Prediction device, prediction method, and program

Abstract

To enable relatively accurate prediction of the decommissioning period for final waste disposal sites. [Solution] The prediction device comprises a measurement data processing unit that calculates parameter values ​​to be set in a model that shows the relationship between the passage of time at a waste final disposal site and the value of the safety evaluation index, based on time-series data of measurement data of a safety evaluation index at a target disposal site which is a waste final disposal site for which the decommissioning period is to be predicted, and a decommissioning period prediction unit that calculates a predicted value for the decommissioning period of the target disposal site based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index, as shown by the model for which the parameter values ​​have been set.

Description

[Technical Field]

[0001] This disclosure relates to a prediction device, a prediction method, and a program. [Background technology]

[0002] If information regarding waste disposal can be obtained in advance, it is expected that this information can be utilized in waste disposal. For example, in the waste information processing method described in Patent Document 1, the waste generator notifies a server device from the generator's terminal of waste information regarding the type and amount of waste to be generated, before the waste is collected by the waste collector and transporter. The server device then predicts the amount of waste to be generated for each waste generator in the near future, such as on the day of collection and transport or the following day, based on the waste information. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Patent No. 6941378 [Overview of the project] [Problems that the invention aims to solve]

[0004] Regarding the management of final waste disposal sites, it is desirable to be able to predict the decommissioning period of these sites with the highest possible accuracy.

[0005] One example of the purpose of this disclosure is to provide a prediction device, prediction method, and program that can predict the decommissioning period of a final waste disposal site with relatively high accuracy. [Means for solving the problem]

[0006] According to a first aspect of the present disclosure, the prediction apparatus comprises: a measurement data processing unit that calculates a parameter value to be set in a model indicating a relationship between elapsed time at a waste final disposal site and the value of a safety evaluation indicator, based on time-series data of measurement data of the safety evaluation indicator at a target disposal site which is a waste final disposal site that is a target for closure period prediction; and a closure period prediction unit that calculates a predicted value of the closure period of the target disposal site based on the relationship between elapsed time at the target disposal site and the value of the safety evaluation indicator indicated by the model in which the parameter value is set.

[0007] According to a second aspect of the present disclosure, a prediction method comprises, by a computer: calculating a parameter value to be set in a model indicating a relationship between elapsed time at a waste final disposal site and the value of a safety evaluation indicator, based on time-series data of measurement data of the safety evaluation indicator at a target disposal site which is a waste final disposal site that is a target for closure period prediction; and calculating a predicted value of the closure period of the target disposal site based on the relationship between elapsed time at the target disposal site and the value of the safety evaluation indicator indicated by the model in which the parameter value is set.

[0008] According to a third aspect of the present disclosure, a program causes a computer to execute: calculating a parameter value to be set in a model indicating a relationship between elapsed time at a waste final disposal site and the value of a safety evaluation indicator, based on time-series data of measurement data of the safety evaluation indicator at a target disposal site which is a waste final disposal site that is a target for closure period prediction; and calculating a predicted value of the closure period of the target disposal site based on the relationship between elapsed time at the target disposal site and the value of the safety evaluation indicator indicated by the model in which the parameter value is set. [Effects of the Invention]

[0009] According to the present disclosure, the closure period of a waste final disposal site can be predicted with relatively high accuracy. [Brief Description of Drawings]

[0010] [Figure 1]This figure shows an example of the configuration of a prediction device according to the embodiment. [Figure 2] This figure shows a first example of the display of a physical model by the prediction device according to the embodiment. [Figure 3] This figure shows a second example of the display of a physical model by the prediction device according to the embodiment. [Figure 4] This figure shows a third example of the display of a physical model by the prediction device according to the embodiment. [Figure 5] This figure shows an example of an experiment regarding the parameter values ​​of the physical model calculated by the measurement data processing unit according to the embodiment. [Figure 6] This figure shows an example of how the strength of correlation between items is displayed by the prediction device according to the embodiment. [Figure 7] This figure shows examples of reliability calculated by the measurement data processing unit according to the embodiment, for each type of machine learning model. [Figure 8] This figure shows an example of a procedure for a process in which a prediction device according to the embodiment predicts the decommissioning period of a final waste disposal site. [Figure 9] This figure shows an example of the procedure for the measurement data processing unit according to the embodiment to obtain the calculation formula for the parameter values ​​of the physical model. [Figure 10] This figure shows an example of the procedure for selecting items to be used in the calculation formula for the parameter values ​​of the physical model by the measurement data processing unit according to the embodiment. [Figure 11] This figure shows an example of the procedure for selecting a machine learning model to be used to calculate the parameter values ​​of a physical model, according to the embodiment of the measurement data processing unit. [Figure 12] This figure shows an example of a computer configuration according to at least one embodiment. [Modes for carrying out the invention]

[0011] The following describes embodiments of the present invention, but these embodiments are not intended to limit the invention as defined in the claims. Furthermore, not all combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0012] Figure 1 is a diagram showing an example of the configuration of a prediction device according to an embodiment. In the configuration shown in Figure 1, the prediction device 100 comprises a communication unit 110, a display unit 120, an operation input unit 130, a storage unit 180, and a processing unit 190. The processing unit 190 comprises a physical model application unit 191, a measurement data processing unit 192, and an obsolescence period prediction unit 193.

[0013] The prediction device 100 predicts the decommissioning period of a final waste disposal site. The prediction device 100 may be configured using a computer such as a personal computer (PC) or workstation (WS). Hereinafter, the final waste disposal site will also be referred to simply as the disposal site. The waste final disposal sites that the prediction device 100 targets for its decommissioning period prediction are also referred to as target disposal sites.

[0014] The term "decommissioning period" for a final waste disposal site, as used here, refers to the period from the completion of waste landfilling at that site until the termination of management of that site. A final waste disposal site remains under management until safety standards are met, including leachate standards (for example, standards related to water pollution), no-abnormality temperature standards, and no-increase gas generation standards, after waste landfilling is complete.

[0015] In the case of general waste disposal sites managed by local governments, a problem may arise such as difficulty in constructing new disposal sites due to a lack of land. In response to this, the prediction device 100 predicts the decommissioning period of final waste disposal sites, and it is expected that by utilizing the disposal sites after the decommissioning period ends (i.e., by repurposing them for other uses), the land shortage can be alleviated and new disposal sites can be constructed.

[0016] Furthermore, in the case of general waste disposal sites managed by local governments, if maintenance costs during the decommissioning period exceed expectations, financial collapse is possible, leading to an increase in dysfunctional disposal sites and, consequently, an increased risk to the health and living environment of residents near these sites. In contrast, by predicting the decommissioning period of final waste disposal sites, the prediction device 100 can predict maintenance costs during the decommissioning period with relatively high accuracy, thereby reducing the risk of financial collapse and enabling the maintenance of sound disposal sites.

[0017] Furthermore, in the case of industrial waste disposal sites managed by private companies, if maintenance costs during the decommissioning period exceed expectations, it could lead to business failure, resulting in a decrease in the number of operators and, ultimately, illegal dumping of industrial waste. In contrast, by predicting the decommissioning period of final waste disposal sites using the prediction device 100, it is expected that maintenance costs during the decommissioning period can be predicted with relatively high accuracy, thereby reducing the risk of business failure.

[0018] The communication unit 110 communicates with other devices. For example, the communication unit 110 may receive measurement data from sensors that perform measurements at the waste final disposal site. Also, if the prediction device 100 is operated as a server device, the communication unit 110 may transmit the prediction results of the decommissioning period to the terminal device of the waste final disposal site administrator.

[0019] The display unit 120 has a display screen such as a liquid crystal panel or an LED (Light Emitting Diode) panel, and displays various images. For example, the display unit 120 may display various information related to the decommissioning period forecast, such as the forecast results for the decommissioning period and the data used to forecast the decommissioning period.

