Circular production control device, circular production control method, and program

Abstract

The circular production control device (1) includes an input unit (11) that accepts input of predictive data indicating constraints on the manufacture, sale, use, or disposal of products and user decision trends regarding reuse; a prediction unit (12) that uses the predictive data to predict the amount of recycled products to be supplied from products; and an output unit (13) that outputs the prediction result of the amount of recycled products to be supplied.

Description

Technical Field

[0001] The present disclosure relates to a cyclic production control device, a cyclic production control method, and a program.

Background Art

[0002] Cyclic production is a production method that efficiently uses resources or energy and suppresses the generation of waste in the manufacture, use, or disposal of products. In order to suppress the generation of waste, it is necessary to control the supply of reusable products that can be recovered from waste to match the actual situation. For example, Patent Document 1 describes an environmental impact assessment method that can accurately predict and evaluate the environmental impact or generation cost generated from a product group of a multi-generation product series and perform reuse and recycling. The manufacture of a multi-generation product series means recovering used products that have become unnecessary at the user's hand and reusing their components and the like.

[0003] In the method described in Patent Document 1, life cycle modeling is performed to define what of the previous generation products that are used and recovered is diverted to which product of the next generation products to be produced. After completing the life cycle modeling, an approximate model is used to predict the supply amount of reusable products for the reuse of parts or the recycling of materials. Patent Document 1 describes the following approximate models (1), (2), and (3).

[0004] Approximate model (1) is a model for predicting the distribution of the production quantity of products to be recovered. In approximate model (1), the time required for product recovery and the like is considered negligible, and the product recovery rate is set to an empirically obtained value. Approximate model (2) is a model for predicting the distribution of the production quantity of products determined by the convenience of the product manufacturing company and predicting the distribution of the usage period of products and the product recovery rate determined by the convenience of the product users. In approximate model (2), the product recovery period is the same as the production period, and the total number of recovered products is calculated using a fixed product recovery rate. Approximate model (3) is a model that predicts the amount of product recovered over a period shifted in time by the product's lifespan. Approximate model (3) assumes that the time required for product recovery and the product recovery rate do not fluctuate. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2003-203138 [Overview of the project] [Problems that the invention aims to solve]

[0006] In the conventional technology described in Patent Document 1, it is assumed that the product recovery rate is constant in all approximation models. Therefore, there is a problem in that it is not possible to predict the amount of recycled products to be supplied from the recovered products, taking into account fluctuations in the amount of recovered products.

[0007] This disclosure aims to solve the above-mentioned problems and to provide a circular production control device that can predict the supply volume of recycled products while taking into account fluctuations in the amount of products recovered. [Means for solving the problem]

[0008] The circular production control system relating to this disclosure is subject to restrictions on the manufacture, sale, use, or disposal of products. This includes at least two of the following: purpose of use, anomaly detection history, degree of use, performance data, usage cost, location of use, and trouble history. An input unit that accepts input of predictive data showing the user's decision-making tendencies regarding reuse, and the predictive data Based on the above, the degree of deterioration of the product is calculated, the remaining lifespan of the product is calculated based on the degree of deterioration and the product's service life data, the recovery period of the product is predicted based on the remaining lifespan, and based on the recovery period The aforementioned Obtained from the product It comprises a prediction unit that predicts the supply volume of reusable materials, and an output unit that outputs the prediction results for the supply volume of reusable materials. [Effects of the Invention]

[0009] The circular production control system described herein predicts the supply volume of recycled materials from products using predictive data that shows constraints on the manufacture, sale, use, or disposal of products, as well as user decision trends regarding reuse. This allows the circular production control system to predict the supply volume of recycled materials, taking into account fluctuations in the amount of products recovered due to the aforementioned constraints and decision trends. [Brief explanation of the drawing]

[0010] [Figure 1] This is a functional block diagram showing an example configuration of a cyclic production control device according to Embodiment 1. [Figure 2] This is a functional block diagram showing an example of the configuration of the prediction unit in Embodiment 1. [Figure 3] This is a flowchart showing the circular production control method according to Embodiment 1. [Figure 4] Figures 4A and 4B are functional block diagrams showing the hardware configuration that realizes the functions of the circular production control device according to Embodiment 1. [Figure 5] This is a functional block diagram showing an example configuration of a cyclic production control device according to Embodiment 2. [Figure 6] This is a functional block diagram showing an example of the configuration of the prediction unit in Embodiment 2. [Figure 7] This is a flowchart showing the circular production control method according to Embodiment 2. [Figure 8] This is a functional block diagram showing an example configuration of a cyclic production control device according to Embodiment 3. [Figure 9] This is a functional block diagram showing an example of the configuration of the prediction unit in Embodiment 3. [Figure 10] This is a flowchart showing the circular production control method according to Embodiment 3. [Figure 11] This is a functional block diagram showing an example configuration of a cyclic production control device according to Embodiment 4. [Figure 12] This is a functional block diagram showing an example of the configuration of the prediction unit in Embodiment 4. [Figure 13]It is a flowchart showing the cyclic production control method according to Embodiment 4. [Figure 14] It is a functional block diagram showing a configuration example of the cyclic production control device according to Embodiment 5. [Figure 15] It is a functional block diagram showing a configuration example of the prediction unit in Embodiment 5. [Figure 16] It is a flowchart showing the cyclic production control method according to Embodiment 5.

Embodiments for Carrying out the Invention

[0011] Embodiment 1. In the adjustment of the supply plan of recycled products in cyclic production, the plan is optimized by predicting the supply amount of recycled products. The cyclic production control device according to Embodiment 1 is a device that predicts the supply amount of recycled products, and the prediction result of the supply amount of recycled products is used for adjusting the supply plan of recycled products. The amount of products recovered as recycled products (recovery amount) varies depending on the constraints on the manufacture, sale, use or disposal of the products, and also varies depending on the judgment tendency of users related to recycling. The cyclic production control device according to Embodiment 1 uses prediction data including the above constraints and the above judgment tendency to predict the supply amount of recycled products from products in order to grasp such fluctuations in the recovery amount of products. Here, the case where the prediction data is "usage record data of products by users" will be described.

[0012] (Basic Configuration of Cyclic Production Control Device) FIG. 1 is a functional block diagram showing a configuration example of the cyclic production control device 1 according to Embodiment 1. The cyclic production control device 1 inputs usage record data of products by users and product durability data, predicts the supply amount of recycled products using the input data, and outputs this prediction result to an external device. The external device is a display device that displays the prediction result of the supply amount of recycled products. As shown in FIG. 1, the cyclic production control device 1 includes an input unit 11, a prediction unit 12, and an output unit 13.

[0013] (Input Unit) The input unit 11 accepts input of predictive data that shows the constraints on the manufacture, sale, use, or disposal of products, as well as the user's decision-making tendencies regarding reuse. For example, prediction data is stored in a memory device as a database. The input unit 11 retrieves prediction data from the database stored in the memory device. If the storage device is an external storage device that is connected to the circular production control device 1 via wired or wireless communication, the input unit 11 acquires the prediction data stored in the external storage device via wired or wireless communication. Alternatively, the memory device may be a memory unit provided by the circular production control device 1. In this case, the input unit 11 acquires prediction data stored in the memory unit via a signal line. Furthermore, the input unit 11 may accept prediction data input from a user using an input device not shown in Figure 1.

[0014] (Product usage data) The prediction data in Embodiment 1 is "user usage data of the product." For example, usage data includes the "start date of use," "end date of use," "usage time," "usage status," "usage environment," "maintenance history," and "data related to the target product" used by the user. "Usage time" refers to the cumulative usage time of the product. "Usage conditions" refer to the product's operating volume, load, or power consumption. "Usage environment" includes the temperature, humidity, and physical conditions such as vibration at the location where the product was used. "Maintenance history" includes data showing the date and time, details, and replacement parts of the product maintenance performed. "Data related to the target product" includes the target product's "intended use," "anomaly detection history," "degree of use," "performance data," "cost of use," "location of use," "trouble history," "usage conditions," "post-use condition," "energy efficiency," "insurance information," and "regulatory compliance status."

[0015] "Purpose of Use" is data indicating the purpose or project for which the product was used by the user. The product may be not only equipment but also facilities, and the product may also include the components that make up such equipment or facilities. The purpose or project may determine whether the user decides to provide the product as a reusable item rather than discarding it when they stop using it or repair it. For example, products that were initially used for reuse purposes or reuse projects are more likely to be provided as reusable items rather than discarded.

[0016] The "anomaly detection history" is data that shows the history of anomalies detected or alarms (warnings) issued while the user was using the product. This "anomaly detection history" influences the user's decision-making tendencies regarding reuse. For example, if the number of times anomalies were detected or alarms were triggered within a given period while the user was using the product exceeds a threshold, the user is likely to discard the product rather than offer it for reuse. Conversely, if the number of times anomalies were detected or alarms were triggered within a given period is below the threshold, the user is likely to determine that the product is reusable and offer it for reuse rather than discarding it.

[0017] "Usage level" is data indicating the frequency or utilization rate of a product's use by a user. This "usage level" reflects the user's decision-making tendencies regarding reuse. For example, if a user's usage frequency or utilization rate of a product exceeds a threshold, the user is likely to want to reuse its parts in order to continue using the product. Conversely, if the usage frequency or utilization rate of a product falls below the threshold during the target period, the user's use of the product is limited, and if the product malfunctions, it is likely to be discarded rather than reused.

[0018] "Performance data" refers to data showing the performance indicators, operating efficiency, or output content of the target product used by the user. This "performance data" indicates the user's decision-making tendencies regarding reuse. For example, if the performance indicator value of the target product is above a threshold, the user is likely to want to reuse its parts in order to continue using the product. Conversely, if the performance indicator value of the target product is below the threshold within the target period, the user's use of the product is likely to be limited, and if the product fails, it is likely to be discarded and not reused.

[0019] "Usage cost" is data that shows the expenses incurred by the user in using the product or the cost of the product itself. This "usage cost" influences the user's decision-making tendencies regarding reuse. For example, if the usage cost of the product is above a threshold, the user is likely to want to reuse its parts in order to continue using the product. Conversely, if the usage cost of the product within the target period is below the threshold, the user's use of the product is likely to be limited, and if the product malfunctions, it is likely to be discarded and not reused.

[0020] "Location of Use" is data indicating the location and environment in which the product was used by the user. "Location of Use" influences the user's decision-making tendencies regarding reuse. For example, if the product is used in a hot and humid environment, the user is likely to consider the product prone to failure and therefore want its parts to be reused. Conversely, if the environment in which the product was used is low humidity and a constant temperature, where failures are less likely to occur, the user is likely to consider the product to be less prone to failure and therefore do not want to reuse it.

[0021] The "trouble history" is data that shows the history of defects or failures that occurred in the product while the user was using the product. The "trouble history" indicates the user's decision-making tendencies regarding reuse. For example, if the number of defects or failures that occurred within the target period during the user's use of the product exceeds a threshold, the user is likely to discard the product rather than offer it for reuse. Conversely, if the number of defects or failures that occurred within the target period is below the threshold, the user is likely to determine that the product is reusable and offer it for reuse rather than discarding it.

[0022] "Usage conditions" refer to data indicating the user's usage conditions, product settings, or operating mode of the product. For example, usage conditions such as frequency of use, durability, degree of soiling, repairability, and ease of recycling influence the user's decision regarding reuse. Products judged to be highly durable and easy to reuse based on usage conditions are more likely to be repaired or reused by users, while products expected to become heavily soiled or deteriorated based on usage conditions are more likely to be discarded.

[0023] "Condition after use" refers to data indicating the condition or evaluation of the product after it has been used by a user. For example, "condition after use" may include the degree of soiling, presence or absence of damage, deterioration of parts, and the need for cleaning or repair. In this case, if the product is in relatively good condition, the user is more likely to choose to clean or make minor repairs and reuse it. On the other hand, if it is severely damaged or important parts are deteriorated, the user is more likely to choose to discard or recycle it.

