Experimental condition determining device, program, and experimental condition determining method

Abstract

Provided is an experimental condition determining device for determining experimental conditions for an experiment including a plurality of steps in which a plurality of experimental conditions can be processed in parallel and the amount of processing for each experimental condition is limited, the experimental condition determining device comprising: an option identifying unit that identifies a plurality of experimental condition options for each step on the basis of an evaluation value of an evaluation function that accepts input of an experimental condition and outputs an evaluation value; and an experimental condition acquiring unit that acquires a combined experimental condition, which is a combination of the experimental conditions for the plurality of steps, from among the plurality of experimental condition options, on the basis of the amount of processing for each experimental condition and the evaluation value obtained by inputting the experimental condition options into the evaluation function.

Description

Experimental condition determination device, program, and experimental condition determination method

[0001] The present invention relates to an apparatus, a program, and a method for determining experimental conditions.

[0002] Patent Document 1 describes a fabrication evaluation system including: a fabrication device that fabricates a sample; a measurement device that measures material information that represents physical properties or a structure of the sample fabricated by the fabrication device; and an estimation device connected to the fabrication device and the measurement device, wherein the estimation device includes an estimation unit that estimates fabrication conditions that optimize the material information based on a data set including fabrication conditions of the sample and the material information of the sample, the fabrication device fabricates the sample according to the fabrication conditions estimated by the estimation unit, and the estimation device includes a data addition unit that adds the material information of the sample measured by the measurement device and the fabrication conditions of the sample to the data set, and the estimation unit sequentially estimates the fabrication conditions based on the data set to which the fabrication conditions and material information have been added (claim 1).Patent Document 2 describes a design support device that determines a plurality of design parameters that satisfy a target value set for each of a plurality of characteristic items that indicate the characteristics of a product, work in progress, semi-finished product, part, or prototype, in order to apply the device to a method of optimizing the design parameters by repeating the process of determining the design parameters and producing the product, work in progress, semi-finished product, part, or prototype based on a group of design parameters consisting of a plurality of design parameters, the device comprising: a data acquisition unit that acquires a plurality of performance data consisting of the group of design parameters and observed values ​​of each of the plurality of characteristic items for the product, work in progress, semi-finished product, part, or prototype that has already been produced; and a model that constructs, based on the performance data, a prediction model that predicts the observed values ​​of the characteristic items as a probability distribution or an approximation or alternative index thereof based on the group of design parameters. The present invention discloses a design support device including: a construction unit; an acquisition function construction unit that constructs, for each characteristic item based on at least the prediction model, an acquisition function that takes the design parameter group as input and outputs an index value of the design parameter group related to improvement of the characteristic indicated by the characteristic item; a design parameter group candidate generation unit that generates a plurality of design parameter group candidates by multi-objective optimization of the design parameter group using the outputs of the plurality of acquisition functions as objective variables; a selection unit that calculates, for each design parameter group candidate, an overall achievement probability, which is the probability that the target values ​​of all characteristic items will be achieved, based on the probability distribution of the observation values ​​obtained by inputting the design parameter group candidates to the prediction model or its approximation or alternative index, and selects at least one design parameter group candidate with the highest overall achievement probability; and an output unit that outputs the selected design parameter group candidate (see claim 1). [Prior art documents] [Patent documents] [Patent document 1] JP 2021-43959 A [Patent document 2] WO 2022 / 124075 A General disclosure

[0003] In a first aspect of the present invention, there is provided an experimental condition determination device that can process multiple experimental conditions in parallel and determines experimental conditions for an experiment that includes multiple steps in which the processing volume for each experimental condition is limited. The experimental condition determination device may include an option identification unit and an experimental condition acquisition unit. The option identification unit may identify multiple experimental condition options for each step based on an evaluation value of an evaluation function that inputs the experimental conditions and outputs an evaluation value. The experimental condition acquisition unit may acquire a combined experimental condition, which is a combination of experimental conditions for multiple steps, from the multiple experimental condition options based on the processing volume for each experimental condition and the evaluation value obtained by inputting the experimental condition options into the evaluation function.

[0004] In the above, the experimental condition determination device may further include an experimental throughput specification unit. The experimental throughput specification unit may specify an experimental throughput, which is the throughput that can be processed at one time for the entire experiment, based on the number of experimental conditions that can be processed in parallel in each step and the throughput for each experimental condition. The experimental condition acquisition unit may acquire combined experimental conditions so that the total number of combined experimental conditions is the experimental throughput or a multiple thereof.

[0005] In the above, the experimental throughput specifying unit may calculate the process throughput, which is the throughput that can be processed at one time in each process, from the product of the number of experimental conditions that can be processed in parallel in each process and the throughput for each experimental condition. The experimental throughput specifying unit may specify the smallest process throughput among all processes as the experimental throughput.

[0006] In the above, the experiment condition acquisition unit may input any combination of experiment conditions from the user.

[0007] In the above, the option specifying unit may specify a combination of experimental conditions for which the evaluation value of the evaluation function satisfies a predetermined criterion, and may specify a first experimental condition for a first step included in the specified combination of experimental conditions as a first experimental condition option.

[0008] In the above, the option specifying unit may specify a plurality of combination experimental conditions whose evaluation values ​​of the evaluation function are at a predetermined top level, and may specify first experimental conditions included in each of the plurality of combination experimental conditions as a plurality of first experimental condition options.

[0009] In the above, the option specifying unit may specify combination experimental conditions whose evaluation values ​​satisfy a predetermined standard in each of the plurality of evaluation functions, and may specify first experimental conditions included in each of the specified combination experimental conditions as first experimental condition options.

[0010] In the above, the option specifying unit may specify a combination of experimental conditions that includes the first experimental condition option as part of the experimental conditions and whose evaluation value of the evaluation function satisfies a predetermined criterion. The option specifying unit may specify, as the second experimental condition option, a second experimental condition for a second step included in the specified combination of experimental conditions.

[0011] In the above, the option specifying unit may output an overall evaluation of the evaluation values ​​of the evaluation function for each of the plurality of first experimental condition options for each second experimental condition. The option specifying unit may specify, as the second experimental condition option, a second experimental condition whose overall evaluation satisfies a predetermined criterion.

[0012] In the above, the option specifying unit may output, as the overall evaluation for the second experimental condition, the average or median of the evaluation values ​​of the evaluation function including the second experimental condition and having each of the multiple first experimental condition options.

[0013] In the above, the option identification unit may output the maximum or minimum value of the evaluation values ​​of the evaluation function including the second experimental condition and having each of the multiple first experimental condition options as the overall evaluation for the second experimental condition.

[0014] In the above, the experimental condition acquisition unit may acquire an evaluation value of an evaluation function for a combination of experimental condition options related to a plurality of processes. The experimental condition acquisition unit may select and acquire a combined experimental condition from the combination of experimental condition options so as to satisfy a processing amount limit for each experimental condition and such that the evaluation value of the evaluation function satisfies a predetermined standard.

[0015] In the above, the evaluation function may output a numerical value relating to the expected quality of the experimental result when the experiment is carried out under the input experimental conditions.

[0016] In the above, the evaluation function may output a numerical value relating to the certainty of the predicted experimental result when the experiment is conducted under the input experimental conditions.

[0017] In a second aspect of the present invention, there is provided a program that, when executed by a computer, causes the computer to function as the experimental condition determination device of the first aspect.

[0018] In a third aspect of the present invention, there is provided an experimental condition determination method executed by the experimental condition determination device of the first aspect. The experimental condition determination method includes an option identification step and an experimental condition acquisition step. In the option identification step, a plurality of experimental condition options for each process may be identified based on an evaluation value of an evaluation function that inputs experimental conditions and outputs an evaluation value. In the experimental condition acquisition step, a combined experimental condition, which is a combination of experimental conditions for a plurality of processes, may be acquired from the plurality of experimental condition options based on the processing volume for each experimental condition and the evaluation value of the evaluation function obtained by inputting the experimental condition options.