[0020] The operation input unit 130 is configured to include, for example, input devices such as a keyboard and a mouse, and accepts user operations. For example, the operation input unit 130 may be configured to accept user operations that perform various settings related to the prediction of the decommissioning period, such as user operations that set coefficient values ​​for a formula for predicting the decommissioning period.

[0021] The memory unit 180 stores various types of data. For example, the memory unit 180 may store data used to predict the decommissioning period for each waste final disposal site. The processing unit 190 controls various parts of the prediction device 100 to perform various processes. The functions of the processing unit 190 are performed, for example, by the CPU (Central Processing Unit) of the prediction device 100 reading a program from the storage unit 180 and executing it.

[0022] The physical model application unit 191 calculates predicted values ​​for safety evaluation indicators at the waste final disposal site based on information regarding the waste final disposal site. The safety evaluation indicators referred to here are the values ​​of items defined as indicators for evaluating the environmental impact of the waste final disposal site. The following explanation uses the chloride ion concentration of leachate as a safety evaluation index as an example. However, the safety evaluation index used by the prediction device 100 is not limited to a specific one. Various indicators that serve as criteria for the closure of final waste disposal sites can be used as safety evaluation indexes.

[0023] The physical model application unit 191 calculates predicted chloride ion concentrations for each elapsed time, for example, using a model that shows the relationship between the elapsed time (years and months) since the completion of landfilling of waste at a final waste disposal site and the chloride ion concentration of leachate at that final waste disposal site. The predicted chloride ion concentrations for each elapsed time calculated by the physical model application unit 191 correspond to an example of time-series data of predicted safety evaluation index values ​​at the target disposal site.

[0024] The model used by the physical model application unit 191 is also referred to as the physical model. A model of a final waste disposal site can be expressed, for example, by a system of differential equations like equation (1).

[0025]

number

[0026] C m This indicates the chloride ion concentration in the movable aqueous phase within the interstitial gaps of the reclaimed land layer. C im This indicates the chloride ion concentration in the static water within the interstitial gaps of the reclaimed land layer. τ represents the elapsed time since the completion of landfilling of waste into the final waste disposal site. In equation (1), the elapsed time τ is expressed as a dimensionless value obtained by dividing the actual elapsed time by the time required to pass through the landfill layer of the disposal site. Here, the time required to pass through the landfill layer is the value obtained by dividing the amount of leachate by the average infiltration rate, which is the thickness of the landfill layer, by the cross-sectional area of ​​the disposal site. C m The value and C im The value of is calculated for each value of τ as the solution to the system of equations given by equations (1) and (2). In particular, the chloride ion concentration C in the static aqueous phase. im This corresponds to the chloride ion concentration of the leachate. The chloride ion concentration of the leachate is also called the leachate concentration.

[0027] ξ represents the depth vector of the reclaimed layer. In equation (1), the depth vector ξ of the reclaimed layer is expressed as a dimensionless value obtained by dividing the depth vector of the reclaimed layer by the thickness of the reclaimed layer.

[0028] Here, from the perspective of predicting leachate concentration, it is conceivable to predict the time change of concentration at only one point in the deepest part of the landfill layer. On the other hand, the concentration at the deepest part is affected by the landfill waste above it. Therefore, let ξ be a vector whose elements are discrete values ​​representing the depth of the landfill layer, and let equation (1) be a differential equation in the direction from the surface to the deepest part. It is expected that the physical model application unit 191 can predict leachate concentration with higher accuracy by calculating the sewage concentration generated from each depth portion of the landfill layer from the surface to the deepest part using equation (1). Furthermore, by dividing the depth of the reclaimed layer by its thickness to make it dimensionless, the vector ξ has discrete values ​​as elements within the range of 0 to 1. The surface is also denoted as ξ=0. The deepest part of the reclaimed layer is also denoted as ξ=1.

[0029] Pe represents the Peclet number. Here, the Peclet number Pe indicates the ratio of advection to diffusion of water in the pores of a landfill layer. Specifically, as the Peclet number Pe, a value obtained by dividing the product of the thickness of the landfill layer and the seepage velocity by the dispersion coefficient is used. Here, the dispersion coefficient is given as the sum of the mechanical dispersion coefficient and the effective molecular diffusion coefficient. The mechanical dispersion coefficient is the product of the seepage velocity and the dispersion length. The effective molecular diffusion coefficient is the product of the molecular diffusion coefficient in water (a constant value) and the tortuosity. The values of the dispersion length and the tortuosity can be estimated using empirical formulas from previous studies, for example.

[0030] K * represents the elution rate. In formula (1), K * = 0. This is because, in the process from the start of landfilling to the closure of the disposal site, all pollutants have been completely eluted over the sufficient years from the start of landfilling to the completion of landfilling, and it is considered that the influence of further elution of pollutants on the estimation of the closure period after the completion of landfilling can be ignored.

[0031] a represents the elution index. As described above, when K * = 0, the value of the elution index a does not affect the estimation of the closure period. Sh represents the Sherwood number. The Sherwood number Sh may be treated as a parameter whose value is determined such that formula (1) fits the actually measured data through data fitting.

[0032] θ represents the volumetric water content. The value of the volumetric water content θ can be obtained through on-site surveys. Alternatively, the value of the volumetric water content θ may be set through literature research or the like. θ m represents the proportion of movable water in pore water. θ im represents the proportion of immovable water in pore water. θ im = θ - θ m can be expressed as follows. For practical purposes, θ m = θ / 2, θ im = θ / 2 can be set. Note that the value of θ m and the value of θ imThe error between the actual value and the value of is reflected in the Sherwood number Sh during data fitting and is not expected to have a significant impact on the estimation of the decommissioning period.

[0033] Here, we approximate equation (1) bilinearly using equation (2).

[0034]

number

[0035] In this context, a bilinear function is a polyline formed by connecting two straight lines. The point where the two lines connect corresponds to the inflection point of the polyline. Equation (2) is a1(τ-τ p )=a²(τ-τ) p This is a bilinear model (two-line approximation model) that shows bilinearity with an inflection point at τ such that ). In the graph showing bilinearity of equation (2), τ is taken as the horizontal axis and log(C m / C m_peak ) is taken as the vertical axis. In equation (2), C im (τ,ξ=1) represents the leachate concentration at the deepest part of the reclaimed layer (ξ=1) at dimensionless time τ. C m_peak This is the peak concentration (C) in the measured data to date. m It shows the maximum value of C. m_peak This is the leachate concentration C at approximately the time of completion of landfill. m It corresponds to this.

[0036] ξ p ξ indicates the position of the inflection point. p This corresponds to a parameter that gives the bilinear shape. a1 represents the slope of the straight line in the first stage of the bilinear circuit. Here, the first stage refers to the point in time before the inflection point (the point in time when the value of elapsed time τ is smaller than the value of τ at the inflection point). a1 is also called the first decay constant or the first half-life. a2 represents the slope of the straight line in the later stage of the bilinear signal. Here, the later stage refers to a point in the future beyond the inflection point (a point in time when the value of elapsed time τ is greater than the value of τ at the inflection point). a2 is also called the second decay constant or the second half-life.

[0037] Position of the inflection point ξ p The first damping constant a1 and the second damping constant a2 correspond to examples of parameters in a physical model (a model of a final waste disposal site). The position of the inflection point ξ p The first damping constant a1 and the second damping constant a2 are also referred to as the parameters of equation (2). ξ p The values ​​of , a2, and a2 correspond to examples of parameter values ​​to be set in the waste final disposal site model. In data fitting, equation (2) approximates the measured data, ξ p Determine the value of a2, the value of a2, and the value of a2.