[0024] "Energy efficiency" is data that indicates the energy efficiency or energy-saving effect of a product used by a user. For example, products with high energy efficiency have lower operating costs and less environmental impact even after reuse, so users are more likely to choose to reuse them. On the other hand, if energy efficiency is low, or if the latest products are significantly more efficient, users are more likely to choose to purchase new products. In particular, for electrical appliances or machinery, the balance between energy efficiency and operating costs is an important decision criterion.

[0025] "Insurance information" refers to insurance information related to the product being used by the user. For example, if the reuse of the product is not covered by the insurance, the user is more likely to choose to purchase a new product rather than reuse it. On the other hand, if the insurance covers the cost of repairing the product or part of the reuse process, the user is more likely to choose to reuse it. Thus, insurance information can be an important factor in shaping the user's decision-making tendencies because it influences their perception of the costs associated with reusing or discarding a product.

[0026] "Regulatory compliance" is data indicating the user's adherence to laws or regulations regarding the use of the product in question. "Regulatory compliance" indicates constraints on the manufacture, sale, use, or disposal of the product in question. For example, whether or not the product in question meets legal standards or regulations influences decisions or actions throughout those processes. During the manufacturing process, if restrictions on the content of hazardous substances or energy efficiency standards are not met, the production or distribution of the product may be restricted. For sales purposes, products must comply with labeling requirements or import / export regulations, and non-compliant products cannot be put on the market. During the usage phase, products that do not meet safety or environmental standards pose a higher risk to users and tend to be avoided. Regarding disposal, if proper disposal or recycling obligations are regulated, violations will be subject to penalties. Thus, data on regulatory compliance is crucial information supporting decision-making throughout the entire product lifecycle.

[0027] For example, if the product in question is an elevator, the user usage data would include the construction date of the building where the elevator was installed and its maintenance plan data. If the product in question is a factory system, user usage data would include, for example, sales timing and sales volume data for products produced by the factory system. If the product in question is a consumer electronics product, the user usage data would include, for example, the product's release date and market demand data. If the product in question is a railway vehicle, the user usage data for the product may include, for example, train schedule information and passenger count data. If the product in question is a passenger aircraft, user usage data would include, for example, flight schedules and passenger count data. Furthermore, if the product in question is a motor within a manufacturing device, information such as the motor being used for 2,000 hours in a year and undergoing 5 maintenance cycles will be recorded as user data on the product's usage history.

[0028] As mentioned above, user usage data is predictive data that shows user decision-making trends regarding constraints on the manufacture, sale, use, or disposal of products, as well as reuse. Furthermore, user usage data of the product is collected using various means and stored as a database in the storage device mentioned above. For example, usage data can be collected by incorporating sensors or IoT devices into the product and through the cloud or a dedicated application, and stored as a database in the storage device. The input unit 11 may also accept information entered by the user as questionnaires or feedback as usage data.

[0029] (Prediction section) The prediction unit 12 uses the prediction data received by the input unit 11 to predict the supply of reusable products from the product. For example, the prediction unit 12 calculates the degree of deterioration of the target product based on usage data, calculates the remaining lifespan of the product based on the degree of deterioration and the product's service life data, predicts the product's collection timing based on the remaining lifespan, and predicts the supply of reusable products based on the collection timing. Specifically, the prediction unit 12 calculates the amount of product circulating in the market based on the manufacturing data and sales data included in the usage data. Next, the prediction unit 12 numerically evaluates the reusability of the product as remaining lifespan based on the user's judgment tendencies. At the disposal stage, the prediction unit 12 predicts the recovery rate of products that can be recovered as reusable products, taking into account constraints such as legal regulations or the capacity of processing facilities. The prediction unit 12 integrates this data to calculate the supply of reusable products.

[0030] (Output section) The output unit 13 outputs the predicted supply quantity of reusable materials to an external device. Examples of external devices include displays, printers, storage devices, mobile devices, and digital signage connected to a personal computer (PC) or server. Forecast results are primarily provided to the user in a visualized form, displayed on the screen in graph or table format. For example, monthly trends in supply or regional distribution may be graphed and made available for real-time viewing. Alternatively, the report can be saved as data in PDF or CSV format and printed out, or saved to cloud storage for sharing. Furthermore, the forecast results may be accessible from outside the office using a smartphone or tablet PC. In industrial settings, digital signage may be used to share supply forecast results with all relevant parties within the factory or facility.

[0031] Figure 2 is a functional block diagram showing an example configuration of the prediction unit 12. As shown in Figure 2, the prediction unit 12 comprises an acquisition unit 121, a degradation degree calculation unit 122, a component lifespan calculation unit 123, a timing prediction unit 124, and a supply quantity calculation unit 125. The acquisition unit 121 acquires lifespan data indicating the service life of each component constituting the target product. The degradation degree calculation unit 122 calculates the degradation degree of the target product based on user usage data of the product. The component lifespan calculation unit 123 calculates the remaining lifespan of each component constituting the target product based on the degradation degree and lifespan data. The timing prediction unit 124 calculates the collection timing of each component based on the remaining lifespan of the component. The supply quantity calculation unit 125 calculates the supply quantity of reusable components, which are recyclable parts, based on the degradation degree of each component and the lifespan data of each component.

[0032] Next, a circular production control method according to Embodiment 1 will be described. Figure 3 is a flowchart illustrating the cyclic production control method according to Embodiment 1, and shows in detail the processing of the prediction unit 12 in Figure 2. The input unit 11 accepts predictive data that includes constraints on the manufacture, sale, use, or disposal of the product, as well as user decision trends regarding reuse (step ST1). The predictive data is stored in the database described above by the input unit 11. Next, the prediction unit 12 uses the prediction data read from the database to predict the supply volume of recycled products from the product.

[0033] In this prediction process, the acquisition unit 121 acquires service life data indicating the service life of each component constituting the target product (step ST2). The service life of a component is the period during which the component can safely and efficiently perform its designed performance. The material properties of the component, the operating environment, and load conditions all affect the service life. Standard values ​​for service life are set for each type of component. For example, component types include housings, power supply units, control devices, motors, cables, displays, batteries, communication modules, mechanical parts such as bearings, interfaces, storage devices, and antennas.

[0034] The service life data may be linked to the usage data and stored in a storage device as a database. In this case, the input unit 11 retrieves the usage data from the database stored in the storage device, and the retrieval unit 121 retrieves the service life data linked to this usage data. The storage device may be an external storage device connected to the circular production control device 1 via wired or wireless communication, or it may be a storage unit provided by the circular production control device 1. Furthermore, the service life data may be input by the user using an input device not shown in Figure 1.

[0035] Next, the degradation degree calculation unit 122 acquires the usage data received by the input unit 11 and calculates the degradation degree of each part constituting the target product based on the acquired usage data (step ST3). For example, the degradation degree calculation unit 122 evaluates the cumulative usage time, usage conditions, or usage environment of the target product included in the usage data to determine the degradation degree of each component constituting the target product. The evaluation method is, for example, to evaluate the degradation degree of a component whose cumulative usage time exceeds 50% of its service life as 50%.

[0036] The component life calculation unit 123 calculates the remaining life of each component based on the degree of degradation and the service life data (step ST4). For example, the component life calculation unit 123 calculates the remaining life of a component by subtracting the time corresponding to the degree of degradation from the service life. For example, if the service life is 5000 hours and the degree of degradation is 50%, the time corresponding to the degree of degradation is 2500 hours, which is 50% of the service life. Therefore, the remaining life of that component is 2500 hours. The remaining life of a component can be used to numerically evaluate the reusability of the component. For example, if the remaining life of a component is 50% of its service life, the reusability may also be set to 50%.

[0037] Next, the timing prediction unit 124 calculates the recovery time for each component based on the remaining lifespan of the component (step ST5). For example, the timing prediction unit 124 compares the remaining lifespan of a component with a preset threshold and determines the recovery time for the component to be a period in the future corresponding to that threshold, starting from the present time. If the threshold is 1000 hours and the period corresponding to this threshold is one year, the timing prediction unit 124 predicts that components with a remaining lifespan of less than 1000 hours will need to be recovered one year from the present time.

[0038] Next, the supply quantity calculation unit 125 calculates the supply quantity of reusable parts, which are recyclable parts, based on the degree of deterioration and service life data of each part (step ST6). For example, the supply quantity calculation unit 125 calculates the total amount of each part to be collected by the predicted collection period as the supply quantity of reusable parts from the target product. The supply quantity calculation unit 125 also determines the supply quantity of reusable parts for each target region, for example, for each city, and outputs the supply quantity of reusable parts for each city to the output unit 13.

[0039] Finally, the output unit 13 outputs the predicted supply quantity of reusable materials (step ST7). For example, the output unit 13 generates output data including the predicted supply quantity of reusable materials obtained from the supply quantity calculation unit 125, and outputs the output data to an external device. The output data includes the supply quantity of reusable materials for each city, the types of parts that can be supplied, and the supply timing, and this data is presented in report format. Furthermore, a feedback loop may be provided between the cyclic production control device 1 and the external device to adjust the supply plan as needed. A feedback loop for adjusting supply plans is a process for modifying or optimizing the supply plan for reusable products based on actual usage data or forecast results in the supply chain. For example, it involves collecting actual usage data such as the supply or demand volume of reusable products, product usage status, and waste recovery rates, and analyzing the collected data to identify the difference between forecast and actual values ​​at the planning stage. This makes it possible to identify problems such as supply shortages or oversupplies and take specific countermeasures such as improving supply efficiency or recovery rates, or reviewing waste treatment processes.

[0040] Furthermore, the prediction unit 12 may use a Kalman filter to predict the supply quantity of reusable products from the prediction data. For example, the prediction unit 12 predicts the supply quantity of reusable products by applying data assimilation using a Kalman filter to the usage data. In this process, the prediction of the supply quantity of reusable products can be adjusted in real time based on the usage data of the target product. Specifically, the prediction unit 12 uses a Kalman filter, which is a dynamic model (a prediction model for waste generation, collection, and supply), to calculate a predicted value for the supply of reusable materials. Next, the prediction unit 12 calculates the error by comparing the supply quantity obtained from the actual usage data of the target product with the predicted value. Based on this error, the prediction unit 12 corrects the predicted value and outputs the updated supply quantity. This method makes it possible to accurately predict the supply quantity of reusable products while minimizing the impact of usage data or observation data that may contain noise. Furthermore, by incorporating the prediction results into a feedback loop, the supply plan can be continuously improved.

[0041] The prediction unit 12 may use a machine learning model to predict the supply quantity of reusable products from the prediction data. For example, the prediction unit 12 trains a machine learning algorithm using past usage data as training data. The prediction data is set as the explanatory variable, and the supply quantity of reusable products is set as the dependent variable. Furthermore, machine learning algorithms may include regression models (linear regression, ridge regression, etc.), nonlinear models such as random forests and XGBoost, and deep learning methods specifically designed for time series data, such as LSTM and GRU. The machine learning model learns complex supply patterns by considering waste type, recovery rate, and external factors (seasonality or policy changes). By inputting new data into the trained machine learning model in real time or in batch processing, it outputs predicted values ​​for the supply of recyclable materials. Furthermore, when the prediction unit 12 predicts the supply volume of reusable products, it may use a large-scale language model that performs language recognition on prediction data to make predictions, or it may use various generative AIs.

[0042] Next, we will describe the hardware configuration that realizes the functions of the circular production control system 1. The functions of the input unit 11, prediction unit 12, and output unit 13 of the circular production control device 1 are realized by a processing circuit. That is, the circular production control device 1 includes a processing circuit for executing the processes from step ST1 to step ST7 shown in Figure 3. The processing circuit may be dedicated hardware, or it may be a CPU (Central Processing Unit) that executes a program stored in memory.