[0019] The above summary of the invention does not list all of the features of the present invention, and subcombinations of these features may also constitute inventions.

[0020] 1 shows an example of an experiment including multiple steps according to the present embodiment. An example of a preparation step 22 (step 1) according to the present embodiment. An example of a drying step 24 (step 2) according to the present embodiment. An example of an evaluation step 26 (step 3) according to the present embodiment. A configuration of an experimental condition determination device 10 according to the present embodiment. A flow of experimental condition determination by the experimental condition determination device 10 according to the present embodiment. An example of specifying the experimental throughput by the experimental throughput specification unit 102. Another example of specifying the experimental throughput by the experimental throughput specification unit 102. An example of wasted experimental slots. An example of combined experimental conditions input by a user in S200. An example of a subflow of S300 when n=1. An example of a subflow of S300 when n≧2. An example of combined experimental conditions specified in S310. An example of a first experimental condition option specified in S330. An example of a second experimental condition searched for in S312. An example of a second experimental condition option specified in S332. 22 shows an example of all combinations of experimental condition options that are candidates for the combined experimental condition in S600. 23 shows an example of the combined experimental condition selected in S600. 24 shows an example of a method for identifying the combined experimental condition in S600. 25 shows an example of a computer 2200 in which multiple aspects of the present invention may be embodied in whole or in part.

[0021] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention as claimed. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0022] In this embodiment, the experimental condition determination device determines the experimental conditions to be adopted in an experiment. In particular, the experimental condition determination device can process multiple experimental conditions in parallel and determines the experimental conditions for an experiment that includes multiple steps in which the processing volume for each experimental condition is limited. Here, examples of experiments and steps in this embodiment will be described using the examples in Figures 1 to 4.

[0023] FIG. 1 shows an example of an experiment including multiple steps according to this embodiment. The experiment according to this embodiment may be any experiment in which experimental conditions are set, and may include, but is not limited to, experiments in chemistry, physics, biology, computational engineering, and various other fields. For example, the experiment may involve preparing a coating liquid sample for a specific application (e.g., paint, electronic materials, etc.), applying and drying the sample, and evaluating the dried product. In such a case, as shown in FIG. 1 , the experiment includes three steps: a preparation step 22 (step 1), a drying step 24 (step 2), and an evaluation step 26 (step 3).

[0024] The preparation step 22 may be a step of mixing raw materials to produce a coating liquid sample. Many different experimental conditions are possible in the preparation step 22 depending on the type of raw materials, the mixing ratio, and / or whether or not pretreatment is performed.

[0025] The drying step 24 may be a step in which the coating liquid of the coating liquid sample is applied to an object and then dried. Many different experimental conditions are assumed for the drying step 24 as well, depending on the drying temperature, drying time, gas atmosphere, etc.

[0026] The evaluation step 26 may be a step of evaluating the dried product that has been subjected to the drying step 24. For example, the dried product may be observed and / or its physical properties may be measured in the evaluation step 26. In the evaluation step 26, no particular experimental conditions may be set.

[0027] FIG. 2 shows an example of the preparation step 22 (step 1) according to this embodiment. The preparation step includes seven experimental conditions, Preparation A to Preparation G. For example, Preparation A may be an experimental condition for producing a coating liquid sample containing a certain raw material at 0 wt %. Preparation B may be an experimental condition in which the raw material is contained at 5 wt %, Preparation C may be an experimental condition in which the raw material is contained at 10 wt %, Preparation D may be an experimental condition in which the raw material is contained at 20 wt %, Preparation E may be an experimental condition in which the raw material is contained at 30 wt %, Preparation F may be an experimental condition in which the raw material is contained at 40 wt %, and Preparation G may be an experimental condition in which the raw material is contained at 50 wt %. For the purpose of explanation, the above example shows a simple example in which only the content of one component is different. However, actual experimental conditions may be more complex, for example, by varying the types and contents of multiple components.

[0028] The amount that can be processed at one time for one experimental condition (hereinafter also referred to as "processing amount") is usually limited. For example, in preparation step 22 (step 1), it is assumed that only a maximum of three coating liquid samples can be produced at one time for each experimental condition due to limitations on the capacity of the mixing device and container. For example, a maximum of three coating liquid samples are produced under the conditions of preparation A. Similarly, a maximum of three coating liquid samples are produced under the conditions of preparations B to G. "One" may mean an amount that can be used once in a subsequent experiment.

[0029] Thus, in the preparation step 22 (step 1), the processing volume for each of the seven experimental conditions is three. Therefore, the preparation step 22 (step 1) can produce a maximum of 7 x 3 = 21 coating liquid samples at one time. In other words, the processing volume for the entire preparation step 22 (step 1) is 21.

[0030] Therefore, 21 or fewer coating liquid samples can be produced in one preparation step 22 (step 1). To produce 22 or more coating liquid samples in preparation step 22 (step 1), two or more repetitions are required. From the viewpoint of completing preparation step 22 (step 1) in one run, it is desirable to produce 21 or fewer coating liquid samples.

[0031] In the example of this embodiment, the processing amount has been described as a number expressed as an integer. However, the processing amount is not limited to this, and may be an amount expressed as a real number such as weight, volume, time, etc. The processing amount may be the amount of sample that can be processed under each experimental condition.

[0032] 3 shows an example of the drying step 24 (step 2) according to this embodiment. The drying step includes six experimental conditions, Drying A to Drying F. For example, Drying A may be a condition in which a coating liquid sample is dried at 130°C for a predetermined time. Drying B may be a condition in which the drying temperature is changed to 110°C, Drying C may be a condition in which the drying temperature is changed to 90°C, Drying D may be a condition in which the drying temperature is changed to 70°C, Drying E may be a condition in which the drying temperature is changed to 50°C, and Drying F may be a condition in which the drying temperature is changed to 30°C.

[0033] It is assumed that the drying step 24 (step 2) can dry a maximum of four coating liquid samples at a time for each experimental condition. In the drying step 24 (step 2), each of the six experimental conditions has a processing capacity of four samples. Therefore, the preparation step 24 (step 2) can dry a maximum of 6 x 4 = 24 coating liquid samples at a time. Therefore, the processing capacity of the drying step 24 (step 2) is 24 samples.

[0034] Therefore, drying of 24 or fewer coating liquid samples can be completed in one drying step 24 (step 2). When drying 25 or more coating liquid samples in the drying step 24 (step 2), the drying step must be repeated two or more times. From the viewpoint of completing the drying step 24 (step 2) in one step, it is desirable to limit the number of coating liquid samples to be dried to 24 or fewer.

[0035] 4 shows an example of the evaluation step 26 (step 3) according to this embodiment. In the evaluation step 26 (step 3), 18 dried products (coating liquid samples after drying) are evaluated at one time. That is, the processing volume of the evaluation step 26 (step 3) is 18. From the viewpoint of completing the evaluation step 26 (step 3) in one run, it is desirable to limit the number of evaluation targets to 18 or less.

[0036] 2 to 4, there are seven experimental conditions in the preparation process 22 and six experimental conditions in the drying process 24, and by simple calculation, there are 7 x 6 = 42 combinations of experimental conditions throughout the entire experiment. Hereinafter, such a combination of experimental conditions for multiple processes (e.g., a combination of preparation A and drying A) will also be referred to as a "combined experimental condition."

[0037] It would be most efficient if all 42 experiments could be completed in one experiment, but this is often difficult due to limitations on the experimental conditions. For example, even if all seven coating liquid samples prepared in Preparations A to G were to be applied to Drying A, only four samples of the processing volume could be tested at a time.

[0038] As described above, it is not possible to execute all combinations of experimental conditions at once, and only a limited number of combinations of experimental conditions can be executed at once. Therefore, it becomes necessary to determine which combinations of combinations of experimental conditions should be prioritized as the subject of the experiment. The experimental condition determination device according to this embodiment can determine which combinations of experimental conditions should be prioritized.

[0039] 5 shows the configuration of the experiment condition determination device 10 according to this embodiment. As shown in FIG. 5, the experiment condition determination device 10 includes an experiment processing amount specifying unit 102, an option specifying unit 104, an experiment condition acquisition unit 106, and a storage unit 108.

