Information processing device, optimization support method, and optimization support program

Abstract

This information processing device is provided with: a strategy acquisition unit that acquires a strategy for calculating an optimal price and procurement amount for a product, said strategy being generated by optimization calculation using a predicted value of a first parameter representing the probability distribution of demand for the product with respect to the price thereof and a predicted value of a second parameter representing the transition probability of the procurement cost of the product; and an optimal value calculation unit that calculates a price and a procurement amount for the product using the strategy.

Description

Information processing device, optimization support method, and optimization support program

[0001] The present disclosure relates to an information processing device, an optimization support method, and an optimization support program.

[0002] There is known a technique for determining an optimal price for a product through optimization calculation using a function that represents customer demand. For example, Patent Document 1 listed below discloses a method for determining an optimal price, which includes the steps of obtaining a demand function, forming an objective function using the demand function and the net profit of the product, and automatically optimizing the objective function with respect to the net profit to generate an optimal price proposal.

[0003] Japan Special Table Publication No. 2005-515531

[0004] The method described in Patent Document 1 performs calculations assuming that all direct costs incurred in the production process of a product or service are known, but in reality, various direct costs can fluctuate. In particular, when purchasing or manufacturing a product, fluctuations in the procurement costs of that product have a significant impact on profits. However, the method described in Patent Document 1 has the problem of being unable to calculate the optimal price while taking fluctuations in the procurement costs of the product into account.

[0005] The present disclosure has been made in consideration of the above-mentioned problems, and one exemplary purpose thereof is to provide a technology that enables calculation of the optimal price and procurement quantity of a product taking into account fluctuations in the procurement cost of the product.

[0006] An information processing device according to an exemplary aspect of the present disclosure includes a strategy acquisition means for acquiring a strategy for calculating the optimal price and procurement quantity of the product, the strategy being generated by optimization calculation using a predicted value of a first parameter representing the probability distribution of demand relative to the price of the product and a predicted value of a second parameter representing the transition probability of the procurement cost of the product, and an optimal value calculation means for calculating the price and procurement quantity of the product using the strategy.

[0007] An optimization support method according to an exemplary aspect of the present disclosure includes a strategy acquisition process in which at least one processor acquires a strategy for calculating the optimal price and procurement quantity of the product, the strategy being generated by an optimization calculation using a predicted value of a first parameter representing the probability distribution of demand relative to the price of the product and a predicted value of a second parameter representing the transition probability of the procurement cost of the product, and an optimal value calculation process in which the strategy is used to calculate the price and procurement quantity of the product.

[0008] An optimization support program according to an exemplary aspect of the present disclosure causes a computer to function as a strategy acquisition means for acquiring a strategy for calculating the optimal price and procurement quantity of a product, the strategy being generated by an optimization calculation using a predicted value of a first parameter representing the probability distribution of demand relative to the price of the product and a predicted value of a second parameter representing the transition probability of the procurement cost of the product, and an optimal value calculation means for calculating the price and procurement quantity of the product using the strategy.

[0009] According to one exemplary aspect of the present disclosure, an exemplary effect is achieved in that a technology can be provided that enables calculation of the optimal price and procurement quantity of a product taking into account fluctuations in the procurement cost of the product.

[0010] 1 is a block diagram showing the configuration of an information processing device according to the present disclosure; FIG. 2 is a flow diagram showing the flow of an optimization support method according to the present disclosure; FIG. 3 is a diagram showing an overview of an optimization support system according to the present disclosure; FIG. 4 is a block diagram showing the configuration of an information processing device included in the optimization support system shown in FIG. 3; FIG. 5 is a diagram showing a specific example of processing executed by the information processing device shown in FIG. 4; FIG. 5 is a flow diagram showing the flow of processing executed by the information processing device shown in FIG. 4; FIG. 6 is a block diagram showing the configuration of a computer that functions as an information processing device according to the present disclosure.

[0011] The following are examples of embodiments of the present invention. However, the present invention is not limited to the exemplary embodiments shown below, and various modifications are possible within the scope of the claims. For example, embodiments obtained by appropriately combining the technologies (part or all of the products or methods) employed in the exemplary embodiments shown below may also be included in the scope of the present invention. Furthermore, embodiments obtained by appropriately omitting some of the technologies employed in the exemplary embodiments shown below may also be included in the scope of the present invention. Furthermore, the effects mentioned in the exemplary embodiments shown below are examples of effects expected in the exemplary embodiments, and do not define the scope of the present invention. In other words, embodiments that do not exhibit the effects mentioned in the exemplary embodiments shown below may also be included in the scope of the present invention.

[0012] [First Exemplary Embodiment] A first exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. This exemplary embodiment is a basic form for each of the exemplary embodiments described below. Note that the scope of application of each technique employed in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technique employed in this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise. Furthermore, each technique shown in the drawings referenced to explain this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise.

[0013] (Configuration of information processing device 1) The configuration of the information processing device 1 will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of the information processing device 1. As shown in Fig. 1, the information processing device 1 includes a measure acquisition unit 101 and an optimal value calculation unit 102.

[0014] The policy acquisition unit 101 acquires a policy for calculating the optimal price and procurement quantity of a product. More specifically, the policy acquisition unit 101 acquires a policy generated by an optimization calculation using a predicted value of a first parameter representing a probability distribution of demand for the product price and a predicted value of a second parameter representing a transition probability of the procurement cost of the product.

[0015] The "product" may be a final product or an intermediate product (such as a part). The "product" may be either an article or a service. The "procurement cost" is the cost of procuring the product. For example, when purchasing products from a wholesaler or manufacturer and selling them, the purchase price is the "procurement cost." When manufacturing and selling products, the manufacturing cost is the "procurement cost." When providing a service to transport cargo, the cost of renting a means of transportation such as an aircraft or purchasing cargo space is the "procurement cost."

[0016] In addition, the "transition probability of procurement costs" is the transition probability of procurement costs at a certain point in time. ｓ , the procurement cost at other times is c ｓ’ When this is done, the procurement cost c ｓ is c ｓ’ If the transition probability of procurement costs is known, it is possible to predict, for example, the procurement costs required for the next procurement of a product.

[0017] Here, the criteria for determining the "optimum" in the "optimum price and procurement quantity of the product" can be arbitrary. For example, the "optimum" may be determined to be the maximum expected profit obtained from the sale of the product.

[0018] The "policy" is a function for calculating the optimal price and procurement quantity of the product. The "policy" can also be referred to as an estimation model for estimating the optimal price and procurement quantity of the product. The optimization calculation method is not particularly limited. For example, the policy acquisition unit 101 may acquire a policy generated by dynamic programming. The policy acquisition unit 101 may also acquire a policy by generating it, or may acquire a policy generated by another device.

[0019] The optimal value calculation unit 102 calculates the price and procurement quantity of the above product using the policy acquired by the policy acquisition unit 101. For example, assume that the policy acquired by the policy acquisition unit 101 is a function that outputs the optimal price and procurement quantity at a time based on the current inventory of the above product and the time. In this case, the optimal value calculation unit 102 can calculate the optimal price and procurement quantity by inputting the current inventory and time of the above product into the function. In this case, the optimal price and procurement quantity can also be calculated taking inventory into consideration.