[0043] Figure 4A is a functional block diagram showing the hardware configuration that realizes the functions of the circular production control device 1. Figure 4B is a functional block diagram showing the hardware configuration that executes the software that realizes the functions of the circular production control device 1. In Figures 4A and 4B, the input unit 11 receives prediction data input using the input device 104 via the input interface 101. The output unit 13 outputs the prediction results to the display 105 via the output interface 102. The input unit 11 stores the prediction data in the storage device 106 via the input / output interface 103, and the prediction unit 12 acquires the prediction data from the storage device 106 via the input / output interface 103.

[0044] If the processing circuit is a dedicated hardware processing circuit 100 as shown in Figure 4A, the processing circuit 100 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof. The input unit 11, prediction unit 12, and output unit 13 of the cyclic production control device 1 may be implemented by separate processing circuits, or these functions may be implemented together by a single processing circuit.

[0045] When the processing circuit is the processor 107 shown in Figure 4B, the functions of the input unit 11, prediction unit 12, and output unit 13 of the circular production control device 1 are realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in memory 108.

[0046] The processor 107 reads and executes a program stored in the memory 108, thereby realizing the functions of the input unit 11, prediction unit 12, and output unit 13 of the circular production control device 1. For example, the circular production control device 1 includes a memory 108 for storing a program that, when executed by the processor 107, will result in the execution of the processes from step ST1 to step ST7 shown in Figure 3. These programs cause the computer to execute the procedures or methods of processing performed by the input unit 11, prediction unit 12, and output unit 13. The memory 108 may be a computer-readable storage medium that stores a program for causing the computer to function as the input unit 11, prediction unit 12, and output unit 13.

[0047] Memory 108 includes, for example, non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), and EEPROM (Electrically-EPROM) (registered trademark), as well as magnetic disks, flexible disks, optical disks, compact disks, minidiscs, and DVDs.

[0048] Some of the functions of the input unit 11, prediction unit 12, and output unit 13 of the circular production control device 1 may be implemented by dedicated hardware, while other parts may be implemented by software or firmware. For example, the functions of the input unit 11 may be implemented by a processing circuit 100 which is dedicated hardware, and the functions of the prediction unit 12 and output unit 13 may be implemented by a processor 107 reading and executing a program stored in memory 108. Thus, the processing circuit can achieve the above functions through hardware, software, firmware, or a combination thereof.

[0049] As described above, the circular production control device 1 according to Embodiment 1 includes an input unit that receives input of predictive data indicating constraints on the manufacture, sale, use, or disposal of products and user decision trends regarding reuse, a prediction unit that uses the predictive data to predict the amount of recycled products to be supplied from products, and an output unit 13 that outputs the prediction result of the amount of recycled products to be supplied. The amount of reusable material supplied from a product is predicted using predictive data that includes constraints on the manufacture, sale, use, or disposal of the product, as well as user decision-making trends regarding reuse. For example, the amount of products collected for reuse largely depends on the user's decision to provide the product for reuse rather than discarding it when they stop using it or have it repaired. Furthermore, the amount of recyclable materials collected is also affected by policy constraints imposed by international organizations, national governments, or cities. In contrast, the circular production control device 1 can predict the supply volume of recycled products by taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends. This makes it possible to accurately predict the supply volume.

[0050] In the circular production control device 1 according to Embodiment 1, the prediction data is user usage data of the product. The prediction unit 12 calculates the degree of product deterioration based on the usage data, calculates the remaining lifespan of the product based on the degree of deterioration and the product's service life data, predicts the product collection time based on the remaining lifespan, and predicts the supply quantity of recycled products based on the collection time. As a result, the circular production control device 1 can accurately predict the supply quantity, taking into account the above constraints and fluctuations in the amount of product recovered due to the above judgment tendencies.

[0051] In the circular production control device 1 according to Embodiment 1, the prediction unit 12 uses a Kalman filter to predict the supply amount of reusable products from prediction data. As a result, the circular production control device 1 can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends.

[0052] In the circular production control device 1 according to Embodiment 1, the prediction unit 12 predicts the supply amount of reusable products from prediction data using a machine learning model. As a result, the circular production control device 1 can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends.

[0053] The circular production control method according to Embodiment 1 includes the steps of: an input unit 11 receiving input of prediction data that includes constraints on the manufacture, sale, use, or disposal of products and user decision trends regarding reuse; a prediction unit 12 predicting the amount of recycled products to be supplied from products using the prediction data; and an output unit 13 outputting the prediction result of the amount of recycled products to be supplied. By executing this method, the circular production control device 1 can predict the amount of recycled products to be supplied, taking into account constraints on the manufacture, sale, use, or disposal of products and fluctuations in the amount of products recovered due to user decision trends regarding reuse.

[0054] A computer executing the program according to Embodiment 1 functions as an input unit 11 that accepts input of predictive data that includes constraints on the manufacture, sale, use, or disposal of products and user decision-making tendencies regarding reuse; a prediction unit 12 that uses the predictive data to predict the amount of recycled products to be supplied from products; and an output unit 13 that outputs the prediction result for the amount of recycled products to be supplied. This allows the computer to predict the supply of reusable products, taking into account constraints on the manufacturing, sale, use, or disposal of products, as well as fluctuations in the amount of products collected based on user decision-making trends regarding reuse.

[0055] Embodiment 2. The circular production control device according to Embodiment 2 is a device that predicts the supply volume of reusable products, similar to Embodiment 1, and the prediction results of the supply volume of reusable products are used to adjust the supply plan for reusable products. The amount of product recovered as reusable products (recovery volume) fluctuates depending on constraints on the manufacturing, sale, use, or disposal of the product, and also fluctuates depending on the user's decision-making tendencies regarding reuse. In order to grasp such fluctuations in the recovery volume of products, the circular production control device according to Embodiment 2 predicts the supply volume of reusable products from products using prediction data that includes the above constraints and decision-making tendencies. Here, we will explain the case where the prediction data is "user maintenance performance data for products".

[0056] (Basic configuration of a circular production control system) Figure 5 is a functional block diagram showing an example configuration of the cyclic production control device 1A according to Embodiment 2. The cyclic production control device 1A receives user-provided product maintenance data and product service life data, uses the input data to predict the supply quantity of reusable products, and outputs this prediction result to an external device. The external device is a display device that displays the prediction result of the supply quantity of reusable products. As shown in Figure 5, the cyclic production control device 1A comprises an input unit 11A, a prediction unit 12A, and an output unit 13A.

[0057] For example, the circular production control system 1A is implemented by a computer. The computer's memory stores programs that constitute information processing applications for realizing the functions of the input unit 11A, the prediction unit 12A, and the output unit 13A. The functions of the input unit 11A, prediction unit 12A, and output unit 13A are realized when the computer's processor executes an information processing application read from memory.

[0058] (Input section) The input unit 11A accepts input of predictive data that includes restrictions on the manufacture, sale, use, or disposal of the product, as well as user decision trends regarding reuse. For example, prediction data is stored in a memory device as a database. The input unit 11A retrieves prediction data from the database stored in the memory device. If the storage device is an external storage device that is connected to the circular production control device 1A via wired or wireless communication, the input unit 11A acquires the prediction data stored in the external storage device via wired or wireless communication. The memory device may be the memory unit of the circular production control device 1A. In this case, the input unit 11A acquires the prediction data stored in the memory unit via a signal line. Furthermore, the input unit 11A may accept prediction data input from a user using an input device not shown in Figure 5.

[0059] (Product maintenance performance data) The prediction data in Embodiment 2 is "user-provided product maintenance data." For example, maintenance history data includes the date and time maintenance was performed, the details of the maintenance, the parts replaced, the condition of the parts, the type of maintenance, the results of the maintenance, the next scheduled maintenance, maintenance history information, warranty information, and data related to the product in question. "Type of maintenance" refers to classifications such as scheduled maintenance, emergency maintenance, or preventive maintenance. "Maintenance results" include the results of the maintenance work and its evaluation. "Next maintenance scheduled" includes the next scheduled maintenance date or maintenance plan. "Maintenance history information" includes records of past maintenance history or matters related to maintenance. "Warranty information" is information regarding warranties or insurance related to the maintenance work.

[0060] The "data related to the target product" includes the "intended use," "anomaly detection history," "degree of use," "performance data," "usage cost," "location of use," "trouble history," "usage conditions," "post-use condition," "energy efficiency," "insurance information," and "regulatory compliance status" of the target product. The details of each piece of information are the same as in Embodiment 1.

[0061] For example, if the product in question is an elevator, the user's maintenance history data for the product would include the construction date of the building where the elevator was installed and its maintenance plan data. If the product in question is a factory system, the user's maintenance history data for the product would include, for example, data on the sales date and sales volume of products produced by the factory system. If the product in question is a consumer electronics product, the user's maintenance history data would include, for example, the product's release date and its market demand data. If the product in question is a railway vehicle, the user's maintenance history data would include, for example, train schedule information and passenger count data. If the product in question is a passenger aircraft, the user's maintenance history data would include, for example, flight schedules and passenger count data. Furthermore, detailed information about the maintenance work performed on the target product over the past year will be recorded as user maintenance history data for the product.

[0062] As mentioned above, user maintenance data for products is predictive data that shows user decision-making trends regarding constraints on the manufacture, sale, use, or disposal of products, as well as reuse. Furthermore, user maintenance data for the product is collected using various means and stored as a database in the storage device mentioned above. For example, maintenance data can be collected by incorporating sensors or IoT devices into the product and using the cloud or a dedicated application, and then stored as a database in the storage device. The input unit 11A may also accept information entered by the user as questionnaires or feedback as maintenance data.

[0063] (Prediction section) The prediction unit 12A uses the prediction data received by the input unit 11A to predict the amount of recycled materials to be supplied from the product. For example, the prediction unit 12A calculates the degree of deterioration of the target product based on maintenance performance data, calculates the remaining lifespan of the product based on the degree of deterioration and the product's service life data, predicts the product's collection date based on the remaining lifespan, and predicts the supply quantity of reusable products based on the collection date.

[0064] (Output section) The output unit 13A outputs the predicted supply quantity of recycled materials to an external device. For example, external devices may include displays, printers, storage devices, mobile devices, and digital signage connected to a PC or server. The forecast results are primarily provided to the user in a visualized form, displayed on the screen in graph or table format. For instance, monthly trends in supply or regional distribution may be graphed and made available for real-time viewing. Alternatively, the report can be saved as data in PDF or CSV format and printed out, or saved to cloud storage for sharing. Furthermore, the forecast results may be accessible from outside the office using a smartphone or tablet PC. In industrial settings, digital signage may be used to share supply forecast results with all relevant parties within the factory or facility.

[0065] Figure 6 is a functional block diagram showing an example configuration of the prediction unit 12A. As shown in Figure 6, the prediction unit 12A comprises an acquisition unit 121A, a degradation degree calculation unit 122A, a component lifespan calculation unit 123, a timing prediction unit 124, and a supply quantity calculation unit 125. The acquisition unit 121A acquires lifespan data indicating the service life of each component constituting the target product. The degradation degree calculation unit 122A calculates the degradation degree of the target product based on user maintenance history data. The component lifespan calculation unit 123 calculates the remaining lifespan of each component constituting the target product based on the degradation degree and lifespan data. The timing prediction unit 124 calculates the collection timing for each component based on the remaining lifespan of the component. The supply quantity calculation unit 125 calculates the supply quantity of reusable parts, which are recyclable components, based on the degradation degree of each component and the lifespan data of each component.

[0066] Next, a circular production control method according to Embodiment 2 will be described. Figure 7 is a flowchart illustrating the cyclic production control method according to Embodiment 2, and shows in detail the processing of the prediction unit 12A in Figure 6. The input unit 11A accepts predictive data that includes constraints on the manufacture, sale, use, or disposal of the product, as well as user decision trends regarding reuse (step ST1A). The predictive data is stored in the database described above by the input unit 11A. Next, the prediction unit 12A uses the prediction data read from the database to predict the supply volume of recycled products from the product.