[0040] The experiment throughput specifying unit 102 specifies an experiment throughput, which is the amount of processing that can be performed at one time for the entire experiment. The amount of processing that can be performed at one time for the entire experiment may mean the number of experimental conditions that can be completed from the first step to the last step in one experiment.

[0041] For example, in the examples of Figures 2 to 4, it was possible to process 7 x 3 = 21 samples in step 1, 6 x 4 = 24 samples in step 2, and 18 samples in step 3. In such a case, the maximum amount that can be processed throughout all steps (i.e., the experimental processing amount) is 18 samples. This is because step 3 can process up to 18 samples at a time, so it is ultimately not possible to process more than 19 samples at a time throughout the entire experiment. For example, even if 19 or more samples are supplied, the 19th and subsequent samples cannot be processed in one step 3. Details of the operation of the experimental processing amount specifying unit 102 will be described later.

[0042] The option identifying unit 104 may identify multiple experimental condition options for each process. The experimental condition options are options that are candidates for the experimental conditions of each process included in the combined experimental conditions that will be determined later. The experimental conditions of the combined experimental conditions will be selected later from the experimental condition options. The option identifying unit 104 may identify multiple experimental condition options based on the evaluation value of an evaluation function that inputs the experimental conditions and outputs an evaluation value. Details of the operation of the option identifying unit 104 will be described later.

[0043] The experimental condition acquisition unit 106 acquires the combined experimental conditions from the multiple experimental condition options identified by the option identification unit 104. The experimental condition acquisition unit 106 may select one from the multiple experimental condition options for each process based on the processing volume for each experimental condition and an evaluation value obtained by inputting the experimental condition options into an evaluation function, thereby acquiring the combined experimental conditions.

[0044] The evaluation function may be a function that outputs a corresponding evaluation value when experimental conditions are input. For example, the evaluation value may be a numerical value relating to the expected pass / fail of the experimental result or a numerical value relating to the certainty of the expected experimental result. Details of the operation of the experimental condition acquisition unit 106 will be described later.

[0045] The storage unit 108 stores programs, commands, information, data, etc. necessary for the operation of the experimental condition determination device 10. For example, the storage unit 108 stores programs corresponding to the experimental processing amount specifying unit 102, the option specifying unit 104, and the experimental condition acquisition unit 106, etc., as well as necessary information such as evaluation functions.

[0046] The experimental condition determination device 10 may be a computer such as a PC (personal computer), tablet computer, smartphone, workstation, server computer, or general-purpose computer, or may be a computer system in which multiple computers are connected.

[0047] Alternatively, the experimental condition determination apparatus 10 may be a dedicated computer designed for each application, or dedicated hardware realized by a dedicated circuit. The experimental condition determination apparatus 10 may be implemented by a single device (computer), or may be realized by multiple devices with different roles. Although not specifically described below, the experimental condition determination apparatus 10 may be equipped with a memory / hard disk or the like and function as a storage unit 108, in which information necessary for processing is appropriately stored, and information is transmitted between each processing module of the experimental condition determination apparatus 10, such as the experimental processing amount determination unit 102, the option determination unit 104, and the experimental condition acquisition unit 106.

[0048] 6 shows the flow of experimental condition determination by the experimental condition determination device 10 according to this embodiment. The experimental condition determination device 10 determines and outputs combined experimental conditions by, for example, executing the processes of S100 to S600. Some of S100 to S600 may be omitted. Other operations may be performed in addition to S100 to S600.

[0049] First, in S100, the experimental throughput specifying unit 102 specifies the experimental throughput. The experimental throughput specifying unit 102 may specify the experimental throughput from the number of experimental conditions that can be processed in parallel in each step and the throughput for each experimental condition.

[0050] The experimental processing amount specifying unit 102 acquires, for each step included in the experiment, information on the number of experimental conditions that can be processed in parallel and the processing amount for each experimental condition. For example, in the case of step 1 (preparation) shown in FIG. 2, seven experimental conditions (A to G) may be acquired as the number of experimental conditions, and three may be acquired as the processing amount for each experimental condition. In the case of step 2 (drying) shown in FIG. 3, six experimental conditions (A to F) may be acquired as the number of experimental conditions, and four may be acquired as the processing amount for each experimental condition. In the case of step 3 (evaluation) shown in FIG. 4, 18 may be acquired as the processing amount for each experimental condition.

[0051] The experimental throughput specifying unit 102 may calculate the throughput that can be processed at one time for each step (hereinafter also referred to as "step throughput") from the product of the number of experimental conditions that can be processed in parallel for each step and the throughput for each experimental condition. For example, the experimental throughput specifying unit 102 may calculate a step throughput of 7 x 3 = 21 for step 1 (preparation) in Figure 2, a step throughput of 6 x 4 = 24 for step 2 (drying) in Figure 3, and a step throughput of 1 x 18 = 18 for step 3 (evaluation) in Figure 4.

[0052] 7 shows an example of the experimental throughput specified by the experimental throughput specifying unit 102. The experimental throughput specifying unit 102 may specify the smallest process throughput among all processes included in the experiment as the experimental throughput. For example, for an experiment including processes 1 to 3 in FIGS. 2 to 4, the experimental throughput specifying unit 102 may execute min (21, 24, 18) and specify 18 as the experimental throughput.

[0053] 8 shows another example of the determination of the experimental throughput by the experimental throughput determination unit 102. The experimental throughput determination unit 102 may modify the process throughput calculated by multiplying the number of experimental conditions and the throughput based on the number of experimental conditions and the throughput of other processes. FIG. 8 shows a case where the number of experimental conditions in process 2 is two and the throughput of each experimental condition is eight.

[0054] In this case, the process throughput for Step 2 is calculated as 2 × 8 = 16. However, since there are seven experimental conditions in Step 1, one of the eight experimental conditions for Step 2 is wasted. As shown in FIG. 9 , coating liquid samples of Preparations A to G can be applied to each of the seven throughputs for each experimental condition (Drying A and Drying B) in Step 2, but there is no coating liquid sample to be applied to the eighth sample (experimental frame 90 indicated by the dotted square in the figure) for each experimental condition in Step 2. For example, if Preparation A is applied to the eighth sample of Drying A, there will be two samples of Preparation A in Drying A (the first and eighth samples), resulting in a wasted experiment for Drying A × Preparation A.

[0055] Therefore, when the throughput of a certain process (e.g., 8 for process 2 in FIG. 9 ) is greater than the number of experimental conditions of another process (e.g., 7 for process 1 in FIG. 9 ), the experimental throughput specifying unit 102 may calculate the process throughput after reducing the "throughput of the certain process" to the "number of experimental conditions of the other process." For example, as shown in FIG. 8 , the experimental throughput specifying unit 102 may calculate 2×7=14 as the process throughput of process 2, and then calculate the experimental throughput.

[0056] In S200, the experimental condition acquisition unit 106 inputs arbitrary combination experimental conditions from the user. In the experimental condition determination method of this embodiment, as will be described later, the experimental condition acquisition unit 106 determines the combination experimental conditions. However, it is also possible that the user may wish to specify the experimental conditions they wish to try in the experiment. In such a case, the experimental condition acquisition unit 106 may first acquire arbitrary combination experimental conditions from the user. This makes it possible to generate conditions based on both the user's preferences and evaluation values. For example, it is possible to set some of the combination experimental conditions as conditions acquired from the user, and generate the other conditions based on the evaluation values.

[0057] 10 shows an example of combination experimental conditions input by the user in S200. For example, the experimental condition acquisition unit 106 may acquire three combination experimental conditions, combination experimental condition 110, combination experimental condition 120, and combination experimental condition 130, in S200. The horizontal axis of the graph in FIG. 10 indicates the numerical value of the experimental condition for step 1 included in the combination experimental conditions (e.g., the concentration of the raw materials in the preparation step), and the vertical axis indicates the numerical value of the experimental condition for step 2 (e.g., the drying temperature in the drying step).