[0020] (Effects of information processing device 1) As described above, the information processing device 1 includes a strategy acquisition unit 101 that acquires a strategy for calculating the optimal price and procurement quantity of the product, generated by optimization calculation using a predicted value of a first parameter that represents the probability distribution of demand for the product relative to its price and a predicted value of a second parameter that represents the transition probability of the procurement cost of the product, and an optimal value calculation unit 102 that calculates the price and procurement quantity of the product using the acquired strategy.

[0021] According to the above configuration, the price and procurement quantity of the product are calculated using a strategy generated by optimization calculation using a predicted value of a first parameter representing the probability distribution of demand for the product price and a predicted value of a second parameter representing the transition probability of the procurement cost of the product, thereby achieving the effect of being able to calculate the optimal price and procurement quantity of the product taking into account fluctuations in the procurement cost of the product.

[0022] (Optimization Support Program) The functions of the information processing device 1 described above can also be realized by a program. The optimization support program according to this exemplary embodiment causes a computer to function as: a policy acquisition means for acquiring a policy for calculating the optimal price and procurement quantity of a product, the policy being generated by an optimization calculation using a predicted value of a first parameter representing the probability distribution of demand for the product relative to its price and a predicted value of a second parameter representing the transition probability of the product's procurement cost; and an optimal value calculation means for calculating the price and procurement quantity of the product using the policy. Therefore, the optimization support program according to this exemplary embodiment has the effect of enabling the calculation of the optimal price and procurement quantity of the product taking into account fluctuations in the product's procurement cost.

[0023] (Flow of the optimization support method) The flow of the optimization support method will be described with reference to Fig. 2. Fig. 2 is a flow diagram showing the flow of the optimization support method. Note that the execution entity of each step in this optimization support method may be a processor provided in the information processing device 1, or may be a processor provided in another device. In other words, the execution entities of each step may be processors provided in different devices.

[0024] In S1 (strategy acquisition process), at least one processor acquires a strategy for calculating the optimal price and procurement quantity of the product, which is generated by optimization calculation using a predicted value of a first parameter representing the probability distribution of demand for the product relative to its price and a predicted value of a second parameter representing the transition probability of the procurement cost of the product.

[0025] In S2 (optimal value calculation process), at least one processor calculates the price and procurement amount of the product using the strategy acquired in S1.

[0026] (Effects of the Optimization Support Method) As described above, the optimization support method according to this exemplary embodiment includes a policy acquisition process in which at least one processor acquires a policy for calculating an optimal price and procurement quantity of the product, the policy being generated by an optimization calculation using a predicted value of a first parameter representing a probability distribution of demand for the product relative to its price and a predicted value of a second parameter representing a transition probability of the procurement cost of the product, and an optimal value calculation process in which the price and procurement quantity of the product are calculated using the predicted value. This provides the effect of enabling the calculation of an optimal price and procurement quantity of the product taking into account fluctuations in the procurement cost of the product.

[0027] [Second Exemplary Embodiment] A second exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. Note that the scope of application of each technique employed in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technique employed in this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical hindrance occurs. Furthermore, each technique shown in each drawing referenced to explain this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical hindrance occurs.

[0028] (Configuration of Optimization Support System 5A) The configuration of the optimization support system 5A will be described with reference to FIG. 3. FIG. 3 is a diagram showing an overview of the optimization support system 5A. The optimization support system 5A is a system equipped with a function of presenting the optimal price and procurement quantity of a target product to a user. As shown in the figure, the optimization support system 5A includes an information processing device 1A that performs various processes to realize the above function, a database 2A that records various data used in the optimization support system 5A, and an output device 3A that presents various information to the user. Note that the database 2A may be stored inside the information processing device 1A. Furthermore, if the information processing device 1A has a function of outputting information, the output device 3A may be omitted.

[0029] In order to present the optimal price and procurement quantity of a product to the user, the information processing device 1A generates a policy for calculating the optimal price and procurement quantity of the product. Below, a method for generating the policy will be described, and then a method for calculating the optimal price and procurement quantity using the generated policy will be described.

[0030] In generating the policy, the information processing device 1A first calculates a first parameter θ ｄ The predicted value of the above product procurement cost c ｓ The second parameter θ represents the transition probability of ｃ The predicted value of and are obtained. The demand distribution of the above product is θ ｄ Using D(θ ｄ ) is expressed as D(θ ｄ ) may be a function of time. In this case, D(θ ｄ ) by inputting a price and time, it outputs the probability distribution that demand will follow when that price is offered at that time.

[0031] Here, the price p offered at each time t ｔ Depending on the demand, the demand distribution D ｔ (p ｔ , θ ｄ ) is assumed to be determined probabilistically according to the demand distribution D ｔ (p ｔ , θ ｄ ) is (p ｔ , θ ｄ Any distribution may be used as long as it is determined by the given θ ｄ By identifying the demand distribution D ｔ is identified.

[0032] Assume that it is known that there are S types of procurement costs for a product. Assume also that the probability of the procurement cost type transitioning from s to s' is expressed as P(s'|s). In this case, for a given s, the following formula holds, and s' follows the categorical distribution of S values:

[0033] In this case, the probability of the categorical distribution of S values ​​for all s is estimated, as expressed by the following equation (1).

[0034] In other words, in this case, the parameter θ that determines the transition probability of procurement costs ｃ is the total probability of the categorical distribution. ｓ，ｓ’ =P(s'|s), and by estimating this for all s and s', the estimated θ ｓ，ｓ’ The whole θ ｃ In the following, the procurement cost of the product is ｓ From C ｓ’ The probability of transitioning to θ ｃ Using P(s'|s,θ ｃ ) is expressed as

[0035] Here, the parameter θ ｄ and θ ｃ There is no particular limitation on the method for obtaining the predicted value of the parameter θ ｄ and θ ｃ As information relating to the above, it is assumed that at least one of demand information relating to past demand for the product, procurement cost information relating to past procurement costs for the product, and inventory information relating to past inventory of the product is recorded in the database 2A. In this case, the information processing device 1A may obtain the above-mentioned predicted value using such information.

[0036] Specifically, the information processing device 1A calculates a first parameter θ using at least one of demand information, procurement cost information, and inventory information. ｄ and the second parameter θ ｃ In this case, the information processing device 1A generates a posterior distribution from each of the prior distributions. ｄ The posterior distribution of f ｄ , the second parameter θ ｃ The posterior distribution of f ｃ It is expressed as:

[0037] The information processing device 1A calculates the posterior distribution f ｄ From θ ｄ We can obtain (in other words, sample) the predicted value of the posterior distribution f ｄSampling from the first parameter θ ｄ Once the search is done, the parameter θ ｄ Even if the known information about θ is biased, various parameters are probabilistically acquired. As a result, the first parameter θ obtained due to the bias of the known information at a certain point in time ｄ Even if the value of θ is not valid, the first parameter θ ｄ The value of the first parameter θ can be made closer to the true value. ｄ The predicted value of may be expressed as follows:

[0038] Similarly, the information processing device 1A calculates the posterior distribution f ｃ From θ ｃ It is possible to obtain a predicted value of . In the following, this predicted value may be expressed as follows:

[0039] Next, the information processing device 1A acquires a strategy for calculating the optimal price and procurement quantity of the product by optimization calculation using the predicted values ​​of the first and second parameters acquired as described above. More specifically, the information processing device 1A acquires a strategy for calculating the optimal price and procurement quantity of the product by optimization calculation using the predicted values ​​of the first and second parameters acquired as described above. and the transition probability of the procurement cost of the product determined by the predicted value of the second parameter. The above-mentioned policy is obtained by solving an optimization problem of the price p and the procurement amount m using the above. For example, the information processing device 1A obtains a policy for finding the price p and the procurement amount m that maximizes the expected value of profit. Note that the information processing device 1A may also perform optimization calculations taking into consideration information such as inventory amount and inventory holding costs.