[0067] In this prediction process, the acquisition unit 121A acquires service life data indicating the service life of each component constituting the target product (step ST2A). The service life of a component is the period during which the component can safely and efficiently perform its designed performance. The material properties of the component, the operating environment, and load conditions all affect the service life. Standard service life values ​​are set for each type of component. For example, component types include housings, power supply units, control devices, motors, cables, displays, batteries, communication modules, mechanical parts such as bearings, interfaces, storage devices, and antennas.

[0068] The service life data may be linked to the usage data and stored in a storage device as a database. In this case, the input unit 11A retrieves the maintenance data from the database stored in the storage device, and the acquisition unit 121A retrieves the service life data linked to this maintenance data. The storage device may be an external storage device connected to the circular production control device 1A via wired or wireless communication, or it may be a storage unit provided by the circular production control device 1A. Furthermore, the service life data may be input by the user using an input device not shown in Figure 5.

[0069] Next, the degradation degree calculation unit 122A acquires the maintenance performance data received by the input unit 11A and calculates the degradation degree of each part constituting the target product based on the acquired usage performance data (step ST3A). For example, the degradation degree calculation unit 122A determines the degradation degree of each component constituting the target product based on the usage period or usage condition of the target product included in the maintenance performance data. The evaluation method is, for example, if the usage period of the target product is above a threshold or the evaluation value of the usage condition is above a threshold, the degradation degree of that component is evaluated as high.

[0070] The component life calculation unit 123 calculates the remaining life of each component based on the degree of degradation and the service life data (step ST4A). For example, the component life calculation unit 123 calculates the remaining life of a component by subtracting the time corresponding to the degree of degradation from the service life. For example, if the service life is 10 years and the degree of degradation is 2 years, the remaining life of that component will be 8 years.

[0071] Next, the timing prediction unit 124 calculates the recovery time for each component based on the remaining lifespan of the component (step ST5A). For example, the timing prediction unit 124 compares the remaining lifespan of a component with a preset threshold and determines the recovery time for the component to be a period in the future corresponding to that threshold, starting from the present time. If the threshold is 1000 hours and the period corresponding to this threshold is one year, the timing prediction unit 124 predicts that components with a remaining lifespan of less than 1000 hours will need to be recovered one year from the present time.

[0072] Next, the supply quantity calculation unit 125 calculates the supply quantity of reusable parts, which are recyclable components, based on the degree of deterioration and service life data of each component (step ST6A). For example, the supply quantity calculation unit 125 calculates the total amount of each component to be collected by the predicted collection period as the supply quantity of reusable parts from the target product. The supply quantity calculation unit 125 also determines the supply quantity of reusable parts for each target region, for example, for each city, and outputs the supply quantity of reusable parts for each city to the output unit 13A.

[0073] The output unit 13A outputs the predicted supply quantity of reusable materials (step ST7A). For example, the output unit 13A generates output data including the predicted supply quantity of reusable materials obtained from the supply quantity calculation unit 125, and outputs the output data to an external device. The output data includes the supply quantity of reusable materials for each city, the types of parts that can be supplied, and the supply timing, and this data is presented in report format. Furthermore, a feedback loop may be provided between the circulating production control device 1A and the external device to adjust the supply plan as needed. A feedback loop for adjusting supply plans is a process for modifying or optimizing the supply plan for reusable products based on actual usage data or forecast results in the supply chain. For example, it involves collecting actual usage data such as the supply or demand volume of reusable products, product usage status, and waste recovery rates, and analyzing the collected data to identify the difference between forecast and actual values ​​at the planning stage. This makes it possible to identify problems such as supply shortages or oversupplies and take specific countermeasures such as improving supply efficiency or recovery rates, or reviewing waste treatment processes.

[0074] The prediction unit 12A may use a Kalman filter to predict the supply quantity of reusable parts from the prediction data. For example, the prediction unit 12A predicts the supply quantity of reusable parts by applying data assimilation using a Kalman filter to the maintenance performance data. In this process, the prediction of the supply quantity of reusable parts can be adjusted in real time based on the maintenance performance data of the target product. Specifically, the prediction unit 12A uses a dynamic model (a prediction model for waste generation, collection, and supply) which is a Kalman filter to calculate a predicted value for the supply of reusable materials. Next, the prediction unit 12A calculates the error by comparing the supply quantity obtained from the maintenance history data of the target product with the predicted value. Based on this error, the prediction unit 12A corrects the predicted value and outputs the updated supply quantity. This method makes it possible to accurately predict the supply quantity of reusable parts while minimizing the impact of maintenance history data or observation data that may contain noise. Furthermore, by incorporating the prediction results into a feedback loop, the supply plan can be continuously improved.

[0075] The prediction unit 12A may use a machine learning model to predict the supply quantity of reusable parts from the prediction data. For example, the prediction unit 12A trains a machine learning algorithm using past maintenance performance data as training data. The prediction data is set as the explanatory variable, and the supply quantity of reusable parts is set as the target variable. In addition, the machine learning algorithm may employ regression models (linear regression, ridge regression, etc.), nonlinear models such as random forest and XGBoost, or deep learning methods specialized for time series data such as LSTM and GRU. The machine learning model learns complex supply patterns by considering waste type, recovery rate, and external factors (seasonality or policy changes). By inputting new data into the trained machine learning model in real time or in batch processing, it outputs predicted values ​​for the supply of recyclable materials. Furthermore, when predicting the supply volume of recycled materials, the prediction unit 12A may use a large-scale language model that performs language recognition on prediction data to make predictions, or it may use various generative AIs.

[0076] As described above, in the circular production control device 1A according to Embodiment 2, the prediction data is user maintenance performance data for the product. The prediction unit 12A calculates the degree of product deterioration based on the maintenance performance data, calculates the remaining lifespan of the product based on the degree of deterioration and the product's service life data, predicts the product collection time based on the remaining lifespan, and predicts the supply quantity of reusable products based on the collection time. As a result, the circular production control device 1A can accurately predict the supply quantity, taking into account constraints on the manufacturing, sale, use, or disposal of the product, and fluctuations in the amount of product collected due to user decision trends regarding reuse.

[0077] In the circular production control device 1A according to Embodiment 2, the prediction unit 12A uses a Kalman filter to predict the supply amount of reusable products from prediction data. As a result, the circular production control device 1A can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends.

[0078] In the circular production control device 1A according to Embodiment 2, the prediction unit 12A predicts the supply amount of reusable products from prediction data using a machine learning model. As a result, the circular production control device 1A can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends.

[0079] Embodiment 3. The circular production control device according to Embodiment 3 is a device that predicts the supply volume of reusable products, similar to Embodiment 1, and the prediction results of the supply volume of reusable products are used to adjust the supply plan for reusable products. The amount of product recovered as reusable products (recovery volume) fluctuates depending on constraints on the manufacturing, sale, use, or disposal of the product, and also fluctuates depending on the user's decision-making tendencies regarding reuse. In order to grasp such fluctuations in the recovery volume of products, the circular production control device according to Embodiment 3 predicts the supply volume of reusable products from products using prediction data that includes the above constraints and decision-making tendencies. The data used for prediction includes "target values ​​for recycled material use" and "historical data on recycled material use." Recycled material is a reusable material obtained by collecting and processing used products or waste.

[0080] (Basic configuration of a circular production control system) Figure 8 is a functional block diagram showing an example configuration of the circular production control device 1B according to Embodiment 3. The circular production control device 1B receives data on target usage values ​​of recycled materials and past usage data of recycled materials, predicts the supply quantity of recycled products using the input data, and outputs this prediction result to an external device. The external device is a display device that displays the prediction result of the supply quantity of recycled products. As shown in Figure 8, the circular production control device 1B comprises an input unit 11B, a prediction unit 12B, and an output unit 13B.

[0081] For example, the circular production control system 1B is implemented by a computer. The computer's memory stores programs that constitute information processing applications for realizing the functions of the input unit 11B, the prediction unit 12B, and the output unit 13B. The functions of the input unit 11B, prediction unit 12B, and output unit 13B are realized when the computer's processor executes an information processing application read from memory.

[0082] (Input section) The input unit 11B accepts input of predictive data that includes restrictions on the manufacture, sale, use, or disposal of the product, as well as user decision trends regarding reuse. For example, prediction data is stored in a memory device as a database. The input unit 11B retrieves prediction data from the database stored in the memory device. If the storage device is an external storage device that is connected to the circular production control device 1B via wired or wireless communication, the input unit 11B acquires the prediction data stored in the external storage device via wired or wireless communication. The memory device may be the memory unit of the circular production control device 1B. In this case, the input unit 11B acquires the prediction data stored in the memory unit via a signal line. Furthermore, the input unit 11B may accept prediction data input from a user using an input device not shown in Figure 8.

[0083] (Target values ​​for the use of recycled materials) The "Target Value for Recycled Materials" is data that includes future target values ​​for recycled materials. For example, it sets a target value for recycled materials to be used in the next year. The data for the target value for recycled materials includes "Type of Recycled Material," "Target Usage Amount," and "Usage Period." "Type of Recycled Material" is a classification of recycled material, such as recycled plastic, recycled metal, recycled paper, or recycled glass. "Target Usage Amount" is the target amount of recycled material set in the user's production plan. "Usage Period" is the target period for using the recycled material. Furthermore, the data on target values ​​for the use of recycled materials includes "target achievement rate," "reason for setting the target," "projects using the material," "budget," "environmental impact assessment," "legal compliance status," "quality standards," "supply plan," "progress management," "stakeholder information," "cost reduction effect," and "risk assessment."

[0084] The "target achievement rate" is the rate at which the target value for recycled material usage is achieved, and is calculated, for example, as the ratio of the actual amount used to the target. If the "target achievement rate" indicates that users are conscious of bringing the amount of recycled material used closer to the target value and are able to avoid overuse or waste, the supply of recycled material tends to be maintained at a good level. On the other hand, if the actual amount used significantly exceeds the target, problems due to overuse are more likely to occur, and the supply of recycled material may decrease.

[0085] "Target setting reasons" is data that shows the background and reasons for setting target values ​​for the use of recycled materials. "Target setting reasons" influence users' decision-making tendencies regarding reuse. For example, the reasons for setting usage targets can raise users' environmental awareness and lead them to reflect compliance with regulations in their reuse behavior. Also, if a company or government sets recycled material usage targets with the aim of realizing a circular economy, users will be aware of these efforts and will be more inclined to prioritize reuse when selecting products or making decisions about disposal. Furthermore, if the benefits are clearly communicated through the reasons for setting targets, users will understand the environmental improvement or resource conservation effects that reuse brings and will take proactive action.

[0086] "Usage Projects" is data that indicates the specific projects or uses in which recycled materials are used. "Usage Projects" influence users' decision-making tendencies regarding reuse. For example, when recycled materials are used in public infrastructure projects, environmental conservation activities, or products that contribute to the local community, users are more likely to recognize their significance and social impact, and are more inclined to choose reuse.

[0087] "Budget" is data indicating the budget allocated to the use of recycled materials. "Budget" influences users' decision-making tendencies regarding reuse. For example, if the budget supports price subsidies or incentive programs to promote the use of recycled materials, users' resistance to cost will be reduced, and the choice of recycled materials will be promoted. On the other hand, if the budget is insufficient, the quality or supply stability of recycled materials cannot be ensured, and users may be more inclined to rely on new materials.

[0088] An "environmental impact assessment" is data that shows the environmental impact of achieving a target. If the impact assessment is positive, users can feel satisfied that they are fulfilling their social responsibility by choosing recycled materials, and their motivation to reuse will increase. On the other hand, if the assessment is insufficient or the target is not met, confidence in the effectiveness of reuse will decrease, and reuse may decline.