[0058] For example, the combined experimental conditions 110 may include a raw material concentration of 0 wt% as an experimental condition for step 1 (preparation) and a drying temperature of 30°C as an experimental condition for step 2 (drying). For example, the combined experimental conditions 120 may include a raw material concentration of 5 wt% as an experimental condition for step 1 (preparation) and a drying temperature of 50°C as an experimental condition for step 2 (drying). For example, the combined experimental conditions 130 may include a raw material concentration of 10 wt% as an experimental condition for step 1 (preparation) and a drying temperature of 50°C as an experimental condition for step 2 (drying).

[0059] After S200, the experimental condition determination apparatus 10 performs a combination of S300 to S500 a number of times (N times) equal to the number of steps. In the first repetition, the experimental condition determination apparatus 10 sets n to 1, and may add 1 to n each time S300 to S500 are repeated, repeating the process until n = N. In S300, the option specifying unit 104 specifies multiple experimental condition options for the nth step.

[0060] Fig. 11 shows an example of the subflow of S300 when n = 1. The option specifying unit 104 may specify the nth (i.e., first) experimental condition option by executing the flow from S310 to S330 in Fig. 11 .

[0061] In S310, the option specifying unit 104 specifies a combination of experimental conditions in which the evaluation value of the evaluation function satisfies a predetermined standard without fixing the experimental conditions of all the steps. For example, the option specifying unit 104 specifies a combination of experimental conditions in which the evaluation value of the evaluation function f(x １ , x ２ …x Ｎ ) (x ｉ indicates the experimental conditions of the i-th step) satisfies a predetermined criterion.１ , x ２ …x Ｎ ) variable set (x １ , x ２ …x Ｎ ) to explore.

[0062] The evaluation function may output a numerical value relating to the predicted quality of the experimental result as an evaluation value, or a numerical value relating to the predicted quality of the experimental result. For example, the evaluation function may be a prediction function that inputs experimental conditions and outputs a predicted value relating to the quality of the experimental result itself. As an example, the evaluation function is expressed as f(x １ , x ２ ) and x １ is the experimental condition of step 1 (e.g., "0.05" for 5% of preparation B), x ２ may be the experimental condition of step 2 (for example, "130" for 130°C in drying A), and the predicted value of the film strength (for example, the breaking strength in MPa) may be output as the evaluation value.

[0063] Alternatively, the evaluation function may utilize a prediction function that inputs experimental conditions and outputs the accuracy of prediction regarding the pass / fail of the experimental result (e.g., a probability distribution of predicted values). When the pass / fail of the experimental result is expressed as a probability distribution by the prediction function, the evaluation function may output the predicted value with the highest probability in the prediction function as the evaluation value. Such a prediction function may be trained in advance using a known method.

[0064] The evaluation function may output a numerical value related to the certainty of the predicted experimental result as an evaluation value. The evaluation function may output the numerical value representing the certainty itself, or a numerical value positively correlated with the numerical value representing the certainty. In this case, the higher the accuracy of the prediction, the higher the evaluation value, since the result can be expected to be certain.

[0065] For example, if the success or failure of an experimental result is expressed as a probability distribution in the prediction function, the numerical value relating to the certainty of the experimental result may be the probability corresponding to the predicted value in the probability distribution of the film strength, e.g., the numerical value relating to the certainty of the experimental result may be the probability of the predicted value (the most probable value) of the film strength.

[0066] The evaluation function may output a numerical value related to the certainty of obtaining experimental results that meet or exceed the standard as an evaluation value. For example, the evaluation function may output an expected improvement or an upper confidence bound (UCB) associated with the predicted value as an evaluation value.

[0067] The success or failure of an experimental result may be predicted by multiple prediction functions, in which case the numerical value relating to the certainty of the experimental result may be the dispersion (e.g., variance or standard deviation) of the multiple predicted results.

[0068] The numerical value relating to the certainty of the experimental result may be the density of a combination of experimental conditions for which the experimental result is already known. For example, the numerical value relating to the certainty of the experimental result may be a data density or kernel density calculated from the known combination of experimental conditions at the coordinates corresponding to the input combination of experimental conditions.

[0069] The evaluation function may output a numerical value that is negatively correlated with the certainty of the experimental results in the probability distribution of film strength (for example, the inverse of the certainty of the experimental results). In this case, a low prediction certainty is highly evaluated as being worth exploring.

[0070] The evaluation function may be a combination of multiple functions with the same or different properties. For example, the evaluation function may be a function that outputs a numerical value relating to the pass / fail of the predicted experimental result and a function that outputs a numerical value relating to the certainty of the predicted experimental result, which are connected by the sum and / or product, etc.

[0071] For example, the evaluation function may be the sum of a function that outputs the predicted value of the experimental result itself and a function that outputs the certainty of the experimental result. In this case, the evaluation function will highly evaluate conditions that are more likely to produce good results.

[0072] As another example, the evaluation function may be the sum of a function that outputs the predicted value of the experimental result itself and a function that outputs the reciprocal of the certainty of the experimental result. In this case, the evaluation function will highly evaluate conditions that are likely to produce good results but whose results are difficult to predict without actually conducting the experiment.

[0073] The predetermined criterion may be "ranking in the top of a predetermined number of combination experiment conditions (for example, within the top 10)." In this case, the option specifying unit 104 specifies a plurality of combination experiment conditions whose evaluation values ​​of the evaluation function are in the top of a predetermined number of combination experiment conditions. The predetermined criterion may be "exceeding a threshold." In this case, the option specifying unit 104 may specify a plurality of combination experiment conditions whose evaluation values ​​of the evaluation function exceed a threshold.

[0074] The option specification unit 104 may specify combination experimental conditions whose evaluation values ​​satisfy a predetermined standard using one type of evaluation function. Alternatively, the option specification unit 104 may specify combination experimental conditions whose evaluation values ​​satisfy a predetermined standard using two or more types of evaluation functions. In this case, the option specification unit 104 may specify combination experimental conditions whose evaluation values ​​satisfy a predetermined standard using each of the multiple evaluation functions.

[0075] The option specifying unit 104 determines the combination experiment conditions (x １ , x ２ …x ｍ ) may be searched for by a known method. For example, the option specifying unit 104 may perform the search by random sampling, grid search, genetic algorithm, stochastic gradient descent, or the like.

[0076] 13 shows an example of combination experimental conditions identified in S310. In addition to combination experimental conditions 110 to 130 acquired in S200, the option identification unit 104 may acquire combination experimental conditions 140, 150, 160, and 170 in S310. As an example, combination experimental conditions 140 to 170 may be combination experimental conditions that give the top 1 to 4 evaluation values ​​from the search results when input into a certain evaluation function (for example, a function that outputs a predicted value of film strength as an evaluation value).

[0077] Next, in S330, the option specifying unit 104 specifies the experimental conditions for the first process included in the combination experimental conditions specified in S310 (hereinafter also referred to as "first experimental conditions") as the first experimental condition option.

[0078] When a plurality of combination experimental conditions are identified, the option identifying unit 104 may identify the first experimental conditions included in each of the plurality of combination experimental conditions as a plurality of first experimental condition options. When a combination experimental condition is identified for each of a plurality of evaluation functions, the option identifying unit 104 may identify the first experimental conditions included in each of the identified combination experimental conditions as a plurality of first experimental condition options.

[0079] For example, the option specifying unit 104 may specify the combination experiment conditions (x １ａ , x ２ａ …x Ｎａ ), combined experimental conditions (x １ｂ , x ２ｂ …x Ｎｂ ), combined experimental conditions (x １ｃ , x ２ｃ …x Ｎｃ ) ... and the combined experimental conditions (x １Ｍ , x ２Ｍ …x ＮＭ ) are identified, the option identifying unit 104 identifies x as the first experimental condition option. １ａ , x １ｂ , x １ｃ …x １Ｍ may be specified.

