time-transition data with different load mode assumptions are compared with past measured load data, and one time-transition data to be adopted is selected from the multiple time-transition data with different load mode assumptions. By using this selection method, even if the tendency of load occurrence changes due to environmental changes at the location where the product is located, it is possible to estimate the load with high accuracy and generate a replacement plan for product 90.

[0071] (Third embodiment) As shown in Figures 7 and 8, the third embodiment is a modification of the first embodiment. The third embodiment will be described focusing on the differences from the first embodiment.

[0072] The processing system 310 of the third embodiment is connected to multiple customer terminals 30 via the Internet. The processing system 310 can provide multiple customer terminals with replacement plans for products 90 managed by the customers. Since the services provided to each customer are substantially the same, communication with one representative customer terminal 30 will be described below.

[0073] The processing system 310 has a management database (hereinafter referred to as the management DB) 10g. The management DB 10g is configured to include at least one type of non-transitional physical storage medium, such as semiconductor memory, magnetic media, and optical media. The management DB 10g stores data about customers to whom the processing system 310 provides exchange plans. The data about customers may include the IP address of the customer-side terminal 30 for communication with the customer, information about products 90a, 90b, and 90c managed by the customer, and records of bids made by the customer. The data about customers may also include information on whether or not the customer permits automated bidding.

[0074] Furthermore, management DB10g stores data about contractors who perform the replacement of product 90. If the contractor is a seller of the product, the data about the contractor may be the latest data on the unit price and delivery date of the product sold by that contractor. If the contractor is a contractor that performs the replacement work, the data about the contractor may be the latest data on the contractor's labor costs and availability of manpower. In addition, the data about the contractor may include records of orders placed with the contractor.

[0075] The processing system 310 should be connected to each vendor's terminal 80a and 80b in a communication-enabled manner. This allows the processing system 310 to obtain the latest data about the vendors from the vendor terminals 80a and 80b. In addition, the processing system 310 can execute order processing to the vendors.

[0076] The customer-side terminal 30 is a system for managing products 90a, 90b, and 90c. The customer-side terminal 30 is primarily composed of, for example, at least one computer. The computer comprising the customer-side terminal 30 has at least one CPU 30a and at least one memory 30b. The memory 30b may be a non-transitional, tangible storage medium that non-temporarily stores computer programs, data, artificial intelligence models, etc., that can be read by the CPU 30a. Furthermore, the computer may have a rewritable, volatile storage medium such as RAM 30c. The CPU 30a can execute various processes according to the computer programs stored in the memory 30b.

[0077] Furthermore, the computer includes an interface 30d for exchanging data with the outside world. Interface 30d may include a communication interface for connecting the computer to the internet or external devices. Interface 30d may also include an operation interface such as a keyboard or mouse for accepting human input.

[0078] The computer may also include a display 30e. The display 30e may be a liquid crystal display or an OLED display that displays the results of computer processing as an image. The display 30e may also be a projection-type display such as a projector.

[0079] Furthermore, the customer terminal 30 may be equipped with a management database 30g. The management database 30g stores data on the products 90a, 90b, and 90c managed by the customer. The data on the managed products includes information on the number, status, specifications, location, purchase date, and contracts of the managed products 90a, 90b, and 90c. The contract information includes whether or not the customer has permitted automated bidding, i.e., information on the automated bidding permission settings. The customer terminal 30 should be able to obtain the latest data on each product 90a, 90b, and 90c via communication or other means.

[0080] The customer terminal 30 then periodically, or based on customer operations, requests the processing system 310 to formulate a replacement plan for the requested product among the managed products 90a, 90b, and 90c. The processing system 310 then uses an algorithm similar to that of the first embodiment to estimate the remaining lifespan of the requested product and formulate a replacement plan.

[0081] Specifically, the information gathering unit P1 acquires the location and environmental data of the requested product, which will be used by the artificial intelligence I1 to output time-trace data of the load. Here, the information gathering unit P1 only needs to acquire the location of the requested product from the customer terminal 30. Then, the time-trace data of the load on the requested product is estimated through learning using the artificial intelligence I1. Next, based on this time-trace data, the life estimation unit P2 estimates the remaining life of the requested product.

[0082] If the customer has permitted automated bidding, the planning unit P3, after formulating a replacement plan as in the first embodiment, transmits the replacement plan and a bidding command for the replacement of the products based on that plan to the customer-side terminal 30. However, the planning unit P3 also selects the vendor that will sell the replaced products and the vendor that will carry out the replacement work.