[0089] "Compliance with regulations" is data that shows compliance with laws and regulations regarding the setting of targets for the use of recycled materials and their use. "Compliance with regulations" indicates constraints on the manufacture, sale, use, or disposal of recycled materials. For example, whether the setting of targets for the use of recycled materials and their use meet legal standards or regulations influences decisions or actions throughout those processes. During the manufacturing process, if restrictions on the content of hazardous substances or energy efficiency standards are not met, the production or distribution of recycled materials may be restricted. For sales purposes, products must comply with labeling requirements or import / export regulations, and non-compliant products cannot be put on the market. During the usage phase, recycled materials that do not meet safety or environmental standards tend to pose a higher risk to users and are therefore often avoided. Regarding disposal, if proper disposal or recycling obligations are regulated, violations will be subject to penalties. Thus, data on compliance with regulations is crucial information supporting decision-making throughout the entire product lifecycle.

[0090] "Quality standards" are data that outlines the standards or requirements for the quality of recycled materials. Quality standards also impose constraints on the manufacture, sale, use, or disposal of recycled materials. For example, whether recycled materials meet the quality standards or requirements influences decisions or actions throughout the entire process of manufacturing, selling, using, or disposing of them.

[0091] The "supply plan" is data that includes information about the supply schedule and source of recycled materials. When the supply schedule is stable, users have confidence that they can reliably obtain recycled materials and are more likely to actively adopt recycled materials. On the other hand, if the supply is unstable or delayed, users may perceive risks in their manufacturing plans or operations and may choose to rely on new materials. In addition, when the source of the recycled material is clearly indicated, such as what kind of waste it is made from and which region or process it is produced through, users can more easily evaluate its environmental value or ethical implications.

[0092] "Progress management" refers to data used to track progress toward achieving a goal. When this management data is properly collected and presented, users can gain a concrete understanding of their contribution to or progress toward achieving the goal, making it easier to understand how their actions contribute to the overall objective. Furthermore, if the progress data indicates delays or deficiencies, sharing the causes within the management data will encourage users to consider actions to resolve the issues.

[0093] "Stakeholder information" refers to information about the individuals and departments involved in achieving the goals. Clearly presenting stakeholder information helps users recognize that the use of recycled materials is being promoted systematically and responsibly throughout the organization, increasing their confidence in recycled materials. In particular, when the person in charge or their department proactively shares progress or challenges and clearly outlines specific actions toward achieving the goals, users understand that the use of recycled materials is being effectively managed and are more likely to choose reuse.

[0094] "Cost reduction effect" is data that shows an estimate of the cost reduction effect of using recycled materials. When specific cost reduction effects are clearly shown, users can directly understand the economic benefits of choosing recycled materials, and their motivation to take recycled actions will be strengthened.

[0095] "Risk assessment" is data that shows the results of an evaluation of the risks or challenges associated with achieving a goal. When these evaluation results are presented appropriately, users can specifically understand the potential risks or challenges to overcome associated with recycling activities, and it helps them make rational choices. For example, if the risks of quality or supply stability of recycled materials, or increased costs, are clearly evaluated, users are more likely to consider ways to avoid those risks, judge the risks to be within an acceptable range, and choose to reuse. On the other hand, if the risk assessment is insufficient or overly negative, users will feel uneasy about recycling activities and are more likely to refrain from using recycled materials.

[0096] (Data on past usage of recycled materials) "Past usage data for recycled materials" refers to data corresponding to past usage targets for recycled materials, for example. Specifically, past usage data for recycled materials includes, like usage target data for recycled materials, "type of recycled material," "target usage amount," "usage period," "target achievement rate," "reason for setting the target," "usage project," "budget," "environmental impact assessment," "legal compliance status," "quality standards," "supply plan," "progress management," "stakeholder information," "cost reduction effect," and "risk assessment."

[0097] As mentioned above, the target values ​​for recycled material use and historical data on recycled material use are predictive data that show constraints on the manufacturing, sale, use, or disposal of products, as well as user decision-making trends related to reuse. Furthermore, the "target values ​​for recycled material usage" and "past usage data of recycled materials" are collected using various means and stored as a database in the storage device. For example, usage data can be collected by incorporating sensors or IoT devices into the product and using the cloud or a dedicated application, and then stored as a database in the storage device. The input unit 11B may also accept information entered by the user as a questionnaire or feedback as the "target values ​​for recycled material usage" and "past usage data of recycled materials."

[0098] (Prediction section) The prediction unit 12B uses the prediction data received by the input unit 11B to predict the amount of recycled materials to be supplied from the product. For example, the prediction unit 12B calculates the difference between the target usage value of recycled materials and past usage data of recycled materials, predicts the future usage amount of recycled materials based on the calculated difference, and predicts the supply amount of recycled materials using the usage amount.

[0099] (Output section) The output unit 13B outputs the predicted supply quantity of recycled materials to an external device. For example, external devices may include displays, printers, storage devices, mobile devices, and digital signage connected to a PC or server. The forecast results are primarily provided to the user in a visualized form, displayed on the screen in graph or table format. For instance, monthly trends in supply or regional distribution may be graphed and made available for real-time viewing. Alternatively, the report can be saved as data in PDF or CSV format and printed out, or saved to cloud storage for sharing. Furthermore, the forecast results may be accessible from outside the office using a smartphone or tablet PC. In industrial settings, digital signage may be used to share supply forecast results with all relevant parties within the factory or facility.

[0100] Figure 9 is a functional block diagram showing an example configuration of the prediction unit 12B. In Figure 9, the prediction unit 12B includes an acquisition unit 121B, a difference calculation unit 126, and a usage prediction unit 127. The acquisition unit 121B acquires usage target value data set in the most recent production plan for recycled materials. The difference calculation unit 126 calculates the difference between the future usage target value of recycled materials and the past usage data of recycled materials. The usage amount prediction unit 127 predicts the future usage amount of recycled materials based on the difference calculated by the difference calculation unit 126 and calculates the supply amount of recycled materials using the usage amount.

[0101] Next, a circular production control method according to Embodiment 3 will be described. Figure 10 is a flowchart illustrating the cyclic production control method according to Embodiment 3, and shows in detail the processing of the prediction unit 12B in Figure 8. The input unit 11A accepts input of predictive data that includes restrictions on the manufacture, sale, use, or disposal of the product, and user decision trends regarding reuse (step ST1B). Here, the forecasting data consists of "target values ​​for recycled material usage" and "past usage data for recycled material." "Target values ​​for recycled material usage" refers to, for example, the target usage data for recycled material to be used in the next year. The forecasting data is stored in the database mentioned above. "Past usage data for recycled material" refers to the actual usage data of recycled material used in the past year.

[0102] Next, the prediction unit 12B uses the prediction data read from the database to predict the supply volume of recycled materials from the product. The acquisition unit 121B acquires future usage target value data for recycled materials from the database, from the data accepted by the input unit 11B, and outputs it to the difference calculation unit 126.

[0103] The difference calculation unit 126 obtains past usage data of recycled materials from the database (step ST2B). Then, the difference calculation unit 126 calculates the difference between the future usage target value of recycled materials and the past usage data of recycled materials (step ST3B). For example, the difference calculation unit 126 compares data on future usage targets for recycled materials with data on past usage of recycled materials to calculate the difference between the usage target for recycled materials for the next year and the actual usage value of recycled materials for the past year.

[0104] The usage prediction unit 127 predicts the future usage of the recycled product based on the difference calculated by the difference calculation unit 126 (step ST4B). For example, the usage prediction unit 127 predicts the future usage of reusable products using a prediction model. The prediction model is a model that, when a calculated difference is input, outputs a predicted value for the future usage of reusable products. The procedure for predicting the future use of recycled materials is as follows: (1) Data collection and preprocessing The usage forecasting unit 127 collects past usage data of recycled materials, identifies missing and abnormal values ​​in the collected usage data, and performs preprocessing such as imputing missing values ​​and removing abnormal values. (2) Feature extraction The usage forecasting unit 127 extracts feature quantities from the future usage target value data and the past usage performance data of the recycled material. These feature quantities are, for example, the contents of various data items included in the data, such as "usage amount," "usage date," "usage location," "target achievement rate," and "budget," and are extracted in numerical form. (3) Model Selection For prediction, a statistical model or a machine learning algorithm is selected. Examples of statistical models include regression analysis and time series analysis models. Examples of machine learning algorithms include random forests and support vector machines. (4) Model training The usage prediction unit 127 trains a selected model using the difference between usage data of recycled materials obtained over different periods as training data. For example, it constructs a predictive model that predicts the amount of recycled materials used over a fixed period in the future using the difference between usage data of recycled materials obtained over different periods within the past year. Furthermore, (3) and (4) may be performed when predicting the amount of recycled material to be used, and the prediction model may be constructed in advance. (5) Prediction The usage prediction unit 127 uses a trained prediction model to predict the future usage of recycled materials. For example, by inputting the difference between the target usage value of recycled materials for the next year and the actual usage value of recycled materials for the past year into the prediction model, the prediction model outputs the usage amount of recycled materials for the next year. This makes it possible to accurately predict the usage amount of recycled materials, taking into account constraints on the manufacturing, sale, use, or disposal of products, as well as fluctuations in the amount of products recovered due to user decision-making trends regarding reuse.

[0105] Alternatively, the prediction unit 12B may not include the difference calculation unit 126, and only the usage prediction unit 127 may predict the amount of recycled material used using future usage target value data and past usage data of recycled material. In (4), the usage prediction unit 127 constructs a prediction model that predicts the future amount of recycled material used based on past usage data of recycled material. For example, when predicting the amount of recycled material used, the usage prediction unit 127 uses a prediction model that outputs a prediction result for the amount of recycled material used in the next year when usage data of recycled material for the past year is input. Even in this way, it is possible to accurately predict the amount of recycled material used, taking into account constraints on the manufacture, sale, use, or disposal of the product, and fluctuations in the amount of product recovered due to user decision trends regarding reuse.

[0106] Next, the usage prediction unit 127 predicts the amount of recycled materials to be supplied, indicating how much of the recycled material should be supplied from recycled materials, based on the predicted future usage of the recycled materials. In supply forecasting, using regression analysis or machine learning models based on historical data allows for highly accurate predictions that reflect the correlations between factors influencing supply.

[0107] The output unit 13B outputs the predicted supply quantity of recycled materials (step ST5B). For example, the output unit 13B generates output data that includes the predicted supply quantity of reusable materials obtained from the usage prediction unit 127, and outputs this output data to an external device. The output data includes the supply quantity of reusable materials for each city, the types of parts that can be supplied, and the supply timing, and this data is presented in report format. Furthermore, a feedback loop may be provided between the circulating production control device 1B and the external device to adjust the supply plan as needed. A feedback loop for adjusting supply plans is a process for modifying or optimizing the supply plan for reusable items based on actual usage data or forecasts in the supply chain. For example, by collecting usage data such as the supply or demand for reusable materials, product usage status, and waste recovery rates, and analyzing the collected data, it is possible to identify the difference between the predicted values ​​at the planning stage and the actual values. This makes it possible to identify problems such as supply shortages or oversupply and to take specific countermeasures such as improving supply efficiency or recovery rates, or reviewing waste treatment processes.

[0108] The prediction unit 12B may use a Kalman filter to predict the supply quantity of recycled materials from the prediction data. For example, the prediction unit 12B predicts the supply quantity of recycled materials by applying data assimilation using a Kalman filter to future usage target data and past usage data of recycled materials. This allows for real-time adjustment of the prediction of the supply quantity of recycled materials. Furthermore, when predicting the supply volume of recycled materials, the prediction unit 12B may use a large-scale language model that performs predictions by recognizing language data for prediction, or it may use various generative AIs.

[0109] As described above, in the circular production control device 1B according to Embodiment 3, the prediction data consists of the future usage target value of recycled material and past usage data of recycled material. The prediction unit 12B calculates the difference between the usage target value and the usage data, predicts the future usage amount of recycled material based on the calculated difference, and predicts the supply amount of recycled material using the usage amount. As a result, the circular production control device 1B can accurately predict the supply amount, taking into account constraints on the manufacturing, sale, use, or disposal of products, and fluctuations in the amount of product recovered due to user decision trends regarding reuse.

[0110] In the circular production control device 1B according to Embodiment 3, the prediction unit 12B uses a Kalman filter to predict the supply amount of reusable products from prediction data. As a result, the circular production control device 1B can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends.