[0040] Here, if the distribution of product demand and the transition probability of the procurement cost of the product are known, it is possible to solve the optimization problem and generate a policy for calculating the optimal price and procurement quantity. However, both the distribution of product demand and the transition probability of the procurement cost of the product are unknown at the initial stage. For this reason, the information processing device 1A calculates the parameter θ ｄand θ ｃ The distribution of product demand and the transition probability of the procurement cost of that product are represented by this, and optimization calculations are performed.

[0041] For example, the expected profit maximization problem can be expressed as follows:

[0042] The cost function h(·) is a function of remaining stock and corresponds to the stock storage cost, etc. It is not essential to include the cost function h(·). ｔ is the procurement amount at time t, n ｔ is the inventory amount at the beginning of time t, D ｔ is the demand at time t, p ｔ is the price at time t, c ｓｔ denotes the procurement cost at time t.

[0043] Also, is determined by the first and second parameters obtained, and is the transition probability of the demand distribution and procurement cost. ｔ , p ｔ ) is determined. The policy set Π represents the expected profit when (t, n ｔ ,c ｓｔ ) as input, (m ｔ , p ｔ )

[0044] By solving this maximization problem, the following policy π can be obtained. This policy π is a function based on the predicted values ​​of the first and second parameters. This policy π is combined with the time t and the inventory amount n at that time. ｔ and procurement cost c ｓｔ By inputting the optimal procurement amount m ｔ and the optimal price p at time t ｔ can be calculated.

[0045] Any method, including known techniques, can be applied to solve the above maximization problem, and for example, dynamic programming may be applied.

[0046] By recording the above-mentioned measures in the database 2A or the storage unit of the information processing device 1A, the information processing device 1A can calculate the optimal price and procurement amount using the measures.

[0047] For example, when a user of the optimization support system 5A determines the procurement cost c of a product at a time t when the user wants to determine the price and procurement quantity, ｓｔ is observed and recorded as procurement cost information in the database 2A. In this case, the information processing device 1A acquires the recorded measures as described above. Then, the information processing device 1A uses the acquired measures to calculate the procurement cost c at time t shown in the procurement cost information. ｓｔ and the inventory quantity n at time t ｔ From this, the optimal price p ｔ and procurement amount m ｔ is calculated and output to the output device 3A.

[0048] As a result, the user of the optimization support system 5A can ｔ and procurement amount m ｔ The user can check the actual demand for the product at that time and use it as a reference to determine the procurement amount and sales price of the product at that time. Then, the user purchases the product in the determined procurement amount and sells it at the determined sales price. This allows the actual demand for the product at that time, d ｔ That is, how much of the product was sold is observed. The observed results of actual demand are recorded in the database 2A as demand information.

[0049] In this way, when the procurement cost information and the actual demand information at a certain time point t are recorded in the database 2A, the information processing device 1A uses this information to calculate the posterior distribution f of the first parameter. ｄ and the posterior distribution of the second parameter f ｃ Since this update is based on actual observations, it is possible to bring each posterior distribution closer to the true distribution.

[0050] Then, the information processing device 1A obtains the predicted values ​​of the first parameter and the second parameter from each updated posterior distribution. Furthermore, the information processing device 1A recalculates the measures for calculating the optimal price and procurement quantity of the product through optimization calculation using these predicted values. As a result, the information processing device 1A uses the new measures to calculate the optimal price p at another time point (t+1). ｔ＋１ and procurement amount m ｔ＋１ can be calculated and presented to the user.

[0051] In this way, the optimization support system 5A can sequentially present the user with the optimal price and procurement quantity of the product at each point in time. Furthermore, the optimization support system 5A can improve the prediction accuracy of the first and second parameters as the product is repeatedly procured and sold by continually updating the posterior distribution based on the observed results of actual demand and procurement costs. As the prediction accuracy of the first and second parameters improves, the measures generated using those predicted values ​​also improve, and therefore the optimization support system 5A can improve the estimation accuracy of the optimal price and procurement quantity over time.

[0052] (Configuration of Information Processing Device 1A) The configuration of the information processing device 1A will be described with reference to FIG. 4. FIG. 4 is a block diagram showing the configuration of the information processing device 1A. As shown in the figure, the information processing device 1A includes a control unit 10A that controls the various units of the information processing device 1A and a storage unit 11A that stores various data used by the information processing device 1A. The information processing device 1A also includes a communication unit 12A that enables the information processing device 1A to communicate with other devices, an input unit 13A that accepts various data input to the information processing device 1A, and an output unit 14A that enables the information processing device 1A to output various data. As shown in the figure, the control unit 10A of the information processing device 1A includes a policy acquisition unit 101A, an optimal value calculation unit 102A, a predicted value acquisition unit 103A, a presentation control unit 104A, a posterior distribution calculation unit 105A, and an observation value acquisition unit 106A.

[0053] Similar to the policy acquisition unit 101 in the first exemplary embodiment, the policy acquisition unit 101A acquires a policy for calculating an optimal price and procurement quantity of the product, the policy being generated by an optimization calculation using the predicted values ​​of the first parameter and the second parameter. More specifically, the policy acquisition unit 101A acquires the policy by an optimization calculation using the predicted values ​​of the first and second parameters. The method of the optimization calculation is not particularly limited. For example, the policy acquisition unit 101A may acquire the policy by dynamic programming.

[0054] The optimal value calculation unit 102A calculates the price and procurement quantity of the product using the measures acquired by the measure acquisition unit 101A, similar to the optimal value calculation unit 102 in exemplary embodiment 1. For example, if the measures acquired by the measure acquisition unit 101A include time, procurement cost, and inventory quantity as variables, the optimal value calculation unit 102A calculates the price and procurement quantity of the product by inputting the time, procurement cost, and inventory quantity into the measures.

[0055] The predicted value acquisition unit 103A acquires a predicted value of the first parameter and a predicted value of the second parameter. More specifically, the predicted value acquisition unit 103A acquires a posterior distribution f of the first parameter calculated based on the price applied to the product and examples of demand corresponding to that price. ｄ The predicted value acquisition unit 103A also acquires a predicted value of the first parameter from the posterior distribution f ｃ The predicted value of the second parameter is obtained from the above. Note that the above "case" can be rephrased as "observation result" or "observation value". In addition, the posterior distribution f ｄ and f ｃ is calculated and updated by the posterior distribution calculation unit 105A.