[0038] The physical model application unit 191 uses equation (2) to calculate the chloride ion concentration of the leachate (leachate concentration C) at each elapsed time τ. m Calculate the predicted value of ). However, the physical model application unit 191 is the leachate concentration C m The model used to predict the leachate concentration C is not limited to a specific one. For example, the physical model application unit 191 may perform a simulation using equation (1), and use equation (1) to predict the leachate concentration C. m You can also try to predict it.

[0039] On the other hand, by using a model represented by a relatively simple equation as shown in equation (2) in the physical model application unit 191, it is expected that, for example, a user of the prediction device 100 will be able to relatively easily understand the calculations of the physical model application unit 191 (calculations using the physical model) and evaluate the validity of those calculations. In particular, it is expected that even if the user does not have detailed knowledge of the changes in leachate concentration over time, they will be able to relatively easily understand the calculations of the physical model application unit 191 and evaluate the validity of those calculations. The fact that equation (2) is represented by a relatively simple graph such as a bilinear graph means that users can relatively easily understand the calculations of the physical model application unit 191 and evaluate the validity of those calculations.

[0040] The measurement data processing unit 192 uses measurement data of safety evaluation indices at the target disposal site (the waste final disposal site for which the decommissioning period is predicted) to perform data fitting of a waste final disposal site model. For example, the measurement data processing unit 192 determines the values ​​of the parameters in equation (2) so that the measurement data at the waste final disposal site is approximated by equation (2).

[0041] The measurement data processing unit 192 may use regression analysis to determine the parameter values ​​of equation (2) so that the least-squares error between equation (2) and the measurement data is as small as possible. However, the method used by the measurement data processing unit 192 for data fitting is not limited to a specific method.

[0042] When the measurement data processing unit 192 acquires measurement data for safety evaluation indicators at a certain waste final disposal site, it may fit equation (2) to the measurement data for safety evaluation indicators at that waste final disposal site (accumulated measurement data and newly obtained measurement data). This allows the measurement data processing unit 192 to update the parameter values ​​of equation (2) at that waste final disposal site. By updating the parameter values ​​of equation (2) by the measurement data processing unit 192, the prediction accuracy of the decommissioning period of the waste final disposal site by the prediction device 100 is expected to improve.

[0043] Figure 2 shows a first example of the display of a physical model by the prediction device 100. For example, the processing unit 190 may control the display unit 120 to display the graph exemplified in Figure 2. The horizontal axis of the graph in Figure 2 represents the elapsed time since the completion of landfill operations at the target disposal site, as a dimensionless quantity. Specifically, the horizontal axis represents the elapsed time τ in equation (2). The vertical axis represents the leachate concentration at the target disposal site, as a dimensionless quantity. Specifically, the vertical axis represents the log(C) in equation (2). m / Cm_peak ) indicates.

[0044] In the example shown in Figure 2, the measurement data processing unit 192 measures the chloride ion concentration C in the movable water within the interlayer gaps of the landfill, measured at the target disposal site during the period from τ=0 to τ=4. m Using this method, the data points plotted in Figure 2 are calculated, and the parameter values ​​of equation (2) are calculated so that equation (2) approximates the calculated data points. Line L111 shows equation (2) after approximation (data fitting). Line L112 indicates the upper limit of the confidence interval. Line L113 indicates the lower limit of the confidence interval.

[0045] Figure 3 shows a second example of the display of the physical model by the prediction device 100. The horizontal and vertical axes of the graph in Figure 3 are the same as in Figure 2. In the example shown in Figure 3, the measurement data processing unit 192 measures the chloride ion concentration C in the movable water within the interlayer gaps of the landfill, measured at the same target disposal site as in Figure 2, during the period from τ=0 to τ=15. m Using this method, the data points plotted in Figure 3 are calculated, and the parameter values ​​of equation (2) are calculated so that equation (2) approximates the calculated data points. Line L121 shows equation (2) after approximation (data fitting). Line L122 indicates the upper limit of the confidence interval. Line L123 indicates the lower limit of the confidence interval.

[0046] Figure 4 shows a third example of the display of the physical model by the prediction device 100. The horizontal and vertical axes of the graph in Figure 4 are the same as those in Figures 2 and 3. In the example in Figure 4, the measurement data processing unit 192 measures the chloride ion concentration C in the movable water within the interlayer gaps of the landfill, measured at the same target disposal site as in Figures 2 and 3, during the period from τ=0 to τ=28. mUsing this method, the data points plotted in Figure 4 are calculated, and the parameter values ​​of equation (2) are calculated so that equation (2) approximates the calculated data points. Line L131 shows equation (2) after approximation (data fitting). Line L132 indicates the upper limit of the confidence interval. Line L133 indicates the lower limit of the confidence interval.

[0047] As time progresses, the amount of measurement data increases from Figure 2 to Figure 3, and from Figure 3 to Figure 4. The measurement data processing unit 192 recalculates the parameter values ​​of equation (2) using more data, which is expected to improve the accuracy of the physical model based on equation (2) and thus improve the accuracy of the prediction of the decommissioning period by the prediction device 100.

[0048] Furthermore, if measurement data for safety evaluation indicators at the target disposal site is unavailable, the measurement data processing unit 192 calculates the parameter values ​​to be set in the physical model based on the design specifications of the target disposal site. As a result, the prediction device 100 can predict the decommissioning period even for waste final disposal sites that are in the design phase and for which there are no measured values ​​for safety evaluation indicators.

[0049] For example, the measurement data processing unit 192 may calculate the value of the first attenuation constant a1 using an equation that uses values ​​based on the design specifications of the waste final disposal site, as exemplified in equation (3).

[0050]

number

[0051] T represents the transpose of a vector or matrix. In equation (3), the values ​​for the disposal site depth, disposal site volume, disposal site area, and the year of commencement of landfill operations are examples of values ​​based on the design specifications of the final waste disposal site. These values ​​can be obtained without the need to measure the values ​​of the safety evaluation index.

[0052] The year in which landfill operations began can be used as an indicator of the type of landfill waste. The names of landfill waste types are determined by each facility, making it difficult to use the names provided in the data from each facility as an indicator of the type of landfill waste. In contrast, the year in which landfill operations began can be used to indicate whether the landfill is relatively old, primarily consisting of food waste, or relatively new, primarily consisting of incinerated ash.

[0053] Equation (3) shows the weighted sum of the values ​​for each item: the depth of the disposal site, the volume of the disposal site, the area of ​​the disposal site, and the year the landfill started. The vector [0.662, -1.08, 1.463, -0.22] (a vertical vector in equation (3)) is a vector of weight coefficient values ​​for each item. The weight coefficients for each item correspond to the parameters in equation (3).

[0054] Alternatively, the measurement data processing unit 192 may calculate the value of the second decay constant a2 using an equation that uses values ​​based on the design specifications of the waste final disposal site, as exemplified in equation (4).

[0055]

number

[0056] In equation (4), the values ​​for the first decay constant a1, the depth of the disposal site, the area of ​​the disposal site, and the year of commencement of landfill operations are examples of values ​​based on the design specifications of the final waste disposal site. Equation (4) shows the weighted sum of the values ​​for each item: the first decay constant a1, the depth of the disposal site, the area of ​​the disposal site, and the year the landfill started. The vector [0.721, -0.26, 0.078, 0.236] (a vertical vector in equation (4)) is a vector of weight coefficient values ​​for each item. The weight coefficients for each item correspond to the parameters in equation (4).