[0111] In the circular production control device 1B according to Embodiment 3, the prediction unit 12B predicts the supply amount of reusable products from prediction data using a machine learning model. As a result, the circular production control device 1B can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends.

[0112] Embodiment 4. The circular production control device according to Embodiment 4 is a device that predicts the supply volume of recycled materials, similar to Embodiment 1, and the prediction results of the supply volume of recycled materials are used to adjust the supply plan for recycled materials. The amount of product recovered as recycled material (recovery volume) fluctuates depending on constraints on the manufacturing, sale, use, or disposal of the product, and also fluctuates depending on the user's decision-making tendencies regarding reuse. In order to grasp such fluctuations in the recovery volume of products, the circular production control device according to Embodiment 4 predicts the supply volume of recycled materials from the product using prediction data that includes the above constraints and decision-making tendencies. The prediction data consists of "past usage target values ​​of recycled materials" and "actual usage data of recycled materials".

[0113] (Basic configuration of a circular production control system) Figure 11 is a functional block diagram showing an example configuration of the circular production control device 1C according to Embodiment 4. The circular production control device 1C receives past usage target value data and actual usage data of recycled materials as input, predicts the supply quantity of recycled products using the input data, and outputs this prediction result to an external device. The external device is a display device that displays the prediction result of the supply quantity of recycled products. As shown in Figure 11, the circular production control device 1C comprises an input unit 11C, a prediction unit 12C, and an output unit 13C.

[0114] For example, the circular production control system 1C is implemented by a computer. The computer's memory stores programs that constitute information processing applications for realizing the functions of the input unit 11C, the prediction unit 12C, and the output unit 13C. The functions of the input unit 11C, prediction unit 12C, and output unit 13C are realized when the computer's processor executes an information processing application read from memory.

[0115] (Input section) The input unit 11C accepts predictive data that includes restrictions on the manufacture, sale, use, or disposal of the product, as well as user decision trends regarding reuse. For example, prediction data is stored in a memory device as a database. The input unit 11C retrieves prediction data from the database stored in the memory device. If the storage device is an external storage device that is connected to the circular production control device 1C via wired or wireless communication, the input unit 11C acquires the prediction data stored in the external storage device via wired or wireless communication. The memory device may be the memory unit of the circular production control device 1C. In this case, the input unit 11C acquires the prediction data stored in the memory unit via a signal line. Furthermore, the input unit 11C may accept prediction data input from a user using an input device not shown in Figure 11.

[0116] (Past usage targets for recycled materials) "Past usage targets for recycled materials" is data that includes usage targets for recycled materials set in past production plans. For example, it includes target values ​​for recycled materials used in the past year. This usage target data includes "type of recycled material," "target usage amount," and "usage period." As described in Embodiment 3, "type of recycled material" is a classification of recycled material, such as recycled plastic, recycled metal, recycled paper, or recycled glass. "Target usage amount" is the target amount of recycled material set in the user's production plan. "Usage period" is the target period for using the recycled material. Furthermore, the historical usage target data for recycled materials includes "supplier," "purchase cost," "intended use," "quality assessment," "environmental impact assessment," "recycling rate," "post-use condition," "legal compliance status," "user," and "supply chain information."

[0117] "Supplier" refers to data indicating information about the company or facility that supplied the recycled material. "Supplier" influences users' decision-making tendencies regarding reuse. For example, if the supplier is a reliable company or facility, users tend to perceive that the recycled material meets quality and safety standards and are more inclined to choose recycled products. In particular, if the supplier is certified or clearly demonstrates its sustainability efforts, users are more likely to feel that it contributes to environmental protection or ethical considerations, leading to a more positive decision. On the other hand, if information about the supplier is unclear and quality control standards are not provided, users tend to doubt the reliability of the recycled material and are more likely to avoid recycled products.

[0118] "Purchase cost" is data that shows the cost incurred in purchasing recycled materials. "Purchase cost" influences users' decision-making tendencies regarding reuse. For example, if the purchase cost of recycled materials is significantly lower than that of new materials, users are more likely to actively choose recycled materials from a cost-saving perspective. On the other hand, even if the purchase cost is higher than that of new materials, if the reason is explained based on added value such as improved quality or reduced environmental impact, users are more likely to be convinced by the use of recycled materials, and their decision may lean more positively.

[0119] "Purpose of Use" is data that indicates the specific application or project in which the recycled material is used. "Purpose of Use" influences users' decision-making tendencies regarding reuse. For example, when recycled materials are used in public infrastructure projects, environmental conservation activities, or products that contribute to the local community, users are more likely to recognize their significance and social impact and choose to reuse them.

[0120] "Quality evaluation" refers to data that indicates an assessment of the quality or performance of recycled materials. For example, if the evaluation is high, users perceive recycled materials as being of equal or superior quality to new materials, ensuring safety and reliability, and are more likely to actively choose recycled materials. On the other hand, if there are negative evaluations or concerns regarding quality or performance, users are more likely to choose new materials, considering the risks associated with using recycled materials (shorter product life or failure). This may lead to a decrease in the utilization rate of recycled materials.

[0121] An "environmental impact assessment" is data that shows the evaluation of the environmental impact of using recycled materials. If the impact assessment is positive, users can feel satisfied that they are fulfilling their social responsibility by choosing recycled materials, and their motivation to reuse will increase. On the other hand, if the assessment is insufficient or targets are consistently not met, confidence in the effectiveness of reuse will decrease, and reuse may decline.

[0122] The "recycling rate" is data that indicates the recycling rate of recycled materials or the reuse rate of recycled materials. When the recycling rate is high, users are more likely to recognize that recycled materials contribute efficiently to resource circulation and to have a positive attitude towards choosing recycled materials. In particular, if the reuse rate is shown with specific numbers and it is clear that recycled materials can be reused many times, users are more likely to feel the effect of reducing environmental impact and saving resources. On the other hand, when the recycling rate or reuse rate is low, users may have doubts about the environmental value or economic viability of recycled materials and are more likely to choose new materials.

[0123] "Condition after use" refers to data indicating the condition of recycled material after use or how it was processed. For example, "condition after use" may include the degree of soiling, presence or absence of damage, deterioration of parts, and the need for cleaning or repair. In this case, if the recycled material is in relatively good condition, the user is more likely to choose to clean or make minor repairs and reuse it. On the other hand, if it is severely damaged or important parts are deteriorated, the user is more likely to choose to discard or recycle it.

[0124] "Compliance status" is data indicating compliance with laws and regulations regarding the use of recycled materials. It represents constraints on the manufacture, sale, use, or disposal of recycled materials. For example, whether the setting of targets for the use of recycled materials and their use meet legal standards or regulations influences decisions or actions throughout those processes. During the manufacturing process, if restrictions on the content of hazardous substances or energy efficiency standards are not met, the production or distribution of recycled materials may be restricted. For sales purposes, products must comply with labeling requirements or import / export regulations, and non-compliant products cannot be put on the market. During the usage phase, recycled materials that do not meet safety or environmental standards tend to pose a higher risk to users and are therefore often avoided. Regarding disposal, if proper disposal or recycling obligations are regulated, violations will be subject to penalties. Thus, data on compliance with regulations is crucial information supporting decision-making throughout the entire product lifecycle.

[0125] "User information" refers to information about the person responsible for using the recycled material. When it is clearly stated that the person responsible has specific knowledge, experience, or qualifications, users feel that the recycled material has been appropriately selected and that quality or environmental considerations have been adequately taken into account, increasing their confidence in the recycled product. In particular, if the person responsible has expertise in environmental technology or recycling, users perceive that there is a scientific basis for the selection or use of the recycled material, and they tend to choose recycled materials with greater confidence. On the other hand, if the information about the person responsible is unclear or does not convey expertise, users may have doubts about the reliability or effectiveness of the recycled material and hesitate to choose the recycled product.

[0126] "Supply chain information" is data that shows the supply chain through which recycled materials were used. For example, if recycled materials are collected locally and processed at a local recycling facility, users will be more inclined to choose recycled products, as they will be aware of their contribution to the local economy. On the other hand, if the supply chain is opaque or involves processes with a high environmental impact, users will have doubts about the environmental value of recycled materials and are more likely to choose new materials.

[0127] As mentioned above, past usage targets and actual usage data for recycled materials are predictive data that show constraints on the manufacturing, sale, use, or disposal of products, as well as user decision-making trends related to reuse. Past usage target values ​​and actual usage data for recycled materials are collected using various means and stored as a database in the above-mentioned storage device. For example, usage data can be collected by embedding sensors or IoT devices in the product, via the cloud or a dedicated application, and stored as a database in a storage device. The input unit 11C may also accept information entered by the user as a questionnaire or feedback, as "past usage target values ​​for recycled materials" and "recycled material usage data."

[0128] (Prediction section) The prediction unit 12C uses the prediction data received by the input unit 11C to predict the amount of recycled materials to be supplied from the product. For example, the prediction unit 12C calculates the difference between past usage target values ​​for recycled materials and actual usage data for recycled materials, predicts future usage of recycled materials based on the calculated difference, and predicts the supply quantity of recycled materials using the usage amount.

[0129] (Output section) The output unit 13C outputs the predicted supply quantity of recycled materials to an external device. For example, external devices may include displays, printers, storage devices, mobile devices, and digital signage connected to a PC or server. The forecast results are primarily provided to the user in a visualized form, displayed on the screen in graph or table format. For instance, monthly trends in supply or regional distribution may be graphed and made available for real-time viewing. Alternatively, the report can be saved as data in PDF or CSV format and printed out, or saved to cloud storage for sharing. Furthermore, the forecast results may be accessible from outside the office using a smartphone or tablet PC. In industrial settings, digital signage may be used to share supply forecast results with all relevant parties within the factory or facility.

[0130] Figure 12 is a functional block diagram showing an example configuration of the prediction unit 12C. In Figure 12, the prediction unit 12C comprises an acquisition unit 121C, a difference calculation unit 126, and a usage prediction unit 127. The acquisition unit 121C acquires usage target value data set in past production plans for recycled materials. The difference calculation unit 126 calculates the difference between the past usage target value of recycled materials and the actual usage data of recycled materials used by users. The usage amount prediction unit 127 predicts the future usage amount of recycled materials based on the difference calculated by the difference calculation unit 126 and calculates the supply amount of recycled materials using the usage amount.

[0131] Next, a circular production control method according to Embodiment 4 will be described. Figure 13 is a flowchart illustrating the cyclic production control method according to Embodiment 4, and shows in detail the processing of the prediction unit 12C in Figure 11. The input unit 11C accepts predictive data that includes restrictions on the manufacture, sale, use, or disposal of the product, as well as user decision trends regarding reuse (step ST1C). The forecasting data consists of "past usage target values ​​for recycled materials" and "actual usage data for recycled materials." "Past usage target values ​​for recycled materials" are, for example, data on the usage target values ​​for recycled materials set in past production plans. "Past usage data for recycled materials" are, for example, data on the usage target values ​​for recycled materials over the past year. "Actual usage data for recycled materials" are the actual usage data for recycled materials. The forecasting data is stored in the database mentioned above.

[0132] Next, the prediction unit 12C uses the prediction data read from the database to predict the supply volume of recycled materials from the product. The acquisition unit 121C acquires the usage data of recycled materials from the database, which was received as input by the input unit 11C, and outputs it to the difference calculation unit 126.

[0133] The difference calculation unit 126 obtains past usage target value data for recycled materials from the database (step ST2C). Then, the difference calculation unit 126 calculates the difference between the past usage target value for recycled materials and the actual usage data for recycled materials (step ST3C). For example, the difference calculation unit 126 compares past usage target data for recycled materials with actual usage data for recycled materials to calculate the difference between the usage target value for recycled materials over the past year and the actual usage value of recycled materials used.