[0056] posterior distribution f ｄ The predicted value of the first parameter may be obtained in any manner from the posterior distribution f ｄ It is preferable to use a value randomly sampled from the posterior distribution f as the predicted value of the first parameter. ｄEven if the observation values ​​used to update f include invalid ones, the posterior distribution f ｄ It is possible to bring the posterior distribution f closer to the true distribution. ｃ The same applies to the posterior distribution f ｃ Preferably, a value randomly sampled from is used as the predicted value of the second parameter.

[0057] The presentation control unit 104A presents the price and procurement quantity calculated by the optimum value calculation unit 102A to the user. The presentation may be performed in any manner. For example, the presentation control unit 104A may present the price and procurement quantity to the user by causing the output device 3A or the output unit 14A shown in FIG. 3 to output the price and procurement quantity. The manner of output is not particularly limited, and the presentation control unit 104A may, for example, display, output audibly, or print out the information, or may present the information by a combination of these.

[0058] The posterior distribution calculation unit 105A calculates the posterior distribution f ｄ and f ｃ More specifically, the posterior distribution calculation unit 105A generates a prior distribution from known information regarding the distribution of the first parameter (for example, information indicating examples of prices applied to products and demand corresponding to those prices), and calculates a posterior distribution f ｄ Similarly, the posterior distribution calculation unit 105A generates a prior distribution from known information regarding the distribution of the second parameter, and calculates the posterior distribution f ｃ In addition, the posterior distribution calculation unit 105A calculates the observed value (specifically, the actual procurement cost c ｓｔ , the actual asking price p ｔ and the actual demand for it d ｔ ) to obtain the posterior distribution f ｄ and f ｃ Update.

[0059] The observation value acquisition unit 106A acquires an observation value of the actual demand for the product. More specifically, the observation value acquisition unit 106A acquires an observation value of the actual demand for the product when the product is procured in the procurement amount calculated by the optimal value calculation unit 102A and sold at the price calculated by the optimal value calculation unit 102A. The observation value acquisition unit 106A also acquires an observation value of the procurement cost of the product. Note that the method for acquiring these observation values ​​is arbitrary and is not particularly limited. For example, the observation value acquisition unit 106A may acquire an observation value input by a user via the communication unit 12A or the input unit 13A.

[0060] As described above, the information processing device 1A includes the presentation control unit 104A that presents to the user the price and procurement quantity calculated by the optimal value calculation unit 102A. Therefore, in addition to the effects of the information processing device 1, the information processing device 1A provides the effect that the user can easily determine an appropriate price and procurement quantity for the product by referring to the price and procurement quantity calculated by the optimal value calculation unit 102A.

[0061] As described above, the information processing device 1A includes the predicted value acquisition unit 103A that acquires a predicted value of a first parameter from the posterior distribution of the first parameter calculated based on examples of prices applied to products and demand corresponding to those prices. The policy acquisition unit 101A then acquires a policy by optimization calculation using the predicted value of the first parameter acquired by the predicted value acquisition unit 103A.

[0062] The above configuration makes it possible to obtain a reasonable predicted value based on the price applied to a product and the demand corresponding to that price. Furthermore, by using this predicted value, it becomes possible to obtain a measure that can estimate the optimal price and procurement quantity with high accuracy. Therefore, in addition to the effects of the information processing device 1, the information processing device 1A also provides the effect of being able to estimate the optimal price and procurement quantity with high accuracy.

[0063] As described above, the information processing device 1A also includes the observed value acquiring unit 106A that acquires an observed value of actual demand for the product when the product is procured in the procurement amount calculated by the optimal value calculating unit 102A and sold at the price calculated by the optimal value calculating unit 102A. The predicted value acquiring unit 103A then acquires a predicted value of the first parameter from the posterior distribution of the first parameter that has been updated using the observed value acquired by the observed value acquiring unit 106A.

[0064] According to the above configuration, when the product is procured in the procurement quantity calculated by the optimal value calculation unit 102A and sold at the price calculated by the optimal value calculation unit 102A, the actual demand for the product is reflected in the posterior distribution of the first parameter. Therefore, by repeatedly determining the price and procurement quantity using the information processing device 1A and observing the actual demand for that price, the posterior distribution of the first parameter can be brought closer to the true distribution. Therefore, in addition to the effects of the information processing device 1, the information processing device 1A can also provide the effect of improving the accuracy of estimating the optimal procurement quantity and price through repeated procurement and sales of the product.

[0065] As described above, the information processing device 1A includes the predicted value acquisition unit 103A that acquires a predicted value of the second parameter from the posterior distribution of the second parameter calculated based on transition cases of the product procurement cost. The measure acquisition unit 101A then acquires a measure by an optimization calculation using the predicted value of the second parameter acquired by the predicted value acquisition unit 103A.

[0066] The above configuration makes it possible to obtain a reasonable predicted value corresponding to the transition example of the product procurement cost. Furthermore, by using this predicted value, it becomes possible to obtain a measure that can estimate the optimal price and procurement quantity with high accuracy. Therefore, in addition to the effects achieved by the information processing device 1, the information processing device 1A also has the effect of being able to estimate the optimal price and procurement quantity with high accuracy.

[0067] Note that the predicted value acquisition unit 103A does not necessarily need to acquire both the predicted value of the first parameter and the predicted value of the second parameter from each posterior distribution. For example, the predicted value acquisition unit 103A may acquire the predicted value of the first parameter from the posterior distribution of the first parameter, and acquire the predicted value of the second parameter using another method. Alternatively, the predicted value acquisition unit 103A may acquire the predicted value of the second parameter from the posterior distribution of the second parameter, and acquire the predicted value of the first parameter using another method.

[0068] As described above, the information processing device 1A includes the observed value acquiring unit 106A that acquires observed values ​​of the procurement cost of the product. The predicted value acquiring unit 103A then acquires a predicted value of the second parameter from the posterior distribution of the second parameter that has been updated using the observed values ​​acquired by the observed value acquiring unit 106A.

[0069] According to the above configuration, the observed value of the procurement cost of the product is reflected in the posterior distribution of the second parameter. Therefore, by repeatedly determining the price and procurement quantity using the information processing device 1A and observing the procurement cost at each stage, the posterior distribution of the second parameter can be brought closer to the true distribution. Therefore, in addition to the effects of the information processing device 1, the information processing device 1A can also achieve the effect of improving the accuracy of estimating the optimal procurement quantity and price through repeated procurement and sales of the product.

[0070] Note that the observation value acquiring unit 106A does not necessarily need to acquire both the observed value of the procurement cost of the product and the observed value of the actual demand for the product. For example, if the predicted value of the first parameter is not acquired from the posterior distribution (e.g., if the first parameter is predicted using a prediction model), the observation value acquiring unit 106A does not need to acquire the observed value of the actual demand for the product. Similarly, if the predicted value of the second parameter is not acquired from the posterior distribution (e.g., if the second parameter is predicted using a prediction model), the observation value acquiring unit 106A does not need to acquire the observed value of the procurement cost of the product.