[0057] Similarly, the measurement data processing unit 192 uses an equation that uses the item values ​​in the design specifications of the waste final disposal site to determine the position ξ of the inflection point. p You may also calculate the value of . Empirically, Cm It has been found that the inflection point is the time obtained by adding the time required to obtain 1 PV (pore volume) of infiltrating water from the time when the maximum chloride ion concentration (in the movable aqueous phase within the interlayer pores) is observed. This finding is expressed as an equation using the item values ​​in the design specifications of the final waste disposal site, and the position of the inflection point ξ is shown. p It may also be used as a formula to calculate the value of ξ. Alternatively, the position of the inflection point ξ p You may also use statistical analysis methods, such as regression analysis, to obtain the formula for calculating the value of . In this context, 1 PV represents the amount of water required to replace all sewage in the voids in the depth direction within a waste landfill. For example, if rainwater equivalent to the void volume is passed through a waste landfill filled with sewage, the sewage will be replaced by (clean) rainwater. This amount of water corresponds to 1 PV. Formulas for calculating the first damping constant a1, the second damping constant a2, and the position ξ of the inflection point. p The calculation formula is an example of a formula for calculating parameter values ​​in a physical model.

[0058] Ideally, after 1 PV of water is added to the waste landfill, the chloride ion concentration should decrease at the same rate as before, eventually reaching zero. In this case, the time change in chloride ion concentration is represented by a single straight line graph, rather than a bilinear graph (it is approximated by a single straight line). On the other hand, in reality, when wastewater in the landfill layer is replaced by rainwater, chloride ions and other elements from the static water phase may re-leach out. This may act as a turning point, causing the subsequent rate of concentration decrease to slow down (the rate of concentration decrease to become smaller than before). As a result, the time change in chloride ion concentration will be represented bilinearly (approximated bilinearly), as shown in the examples in Figures 2 to 4.

[0059] The measurement data processing unit 192 may calculate the parameter values ​​of the physical model using a predetermined formula. Alternatively, the measurement data processing unit 192 may update the formula for calculating the parameter values ​​of the physical model.

[0060] For example, as described above, the measurement data processing unit 192 may update the parameter values ​​of equation (2) at a certain waste final disposal site when it acquires measurement data for safety evaluation indicators at that waste final disposal site. Furthermore, the measurement data processing unit 192 may update the calculation formula for the parameter values ​​of equation (2) to fit the parameter values ​​of equation (2) to the measurement data of the safety evaluation index and the design specifications of all final waste disposal sites.

[0061] For example, the measurement data processing unit 192 may update the weight coefficient values ​​of equation (3) so that equation (3) fits the first decay constant a1, the depth of the landfill, the volume of the landfill, the area of ​​the landfill, and the year of commencement of landfilling for all waste final disposal sites where equation (2) has been fitted to the measurement data of the safety evaluation index and the parameter values ​​of equation (2) have been calculated. The measurement data processing unit 192 may use a multiple regression analysis method to determine the values ​​of the parameters (weight coefficients for each item) of equation (3) so as to minimize the least-squares error between equation (3) and the data. However, the method used by the measurement data processing unit 192 for data fitting is not limited to a specific method.

[0062] Furthermore, here, the objective variable (first damping constant a1, second damping constant a2, and inflection point ξ) is defined. p The explanation used, as an example, the depth of the disposal site, the volume of the disposal site, the area of ​​the disposal site, and the year of commencement of landfill as explanatory variables to express the dependent variable, but it is not limited to these. The "dependent variable" here is the variable whose value is to be calculated. The explanatory variables here are the variables referenced in order to calculate the value of the dependent variable.

[0063] For example, indicators of organic pollution such as BOD (Biochemical Oxygen Demand) or COD (Chemical Oxygen Demand) are also attenuated by microbial decomposition, which is greatly influenced by the oxygen and temperature conditions in the landfill layer. When using indicators of organic pollution such as BOD or COD as safety assessment indicators, explanatory variables may include not only the depth of the landfill, the volume of the landfill, the area of ​​the landfill, and the year of commencement of landfill operations, but also data that correlates with microbial activity, such as the temperature of the landfill layer.

[0064] Figure 5 shows an example of an experiment regarding the parameter values ​​of the physical model calculated by the measurement data processing unit 192. Figure 5 shows an example of an experiment to evaluate the first decay constant a1 calculated by inputting data from an actual waste final disposal site into equation (3). The horizontal axis of the graph in Figure 5 shows the value of the first damping constant a1 obtained by fitting equation (2) to the measurement data. The vertical axis shows the value of the first damping constant a1 calculated using equation (3). The value of the first damping constant a1 obtained by fitting equation (2) to the measurement data is also referred to as the actual value of the first damping constant a1. The value of the first damping constant a1 calculated using equation (3) is also referred to as the estimated value of the first damping constant a1.

[0065] Line L211 is the line where the actual value and estimated value of the first damping constant a1 are equal. The closer the plot in Figure 5 is to line L211, the higher the estimation accuracy of the first damping constant a1 using equation (3) can be considered. In the example in Figure 5, with the exception of two points (the first decay constant a1 for the two waste final disposal sites), the plotted points are located near line L211. From this, we can conclude that the estimation accuracy of the first decay constant a1 using equation (3) is high. Furthermore, it was confirmed that the waste disposal site corresponding to the point far from line L211 has special circumstances, such as having been submerged in water in the past.

[0066] The measurement data processing unit 192 may use items that have a high correlation with the parameters for which it calculates values, such as the first decay constant a1, for calculating the parameter values ​​of the physical model. "Depth of the disposal site," "Volume of the disposal site," "Area of ​​the disposal site," "Year of commencement of landfill operations," and the first decay constant a1 in equations (3) and (4) are examples of items that the measurement data processing unit 192 uses to calculate the parameter values ​​of the physical model. "Depth of the disposal site," "Volume of the disposal site," "Area of ​​the disposal site," and "Year of commencement of landfill operations" are examples of items related to the design specifications of the final waste disposal site. The first decay constant a1 is an example of other parameters (in this case, parameters other than the second decay constant a2).

[0067] For example, the measurement data processing unit 192 or a human may calculate the correlation coefficient between the parameter whose value is to be calculated (a parameter of the physical model) and each available item, and then select the item to be used to calculate the value of that parameter based on the calculated correlation coefficient. Furthermore, for example, a predetermined number of items may be selected as items used to calculate the value of the parameter, in order of their high correlation with that parameter. Alternatively, items whose correlation coefficient (absolute value) with the parameter to be calculated is greater than or equal to a predetermined threshold may be selected as items used to calculate the value of that parameter.

[0068] Figure 6 shows an example of how the prediction device 100 displays the strength of correlation between items. Figure 6 shows the magnitude of the correlation coefficients between the items "Year of Operation Start," "Number of Years in Landfill," "Area of ​​Disposal Site," "Volume of Disposal Site," "Depth of Disposal Site," "First Half-Life," and "Second Half-Life" as a heat map. For example, the processing unit 190 may control the display unit 120 to display the heat map illustrated in Figure 6.

[0069] In the example in Figure 6, the correlation between the first half-life (first decay constant) a1 and each item is as follows, in order of strongest correlation: "first half-life," "second half-life," "depth of the disposal site," "area of ​​the disposal site," "volume of the disposal site," "year of commencement of landfill," and "years of landfill operation." Here, to avoid reference loops during calculation, the "first half-life" itself is excluded from the items used to calculate the first half-life. Similarly, to avoid reference loops during calculation, the "second half-life" is also excluded from the items used to calculate the first half-life. Then, from the remaining items, the four items with the strongest correlation to the first half-life—"depth of the disposal site," "area of ​​the disposal site," "volume of the disposal site," and "year of commencement of landfill"—are adopted as items to be used to calculate the first half-life, and equation (3) is set up.

[0070] Furthermore, the correlation between the second half-life (second decay constant) a2 and each item is as follows, in order of strongest correlation: "second half-life," "first half-life," "depth of the disposal site," "volume of the disposal site," "year of commencement of landfill," "area of ​​the disposal site," and "number of years of landfill operation." Here, to avoid reference loops during calculation, the "second half-life" itself is excluded from the items used to calculate the second half-life. On the other hand, the "first half-life" is calculated using equation (3) without using the second half-life, so it is included in the items used to calculate the second half-life. Then, the "first half-life" and four of the remaining items that have the strongest correlation with the second half-life—"first half-life," "depth of the disposal site," "volume of the disposal site," and "year of commencement of landfill"—are adopted as items to be used to calculate the second half-life, and equation (4) is set up.