[0134] The usage prediction unit 127 predicts the future usage of the reusable product based on the difference calculated by the difference calculation unit 126 (step ST4C). For example, the usage prediction unit 127 predicts the future usage of reusable products using a prediction model. The prediction model is a model that, when a calculated difference is input, outputs a predicted value for the future usage of reusable products. The procedure for predicting the future use of recycled materials is as follows: (1) Data collection and preprocessing The usage forecasting unit 127 collects usage data of recycled materials, identifies missing and abnormal values ​​in the collected usage data, and performs preprocessing such as imputing missing values ​​and removing abnormal values. (2) Feature extraction The usage forecasting unit 127 extracts feature quantities from past usage target data and actual usage data of recycled materials. These feature quantities are, for example, the contents of various data items included in the data, such as "usage amount," "usage date," "usage location," "supplier," and "purchase cost," and are extracted in numerical form. (3) Model Selection For prediction, a statistical model or a machine learning algorithm is selected. Examples of statistical models include regression analysis and time series analysis models. Examples of machine learning algorithms include random forests and support vector machines. (4) Model training The usage prediction unit 127 trains a selected model using the difference between past usage target value data for recycled materials obtained over different periods as training data. For example, it constructs a predictive model that predicts the amount of recycled materials used in a fixed period in the future by using the difference between usage target value data for recycled materials set in production plans for different periods within the past year. Furthermore, (3) and (4) may be performed when predicting the amount of recycled material to be used, and the prediction model may be constructed in advance. (5) Prediction The usage prediction unit 127 uses a trained prediction model to predict the future usage of recycled materials. For example, by inputting the difference between the target usage value and the actual usage value of recycled materials over the past year into the prediction model, the prediction model outputs the usage amount of recycled materials for the next year. This makes it possible to accurately predict the usage amount of recycled materials, taking into account constraints on the manufacturing, sale, use, or disposal of products, as well as fluctuations in the amount of products recovered due to user decision-making trends regarding reuse.

[0135] Alternatively, the prediction unit 12C may not include the difference calculation unit 126, and only the usage prediction unit 127 may predict the amount of recycled material used using past usage target value data and actual usage data of recycled material. In (4), the usage prediction unit 127 constructs a prediction model that predicts the future amount of recycled material used based on past usage data of recycled material. For example, when predicting the amount of recycled material used, the usage prediction unit 127 uses a prediction model that outputs a prediction result for the amount of recycled material used in the next year when it receives usage data of recycled material for the past year. Even in this way, it is possible to accurately predict the amount of recycled material used, taking into account constraints on the manufacture, sale, use, or disposal of the product, and fluctuations in the amount of product collected due to user decision trends regarding reuse.

[0136] Next, the usage prediction unit 127 predicts the amount of recycled materials to be supplied, indicating how much of the recycled material should be supplied from recycled materials, based on the predicted future usage of the recycled materials. In supply forecasting, using regression analysis or machine learning models based on historical data allows for highly accurate predictions that reflect the correlations between factors influencing supply.

[0137] The output unit 13C outputs the predicted supply quantity of recycled materials (step ST5C). For example, the output unit 13C generates output data that includes the predicted supply quantity of reusable materials obtained from the usage prediction unit 127, and outputs this output data to an external device. The output data includes the supply quantity of reusable materials for each city, the types of parts that can be supplied, and the supply timing, and this data is presented in report format. Furthermore, a feedback loop may be provided between the circulating production control device 1C and the external device to adjust the supply plan as needed. A feedback loop for adjusting supply plans is a process for modifying or optimizing the supply plan for reusable products based on actual usage data or forecast results in the supply chain. For example, it involves collecting actual usage data such as the supply or demand volume of reusable products, product usage status, and waste recovery rates, and analyzing the collected data to identify the difference between forecast and actual values ​​at the planning stage. This makes it possible to identify problems such as supply shortages or oversupplies and take specific countermeasures such as improving supply efficiency or recovery rates, or reviewing waste treatment processes.

[0138] The prediction unit 12C may use a Kalman filter to predict the supply quantity of recycled materials from the prediction data. For example, the prediction unit 12C predicts the supply quantity of recycled materials by applying data assimilation using a Kalman filter to past usage target value data and actual usage data of recycled materials. This allows for real-time adjustment of the prediction of the supply quantity of recycled materials. Furthermore, when the prediction unit 12C predicts the supply volume of recycled products, it may use a large-scale language model that performs predictions by recognizing the prediction data, or it may use various generative AIs.

[0139] As described above, in the circular production control device 1C according to Embodiment 4, the prediction data consists of the usage target value set in past production plans for recycled materials and the actual usage data of recycled materials currently being used by the user. The prediction unit 12C calculates the difference between the usage target value and the actual usage data, predicts the future usage amount of recycled products based on the calculated difference, and predicts the supply amount of recycled products using the usage amount. As a result, the circular production control device 1C can accurately predict the supply amount, taking into account constraints on the manufacturing, sale, use, or disposal of products, and fluctuations in the amount of products recovered due to the user's decision-making tendencies regarding reuse.

[0140] In the circular production control device 1C according to Embodiment 4, the prediction unit 12C uses a Kalman filter to predict the supply amount of reusable products from prediction data. As a result, the circular production control device 1C can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends.

[0141] In the circular production control device 1C according to Embodiment 4, the prediction unit 12C predicts the supply amount of reusable products from prediction data using a machine learning model. As a result, the circular production control device 1C can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends.

[0142] Embodiment 5. The circular production control device according to Embodiment 5 is a device that predicts the supply volume of reusable products, similar to Embodiment 1, and the prediction results of the supply volume of reusable products are used to adjust the supply plan for reusable products. The amount of product recovered as reusable products (recovery volume) fluctuates depending on constraints on the manufacturing, sale, use, or disposal of the product, and also fluctuates depending on the user's decision-making tendencies regarding reuse. In order to grasp such fluctuations in the recovery volume of products, the circular production control device according to Embodiment 5 predicts the supply volume of reusable products from products using prediction data that includes the above constraints and decision-making tendencies. The prediction data is "waste processing performance data".

[0143] (Basic configuration of a circular production control system) Figure 14 is a functional block diagram showing an example configuration of the circular production control device 1D according to Embodiment 5. The circular production control device 1D receives past usage target value data and actual usage data of recycled materials as input, predicts the supply quantity of recycled products using the input data, and outputs this prediction result to an external device. The external device is a display device that displays the prediction result of the supply quantity of recycled products. As shown in Figure 14, the circular production control device 1D comprises an input unit 11D, a prediction unit 12D, and an output unit 13D.

[0144] For example, the circular production control system 1D is implemented by a computer. The computer's memory stores programs that constitute information processing applications for realizing the functions of the input unit 11D, the prediction unit 12D, and the output unit 13D. The input unit 11D, prediction unit 12D, and output unit 13D functions are realized when the computer's processor executes an information processing application read from memory.

[0145] (Input section) The input unit 11D accepts predictive data that includes constraints on the manufacture, sale, use, or disposal of the product, as well as user decision trends regarding reuse. For example, prediction data is stored in a memory device as a database. The input unit 11D retrieves prediction data from the database stored in the memory device. If the storage device is an external storage device that is connected to the circular production control device 1D via wired or wireless communication, the input unit 11D acquires the prediction data stored in the external storage device via wired or wireless communication. The memory device may be a memory unit provided by the circular production control device 1D. In this case, the input unit 11D acquires prediction data stored in the memory unit via a signal line. Furthermore, the input unit 11D may accept prediction data input from a user using an input device not shown in Figure 14.

[0146] (Waste disposal record) The "waste processing performance data" includes information such as "type of waste," "amount processed," "processing date," "processing method," and "sale price." "Type of waste" refers to classifications of waste, such as general waste (household or business waste), industrial waste (construction waste or chemical waste), and hazardous waste (medical waste or heavy metal-containing materials). "Amount processed" is the actual amount of waste that was processed. "Processing method" refers to the method used to process the waste, such as incineration, landfill, or recycling. Furthermore, the waste treatment performance data includes "source of generation," "transporter," "treatment facility," "treatment cost," "environmental impact assessment," "legal compliance status," "recycling rate," "treatment personnel," "treatment time," and "condition of the waste after treatment."

[0147] "Origin" is data indicating the source of waste related to waste disposal records. "Origin" influences users' decision-making tendencies regarding reuse. For example, if the origin is clear, users perceive that waste collection or disposal is properly managed, and their confidence in recycled materials increases. On the other hand, if the origin is unclear or there is a history of improper disposal, users tend to have doubts about the quality or safety of recycled products and are more likely to avoid choosing them.

[0148] "Transporter" refers to information about the company that transported the waste. Information about the waste transporter influences users' decision-making tendencies regarding recycling. For example, if a transporter is shown to be reliable or have a proven track record, users will perceive that the waste has been properly processed and supplied as recycled material in a safe and environmentally friendly manner. On the other hand, if a transporter has a history of improper handling or if the company information is unclear, users may have doubts about the quality or safety of the recycled material and are more likely to choose new materials.

[0149] "Processing facility" refers to information about the facility where the waste was processed. Information about the waste processing facility influences users' decision-making tendencies regarding reuse. For example, if it is clearly stated that the processing facility processes the waste according to appropriate technology or standards, users will perceive that the recycled product is safe and of good quality, and will be more motivated to choose the recycled product. On the other hand, if information about the processing facility is unclear, or if there has been a history of improper processing or violations of environmental standards, users may doubt the reliability or safety of the recycled product and hesitate to choose it.

[0150] "Processing costs" are data that shows the expenses incurred in processing waste. The cost of waste processing influences users' decision-making regarding reuse. For example, clearly stating processing costs allows users to understand the costs involved in proper waste disposal and increases their motivation to choose recycled materials as a means of reducing those costs. On the other hand, if processing costs are too low, concerns may arise that waste is not being processed properly, and doubts about the reliability or quality of recycled materials may cause users to be more cautious in their decision-making.

[0151] An "environmental impact assessment" is data that shows an evaluation of the environmental impact of waste treatment. The evaluation of the environmental impact of waste treatment influences users' decision-making tendencies regarding reuse. For example, if the results of the environmental impact assessment are clear and show to what extent waste treatment contributes to reducing greenhouse gas emissions or conserving resources, users will recognize that choosing recycled materials is an action that directly leads to environmental improvement, and their motivation to reuse will be strengthened. On the other hand, if the environmental impact is not sufficiently evaluated or the evaluation results are negative, users may be more inclined to refrain from choosing recycled products.

[0152] "Compliance status" is data that indicates compliance with laws and regulations regarding waste disposal. It also indicates constraints on the manufacture, sale, use, or disposal of recycled materials. For example, whether waste disposal meets legal standards or regulations influences decisions or actions throughout those processes. Where proper disposal or recycling obligations are regulated, violations are subject to penalties. Thus, compliance data is crucial information supporting decision-making throughout the entire product lifecycle.

[0153] The "recycling rate" is data that shows the percentage of waste that has been recycled. The percentage of recycled waste influences users' decision-making tendencies regarding reuse. For example, if this rate is high, users tend to perceive the recycling system as functioning effectively and have a stronger sense of confidence in choosing reusable items. On the other hand, if the recycling rate is low, users may feel that the supply of reusable items is insufficient or may have doubts about the reliability of the system as a whole.

[0154] "Processing personnel" refers to information about the person responsible for handling the waste. Information about the person responsible for waste handling influences users' decision-making tendencies regarding recycling. For example, if the personnel's expertise, qualifications, or experience are clearly stated, users will perceive that the waste has been handled appropriately and safely, and their confidence in the recycled material will increase. On the other hand, if information about the personnel is unclear or if there is a history of improper handling, users may have doubts about the quality or safety of the recycled product and hesitate to make a choice.

[0155] "Processing time" is data that indicates the time taken to process waste. The time taken to process waste influences users' decision-making tendencies regarding reuse. For example, if the processing time is short, users tend to perceive that the waste is being processed quickly and efficiently, and have confidence that the supply of recycled materials is stable. On the other hand, if the processing time is long, users may be concerned about the risk of delays in the supply of recycled materials or increased costs, and may be more inclined to choose new materials.