[0083] The customer terminal 30 stores the received exchange plan in the management DB 30g and transmits the acceptance information of the bid instruction to the processing system 310. The planning unit P3 stores the bid record (the record that the customer terminal 30 accepted the bid) in the management DB 10g.

[0084] Furthermore, the planning unit P3 executes arrangements based on the replacement plan. It places orders for products and orders for construction work with the contractors selected in the replacement plan, preferably both. In other words, the order information is transmitted to each contractor's terminal 80a, 80b.

[0085] On the other hand, if the customer does not permit automated bidding, Planning Department P3 will formulate multiple replacement plans with different optimization conditions. Planning Department P3 will formulate replacement plans that prioritize reducing environmental impact, replacement plans that prioritize reducing replacement costs, replacement plans that prioritize reducing the probability of trouble occurring, and balanced replacement plans, etc.

[0086] These exchange plans are displayed as a selectable list on the display 30e of the customer terminal 30, and the customer's manual bid is completed by selecting an exchange plan from among the multiple exchange plans. In other words, the customer terminal 30 transmits bid acceptance information corresponding to the exchange plan selected by the customer to the processing system 310.

[0087] Next, an example of how the processing system 310 plans product replacement will be explained using the flowchart in Figure 8. The series of processes in steps S101 to S111 of this flowchart may be realized by the CPU 10a executing a computer program stored in memory 10c.

[0088] Steps S101 to S104 are the same as steps S1 to S4 in the first embodiment. In step S105, after processing in step S104, the planning unit P3 determines whether the customer has permitted automated bidding. If yes, proceed to step S106. If no, proceed to step S109.

[0089] In S106, similar to S5 in the first embodiment, the planning unit P3 formulates a replacement plan based on the remaining lifespan of the product. Next, the planning unit P3 transmits the replacement plan and a bid command. In S107, after processing S106, the planning unit P3 receives bid acceptance information as a reply to the transmission in S106 and records it in the management DB 10g. This confirms the bid. In S108, after processing S107, the planning unit P3 transmits order information to the vendor terminals 80a and 80b based on the replacement plan. The series of processes ends with S108.

[0090] Meanwhile, in S109, the planning unit P3 formulates multiple exchange plans. In S110, after processing S109, the planning unit P3 generates content for the customer to select an exchange plan from the formulated plans and transmits it to be displayed on the display 30e of the customer terminal 30. After processing S110, when the planning unit P3 receives bid acceptance information for the exchange plan selected by the customer, in S111, the planning unit P3 transmits order information to the vendor terminals 80a and 80b based on the exchange plan. The series of processes ends with S111.

[0091] According to the third embodiment described above, the processing system 310 is connected to the customer terminal 30 as a customer-side system and is also connected to the vendor-side terminals 80a and 80b as vendor-side systems. Here, the products are products 90a, 90b, and 90c managed using the customer terminal 30. The CPU 10a of the processing system 310 obtains information regarding the permission settings for automatic bidding from the customer terminal 30. If automatic bidding is permitted, the CPU 10a sends order information to the vendor-side terminals 80a and 80b, including at least one of the following: an order for products 90a, 90b, and 90c and an order for replacement work for products 90a, 90b, and 90c, based on the generated replacement plan. Automatic bidding reduces the effort of the customer while appropriately promoting the replacement of products 90a, 90b, and 90c.

[0092] Furthermore, according to the third embodiment, the processing system 310 is communicably connected to the customer terminal 30, which is a customer-side system. Here, the products are products 90a, 90b, and 90c, which are managed using the customer terminal 30. The CPU 10a of the processing system 310 generates multiple exchange plans with different conditions for each product 90a, 90b, and 90c. The CPU 10a transmits the generated multiple exchange plans to the customer terminal 30. Subsequently, the customer terminal 30 receives the selected exchange plan from among the multiple exchange plans. In this way, by having the customer select the optimal plan from among multiple exchange plans with different conditions, it is possible to promote appropriate exchange of products 90a, 90b, and 90c that reflect the customer's situation and judgment.

[0093] Furthermore, according to the third embodiment, in a configuration in which a selection is made from multiple replacement plans, the processing system 310 is further connected to the contractor-side terminals 80a and 80b, which are contractor-side systems, in a communicative manner. The CPU 10a is further configured to transmit order information, including at least one of the following, to the contractor-side terminals 80a and 80b based on the selected replacement plan: an order for products 90a, 90b, and 90c and an order for replacement work for products 90a, 90b, and 90c. By having the processing system 310 handle the arrangements on behalf of the customer, the replacement of products 90a, 90b, and 90c can be promoted appropriately while reducing the customer's workload.