[0080] 14 shows an example of first experimental condition options identified in S330. The option identification unit 104 acquires the first experimental conditions for each of the combination experimental conditions 110 to 170 acquired up to S310. For example, the option identification unit 104 may acquire the first experimental condition 210 (e.g., concentration 0%) from the combination experimental condition 110, the first experimental condition 212 (e.g., concentration 5%) from the combination experimental condition 120, the first experimental condition 214 (e.g., concentration 10%) from the combination experimental condition 130, the first experimental condition 216 (e.g., concentration 20%) from the combination experimental condition 140, the first experimental condition 218 (e.g., concentration 30%) from the combination experimental condition 150, the first experimental condition 220 (e.g., concentration 40%) from the combination experimental condition 160, and the first experimental condition 222 (e.g., concentration 50%) from the combination experimental condition 170. The option specifying unit 104 may specify the acquired first experimental conditions 210 to 222 as first experimental condition options.

[0081] Fig. 12 shows an example of a subflow of S300 when n is equal to or greater than 2. The option specifying unit 104 may specify the nth (n>1) experimental condition option by executing the flow from S312 to S332 in Fig. 12 .

[0082] In S312, the option specifying unit 104 specifies a combination of experimental conditions that includes the already specified experimental condition options as part of the experimental conditions and whose evaluation value of the evaluation function satisfies a predetermined criterion.

[0083] Here, a specific description will be given of the case where n = 2 (i.e., the first S312). The option specifying unit 104 may specify a combination of experimental conditions that includes the first experimental condition option as part of the experimental conditions and whose evaluation value of the evaluation function satisfies a predetermined criterion.

[0084] The option specifying unit 104 may output, for each second experimental condition, an overall evaluation of the evaluation values ​​of the evaluation function for the plurality of first experimental condition options specified in S310. １ , x ２ …x Ｎ ) satisfies the predetermined criteria. １ , x ２ …x Ｎ ) to explore.

[0085] For example, if an experimental condition option is specified for the i-th step in the previous process of S300, x ｉ In the example of FIG. 14, the first experimental condition option is x １ For example, the option specifying unit 104 selects the concentrations of 0%, 5%, 10%, 20%, 30%, 40%, and 50% for the experimental condition x ２ The evaluation of the evaluation function f(0%, x ２ ), evaluation function f(5%, x ２ ), evaluation function f(10%, x ２ ), evaluation function f(20%, x ２ ), evaluation function f(30%, x ２ ), evaluation function f(40%, x ２ ) and evaluation function f(50%, x ２) may be determined based on a comprehensive evaluation of the above.

[0086] As an example, the option specifying unit 104 may select the second experimental condition x ２ The evaluation of (dry 30°C) may be determined by a comprehensive evaluation of the evaluation function f(0%, 30°C), the evaluation function f(5%, 30°C), the evaluation function f(10%, 30°C), the evaluation function f(20%, 30°C), the evaluation function f(30%, 30°C), the evaluation function f(40%, 30°C), and the evaluation function f(50%, 30°C).

[0087] The option specifying unit 104 may output the average or median of the evaluation values ​​of the evaluation function including the second experimental condition and having each of the multiple first experimental condition options as the overall evaluation for the second experimental condition. That is, the option specifying unit 104 may output the average or median of the evaluation values ​​of the evaluation function f(0%, 30°C), evaluation function f(5%, 30°C), ... evaluation function f(50%, 30°C) for the second experimental condition x ２ This allows the second experimental conditions to be evaluated from a comprehensive perspective, including their advantages and disadvantages.

[0088] Alternatively or additionally, the option specifying unit 104 may output the maximum or minimum value of the evaluation values ​​of the evaluation function including the second experimental condition and having each of the plurality of first experimental condition options as the overall evaluation for the second experimental condition. That is, the option specifying unit 104 may output the maximum or minimum value of the evaluation function f(0%, 30°C), evaluation function f(5%, 30°C), ... evaluation function f(50%, 30°C) for the second experimental condition x ２ This may be output as an overall evaluation of the second experimental conditions (drying at 30°C). This allows the second experimental conditions to be evaluated based on the best evaluation or the minimum evaluation that can be obtained.

[0089] Similarly, the option specifying unit 104 selects the second experimental condition x ２ In addition to (drying 30 ° C), many other second experimental conditions x ２ (For example, drying at 31°C, drying at 32°C, etc.) and search for a combination of all process experimental conditions (x １ , x ２ …x ＮThe predetermined criteria, the type / number of evaluation functions, the search method, etc. may be the same as those described in S310.

[0090] The option identification unit 104 calculates an overall evaluation for each second experimental condition as described above and determines the relative merits of the second experimental conditions. As an example, the option identification unit 104 compares the overall evaluation of the evaluation function f (0%, 30°C) to the evaluation function f (50%, 30°C) with the overall evaluation of the evaluation function f (0%, 40°C) to the evaluation function f (50%, 40°C), and if the former is superior, the second experimental condition (dry 30°C) is deemed to be superior to the second experimental condition (dry 40°C). The option identification unit 104 may identify superior second experimental conditions by performing a search.

[0091] Next, in S332, the nth experimental condition for the nth process included in the combination experimental conditions identified in S312 is identified as the nth experimental condition option. For example, the option identification unit 104 may identify the second experimental condition for the second process included in the combination experimental conditions identified in S312 for the first time as the second experimental condition option. In this way, the option identification unit 104 can identify the nth experimental condition whose overall evaluation satisfies a predetermined standard as the nth experimental condition option.

[0092] 15 shows an example of the second experimental conditions searched for in S312. For example, the option identification unit 104 obtains an overall evaluation of the evaluation values ​​of a group of combined experimental conditions 190 including a combined experimental condition 182, a combined experimental condition 184, a combined experimental condition 186, ..., and a combined experimental condition 188, all of which have the same second experimental condition for process 2 (drying). Similarly, the option identification unit 104 obtains an overall evaluation of the evaluation values ​​of a group of combined experimental conditions 192, which has a different second experimental condition.

[0093] The option specifying unit 104 searches for second experimental conditions based on the overall evaluation. Here, the first experimental conditions in the combined experimental condition group 190, the combined experimental condition group 192, etc. are the first experimental conditions 210 to 222 specified as first experimental condition options.

[0094] 16 shows examples of second experimental condition options identified in S332. As a result of the search, the option identification unit 104 identifies second experimental conditions that satisfy predetermined criteria. For example, as shown in the figure, the option identification unit 104 may identify second experimental condition 310, second experimental condition 312, second experimental condition 314, second experimental condition 316, second experimental condition 318, and second experimental condition 320 as second experimental condition options. This allows the option identification unit 104 to identify experimental condition options in both step 1 and step 2.

[0095] The option identification unit 104 may execute S300 for the nth process, but may omit the processing of S300 for processes that do not require setting experimental conditions, such as process 3 (evaluation), and processes that have only one experimental condition.

[0096] Although the above description is given for the case where n = 2, the option specifying unit 104 specifies multiple n-th experimental conditions whose overall evaluation for the n-th step satisfies a predetermined standard. For example, the option specifying unit 104 may acquire a predetermined number of combination experimental conditions with the highest evaluation values ​​of the evaluation function for the second experimental conditions related to the second step.

[0097] When there are three or more processes that include multiple experimental conditions (i.e., when n≧3 is possible), the option identification unit 104 may set the experimental conditions of processes prior to the nth process where an option has already been identified (e.g., the first process where the first experimental condition option is identified) and processes other than the nth process that is the target of evaluation and search (e.g., the second process where the second experimental condition is identified) to the same experimental conditions, or may set different experimental conditions.