[0156] "Condition of treated waste" is data that indicates the form or state of the treated waste. The form or state of treated waste influences users' decision-making tendencies regarding reuse. For example, if the waste is transformed into a uniform and stable form after treatment, users tend to feel that the reused product is of reliable quality and are more likely to choose it. Also, if the treated state is easy to use and provided in a form suitable for the reuse process, users are more inclined to choose it because it reduces the cost or effort involved in introducing the reused product. On the other hand, if the treated waste is in a non-uniform state or contains foreign matter, users are more likely to have doubts about its quality or safety and choose new materials.

[0157] As mentioned above, waste treatment performance data is predictive data that shows constraints on the manufacture, sale, use, or disposal of products, as well as user decision-making trends regarding reuse. Waste processing performance data is collected using various means and stored as a database in the aforementioned storage device. For example, processing performance data can be collected by embedding sensors or IoT devices in the product, via the cloud or a dedicated application, and stored as a database in a storage device. Alternatively, the input unit 11C may accept information entered by the user as a questionnaire or feedback, as "waste processing performance data."

[0158] (Prediction section) The prediction unit 12D uses the prediction data received as input by the input unit 11D to predict the amount of recycled materials to be supplied from the product. For example, the prediction unit 12D predicts the supply of reusable materials that can be recovered from waste based on waste processing performance data.

[0159] (Output section) The output unit 13D outputs the predicted supply quantity of reusable materials to an external device. For example, external devices may include displays, printers, storage devices, mobile devices, and digital signage connected to a PC or server. The forecast results are primarily provided to the user in a visualized form, displayed on the screen in graph or table format. For instance, monthly trends in supply or regional distribution may be graphed and made available for real-time viewing. Alternatively, the report can be saved as data in PDF or CSV format and printed out, or saved to cloud storage for sharing. Furthermore, the forecast results may be accessible from outside the office using a smartphone or tablet PC. In industrial settings, digital signage may be used to share supply forecast results with all relevant parties within the factory or facility.

[0160] Figure 15 is a functional block diagram showing an example configuration of the prediction unit 12D. In Figure 15, the prediction unit 12D includes an acquisition unit 121D and a supply quantity prediction unit 128. The acquisition unit 121D acquires waste processing performance data. The supply quantity forecasting unit 128 forecasts the supply quantity of reusable materials based on the waste processing performance data acquired by the acquisition unit 121D.

[0161] Next, a circular production control method according to Embodiment 5 will be described. Figure 16 is a flowchart illustrating the cyclic production control method according to Embodiment 5, and shows in detail the processing of the prediction unit 12D in Figure 14. The input unit 11D accepts predictive data that includes restrictions on the manufacture, sale, use, or disposal of the product, as well as user decision trends regarding reuse (step ST1D). The data used for prediction is "waste processing performance data." This prediction data is stored in the database mentioned above.

[0162] Next, the prediction unit 12D uses the prediction data read from the database to predict the supply volume of recycled products from the products. The acquisition unit 121D acquires waste processing performance data from the database, which has been received as input by the input unit 11D, and outputs it to the supply volume prediction unit 128.

[0163] The supply volume forecasting unit 128 predicts the supply volume of reusable materials based on waste processing performance data (step ST2D). For example, the supply quantity forecasting unit 128 uses a forecasting model to predict the supply quantity of reusable products. The prediction model is a model that, when given historical waste processing data, outputs a predicted value for the supply of reusable materials. The procedure for predicting the supply volume of recycled materials is as follows: (1) Data collection and preprocessing The supply volume forecasting unit 128 collects past waste processing data, identifies missing values ​​and outliers in the collected processing data, and performs preprocessing such as imputing missing values ​​and removing outliers. (2) Feature extraction The supply volume forecasting unit 128 extracts feature quantities from past waste processing data. These feature quantities are the contents of various data items included in the data, such as "processing volume," "sale price," "processing date," and "processing method," and are extracted as numerical values. (3) Model Selection For prediction, a statistical model or a machine learning algorithm is selected. Examples of statistical models include regression analysis and time series analysis models. Examples of machine learning algorithms include random forests and support vector machines. (4) Model training The supply volume forecasting unit 128 trains a selected model using past waste processing data obtained over different periods as training data. For example, it constructs a forecasting model that predicts the supply volume of reusable materials for the next year using waste processing data from the past year. (3) and (4) may be performed when forecasting supply quantities, and the forecasting model may be constructed in advance. (5) Prediction The supply quantity forecasting unit 128 predicts the supply quantity of reusable materials using a trained forecasting model. For example, by inputting waste processing data for the past year into the forecasting model, the forecasting model outputs the supply quantity of reusable materials for the next year. This makes it possible to accurately predict the amount of recycled material used, taking into account constraints on the manufacturing, sale, use, or disposal of products, as well as fluctuations in the amount of products recovered due to user decision trends regarding reuse.

[0164] The output unit 13D outputs the predicted supply quantity of recycled materials (step ST3D). For example, the output unit 13D generates output data that includes the predicted supply quantity of reusable items obtained from the supply quantity forecasting unit 128, and outputs the output data to an external device. The output data includes the supply quantity of reusable items for each city, the types of parts that can be supplied, and the supply timing, and this data is presented in report format. Furthermore, a feedback loop may be provided between the circulating production control device 1D and the external device to adjust the supply plan as needed. A feedback loop for adjusting supply plans is a process for modifying or optimizing the supply plan for reusable products based on actual usage data or forecast results in the supply chain. For example, it involves collecting actual usage data such as the supply or demand volume of reusable products, product usage status, and waste recovery rates, and analyzing the collected data to identify the difference between forecast and actual values ​​at the planning stage. This makes it possible to identify problems such as supply shortages or oversupplies and take specific countermeasures such as improving supply efficiency or recovery rates, or reviewing waste treatment processes.

[0165] The prediction unit 12D may use a Kalman filter to predict the supply quantity of reusable materials from the prediction data. For example, the prediction unit 12D predicts the supply quantity of reusable materials by applying data assimilation using a Kalman filter to past waste processing data. This allows for real-time adjustment of the prediction of the supply quantity of reusable materials. Furthermore, when predicting the supply volume of recycled materials, the prediction unit 12D may use a large-scale language model that performs language recognition on prediction data to make predictions, or it may use various generative AIs.

[0166] As described above, in the circular production control device 1D according to Embodiment 5, the prediction data is waste processing performance data, and the prediction unit 12D predicts the supply amount of reusable products that can be recovered from the waste based on the processing performance data. As a result, the circular production control device 1D can accurately predict the supply amount, taking into account constraints on the manufacture, sale, use, or disposal of products, and fluctuations in the amount of products recovered due to user decision trends regarding reuse.

[0167] In the circular production control device 1D according to Embodiment 5, the prediction unit 12D uses a Kalman filter to predict the supply amount of reusable products from prediction data. As a result, the circular production control device 1D can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends.

[0168] In the circular production control device 1D according to Embodiment 5, the prediction unit 12D predicts the supply amount of reusable products from prediction data using a machine learning model. As a result, the circular production control device 1D can accurately predict the supply amount, taking into account the above constraints and fluctuations in the amount of products recovered due to the above judgment trends. [Industrial applicability]

[0169] The circular production control system described herein can be used, for example, to formulate supply plans for various recycled materials.

[0170] Furthermore, it is possible to combine each embodiment, modify any component of each embodiment, or omit any component in each embodiment. [Explanation of Symbols]

[0171] 1,1A~1D Circular production control device, 11,11A~11D Input unit, 12,12A~12D Prediction unit, 13,13A~13D Output unit, 100 Processing circuit, 101 Input interface, 102 Output interface, 103 Input / output interface, 104 Input device, 105 Display, 106 Storage device, 107 Processor, 108 Memory, 121,121A~121D Acquisition unit, 122,122A Degradation degree calculation unit, 123 Component lifespan calculation unit, 124 Timing prediction unit, 125 Supply quantity calculation unit, 126 Difference calculation unit, 127 Usage quantity prediction unit, 128 Supply quantity prediction unit.

Claims

1. An input unit that accepts input of predictive data indicating the user's decision-making tendencies regarding reuse, including restrictions on the manufacture, sale, use, or disposal of the product, and at least two of the following: intended use, anomaly detection history, degree of use, performance data, usage cost, location of use, and trouble history. A prediction unit calculates the degree of deterioration of the product based on the prediction data, calculates the remaining lifespan of the product based on the degree of deterioration and the product's service life data, predicts the collection time of the product based on the remaining lifespan, and predicts the supply amount of reusable products obtained from the product based on the collection time. An output unit that outputs the predicted supply quantity of the aforementioned recycled products, A circular production control system equipped with the following features.

2. The aforementioned prediction data is the user's maintenance history data for the product. The prediction unit predicts the supply quantity of the reusable parts by applying data assimilation using a Kalman filter to the maintenance performance data. The circular production control device according to feature 1.

3. The aforementioned prediction data is data on the user's usage of the product. The prediction unit calculates the degree of deterioration based on the usage data, calculates the remaining lifespan of the product based on the degree of deterioration and the product's service life data, predicts the collection time based on the remaining lifespan, and predicts the supply quantity of the recycled product based on the collection time. The circular production control device according to feature 1.

4. The aforementioned prediction data is the user's maintenance history data for the product. The prediction unit calculates the degree of deterioration based on the maintenance performance data, calculates the remaining lifespan of the product based on the degree of deterioration and the product's service life data, predicts the collection period based on the remaining lifespan, and predicts the supply quantity of the recycled product based on the collection period. The circular production control device according to feature 1.

5. The aforementioned forecasting data consists of future usage target values ​​for recycled materials and past usage data for said recycled materials. The prediction unit calculates the difference between the target usage value and the actual usage data, predicts the future usage amount of the recycled product based on the calculated difference, and predicts the supply amount of the recycled product using the usage amount. The circular production control device according to feature 1.

6. The aforementioned prediction data consists of past usage target values ​​for recycled materials and actual usage data for said recycled materials. The prediction unit calculates the difference between the target usage value and the actual usage data, predicts the future usage amount of the recycled product based on the calculated difference, and predicts the supply amount of the recycled product using the usage amount. The circular production control device according to feature 1.

7. The aforementioned prediction data is waste treatment performance data, The prediction unit predicts the supply volume of reusable materials that can be recovered from the waste, based on the processing performance data. The circular production control device according to feature 1.

8. The prediction unit uses a Kalman filter to predict the supply quantity of the recycled product from the prediction data. A circulating production control device according to any one of claims 3 to 7.

9. The prediction unit uses a machine learning model to predict the supply quantity of the reusable products from the prediction data. A circulating production control device according to any one of claims 3 to 7.

10. A circular production control method using a circular production control device, The input unit receives input for predictive data that indicates the user's decision-making tendencies regarding reuse, including restrictions on the manufacture, sale, use, or disposal of the product, and at least two of the following: intended use, anomaly detection history, degree of use, performance data, usage cost, location of use, and trouble history. The prediction unit calculates the degree of deterioration of the product based on the prediction data, calculates the remaining lifespan of the product based on the degree of deterioration and the product's service life data, predicts the collection time of the product based on the remaining lifespan, and predicts the supply quantity of reusable products obtained from the product based on the collection time. The output unit outputs the predicted supply quantity of the recycled product, A circular production control method equipped with the following features.

11. Computers, An input unit that accepts input of predictive data indicating user decision trends regarding reuse, including restrictions on the manufacture, sale, use, or disposal of the product, and at least two of the following: intended use, anomaly detection history, degree of use, performance data, usage cost, location of use, and trouble history. A prediction unit calculates the degree of deterioration of the product based on the prediction data, calculates the remaining lifespan of the product based on the degree of deterioration and the product's service life data, predicts the collection time of the product based on the remaining lifespan, and predicts the supply amount of reusable products obtained from the product based on the collection time. An output unit that outputs the predicted supply quantity of the aforementioned recycled products. A program designed to function as such.

Images