[0071] (Specific Example) A specific example of processing by the information processing device 1A will be described with reference to Fig. 5. Fig. 5 is a diagram showing a specific example of processing executed by the information processing device 1A. In the figure, a user U is a user of the information processing device 1A. The user U is a retailer who procures products from a supplier Su and sells them to a customer Cu. As shown in A1 of Fig. 5, the inventory n of products held by the user U at an initial point in time (t=1) is １ is zero. Initial stock n １ = 0 is recorded as inventory information in the database 2A.

[0072] Here, user U determines the optimal price p ｔ and procurement amount m ｔ When the information processing device 1A calculates the product procurement cost c ｓｔ In the example of FIG. 5, the procurement cost of the product from the supplier Su at time t is c ｓｔ has been notified by the supplier Su. Therefore, the user U can use the notified procurement cost c ｓｔ and the target time t are input to the information processing device 1A, and the optimal price p ｔ and procurement amount m ｔ The information processing device 1A calculates:

[0073] For example, in A1 of FIG. 5, the procurement cost c ｓｔ It has been notified that the procurement cost c is 1,000 yen. ｓｔ The observation value acquisition unit 106A acquires the procurement cost c ｓｔ is obtained as the observed value.

[0074] The optimal value calculation unit 102A calculates the optimal value by adding the c obtained by the observation value acquisition unit 106A to the measure π obtained in advance by the measure acquisition unit 101A. ｓｔ (= 1000 yen) and the stock amount n recorded in database 2A １ (=0) and the optimal price p ｔ and procurement amount m ｔ As shown in A1 of FIG. 5, the calculated optimal procurement amount m ｔUser U purchases five items from supplier Su according to this estimation result. As a result, the inventory n at time t is ｔ There are five.

[0075] Also, as shown in A2 of the figure, the optimal price p ｔ is 2000 yen. Based on this estimation result, user U sells the product to customer Cu at a suggested price of 2000 yen. In the example of Figure 5, the number of products purchased by the customer at this suggested price is 3. In other words, the observed actual demand d ｔ = 3. User U has observed actual demand d ｔ is input to the information processing device 1A, and the observation value acquisition unit 106A calculates the input actual demand d ｔ is obtained as an observed value. ｔ is recorded in the database 2A.

[0076] In the above example, the profit gained from trading at time t is p ｔ × min (d ｔ , n ｔ +m ｔ )-m ｔ ×c ｓｔ When t=1, d ｔ = 3, n ｔ = 0, m ｔ = 5, c ｓｔ = 1000, so the profit at this time is p ｔ As shown in the figure, the stock price after the transaction is completed is 1,000 yen. ｔ＋１ is max(n ｔ +m ｔ -d ｔ , 0). When t=1, d ｔ = 3, n ｔ = 0, m ｔ = 5, the inventory at time t = 2 is 2 units. The acquisition data acquired by the observation value acquisition unit 106A regarding the purchase and sale at time t is the above-mentioned procurement cost c ｓｔ The price p offered to customer Cu ｔ and the corresponding actual demand d ｔ is added, and {c ｓｔ , (p ｔ , dｔ ).

[0077] The total profit obtained by repeating the above process from t=1 to t=T, in other words, the profit obtained during the period from t=1 to t=T, is expressed by equation (1) shown in A3 of Figure 5.

[0078] The acquired data obtained by this repetition is expressed by Equation (2) shown in A3 of Fig. 5. As described above, the acquired data is used by the posterior distribution calculation unit 105A to update the posterior distribution. Then, the predicted value acquisition unit 103A acquires predicted values ​​of the first and second parameters from the updated posterior distribution, and the policy acquisition unit 101A acquires a new policy π through optimization calculation using the acquired predicted values. By using the new policy π, it becomes possible to estimate the optimal price and procurement quantity with higher accuracy.

[0079] (Processing Flow) The processing flow executed by the information processing device 1A will be described with reference to Fig. 6. Fig. 6 is a flow diagram showing the processing flow executed by the information processing device 1A. This flow diagram includes each step of the optimization support method according to this exemplary embodiment.

[0080] In S11, the posterior distribution calculation unit 105A calculates the posterior distribution for each of the first and second parameters. For example, in the first cycle of the loop from S11 to S18, the posterior distribution calculation unit 105A calculates the posterior distribution f of the first parameter from the prior distribution of the first parameter. ｄ Similarly, the posterior distribution calculation unit 105A may calculate the posterior distribution f of the second parameter from the prior distribution of the second parameter. ｃ The various data required to calculate the posterior distribution may be input by the user of the information processing device 1A via the communication unit 12A or the input unit 13A. In addition, from the second round of the loop of S11 to S18 onwards, the posterior distribution calculation unit 105A calculates the posterior distribution f calculated in the previous loop. ｄ and f ｃ Update the following.

[0081] In S12, the predicted value acquisition unit 103A acquires predicted values ​​of the first and second parameters. More specifically, the predicted value acquisition unit 103A acquires the posterior distribution f ｄ Similarly, the predicted value acquisition unit 103A acquires the predicted value of the first parameter from the posterior distribution f ｃ The predicted value of the second parameter is obtained from

[0082] In S13 (measure acquisition processing), the measure acquisition unit 101A acquires a measure π generated by an optimization calculation using the predicted values ​​of the first and second parameters obtained in S12. In this exemplary embodiment, the measure acquisition unit 101A acquires the measure π by performing an optimization calculation using the predicted values ​​of the first and second parameters. Note that the measure acquisition unit 101A may acquire a measure π generated by another device. In this case, the measure acquisition unit 101A may transmit the predicted values ​​of the first and second parameters acquired in S12 to the other device and acquire the measure π generated using those predicted values. Furthermore, the other device may execute the processes of S11 and S12 and generate the measure π using the predicted values ​​of the first and second parameters acquired in S12.

[0083] In S14, the observation value acquisition unit 106A calculates the procurement cost c of the product at the time t (which can also be referred to as a point in time or a period) for which the price and procurement quantity are to be determined. ｓｔ Obtain the observed value of procurement cost c ｓｔ The observed value may be input by a user of the information processing device 1A via the communication unit 12A or the input unit 13A, or may be obtained from a supplier of the product.

[0084] In S15 (optimal value calculation process), the optimal value calculation unit 102A calculates the optimal value by calculating the policy π acquired in S13 and the procurement cost c acquired in S14. ｓｔ In more detail, the optimal value calculation unit 102A calculates the optimal price and procurement quantity of the product by using the observed value of the procurement cost c ｓｔ The observed value of , the time t for which the price and procurement quantity are to be determined, and the inventory quantity n at time t ｔThis allows us to calculate the optimal price p ｔ and procurement amount m ｔ is calculated.

[0085] In S16, the presentation control unit 104A calculates the price p ｔ and procurement amount m ｔ For example, the presentation control unit 104A presents the price p calculated in S15 to the user. ｔ and procurement amount m ｔ The information may be presented to the user by outputting the above information to the output device 3A. After that, the user determines the price and the amount to be procured, and the product is procured and sold.