[0098] As an example, in addition to the above-mentioned step 1 (preparation) and step 2 (drying), step 3 (pressing) is included instead of step 3 (evaluation), and experimental condition x is used as the pressing condition (pressure). ３ In this case, the option specifying unit 104 calculates the overall evaluation of the experimental condition x (for example, 10 MPa) of the third step. ３ , the evaluation function f(0%, 30°C, x ３ ) ~ Evaluation function f (50%, 30°C, x ２ ) (experimental conditions x １= 0 to 50%, the same applies below), evaluation function f(0%, 50°C, x ３ ) ~ Evaluation function f(50%, 50°C, x ２ ), evaluation function f(0%, 70°C, x ３ ) ~ Evaluation function f (50%, 70 ° C, x ２ ), evaluation function f(0%, 90°C, x ３ ) ~ Evaluation function f (50%, 90 ° C, x ２ ), evaluation function f(0%, 110°C, x ３ ) ~ Evaluation function f (50%, 110 ° C, x ２ ), and the evaluation function f(0%, 130°C, x ３ ) ~ Evaluation function f (50%, 130 ° C, x ２ ) may be evaluated based on the overall evaluation.

[0099] When there are a large number of evaluation functions to be calculated, the option specification unit 104 may calculate the overall evaluation using only the evaluation values ​​of some of the evaluation functions, instead of calculating the evaluation values ​​of all of the evaluation functions, as necessary. For example, the option specification unit 104 may calculate the evaluation values ​​of only some of the evaluation functions by periodically or randomly omitting some of the evaluation functions.

[0100] In S400, following S300, the option identification unit 104 determines whether experimental condition options have been identified for all steps. For example, the option identification unit 104 determines whether n has reached the total number of steps, N. If all steps have been completed (e.g., n=N), the option identification unit 104 proceeds to S600; otherwise, the option identification unit 104 proceeds to S500.

[0101] In S500, the option specifying unit 104 adds 1 to n and proceeds to S300, whereby in the next S300, an experimental condition option can be specified for the new nth step.

[0102] In S600, the experimental condition acquisition unit 106 acquires combination experimental conditions, which are combinations of experimental conditions for multiple processes, from multiple experimental condition options. The experimental condition acquisition unit 106 determines the number of combination experimental conditions to acquire based on the experimental throughput identified in S100.

[0103] For example, the experimental condition acquisition unit 106 may acquire combined experimental conditions so that the total number of processing amounts under all combined experimental conditions does not exceed the experimental processing amount. Furthermore, for example, the experimental condition acquisition unit 106 may acquire combined experimental conditions so that the total number of combined experimental conditions is equal to the experimental processing amount or a multiple thereof. For example, if the experimental processing amount specified in S100 is 18, the experimental condition acquisition unit 106 may acquire 18, 36, 48, etc. combined experimental conditions in S600. The experimental condition acquisition unit 106 may acquire a number of combined experimental conditions that is greater than or equal to k×a−0.2×a (decimals are rounded down, rounded up, or rounded to the nearest integer) and less than or equal to k×a, where a is the experimental processing amount (an integer) and k is an arbitrary integer. This allows the experimental condition acquisition unit 106 to efficiently utilize not only the given experimental environment but also the computer's computational and storage resources.

[0104] The experimental condition acquisition unit 106 acquires combined experimental conditions by identifying experimental conditions from a plurality of experimental condition options for each process. The experimental condition acquisition unit 106 may acquire an evaluation value of an evaluation function for a combination of experimental condition options related to a plurality of processes. The experimental condition acquisition unit 106 may then select and acquire combined experimental conditions from the combination of experimental condition options so as to satisfy the processing volume limit for each experimental condition and the evaluation value meets a predetermined standard.

[0105] The predetermined criterion may be that the evaluation value of the evaluation function is equal to or greater than a threshold value, or that the evaluation value of the evaluation function is equal to or greater than a predetermined rank. For example, the experimental condition acquisition unit 106 may acquire evaluation values ​​of the evaluation function for all combinations of experimental condition options related to multiple processes, and select the combined experimental condition with the highest evaluation value.

[0106] 17 shows an example of all combinations of experimental condition options that are candidates for combined experimental conditions in S600. As described above, by performing S300 multiple times, first experimental condition 210 to first experimental condition 222 are identified as seven first experimental condition options related to process 1 (preparation), and second experimental condition 310 to second experimental condition 320 are identified as six second experimental condition options related to process 2 (drying).

[0107] When generating combined experimental conditions from seven first experimental condition options and six second experimental condition options, it is possible to generate 7 x 6 = 42 combination experimental conditions 194 (indicated by dotted circles in the figure). From these 42 combination experimental conditions, the experimental condition acquisition unit 106 selects combination experimental conditions that are equal to the experimental throughput or a multiple thereof (e.g., 18) based on the evaluation value of the evaluation function.

[0108] The experimental condition acquisition unit 106 may use a penalty function to avoid selecting a combination of experimental conditions that have already been selected and / or executed, and / or a similar combination of experimental conditions. The experimental condition acquisition unit 106 may select a combination of experimental conditions using the sum of the evaluation function and the penalty function.

[0109] The penalty function may be one that outputs a predetermined large negative value when the selected and executed combined experimental conditions are input. Additionally / alternatively, the penalty function may be one that outputs a negative value (e.g., the reciprocal of the negative distance (or a power of the negative distance)) according to the closeness of the selected and executed combined experimental conditions.

[0110] As an example, the penalty function may be a Gaussian function or a ReLU function expressed by the following Equation 1. ...Equation 1 where X is the spatial coordinate of the combination experimental condition to be evaluated, x ｉ may be the spatial coordinates of the combination experimental conditions that have already been selected and / or executed. ０ may be the penalty strength, and r may be the overlap determination distance.

[0111] Here, the experimental condition acquisition unit 106 may select the number of combined experimental conditions obtained by subtracting the combined experimental conditions already input in S200 from the experimental throughput (or a multiple thereof). For example, the experimental condition acquisition unit 106 may select 15 combined experimental conditions from the 18 experimental throughputs by subtracting the three input combined experimental conditions 110 to 130. Alternatively, the experimental condition acquisition unit 106 may select the combined experimental conditions of the experimental throughput or a multiple thereof (e.g., 18), regardless of the input in S200.

[0112] FIG. 18 shows an example of the combined experimental conditions selected in S600. The experimental condition acquisition unit 106 may select 15 combined experimental conditions 196 (indicated by hatched circles in the figure) from 42 candidates. Here, the experimental condition acquisition unit 106 selects the combined experimental conditions 196 so as to satisfy the processing volume limit for each experimental condition. For example, the experimental condition acquisition unit 106 selects the combined experimental conditions 196 so that the number of experimental conditions of another process executed together with one experimental condition of a certain process does not exceed the processing volume of the certain process. The combined experimental condition 198 (indicated by a dotted circle) indicates a combined experimental condition that was not selected.

[0113] 2 and 3, the processing amount of one experimental condition in step 1 (preparation) is limited to 3, and the processing amount of one experimental condition in step 2 (drying) is limited to 4. In such a case, the experimental condition acquisition unit 106 ensures that the number of experimental conditions in step 2 (drying) does not exceed 3 for one experimental condition in step 1 (preparation).

[0114] 18, for example, each vertical column corresponding to the first experimental condition (step 1) does not contain more than three combined experimental conditions, and each horizontal column corresponding to the second experimental condition (step 2) does not contain more than four combined experimental conditions. In this way, the experimental condition acquisition unit 106 identifies combined experimental conditions with high evaluation values ​​while satisfying the processing volume limit.

[0115] The experimental condition acquisition unit 106 may identify the combination experimental conditions using a known method. The experimental condition acquisition unit 106 may acquire the evaluation values ​​of all candidate combination experimental conditions (e.g., 42) and identify the combination experimental condition with the highest evaluation value from among them. Alternatively, the experimental condition acquisition unit 106 may identify the combination experimental conditions based on the evaluation values ​​of only some of the combination experimental conditions.

[0116] 19 shows an example of a method for identifying the combination experiment conditions in S600. For example, the experiment condition acquisition unit 106 may identify the combination experiment conditions in S600 by applying an existing algorithm for solving the minimum cost flow problem. In this case, the selection of the combination experiment conditions is represented by a graph 1900.