[0094] (Fourth Embodiment) As shown in Figure 9, the fourth embodiment is a modification of the first embodiment. The fourth embodiment will be described focusing on the differences from the first embodiment.

[0095] In the fourth embodiment, the product for which the processing system 410 formulates a replacement plan is not the equipment 490, but at least one of the devices 491, 492, and 493 installed in the equipment 490. For example, the product may be an inverter, motor, antenna, or other device 491, 492, or 493 installed in the equipment 490, such as a power generation facility, power supply facility, communication facility, or signaling facility.

[0096] The information gathering unit P1 of the processing system 410 collects not only environmental data but also update data for the equipment 490.

[0097] The update data is data relating to past replacements, repairs, additions, reductions, specification changes, etc., that have occurred in equipment 490. The update data may include at least one of the following: information identifying the product subject to the update, information regarding the specifications of the product, and information regarding the update work period. The information regarding the update work period may include information regarding the start and end dates of the work for the update.

[0098] The information gathering unit P1 may further collect operational data of the equipment 490. The operational data may include information regarding the past operational status of the equipment 490. For example, if multiple devices 491, 492, and 493 are used in the equipment 490, it is preferable to show the operational status of each device 491, 492, and 493. The operational status may simply indicate whether the equipment or device is operating or stopped at a particular time or period, or it may further indicate the operating mode and output of the equipment or device while it is in operation.

[0099] Then, various pieces of information, combining this environmental data, update data, and operational data, are input into the artificial intelligence I1. The artificial intelligence I1 then outputs data showing the time progression of the load on the product.

[0100] For example, suppose that equipment 490 initially had two devices 491 and 492, and at some point an additional device 493 was added. In this case, at the time of the addition, the state changed from one where two devices 491 and 492 shared the output of equipment 490 to one where three devices 491, 492, and 493 shared the output of equipment 490. In other words, at the time of the addition, the load on each device decreased to a discretely fluctuating level. When update data is input, the artificial intelligence I1 reflects this discrete fluctuation over time in the time-series data.

[0101] For example, suppose that after installation, the added device 493 continues trial operation with its output suppressed for a certain period, and then at some point the trial operation ends and the output is increased. In that case, at the time the trial operation ends, the load on devices 491 and 492 decreases in a discrete manner. The artificial intelligence I1, upon receiving operational data, reflects such discrete fluctuations over time in the time-series data.

[0102] According to the fourth embodiment described above, the devices 491, 492, and 493, as products, are used in the equipment 490. In this configuration, when the artificial intelligence model receives update data for the equipment 490 in addition to environmental data, it outputs time-series data of the load received by devices 491, 492, and 493 at their respective locations. That is, this time-series data accurately reflects the product load specific to the location of devices 491, 492, and 493, as well as the update history of the equipment 490, by using a combination of environmental data and update data for the equipment 490. Based on this time-series data, which is an estimated future result, a replacement plan for devices 491, 492, and 493 is generated, making it possible to systematically replace the products. Therefore, product replacement can be appropriately implemented.

[0103] (Other embodiments) Although several embodiments have been described above, this disclosure is not limited to those embodiments and can be applied to various embodiments and combinations without departing from the spirit of this disclosure.

[0104] In another embodiment, the product for which load estimation and replacement planning are generated may be subject to movement of its location. For example, the product may be a mobile object such as an automobile, train, aircraft, drone, robot, ship, or submarine, or it may be a component mounted on such a mobile object. Taking into account the movement of the mobile object, the location input for outputting time-series data may be input to the artificial intelligences I1 and I2 at each time interval.

[0105] In another embodiment, the CPU 10a does not need to acquire environmental data from an external data provision center 20 or the like. The processing system 10 itself may have databases 20e, 20f, and 20g, and the CPU 10a may acquire environmental data from databases located inside the processing system 10.

[0106] In another embodiment, the processing system 10 does not need to be communicatively connected to the product 90, and may acquire data related to the product 90 from a data provision center 20 or the like. Data related to the product 90 may also be acquired through user input operations to the processing system 10.

[0107] In another embodiment, artificial intelligences I1 and I2 may be composed of a single artificial intelligence model.

[0108] In another embodiment, the time-elapsed data may be represented by discrete data rather than a function or curve. For example, the time-elapsed data may consist of pairs of data represented as (numerical value of time, numerical value of load).