[0086] In S17, the observation value acquisition unit 106A calculates the actual demand d for the price determined by the user and presented to the customer. ｔ Obtain the observed value of actual demand d ｔ The observed value of may be input by the user of the information processing device 1A via the communication unit 12A or the input unit 13A.

[0087] In S18, the observation value acquisition unit 106A determines whether or not the observation for the predetermined period has ended. The predetermined period can be set arbitrarily. For example, if the period from t=1 to t=T is set, the observation value acquisition unit 106A calculates the actual demand d ｔ If the observed value has already been acquired, it is determined that the observation for the predetermined period has ended.

[0088] If the determination in S18 is NO, the process returns to S14, and the observation value acquisition unit 106A calculates the procurement cost c ｓｔ For example, the observation value acquisition unit 106A acquires the procurement cost c ｓ１ In the second step of S14, the procurement cost c ｓ２ The optimal value calculation unit 102A also obtains the observed value of the inventory quantity n ｔ The inventory quantity at the next time n ｔ＋１ Update to.

[0089] On the other hand, if the determination in S18 is YES, the process returns to S11. In S11 after transition from S18, the posterior distribution calculation unit 105A calculates the posterior distribution f of the first parameter using the observation results acquired and accumulated by the repeated processing from S14 to S18 during the predetermined period. ｄ and the posterior distribution of the second parameter f ｃ and update the first and second parameters, respectively. In addition, the predicted value acquisition unit 103A initializes the predicted values ​​of the first and second parameters, and the optimal value calculation unit 102A initializes the inventory amount. Note that the observation results can also be referred to as acquired data (see Equation (2) in FIG. 5). Then, in the following S12, the predicted value acquisition unit 103A acquires the predicted values ​​of the first and second parameters from the updated posterior distribution.

[0090] (Regarding Prediction of First and Second Parameters) In the exemplary embodiment 2, the predicted value acquisition unit 103A acquires the predicted value of the first parameter from the posterior distribution of the first parameter. However, any method for acquiring the predicted value of the first parameter may be used, and is not limited to this example.

[0091] For example, the predicted value acquisition unit 103A may acquire the predicted value of the first parameter using a prediction model of the first parameter generated by machine learning based on examples of prices applied to products and demand corresponding to those prices. In this case, the measure acquisition unit 101A acquires a measure by an optimization calculation using the predicted value acquired by the predicted value acquisition unit 103A.

[0092] With the above configuration, it is possible to obtain a reasonable predicted value based on the price applied to the product and the demand corresponding to that price. Furthermore, by using this predicted value, it is possible to obtain a measure that can estimate the optimal price and procurement quantity with high accuracy. Therefore, with the above configuration, in addition to the effects achieved by the information processing device 1, it is possible to obtain the effect of being able to estimate the optimal price and procurement quantity with high accuracy.

[0093] The prediction model for the first parameter may be generated by machine learning the relationship between the objective variable and the explanatory variables, with the first parameter as the objective variable and various data related to the first parameter as the explanatory variables. Examples of data related to the first parameter include prices previously applied to the product and the demand when those prices were applied.

[0094] Furthermore, when the first parameter is predicted by a prediction model, the control unit 10A may be provided with a model update unit that updates the prediction model of the first parameter using the above-mentioned acquired data acquired by procuring and selling the product. This has the effect of enabling the accuracy of estimating the optimal procurement quantity and price to be improved as the procurement and sale of the product are repeated.

[0095] Similarly, the method for acquiring the predicted value of the second parameter may be any method and is not limited to the above example. For example, the predicted value acquisition unit 103A may acquire the predicted value of the second parameter using a prediction model of the second parameter generated by machine learning based on transition cases of the product procurement cost. In this case, the measure acquisition unit 101A acquires the measure by an optimization calculation using the predicted value acquired by the predicted value acquisition unit 103A.

[0096] The above configuration makes it possible to obtain a reasonable predicted value according to the transition example of the product procurement cost. Furthermore, by using this predicted value, it becomes possible to obtain a measure that can estimate the optimal price and procurement quantity with high accuracy. Therefore, in addition to the effects achieved by the information processing device 1, the above configuration also provides the effect of being able to estimate the optimal price and procurement quantity with high accuracy.

[0097] The prediction model for the second parameter may be generated by machine learning the relationship between the objective variable and the explanatory variables, with the second parameter as the objective variable and various data related to the second parameter as the explanatory variables. Examples of data related to the second parameter include time-series data on the procurement costs of a product.

[0098] Furthermore, when the second parameter is predicted using a prediction model, the control unit 10A may be provided with a model update unit that updates the prediction model for the second parameter using the above-mentioned acquired data acquired by procuring and selling the product. This has the effect of enabling the accuracy of estimating the optimal procurement quantity and price to be improved as the product is repeatedly procured and sold.

[0099] [Modifications] The execution entity of each process described in each of the exemplary embodiments above is arbitrary and is not limited to the above examples. In other words, the functions of the information processing devices 1 and 1A can be realized by multiple devices (which can also be called processors) that can communicate with each other. For example, each process described in the flow charts of Figures 2 and 6 can be shared and executed by multiple processors. In other words, the execution entity of the optimization support method in each of the above embodiments may be one processor or multiple processors.

[0100] [Example of Software Implementation] Some or all of the functions of the information processing device 1, 1A may be implemented by hardware such as an integrated circuit (IC chip), or may be implemented by software.

[0101] In the latter case, the information processing device 1 or 1A is realized by, for example, a computer that executes instructions of a program, which is software that realizes each function. An example of such a computer (hereinafter referred to as computer C) is shown in Fig. 7. Fig. 7 is a block diagram showing the hardware configuration of computer C that functions as information processing device 1 or 1A.

[0102] The computer C includes at least one processor C1 and at least one memory C2. The memory C2 stores a program (optimization support program) P for operating the computer C as the information processing device 1 or 1A. In the computer C, the processor C1 reads and executes the program P from the memory C2, thereby realizing each function of the information processing device 1 or 1A.

[0103] The processor C1 may be, for example, a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a micro processing unit (MPU), a floating point number processing unit (FPU), a physics processing unit (PPU), a tensor processing unit (TPU), a quantum processor, a microcontroller, or a combination thereof. The memory C2 may be, for example, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or a combination thereof.

[0104] The computer C may further include a RAM (Random Access Memory) for expanding the program P during execution and for temporarily storing various data. The computer C may also include a communication interface for transmitting and receiving data to and from other devices. The computer C may also include an input / output interface for connecting input / output devices such as a keyboard, a mouse, a display, and a printer.

[0105] The program P can also be recorded on a non-transitory, tangible recording medium M that can be read by the computer C. Such a recording medium M can be, for example, a tape, a disk, a card, a semiconductor memory, or a programmable logic circuit. The computer C can acquire the program P via such a recording medium M. The program P can also be transmitted via a transmission medium. Such a transmission medium can be, for example, a communication network or broadcast waves. The computer C can also acquire the program P via such a transmission medium.