[0071] The measurement data processing unit 192 or a human may re-select the items used in the calculation formula for the parameter values ​​of the physical model at predetermined intervals, such as when measurement data for a final waste disposal site is added. This allows for the selection of more appropriate items as the amount of data increases, and it is expected that the prediction device 100 will be able to estimate the decommissioning period of the final waste disposal site with greater accuracy.

[0072] As shown in the example in Figure 6, the prediction device 100 displays the strength of the correlation between items, allowing the user to verify the basis for the calculation formula of the physical model's parameter values. In particular, by displaying the strength of the correlation between items in the form of a heatmap, the user can intuitively verify the basis for the calculation formula of the physical model's parameter values.

[0073] The prediction device 100 may also select a machine learning model from among several types of machine learning models to be used for calculating the parameter values ​​of the physical model. For example, the measurement data processing unit 192 may use the data on the design specifications of the waste final disposal site, where the physical model has been fitted to the measurement data and the parameter values ​​of the physical model have been calculated, as training data to perform learning and testing of parameter value calculation for each machine learning model. Learning of a machine learning model can also be called training of a machine learning model. Furthermore, the measurement data processing unit 192 may calculate the reliability of each machine learning model in the test and select the machine learning model with the highest calculated reliability as the machine learning model to be used to calculate the parameter values ​​of the physical model.

[0074] Alternatively, the measurement data processing unit 192 may select a machine learning model to be used to calculate the parameter values ​​of the physical model from among highly explainable machine learning models, such as a linear regression model. Here, "high explainability" of a machine learning model means that it is easy for a person to understand the relationship between the machine learning model and the values ​​of the parameters being learned in that machine learning model. For example, in the case of a linear regression model, the machine learning model can be expressed with relatively simple mathematical formulas, as in the examples of equations (3) and (4), and in this respect, it can be evaluated as having high explainability.

[0075] Figure 7 shows examples of reliability calculated by the measurement data processing unit 192 for each type of machine learning model. In the example in Figure 7, a reliability index is used in which a smaller value indicates higher reliability. In the example shown in Figure 7, the reliability index value for Gaussian Process Regression (GPR) is the smallest. The measurement data processing unit 192 may also use Gaussian process regression to calculate the parameter values ​​of the physical model. Alternatively, among the machine learning models shown in Figure 7, linear regression (multiple regression analysis), interaction linear regression, robust linear regression, and stepwise linear regression are considered to have relatively high explainability. The measurement data processing unit 192 may use linear regression (multiple regression analysis), which has the smallest reliability index value among these machine learning models, to calculate the parameter values ​​of the physical model.

[0076] Thus, by having the measurement data processing unit 192 select a machine learning model to be used for calculating the parameter values ​​of the physical model based on its reliability, it is expected that the parameter values ​​of the physical model can be calculated with relatively high accuracy. Furthermore, by using a highly explainable machine learning model as the machine learning model used to calculate the parameter values ​​of the physical model in the measurement data processing unit 192, the user can verify the basis for calculating the parameter values ​​of the physical model.

[0077] The measurement data processing unit 192 may re-select the machine learning model used to calculate the parameter values ​​of the physical model at predetermined intervals, such as when measurement data for a final waste disposal site is added. This allows for the selection of a more appropriate machine learning model in response to the increase in data, and it is expected that the prediction device 100 will be able to estimate the closure period of the final waste disposal site with greater accuracy.

[0078] The decommissioning period prediction unit 193 calculates a predicted value for the decommissioning period of the waste final disposal site. Specifically, the decommissioning period prediction unit 193 calculates the leachate concentration C calculated by the physical model application unit 191 for each elapsed time τ. imRefer to the leachate concentration C. im The elapsed time τ during which the value falls below a safe threshold is calculated as the predicted value for the decommissioning period.

[0079] However, the method by which the decommissioning period prediction unit 193 calculates the predicted value of the decommissioning period of the waste final disposal site is not limited to a specific method, and various methods can be used to calculate the predicted value of the decommissioning period using a physical model. For example, the decommissioning period prediction unit 193 may treat an equation obtained by substituting a threshold value of the safety evaluation index into the physical model, calculate the elapsed time τ, and use the calculated elapsed time τ as the decommissioning period. Alternatively, the decommissioning period prediction unit 193 may convert the calculated elapsed time τ into a decommissioning period as a value having the dimension of time.

[0080] Figure 8 shows an example of the procedure for which the prediction device 100 predicts the decommissioning period of a final waste disposal site. In the process shown in Figure 8, the measurement data processing unit 192 determines whether or not a sufficient amount of measurement data for the safety evaluation index at the target disposal site has been accumulated to calculate the parameter values ​​of the physical model (step S101).

[0081] If the measurement data processing unit 192 determines that there is enough accumulated measurement data to calculate the parameter values ​​of the physical model (step S101: YES), the measurement data processing unit 192 calculates the parameter values ​​of the physical model so as to fit the physical model to the measurement data of the safety evaluation index at the target disposal site (step S111). For example, the measurement data processing unit 192 approximates the measurement data of the leachate concentration at the target disposal site with equation (2), and calculates the position ξ of the inflection point. p The values ​​of the first damping constant a1 and the second damping constant a2 are calculated.

[0082] For waste final disposal sites where measurement data for safety evaluation indicators has not been added, the measurement data processing unit 192 may use the parameter values ​​of the physical model calculated in the previous step. For example, the measurement data processing unit 192 may store the parameter values ​​of the physical model calculated in step S111 in the storage unit 180. The measurement data processing unit 192 may then determine at the start of the process in step S111 whether the conditions are met that there is no additional measurement data for safety evaluation indicators for the target disposal site and that the storage unit 180 has stored the parameter values ​​of the physical model. If it is determined that the conditions are met, the measurement data processing unit 192 may read the parameter values ​​of the physical model from the storage unit 180 instead of calculating the parameter values ​​of the physical model in step S111.

[0083] Next, the physical model application unit 191 calculates predicted values ​​for safety evaluation indicators for each elapsed time using a physical model with set parameter values ​​(step S131). Next, the decommissioning period prediction unit 193 compares the predicted value of the safety evaluation index for each elapsed time with the threshold value of the safety product evaluation index, and calculates the elapsed time at which the predicted value of the safety evaluation index falls below the threshold value as the predicted value of the decommissioning period (step S132).

[0084] Next, the prediction device 100 outputs a predicted value for the decommissioning period (step S133). For example, the processing unit 190 may control the display unit 120 to display the predicted value for the decommissioning period. After step S133, the prediction device 100 terminates the process shown in Figure 8.

[0085] On the other hand, if it is determined in step S101 that there is not enough measurement data accumulated to calculate the parameter values ​​of the physical model (step S101: NO), the measurement data processing unit 192 calculates the parameter values ​​of the physical model using the calculation formula for the parameter values ​​of the physical model (step S121). For example, the measurement data processing unit 192 substitutes the depth of the target disposal site, the volume of the target disposal site, the area of ​​the target disposal site, and the year in which landfill operations began at the target disposal site into equations (3) and (4) to calculate the value of the first decay constant a1 and the value of the second decay constant a2. After step S121, the process proceeds to step S131.

[0086] Figure 9 shows an example of the procedure for the measurement data processing unit 192 to obtain the calculation formula for the parameter values ​​of the physical model. The measurement data processing unit 192 may repeat the process shown in Figure 9. For example, the measurement data processing unit 192 may perform the process shown in Figure 9 each time measurement data is added at the waste final disposal site.

[0087] In the process shown in Figure 9, the measurement data processing unit 192 acquires measurement data for safety evaluation indicators at the waste final disposal site (step S201). Next, the measurement data processing unit 192 starts a loop L11 that processes each waste final disposal site for which there is enough measurement data to calculate the parameter values ​​of the physical model (step S202).