[0117] The experimental condition acquisition unit 106 selects a path that starts from node S, passes through the first layer 50 corresponding to process 1, and the second layer 60 corresponding to process 2, and reaches node D corresponding to process 3. The first layer 50 includes nodes 51, 52, 53, 54, 55, 56, and 57 corresponding to seven first experimental condition options. The second layer 60 includes nodes 61, 62, 63, 64, 65, and 66 corresponding to six first experimental condition options.

[0118] A capacity is assigned to each node, which corresponds to the amount of processing. For example, three capacities are assigned to nodes 51 to 57 on the first layer 50, and four capacities are assigned to nodes 61 to 66 on the second layer 60. If one path has already been set between nodes 51 to 53 in S200, the remaining capacities of nodes 51 to 53 are 3-1 = 2. Similarly, if one path has been set to node 61 and two paths to node 62 in S200, the remaining capacity of node 61 is 4-1 = 3, and the remaining capacity of node 62 is 4-2 = 2.

[0119] The cost of the minimum cost flow problem may be a number (e.g., the reciprocal of the evaluation value, or a number obtained by subtracting the evaluation value from a predetermined constant) that is negatively correlated with the evaluation value of the evaluation function obtained from the combination experimental conditions corresponding to the path from the first layer 50 to the second layer 60. For example, the cost from node 51 to node 61 may be the reciprocal of the evaluation value obtained by inputting the combination experimental conditions corresponding to nodes 1 and 2 into the evaluation function.

[0120] Here, the experimental condition acquisition unit 106 identifies routes for the experimental processing volume (e.g., 18 routes). When adopting the combination experimental conditions input by the user in S200, the number of routes identified is the number excluding the input combination experimental conditions. For example, in the example of FIG. 19, three routes, from node 51 to node 61, from node 52 to node 62, and from node 53 to node 6, are input as routes input in S200. In this case, the experimental condition acquisition unit 106 identifies 18-3=15 routes.

[0121] The experimental condition acquisition unit 106 may finally output combination experimental conditions corresponding to the combination of nodes in the first layer 50 and the second layer 60 corresponding to the identified path. An experiment may be performed under conditions corresponding to the output combination experimental conditions.

[0122] An experiment corresponding to the output combination of experimental conditions may be performed, the results may be measured, and the evaluation function may be updated based on the measurement results. After that, the processes of S100 to S600 may be performed again, using the updated evaluation function. This allows for greater efficiency with each repetition of the experiment.

[0123] As described above, according to this embodiment, in a situation where the number of experiments that can be performed is limited due to constraints such as experimental equipment and time, it is possible to efficiently output a set of combined experimental conditions that predicts results based on evaluation values ​​and minimizes waste. This makes it possible to prevent the waste of computer computational resources and storage resources that would otherwise be incurred by preparing and evaluating wasteful experimental plans. Furthermore, by executing such experimental plans, it is possible to increase the operating rate of each experimental device. As a result, it becomes possible to efficiently acquire experimental data. Furthermore, samples prepared in each experiment can be used without waste.

[0124] In the above example, the first experimental condition option is extracted in S300 from the combination experimental conditions including the combination experimental conditions input by the user in S200. In other words, the option for selection in S600 is extracted from the combination experimental conditions input by the user.

[0125] However, in a modification of this embodiment, it is not necessary to extract the first experimental condition option from the combined experimental conditions input by the user in the first S300. In this case, in S600, the combined experimental conditions input by the user in S200 may be selected first. According to this modification, the influence of the user's input is eliminated from the extraction of experimental condition options, and the experimental condition options can be determined purely based on the evaluation value of the evaluation function.

[0126] In the above example, all of the experimental condition options for each process are identified in a single S300 (e.g., all of the first experimental condition options for process 1 are identified in the first S300), but this is not limited to this. The option identification unit 104 may identify experimental condition options for one process multiple times. For example, after identifying four first experimental condition options for process 1 (preparation) in the first S300, six second experimental condition options for process 2 (drying) in the second S300, and then three additional first experimental condition options for process 1 (preparation) in the third S300. In this case, instead of identifying all seven first experimental condition options in the first S300, a total of seven first experimental condition options are identified in the first S300 and the third S300.

[0127] Various embodiments of the present invention may be described with reference to flowcharts and block diagrams, where the blocks may represent (1) stages of a process in which operations are performed or (2) sections of apparatus responsible for performing the operations. Particular stages and sections may be implemented by dedicated circuitry, programmable circuitry provided with computer-readable instructions stored on a computer-readable medium, and / or a processor provided with computer-readable instructions stored on a computer-readable medium. Dedicated circuitry may include digital and / or analog hardware circuitry, and may include integrated circuits (ICs) and / or discrete circuits. Programmable circuitry may include reconfigurable hardware circuitry including logical AND, OR, XOR, NAND, NOR, and other logic operations, flip-flops, registers, memory elements such as field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), and the like.

[0128] A computer-readable medium may include any tangible device capable of storing instructions that are executed by an appropriate device, such that the computer-readable medium having instructions stored thereon comprises an article of manufacture containing instructions that can be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of computer-readable media may include electronic, magnetic, optical, electromagnetic, and semiconductor storage media. More specific examples of computer-readable media may include floppy disks, diskettes, hard disks, random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), electrically erasable programmable read-only memories (EEPROMs), static random access memories (SRAMs), compact disc read-only memories (CD-ROMs), digital versatile discs (DVDs), Blu-ray discs, memory sticks, integrated circuit cards, and the like.

[0129] The computer readable instructions may include either assembler instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages ​​such as the “C” programming language or similar programming languages.

[0130] The computer-readable instructions may be provided to a processor or programmable circuitry of a programmable data processing apparatus, such as a general-purpose computer, special-purpose computer, or other computer, either locally or over a local area network (LAN), a wide area network (WAN) such as the Internet, etc., which executes the computer-readable instructions to create means for performing the operations specified in the flowcharts or block diagrams. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc.

[0131] 20 illustrates an example of a computer 2200 in which aspects of the present invention may be embodied, in whole or in part. Programs installed on the computer 2200 may cause the computer 2200 to function as or perform operations associated with an apparatus or one or more sections of the apparatus according to embodiments of the present invention, and / or to perform a process or steps of a process according to embodiments of the present invention. Such programs may be executed by the CPU 2212 to cause the computer 2200 to perform specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

[0132] A computer 2200 according to this embodiment includes a CPU 2212, a RAM 2214, a graphics controller 2216, and a display device 2218, which are interconnected by a host controller 2210. The computer 2200 also includes input / output units such as a communication interface 2222, a hard disk drive 2224, a DVD-ROM drive 2226, and an IC card drive, which are connected to the host controller 2210 via an input / output controller 2220. The computer also includes legacy input / output units such as a ROM 2230 and a keyboard 2242, which are connected to the input / output controller 2220 via an input / output chip 2240.

[0133] The CPU 2212 operates according to programs stored in the ROM 2230 and RAM 2214, thereby controlling each unit. The graphics controller 2216 acquires image data generated by the CPU 2212 into a frame buffer or the like provided in the RAM 2214 or into the graphics controller 2216 itself, and causes the image data to be displayed on the display device 2218.

[0134] The communication interface 2222 communicates with other electronic devices via a network. The hard disk drive 2224 stores programs and data used by the CPU 2212 in the computer 2200. The DVD-ROM drive 2226 reads programs or data from the DVD-ROM 2201 and provides the programs or data to the hard disk drive 2224 via the RAM 2214. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0135] ROM 2230 stores therein a boot program or the like that is executed by computer 2200 upon activation, and / or programs that depend on the hardware of computer 2200. I / O chip 2240 may also connect various I / O units to I / O controller 2220 via parallel ports, serial ports, keyboard ports, mouse ports, etc.

[0136] The programs are provided by a computer-readable medium such as a DVD-ROM 2201 or an IC card. The programs are read from the computer-readable medium, installed in the hard disk drive 2224, RAM 2214, or ROM 2230, which are also examples of computer-readable media, and executed by the CPU 2212. Information processing described in these programs is read by the computer 2200, and brings about cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be configured by implementing information manipulation or processing in accordance with the use of the computer 2200.