[0106] Furthermore, each of the above functions of the information processing device 1 or 1A may be realized by a single processor provided in a single computer, by multiple processors provided in a single computer working together, or by multiple processors provided in each of multiple computers working together. Furthermore, the program for causing the information processing device 1 or 1A to realize each of the above functions may be stored in a single memory provided in a single computer, or may be distributed and stored in multiple memories provided in a single computer, or may be distributed and stored in multiple memories provided in each of multiple computers.

[0107] [Appendix 1] This disclosure includes the techniques described in the following appendices. However, the present invention is not limited to the techniques described in the following appendices, and various modifications are possible within the scope of the claims.

[0108] (Appendix A1) An information processing device comprising: a strategy acquisition means for acquiring a strategy for calculating the optimal price and procurement quantity of the product, the strategy being generated by optimization calculation using a predicted value of a first parameter representing the probability distribution of demand for the product's price and a predicted value of a second parameter representing the transition probability of the procurement cost of the product; and an optimal value calculation means for calculating the price and procurement quantity of the product using the strategy.

[0109] (Appendix A2) The information processing device according to appendix A1, further comprising: a presentation control unit that presents the price and procurement amount calculated by the optimum value calculation unit to a user.

[0110] (Appendix A3) An information processing device as described in Appendix A1 or A2, comprising a predicted value acquisition means for acquiring a predicted value of the first parameter from the posterior distribution of the first parameter calculated based on examples of the price applied to the product and demand corresponding to that price, and the measure acquisition means acquires the measure by optimization calculation using the predicted value of the first parameter acquired by the predicted value acquisition means.

[0111] (Appendix A4) An information processing device as described in Appendix A3, comprising an observation value acquisition means for acquiring an observation value of actual demand for the product when the product is procured in the procurement amount calculated by the optimal value calculation means and sold at the price calculated by the optimal value calculation means, and the predicted value acquisition means acquires a predicted value of the first parameter from the posterior distribution of the first parameter updated using the observation value acquired by the observation value acquisition means.

[0112] (Appendix A5) An information processing device described in any of Appendices A1 to A4, comprising a predicted value acquisition means for acquiring a predicted value of the second parameter from the posterior distribution of the second parameter calculated based on transition examples of the procurement cost of the product, and the measure acquisition means acquires the measure by optimization calculation using the predicted value of the second parameter acquired by the predicted value acquisition means.

[0113] (Appendix A6) The information processing device according to Appendix A5, further comprising: an observation value acquisition means for acquiring an observation value of a procurement cost of the product; and the predicted value acquisition means for acquiring a predicted value of the second parameter from a posterior distribution of the second parameter updated using the observation value acquired by the observation value acquisition means.

[0114] (Appendix A7) An information processing device as described in Appendix A1 or A2, comprising a predicted value acquisition means for acquiring a predicted value of the first parameter using a prediction model of the first parameter generated by machine learning based on examples of the price applied to the product and the demand corresponding to that price, and the measure acquisition means acquires the measure by optimization calculation using the predicted value of the first parameter acquired by the predicted value acquisition means.

[0115] (Appendix A8) An information processing device as described in Appendix A1 or A2, comprising a predicted value acquisition means for acquiring a predicted value of the second parameter using a prediction model of the second parameter generated by machine learning based on transition examples of the procurement cost of the product, and the measure acquisition means for acquiring the measure by optimization calculation using the predicted value of the second parameter acquired by the predicted value acquisition means.

[0116] (Appendix B1) An optimization support method including: a strategy acquisition process in which at least one processor acquires a strategy for calculating the optimal price and procurement quantity of the product, the strategy being generated by optimization calculation using a predicted value of a first parameter representing the probability distribution of demand for the product relative to its price and a predicted value of a second parameter representing the transition probability of the procurement cost of the product; and an optimal value calculation process in which the strategy is used to calculate the price and procurement quantity of the product.

[0117] (Supplementary Note B2) The optimization support method according to Supplementary Note B1, further comprising a presentation control process in which the at least one processor presents to a user the price and procurement amount calculated in the optimal value calculation process.

[0118] (Appendix B3) An optimization support method described in Appendix B1 or B2, wherein the at least one processor includes a predicted value acquisition process that acquires a predicted value of the first parameter from a posterior distribution of the first parameter calculated based on examples of the price applied to the product and demand corresponding to that price, and in the measure acquisition process, the at least one processor acquires the measure by optimization calculation using the predicted value of the first parameter acquired in the predicted value acquisition process.

[0119] (Appendix B4) The optimization support method described in Appendix B3 includes an observation value acquisition process in which the at least one processor acquires an observation value of actual demand for the product when the product is procured in the procurement amount calculated in the optimal value calculation process and sold at the price calculated in the optimal value calculation process, and in the predicted value acquisition process, the at least one processor acquires a predicted value of the first parameter from the posterior distribution of the first parameter updated using the observation value acquired in the observation value acquisition process.

[0120] (Appendix B5) An optimization support method described in any of Appendices B1 to B4, wherein the at least one processor includes a predicted value acquisition process that acquires a predicted value of the second parameter from the posterior distribution of the second parameter calculated based on transition examples of the procurement cost of the product, and in the measure acquisition process, the at least one processor acquires the measure by optimization calculation using the predicted value of the second parameter acquired in the predicted value acquisition process.

[0121] (Appendix B6) The optimization support method described in Appendix B5, wherein the at least one processor includes an observation value acquisition process that acquires an observation value of a procurement cost of the product, and in the predicted value acquisition process, the at least one processor acquires a predicted value of the second parameter from a posterior distribution of the second parameter that has been updated using the observation value acquired in the observation value acquisition process.

[0122] (Appendix B7) An optimization support method described in Appendix B1 or B2, wherein the at least one processor includes a predicted value acquisition process for acquiring a predicted value of the first parameter using a prediction model of the first parameter generated by machine learning based on examples of the price applied to the product and the demand corresponding to that price, and in the measure acquisition process, the at least one processor acquires the measure by optimization calculation using the predicted value of the first parameter acquired in the predicted value acquisition process.

[0123] (Appendix B8) An optimization support method described in Appendix B1 or B2, wherein the at least one processor includes a predicted value acquisition process for acquiring a predicted value of the second parameter using a prediction model of the second parameter generated by machine learning based on transition examples of the procurement cost of the product, and in the measure acquisition process, the at least one processor acquires the measure by optimization calculation using the predicted value of the second parameter acquired in the predicted value acquisition process.

[0124] (Appendix C1) An optimization support program that causes a computer to function as a strategy acquisition means that acquires a strategy for calculating the optimal price and procurement quantity of a product, the strategy being generated by optimization calculation using a predicted value of a first parameter that represents the probability distribution of demand for the product's price and a predicted value of a second parameter that represents the transition probability of the procurement cost of the product, and an optimal value calculation means that calculates the price and procurement quantity of the product using the strategy.

[0125] (Appendix C2) The optimization support program according to appendix C1, which causes the computer to function as a presentation control process that presents the price and procurement amount calculated by the optimal value calculation means to a user.

[0126] (Appendix C3) An optimization support program according to appendix C1 or C2, which causes the computer to function as a predicted value acquisition process that acquires a predicted value of the first parameter from a posterior distribution of the first parameter calculated based on examples of prices applied to the product and demand corresponding to those prices, and the measure acquisition means acquires the measure by optimization calculation using the predicted value of the first parameter acquired by the predicted value acquisition means.