[0088] In the processing of loop L11, the measurement data processing unit 192 calculates the parameter values ​​of the physical model so that the physical model fits the measurement data (step S203). Next, the measurement data processing unit 192 performs termination processing for loop L11 (step S204). Specifically, the measurement data processing unit 192 determines whether or not it has performed loop L11 processing for all waste final disposal sites for which there is enough measurement data to calculate the parameter values ​​of the physical model. If it determines that there are waste final disposal sites for which loop L11 processing has not been performed, the measurement data processing unit 192 continues to perform loop L11 processing for those waste final disposal sites for which loop L11 processing has not been performed. On the other hand, if it determines that it has performed loop L11 processing for all waste final disposal sites for which there is enough measurement data to calculate the parameter values ​​of the physical model, the measurement data processing unit 192 terminates loop L11.

[0089] When loop L11 is completed, the measurement data processing unit 192 obtains the calculation formula for the physical model's parameter values ​​by fitting the calculation formula for the physical model's parameter values ​​to the data obtained by combining the data related to the design specifications of the waste final disposal site and the parameter values ​​of the physical model calculated in step S203 (step S205).

[0090] For example, the measurement data processing unit 192 calculates the weight coefficient values ​​for each item of equation (3) so as to fit equation (3) to a dataset for all waste final disposal sites for which there is enough measurement data to calculate the parameter values ​​of a physical model, which is a combination of the depth, volume, area, and year of landfill commencement of the waste final disposal site and the first decay constant a1 calculated in step S203 for that waste final disposal site. As described above, the measurement data processing unit 192 may also use the multiple regression analysis method to calculate the parameter values ​​(weight coefficient values ​​for each item) of equation (3) so as to minimize the least-squares error between equation (3) and the data. After step S205, the measurement data processing unit 192 terminates the process shown in Figure 9.

[0091] Figure 10 shows an example of the procedure for which the measurement data processing unit 192 selects items to be used in the calculation formula for the parameter values ​​of the physical model. The measurement data processing unit 192 may repeat the process shown in Figure 10. For example, the measurement data processing unit 192 may perform the process shown in Figure 10 each time measurement data is added at the waste final disposal site.

[0092] In the process shown in Figure 10, the measurement data processing unit 192 acquires data for each waste final disposal site that has enough measurement data to calculate the parameter values ​​of the physical model, by combining the parameter values ​​of the physical model calculated using the measurement data with data related to the design specifications (step S301).

[0093] Next, the measurement data processing unit 192 uses the dataset of all waste final disposal sites for which there is enough measurement data to calculate the parameter values ​​of the physical model, obtained in step S301, to calculate correlation values ​​between candidate items used to calculate the parameter values ​​of the physical model (step S302). The candidate items used to calculate the parameter values ​​of the physical model may include items related to the design specifications of the waste final disposal site, and either one or both of the parameters of the physical model.

[0094] Next, the measurement data processing unit 192 selects one or more items from the candidates to be used in calculating the parameter values ​​of the physical model based on the correlation values ​​calculated in step S302 (step S303). For example, the measurement data processing unit 192 may select a predetermined number of items for each parameter of the physical model, in order of their correlation with that parameter, as items to be used in calculating the value of that parameter. Alternatively, the measurement data processing unit 192 may select items for each parameter of the physical model, in which the magnitude (absolute value) of the correlation coefficient with that parameter is greater than or equal to a predetermined threshold, as items to be used in calculating the value of that parameter.

[0095] Furthermore, if the parameters of the physical model are included in the list of items used to calculate the parameter values ​​of the physical model, the measurement data processing unit 192 selects the items in a way that prevents reference loops from occurring during the calculation of the parameter values ​​of the physical model, as described above with reference to Figure 6, Equation (3), and Equation (4). After step S303, the measurement data processing unit 192 terminates the processing shown in Figure 10.

[0096] Figure 11 shows an example of the procedure for the measurement data processing unit 192 to select a machine learning model to be used to calculate the parameter values ​​of the physical model. The measurement data processing unit 192 may repeat the process shown in Figure 11. For example, the measurement data processing unit 192 may perform the process shown in Figure 11 each time measurement data is added at the waste final disposal site.

[0097] In the process shown in Figure 11, the measurement data processing unit 192 acquires data related to the design specifications and the calculated parameter values ​​for each waste final disposal site where the physical model has been fitted to the measurement data and the parameter values ​​of the physical model have been calculated (step S401). The measurement data processing unit 192 uses the data obtained by combining the design specification data and the calculated parameter values ​​for all waste final disposal sites where the physical model has been fitted to the measurement data and the parameter values ​​of the physical model have been calculated as a training dataset for machine learning and testing in steps S403 and S404.

[0098] Next, the measurement data processing unit 192 starts a loop L21 that processes each candidate machine learning model used to calculate the parameter values ​​of the physical model (step S402). The machine learning model being processed in loop L21 is also referred to as the target model. In the processing of loop L402, the measurement data processing unit 192 trains the target model using the training dataset acquired in step S401 (step S403). Through machine learning in step S403, the measurement data processing unit 192 obtains the parameter values ​​of the target model (the values ​​of the parameters that are being trained in the target model).

[0099] Next, the measurement data processing unit 192 performs a test on the results of machine learning in step S403 (step S404). Specifically, the measurement data processing unit 192 uses the training dataset obtained in step S401 to calculate a reliability evaluation index value for the target model, which is set to the parameter values ​​obtained in step S403.

[0100] The measurement data processing unit 192 may repeat the machine learning in step S403 and the testing in step S404 for the same target model until a predetermined termination condition is met. In this case, the measurement data processing unit 192 may use the reliability evaluation index value calculated last for each individual machine learning model as the reliability evaluation index value for that machine learning model.

[0101] Next, the measurement data processing unit 192 performs the termination process for loop L21 (step S405). Specifically, the measurement data processing unit 192 determines whether or not the processing for loop L21 has been performed for all machine learning models that are candidates for use in calculating the parameter values ​​of the physical model. If it determines that there are machine learning models that have not been processed for loop L21, the measurement data processing unit 192 continues to perform the processing for loop L21 for those machine learning models that have not been processed for loop L21. On the other hand, if it determines that the processing for loop L21 has been performed for all machine learning models that are candidates for use in calculating the parameter values ​​of the physical model, the measurement data processing unit 192 terminates loop L21.

[0102] When loop L21 is terminated, the measurement data processing unit 192 selects a machine learning model to be used to calculate the parameter values ​​of the physical model based on the reliability evaluation index value calculated for each machine learning model (step S406). The measurement data processing unit 192 may also select the machine learning model with the highest reliability indicated by the reliability evaluation index value as the machine learning model to be used to calculate the parameter values ​​of the physical model.

[0103] Furthermore, the measurement data processing unit 192 may select the machine learning model with the highest reliability indicated by the reliability evaluation index value from among the highly explainable machine learning models to be used for calculating the parameter values ​​of the physical model. In this case, only highly explainable machine learning models (machine learning models selected as highly explainable machine learning models) may be used as candidates for machine learning models to be used for calculating the parameter values ​​of the physical model. After step S406, the measurement data processing unit 192 terminates the process shown in Figure 11.

[0104] As described above, the measurement data processing unit 192 calculates parameter values ​​to be set in a model (physical model) that shows the relationship between the passage of time at a waste final disposal site and the value of the safety evaluation index, based on the time-series data of measurement data of the safety evaluation index at the target disposal site, which is a waste final disposal site for which the decommissioning period is predicted. The decommissioning period prediction unit 193 calculates a predicted value for the decommissioning period of the target disposal site based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index, as shown by a model (physical model) with set parameter values.