[0137] For example, when communication is performed between computer 2200 and an external device, CPU 2212 may execute a communication program loaded in RAM 2214 and instruct communication interface 2222 to perform communication processing based on the processing described in the communication program. Under the control of CPU 2212, communication interface 2222 reads transmission data stored in a transmission buffer processing area provided in RAM 2214, hard disk drive 2224, DVD-ROM 2201, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes received data received from the network to a reception buffer processing area or the like provided on the recording medium.

[0138] The CPU 2212 may cause all or a necessary portion of a file or database stored on an external recording medium such as the hard disk drive 2224, the DVD-ROM drive 2226 (DVD-ROM 2201), an IC card, etc. to be read into the RAM 2214, and may perform various types of processing on the data on the RAM 2214. The CPU 2212 then writes back the processed data to the external recording medium.

[0139] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. The CPU 2212 may perform various types of processing on data read from the RAM 2214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, etc., as described throughout this disclosure and specified by the instruction sequences of the programs, and write the results back to the RAM 2214. The CPU 2212 may also search for information in a file, database, etc. on the recording medium. For example, if multiple entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored on the recording medium, the CPU 2212 may search for an entry that matches a condition specified by the attribute value of the first attribute from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0140] The above-described programs or software modules may be stored in a computer-readable medium on or near the computer 2200. A recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can also be used as a computer-readable medium, thereby providing the programs to the computer 2200 via the network.

[0141] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

[0142] It should be noted that the execution order of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before," "prior to," etc., and can be implemented in any order unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, this does not mean that the process must be implemented in this order. The expression "A and / or B" may mean "A, B, or A and C." The expression "A, B, and / or C" may mean "any one of A, B, and C, or any combination of two or more of these."

[0143] 10 Experimental condition determination device 22 Preparation process 24 Drying process 26 Evaluation process 50 First layer 51 Node 52 Node 53 Node 54 Node 55 Node 56 Node 57 Node 60 Second layer 61 Node 62 Node 63 Node 64 Node 65 Node 66 Node 90 Experimental frame 102 Experimental processing amount specification unit 104 Choice specification unit 106 Experimental condition acquisition unit 108 Storage unit 110 Combined experimental conditions 120 Combined experimental conditions 130 Combined experimental conditions 140 Combined experimental conditions 150 Combined experimental conditions 160 Combined experimental conditions 170 Combined experimental conditions 182 Combined experimental conditions 184 Combined experimental conditions 186 Combined experimental conditions 188 Combined experimental conditions 190 Combined experimental condition group 192 Combined experimental condition group 194 Combined experimental condition 196 Combined experimental condition 198 Combined experimental condition 212 First experimental condition 214 First experimental condition 216 First experimental condition 218 First experimental condition 220 First experimental condition 222 First experimental condition 310 Second experimental condition 312 Second experimental condition 314 Second experimental condition 316 Second experimental condition 318 Second experimental condition 320 Second experimental condition 1900 Graph 2200 Computer 2201 DVD-ROM 2210 Host controller 2212 CPU 2214 RAM 2216 Graphics controller 2218 Display device 2220 Input / output controller 2222 Communication interface 2224 Hard disk drive 2226 DVD-ROM drive 2230 ROM 2240 Input / Output Chip 2242 Keyboard

Claims

1. An apparatus for determining experimental conditions for an experiment that can process multiple experimental conditions in parallel and includes multiple processes in which the processing volume for each experimental condition is limited, comprising: an option identification unit that identifies multiple experimental condition options for each process based on the output value of an evaluation function that inputs the experimental conditions and outputs an evaluation value; and an experimental condition acquisition unit that acquires a combined experimental condition, which is a combination of experimental conditions for multiple processes, from the multiple experimental condition options based on the processing volume for each experimental condition and the output value of the evaluation function obtained by inputting the experimental condition options.

2. The experimental condition determination device of claim 1, further comprising an experimental processing amount determination unit that determines the experimental processing amount, which is the processing amount that can be processed at one time for the entire experiment, from the number of experimental conditions that can be processed in parallel in each process and the processing amount for each of the experimental conditions, and the experimental condition acquisition unit acquires the combined experimental conditions so that the total number of processing amounts for all of the combined experimental conditions does not exceed the experimental processing amount.

3. The experimental condition determination device described in claim 2, wherein the experimental processing volume determination unit calculates the process processing volume, which is the processing volume that can be processed at one time in each process, from the product of the number of experimental conditions that can be processed in parallel in each process and the processing volume for each experimental condition, and identifies the smallest process processing volume among all processes as the experimental processing volume.

4. The experimental condition determination device according to any one of claims 1 to 3, wherein the experimental condition acquisition unit receives input of any combination of experimental conditions from a user.

5. An experimental condition determination device described in any one of claims 1 to 4, wherein the option identification unit identifies a combination experimental condition for which the output value of the evaluation function is better than a predetermined standard, and identifies a first experimental condition for a first process included in the identified combination experimental condition as a first experimental condition option.

6. The experimental condition determination device described in claim 5, wherein the option identification unit identifies a plurality of combination experimental conditions whose output values ​​of the evaluation function are at the top of a predetermined number, and identifies the first experimental conditions included in each of the plurality of combination experimental conditions as a plurality of first experimental condition options.

7. The experimental condition determination device described in claim 5, wherein the option identification unit identifies a combination experimental condition whose output value is better than a predetermined standard for each of the multiple evaluation functions, and identifies the first experimental condition included in each of the identified combination experimental conditions as a first experimental condition option.

8. The experimental condition determination device described in claim 5, wherein the option identification unit identifies a combination experimental condition that includes the first experimental condition option as part of the experimental conditions and has an output value of the evaluation function that is better than a predetermined standard, and identifies a second experimental condition for a second process included in the identified combination experimental condition as a second experimental condition option.

9. The experimental condition determination device described in claim 8, wherein the option identification unit outputs an overall evaluation of the output values ​​of the evaluation function for each of the second experimental conditions for a plurality of the first experimental condition options, and identifies the second experimental condition whose overall evaluation is better than a predetermined standard as the second experimental condition option.

10. The experimental condition determination device described in claim 9, wherein the option identification unit outputs the average or median of the output values ​​of the evaluation function that includes the second experimental condition and has each of the multiple first experimental condition options as the overall evaluation for the second experimental condition.

11. The experimental condition determination device described in claim 9, wherein the option identification unit outputs the maximum or minimum value among the output values ​​of the evaluation function that includes the second experimental condition and has each of the multiple first experimental condition options as the overall evaluation for the second experimental condition.

12. An experimental condition determination device described in any one of claims 1 to 11, wherein the experimental condition acquisition unit acquires the output value of the evaluation function for a combination of the experimental condition options relating to multiple processes, and selects and acquires the combined experimental condition from the combination of the experimental condition options so as to satisfy the processing volume limit for each experimental condition and so that the output value of the evaluation function satisfies a predetermined standard.

13. The experimental condition determination device according to any one of claims 1 to 12, wherein the evaluation function outputs a numerical value relating to the expected quality of the experimental result when an experiment is conducted under input experimental conditions.

14. An experimental condition determination device according to any one of claims 1 to 12, wherein the evaluation function outputs a numerical value relating to the certainty of the predicted experimental results when an experiment is conducted under the input experimental conditions.

15. A program that, when executed by a computer, causes the computer to function as the experimental condition determination device according to any one of claims 1 to 14.

16. A method for determining experimental conditions for an experiment that can process multiple experimental conditions in parallel and includes multiple processes in which the processing volume for each experimental condition is limited, using an experimental condition determination device described in any one of claims 1 to 14, the method comprising: an option identification step for identifying multiple experimental condition options for each process based on the output value of an evaluation function that inputs experimental conditions and outputs an evaluation value; and an experimental condition acquisition step for acquiring a combined experimental condition, which is a combination of experimental conditions for multiple processes, from the multiple experimental condition options based on the processing volume for each experimental condition and the output value of the evaluation function obtained by inputting experimental condition options.

Images