[0109] In another embodiment, the planning of the replacement may include selecting a company to replace product 90.

[0110] As another embodiment relating to the second embodiment, the mode selection model M24 may use a logistic regression model instead of a linear regression model.

[0111] In another embodiment, a processor other than a CPU, such as a GPU, may be used for the artificial intelligence model.

[0112] In other embodiments, at least one of the computers constituting the processing systems 10, 310, and 410, the computer constituting the data provision center 20, the computer constituting the customer-side terminal 30, and the computers constituting the vendor-side terminals 80a and 80b may be configured to implement functions such as load estimation and exchange planning using circuits such as FPGAs together with a processor. [Explanation of Symbols]

[0113] 10: Processing system, 10a: CPU, 90, 90a, 90b, 90c: Products, 491, 492, 493: Equipment (products)

Claims

1. A processing system comprising at least one processing unit (10a) configured to perform processing related to the replacement of products (90, 90a, 90b, 90c, 491, 492, 493), The at least one processing unit is, To obtain the location of the aforementioned product, To obtain environmental data indicating the environment of the aforementioned location, The environmental data is input into an artificial intelligence model, and the time-series data of the load the product receives at the location is output by the artificial intelligence model as a future estimation result. A processing system configured to generate a replacement plan for the product based on the aforementioned time-series data.

2. The processing system according to claim 1, wherein the environmental data includes weather data of the location.

3. The processing system according to claim 1, wherein the environmental data includes space data indicating the space environment.

4. The processing system according to claim 1, wherein the environmental data includes topographic data of the location.

5. Multiple artificial intelligence models (M1, M2, M3) are provided. Obtaining the aforementioned time-series data means The environmental data is input into a plurality of artificial intelligence models, each set up for each load factor, and the time-series data for each factor is obtained from each artificial intelligence model. The processing system according to claim 1, comprising integrating the time-series data for each of the factors and obtaining time-series data that integrates the effects of multiple factors.

6. Multiple artificial intelligence models (M21, M22, M23) are provided. Obtaining the aforementioned time-series data means The environmental data is input into multiple artificial intelligence models, each of which assumes different load modes, which are modes that indicate the tendency of loads to occur on the product, and the time-series data for each load mode is obtained from each of the artificial intelligence models. The processing system according to claim 1, comprising: comparing a plurality of time-transition data sets with different assumed load modes with past load measurement data; and selecting one of the plurality of time-transition data sets with different assumed load modes to be adopted.

7. It is connected to the customer's system (30) in a way that allows communication, and is also connected to the vendor's systems (80a, 80b) in a way that allows communication, The aforementioned product is a product managed using the aforementioned customer-side system, The at least one processing unit is, Obtaining information regarding the permission settings for automated bidding from the customer's system, The processing system according to claim 1, further configured to transmit, when the automated bidding is permitted, order information, including at least one of the following, an order for the product and an order for the replacement work of the product, to the contractor's system based on the generated replacement plan.

8. The customer-side system (30) is connected in a way that allows it to communicate with the customer's system. The aforementioned product is a product managed using the aforementioned customer-side system, Generating a replacement plan for the aforementioned product includes generating multiple replacement plans with different conditions, The at least one processing unit is, Sending the generated multiple exchange plans to the customer-side system, The processing system according to claim 1, further configured to receive and execute the exchange plan selected from the plurality of exchange plans from the customer-side system.

9. It is connected to the vendor's system (80a, 80b) in a way that enables communication. The processing system according to claim 8, further configured to transmit order information, including at least one of the following: an order for the product and an order for replacement work for the product, to the contractor's system, based on the selected replacement plan.

10. The aforementioned product is used in equipment (490), The at least one processing unit is, It is configured to further perform the acquisition of update data for the aforementioned equipment, The processing system according to claim 1, wherein acquiring the aforementioned time-series data includes, in addition to the environmental data, inputting the update data into an artificial intelligence model and acquiring time-series data of the load the product receives at the location, which is output by the artificial intelligence model as a future estimation result.

11. A program used for processing the exchange of products (90, 90a, 90b, 90c, 491, 492, 493), At least one processor (10a) To obtain the location of the aforementioned product, To obtain environmental data indicating the environment of the aforementioned location, The environmental data is input into an artificial intelligence model, and the time-series data of the load the product receives at the location is output by the artificial intelligence model as a future estimation result. A program that generates a replacement plan for the product based on the aforementioned time-series data and performs the following actions.

Images