[0105] The prediction device 100 can predict the decommissioning period of a final waste disposal site with relatively high accuracy. In particular, the prediction device 100 can reflect measurement data of safety evaluation indicators at the target disposal site into the physical model. In this respect, the prediction device 100 can predict the decommissioning period of a final waste disposal site with higher accuracy compared to cases where measurement data of safety evaluation indicators at the target disposal site is not reflected in the physical model. Furthermore, according to the prediction device 100, as the number of measurement data points for safety evaluation indicators at the target disposal site increases, the accuracy of the physical model improves through fitting the physical model to the measurement data, and it is expected that the prediction accuracy of the decommissioning period of the final waste disposal site will improve.

[0106] Furthermore, it is expected that by allowing managers of waste final disposal sites to experience the usefulness of the data using the prediction device 100, data on waste final disposal sites will become more readily available, and the accuracy of the prediction device 100's forecast of the closure period of waste final disposal sites will improve. It is conceivable that data such as leachate concentration is measured and recorded at waste final disposal sites for management purposes. On the other hand, it is generally believed that data related to waste final disposal sites is not provided to others. Furthermore, it takes a long period of time, for example, 50 years, from the completion of landfilling to the closure of a waste final disposal site. Due to reasons such as inadequate data transfer during personnel changes, there is a possibility that data related to waste final disposal sites may be lost without being utilized.

[0107] In response to this, it is expected that managers of waste final disposal sites will be more likely to provide data from these sites by using the prediction device 100 and experiencing the usefulness of the data, leading to expectations of improved accuracy in predicting the closure period. By applying the provided data to the prediction device 100, it is expected that the accuracy of the prediction device 100 in predicting the closure period of waste final disposal sites will improve.

[0108] Furthermore, if the measurement data processing unit 192 determines that it has not obtained enough measurement data for the safety evaluation index at the target disposal site to calculate the parameter values ​​of the physical model, it calculates the parameter values ​​of the physical model based on the design specifications of the target disposal site. According to the prediction device 100, it is possible to predict the decommissioning period even for waste final disposal sites where measurement data for safety evaluation indicators is not available, such as waste final disposal sites in the design phase. By being able to predict the decommissioning period for waste final disposal sites before operation or during operation, it is possible to predict the maintenance costs during the decommissioning period with relatively high accuracy. This makes it possible to take measures such as setting waste acceptance fees that take into account the maintenance costs during the decommissioning period, and it is expected that this will reduce the risk of financial collapse when local governments manage waste final disposal sites, and the risk of business failure when private companies manage waste final disposal sites.

[0109] Furthermore, the physical model application unit 191 calculates time-series data of predicted safety evaluation indicators for the target disposal site using a physical model with set parameter values. The decommissioning period prediction unit 193 calculates a predicted value for the decommissioning period of the target disposal site by comparing time-series data of predicted values ​​of safety evaluation indicators with threshold values ​​of safety evaluation indicators.

[0110] According to the prediction device 100, the decommissioning period can be predicted through a relatively simple process: inputting various elapsed times into a physical model with set parameter values ​​to calculate time-series data of predicted values ​​for safety evaluation indicators, and then comparing the obtained time-series data with a threshold. In this respect, the prediction device 100 is expected to have a relatively light load on predicting the decommissioning period and a relatively short time required for prediction.

[0111] Furthermore, as a physical model, a model is used that approximates the relationship between the elapsed time since the completion of landfilling of waste at a final disposal site and the logarithmic value of the leachate concentration, which is used as a safety evaluation index, with a piecewise linear model formed by connecting two straight lines. The measurement data processing unit 192 calculates the slopes of the two straight lines and the positions of the inflection points of the polyline as parameter values ​​for the physical model.

[0112] According to the prediction device 100, by using a highly explainable model, users can verify the basis for the prediction of the decommissioning period. Furthermore, the inventors of this application have found that the relationship between the elapsed time since the completion of landfilling of waste at a final waste disposal site and the logarithmic value of the leachate concentration, which is used as a safety evaluation index, can be represented by a broken line formed by connecting two straight lines. The prediction device 100 is expected to be able to accurately predict the decommissioning period of a final waste disposal site by reflecting this finding in a physical model. Furthermore, because the prediction device 100 uses a relatively simple model represented by a polyline formed by connecting two straight lines as its physical model, the processing load for predicting the closure period of a final waste disposal site is expected to be relatively light, and the time required to predict the closure period is also expected to be relatively short.

[0113] Furthermore, if the measurement data processing unit 192 determines that it has not obtained enough measurement data for the safety evaluation index at the target disposal site to calculate the parameter values ​​of the physical model, it calculates the slope of each of the two lines based on the depth of the disposal site, the volume of the disposal site, the area of ​​the disposal site, and the year in which landfilling began at the target disposal site.

[0114] The inventors of this application have found that there is a relatively strong correlation between the slopes of the two lines in a bilinear physical model (a polyline formed by connecting two straight lines) and the depth, volume, area, and year of commencement of landfill operations at the target disposal site. The prediction device 100 is expected to be able to predict the decommissioning period with relatively high accuracy, even for waste final disposal sites where measurement data for safety evaluation indicators is not available or where sufficient measurement data for safety evaluation indicators is not available, because this finding can be reflected in the parameter values ​​of the physical model.

[0115] Furthermore, the prediction device 100 can update the parameter values ​​of the physical model in accordance with the accumulation of measurement data at the final waste disposal site. In this respect, the prediction device 100 is expected to be able to predict the decommissioning period with relatively high accuracy, especially for final waste disposal sites where measurement data for safety evaluation indicators is not available or where sufficient measurement data for safety evaluation indicators has not been obtained.

[0116] Furthermore, the measurement data processing unit 192 uses measurement data of safety evaluation indices at the waste final disposal site to evaluate the prediction accuracy of safety evaluation index values ​​for each of several types of models that show the relationship between the passage of time at the waste final disposal site and the value of the safety evaluation index. The decommissioning period prediction unit 193 calculates a predicted value for the decommissioning period of the target disposal site based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index, as shown by the model selected based on the evaluation results.

[0117] According to the prediction device 100, a model with relatively good prediction accuracy for safety evaluation index values ​​can be used as the physical model, and in this respect, it is expected that the decommissioning period of waste final disposal sites can be predicted with relatively high accuracy. In particular, according to the prediction device 100, the model used as the physical model can be changed in accordance with the accumulation of measurement data at the waste final disposal site, and in this respect, it is expected that the decommissioning period of waste final disposal sites can be predicted with relatively high accuracy.

[0118] Figure 12 shows an example of a computer configuration according to at least one embodiment. In the configuration shown in Figure 12, the computer 700 comprises a CPU 710, a main memory 720, an auxiliary memory 730, an interface 740, and a non-volatile recording medium 750.

[0119] The prediction device 100 described above may be implemented in a computer 700. In that case, the operation of each processing unit described above is stored in auxiliary storage device 730 in the form of a program. The CPU 710 reads the program from auxiliary storage device 730, loads it into main memory 720, and executes the above processing according to the program. The CPU 710 also allocates storage areas in main memory 720 corresponding to each of the storage units described above, according to the program.

[0120] When the prediction device 100 is implemented in a computer 700, the operation of the processing unit 190 and each of its components is stored in auxiliary storage device 730 in the form of a program. The CPU 710 reads the program from auxiliary storage device 730, loads it into main memory 720, and executes the above processing according to the program.

[0121] Furthermore, the CPU 710 reserves a memory area in the main memory 720 corresponding to the memory unit 180 according to the program. Communication between the communication unit 110 and other devices is performed by the interface 740 having a communication function and performing communication according to the control of the CPU 710. The display unit 120 performs the display by having an interface 740 that has a display device and displaying various images according to the control of the CPU 710. The reception of user operations by the operation input unit 130 is performed when the interface 740, which has input devices such as a keyboard and mouse, receives user operations and outputs information indicating the received user operations to the CPU 710.

[0122] One or more of the above-mentioned programs may be recorded on the non-volatile recording medium 750. In this case, the interface 740 may read the program from the non-volatile recording medium 750. The CPU 710 may then either directly execute the program read by the interface 740, or temporarily save it in the main memory 720 or auxiliary memory 730 before executing it.

[0123] Alternatively, a program to implement all or part of the functions of the prediction device 100 may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be loaded into a computer system and executed to perform the processing of each part. The term "computer system" here includes hardware such as an operating system (OS) and peripheral devices. Furthermore, "computer-readable recording media" refers to portable media such as flexible disks, magneto-optical disks, ROMs (Read Only Memory), CD-ROMs (Compact Disc Read Only Memory), and storage devices such as hard disks built into computer systems. The above-mentioned program may be intended to implement only a part of the functions described above, and may also be able to implement the above-mentioned functions in combination with programs already recorded in the computer system.

[0124] Although embodiments of the present invention have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments, and design modifications and the like that do not depart from the spirit of the invention are also included.

[0125] Some or all of the above embodiments may also be described as follows, but are not limited to these.

[0126] (Note 1) A measurement data processing unit calculates parameter values ​​to be set in a model that shows the relationship between the passage of time at a waste final disposal site and the value of the safety evaluation index, based on time-series data of measurement data of the safety evaluation index at the target disposal site, which is a waste final disposal site for which the decommissioning period is predicted. A decommissioning period prediction unit calculates a predicted value for the decommissioning period of the target disposal site based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index shown by the model in which the parameter values ​​have been set, A prediction device equipped with the following features.

[0127] (Note 2) If the measurement data processing unit determines that it has not obtained enough measurement data for the safety evaluation index at the target disposal site to calculate the parameter value, it calculates the parameter value based on the design specifications of the target disposal site. The prediction device described in Appendix 1.

[0128] (Note 3) The system includes a physical model application unit that uses the model in which the parameter values ​​are set to calculate time-series data of predicted values ​​of safety evaluation indicators at the target disposal site. The decommissioning period prediction unit calculates a predicted value for the decommissioning period of the target disposal site by comparing the time-series data of the predicted value of the safety evaluation index with the threshold value of the safety evaluation index. The prediction device described in Appendix 1 or Appendix 2.

[0129] (Note 4) As the aforementioned model, a model is used that approximates the relationship between the elapsed time since the completion of landfilling of waste at the final waste disposal site and the logarithmic value of the leachate concentration, which is the safety evaluation index, with a piecewise linear model formed by connecting two straight lines. The measurement data processing unit calculates the slopes of the two straight lines and the positions of the inflection points of the broken line as parameter values. A prediction device described in any one of the appendices 1 to 3.

[0130] (Note 5) If the measurement data processing unit determines that it has not obtained enough measurement data for the safety evaluation index at the target disposal site to calculate the parameter values, it calculates the slope of each of the two lines based on the depth of the disposal site, the volume of the disposal site, the area of ​​the disposal site, and the year in which landfilling began at the target disposal site. The prediction device described in Appendix 4.

[0131] (Note 6) The measurement data processing unit uses measurement data of safety evaluation indicators at the waste final disposal site to evaluate the prediction accuracy of the decommissioning period for each of several types of models that show the relationship between the passage of time at the waste final disposal site and the value of the safety evaluation indicators. The decommissioning period prediction unit calculates a predicted value for the decommissioning period of the target disposal site based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index, as shown by the model selected based on the evaluation results. A prediction device described in any one of the appendices 1 to 5.

[0132] (Note 7) Computers Based on time-series data of measurement data for safety evaluation indicators at the target waste disposal site, which is a final waste disposal site for which the decommissioning period is predicted, parameter values ​​to be set in a model that shows the relationship between the passage of time at the final waste disposal site and the value of the safety evaluation indicator are calculated. Based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index shown by the model in which the parameter values ​​are set, a predicted value for the decommissioning period of the target disposal site is calculated. A prediction method that includes this.

[0133] (Note 8) On the computer, Based on time-series data of measurement data of safety evaluation indicators at the target waste final disposal site, which is subject to closure period prediction, the parameter values ​​to be set in a model that shows the relationship between the passage of time at the waste final disposal site and the value of the safety evaluation indicators are calculated. Based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index shown by the model in which the parameter values ​​are set, a predicted value for the decommissioning period of the target disposal site is calculated. A program that executes the command. [Explanation of Symbols]

[0134] 100 Prediction Devices 110 Communications Department 120 Display section 130 Operation Input Section 180 Storage section 190 Processing Unit 191 Physical Model Application Section 192 Measurement Data Processing Unit 193 Abolition Period Forecast Section

Claims

1. A measurement data processing unit calculates parameter values ​​to be set in a model that shows the relationship between the passage of time at a waste final disposal site and the value of the safety evaluation index, based on time-series data of measurement data of the safety evaluation index at the target disposal site, which is a waste final disposal site for which the decommissioning period is predicted. A decommissioning period prediction unit calculates a predicted value for the decommissioning period of the target disposal site based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index shown by the model in which the parameter values ​​have been set, A prediction device equipped with the following features.

2. If the measurement data processing unit determines that it has not obtained enough measurement data for the safety evaluation index at the target disposal site to calculate the parameter value, it calculates the parameter value based on the design specifications of the target disposal site. The prediction device according to claim 1.

3. The system includes a physical model application unit that uses the model in which the parameter values ​​are set to calculate time-series data of predicted values ​​of safety evaluation indicators at the target disposal site. The decommissioning period prediction unit calculates a predicted value for the decommissioning period of the target disposal site by comparing the time-series data of the predicted value of the safety evaluation index with the threshold value of the safety evaluation index. The prediction device according to claim 1.

4. As the aforementioned model, a model is used that approximates the relationship between the elapsed time since the completion of landfilling of waste at the final waste disposal site and the logarithmic value of the leachate concentration, which is the safety evaluation index, with a piecewise linear model formed by connecting two straight lines. The measurement data processing unit calculates the slopes of the two straight lines and the positions of the inflection points of the broken line as parameter values. The prediction device according to claim 1.

5. If the measurement data processing unit determines that it has not obtained enough measurement data for the safety evaluation index at the target disposal site to calculate the parameter values, it calculates the slope of each of the two lines based on the depth of the disposal site, the volume of the disposal site, the area of ​​the disposal site, and the year in which landfilling began at the target disposal site. The prediction device according to claim 4.

6. The measurement data processing unit uses measurement data of safety evaluation indicators at the waste final disposal site to evaluate the prediction accuracy of the decommissioning period for each of several types of models that show the relationship between the passage of time at the waste final disposal site and the value of the safety evaluation indicators. The decommissioning period prediction unit calculates a predicted value for the decommissioning period of the target disposal site based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index, as shown by the model selected based on the evaluation results. The prediction device according to claim 1.

7. Computers Based on time-series data of measurement data for safety evaluation indicators at the target waste disposal site, which is a final waste disposal site for which the decommissioning period is predicted, parameter values ​​to be set in a model that shows the relationship between the passage of time at the final waste disposal site and the value of the safety evaluation indicator are calculated. Based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index shown by the model in which the parameter values ​​are set, a predicted value for the decommissioning period of the target disposal site is calculated. A prediction method that includes this.

8. On the computer, Based on time-series data of measurement data of safety evaluation indicators at the target waste final disposal site, which is the waste final disposal site for which the decommissioning period is predicted, the parameter values ​​to be set in a model that shows the relationship between the passage of time at the waste final disposal site and the value of the safety evaluation indicators are calculated. Based on the relationship between the passage of time at the target disposal site and the value of the safety evaluation index shown by the model in which the parameter values ​​are set, a predicted value for the decommissioning period of the target disposal site is calculated. A program that executes the command.

Images