[0127] (Appendix C4) An optimization support program as described in Appendix C3, which causes the computer to function as an observation value acquisition process that acquires an observation value of actual demand for the product when the product is procured in the procurement amount calculated by the optimal value calculation means and sold at the price calculated by the optimal value calculation means, and the predicted value acquisition means acquires a predicted value of the first parameter from the posterior distribution of the first parameter updated using the observation value acquired by the observed value acquisition means.

[0128] (Appendix C5) An optimization support program described in any of Appendices C1 to C4, which causes the computer to function as a predicted value acquisition process that acquires a predicted value of the second parameter from the posterior distribution of the second parameter calculated based on transition examples of the procurement cost of the product, and the measure acquisition means acquires the measure by optimization calculation using the predicted value of the second parameter acquired by the predicted value acquisition means.

[0129] (Appendix C6) The optimization support program according to Appendix C5, which causes the computer to function as an observed value acquisition process that acquires an observed value of a procurement cost of the product, and the predicted value acquisition means acquires a predicted value of the second parameter from a posterior distribution of the second parameter updated using the observed value acquired by the observed value acquisition means.

[0130] (Appendix C7) An optimization support program as described in Appendix C1 or C2, which causes the computer to function as a predicted value acquisition process that acquires a predicted value of a first parameter using a prediction model of the first parameter generated by machine learning based on examples of prices applied to the product and demand corresponding to those prices, and the measure acquisition means acquires the measure by optimization calculation using the predicted value of the first parameter acquired by the predicted value acquisition means.

[0131] (Appendix C8) An optimization support program described in Appendix C1 or C2, which causes the computer to function as a predicted value acquisition process that acquires a predicted value of the second parameter using a prediction model of the second parameter generated by machine learning based on transition examples of the procurement cost of the product, and the measure acquisition means acquires the measure by optimization calculation using the predicted value of the second parameter acquired by the predicted value acquisition means.

[0132] (Appendix D1) An information processing device comprising at least one processor, the at least one processor executing a strategy acquisition process to acquire a strategy for calculating the optimal price and procurement quantity of the product, the strategy being generated by optimization calculation using a predicted value of a first parameter representing the probability distribution of demand for the product relative to its price and a predicted value of a second parameter representing the transition probability of the procurement cost of the product, and an optimal value calculation process to calculate the price and procurement quantity of the product using the strategy.

[0133] The information processing device may further include a memory, and the memory may store a program for causing the at least one processor to execute each of the processes.

[0134] (Supplementary Note D2) The information processing device according to Supplementary Note D1, wherein the at least one processor executes a presentation control process of presenting the price and procurement amount calculated in the optimum value calculation process to a user.

[0135] (Appendix D3) An information processing device described in Appendix D1 or D2, which executes a predicted value acquisition process to acquire a predicted value of the first parameter from the posterior distribution of the first parameter calculated based on the price applied to the product and examples of demand corresponding to that price, and in the policy acquisition process, the at least one processor acquires the policy by optimization calculation using the predicted value of the first parameter acquired in the predicted value acquisition process.

[0136] (Appendix D4) An information processing device described in Appendix D3, which executes an observation value acquisition process to acquire an observation value of actual demand for the product when the product is procured in the procurement amount calculated in the optimal value calculation process and sold at the price calculated in the optimal value calculation process, and in the predicted value acquisition process, the at least one processor acquires a predicted value of the first parameter from the posterior distribution of the first parameter updated using the observation value acquired in the observation value acquisition process.

[0137] (Appendix D5) An information processing device described in any of Appendices D1 to D4, which executes a predicted value acquisition process to acquire a predicted value of the second parameter from the posterior distribution of the second parameter calculated based on transition examples of the procurement cost of the product, and in the measure acquisition process, the at least one processor acquires the measure by optimization calculation using the predicted value of the second parameter acquired in the predicted value acquisition process.

[0138] (Appendix D6) An information processing device according to Appendix D5, which executes an observation value acquisition process to acquire an observation value of the procurement cost of the product, and in the predicted value acquisition process, the at least one processor acquires a predicted value of the second parameter from the posterior distribution of the second parameter updated using the observation value acquired in the observation value acquisition process.

[0139] (Appendix D7) An information processing device described in Appendix D1 or D2, which executes a predicted value acquisition process to acquire a predicted value of a first parameter using a prediction model of the first parameter generated by machine learning based on examples of the price applied to the product and demand corresponding to that price, and in the measure acquisition process, the at least one processor acquires the measure by optimization calculation using the predicted value of the first parameter acquired in the predicted value acquisition process.

[0140] (Appendix D8) An information processing device described in Appendix D1 or D2, which executes a predicted value acquisition process to acquire a predicted value of the second parameter using a prediction model of the second parameter generated by machine learning based on transition examples of the procurement cost of the product, and in the measure acquisition process, the at least one processor acquires the measure by optimization calculation using the predicted value of the second parameter acquired in the predicted value acquisition process.

[0141] (Appendix E1) A non-transient recording medium having recorded thereon an optimization support program that causes a computer to execute a strategy acquisition process that acquires a strategy for calculating the optimal price and procurement quantity of the product, generated by optimization calculation using a predicted value of a first parameter that represents the probability distribution of demand for the product's price and a predicted value of a second parameter that represents the transition probability of the procurement cost of the product, and an optimal value calculation process that calculates the price and procurement quantity of the product using the strategy.

[0142] REFERENCE SIGNS LIST 1 Information processing device 101 Measure acquisition unit (measure acquisition means) 102 Optimal value calculation unit (optimal value calculation means) 1A Information processing device 101A Measure acquisition unit (measure acquisition means) 102A Optimal value calculation unit (optimal value calculation means) 103A Predicted value acquisition unit (predicted value acquisition means) 104A Presentation control unit (presentation control means) 106A Observed value acquisition unit (observed value acquisition means)

Claims

1. An information processing device comprising: a policy acquisition means for acquiring a policy for calculating the optimal price and procurement quantity of the product, generated by optimization calculation using a predicted value of a first parameter representing the probability distribution of demand for the product's price and a predicted value of a second parameter representing the transition probability of the product's procurement cost; and an optimal value calculation means for calculating the price and procurement quantity of the product using the policy.

2. The information processing device according to claim 1, further comprising presentation control means for presenting the price and procurement amount calculated by said optimum value calculation means to a user.

3. An information processing device as described in claim 1 or 2, comprising a predicted value acquisition means for acquiring a predicted value of the first parameter from the posterior distribution of the first parameter calculated based on examples of the price applied to the product and the demand corresponding to that price, and the measure acquisition means acquires the measure by an optimization calculation using the predicted value of the first parameter acquired by the predicted value acquisition means.

4. An information processing device as described in claim 3, further comprising an observation value acquisition means for acquiring an observation value of actual demand for the product when the product is procured in the procurement amount calculated by the optimal value calculation means and sold at the price calculated by the optimal value calculation means, and wherein the predicted value acquisition means acquires a predicted value of the first parameter from the posterior distribution of the first parameter updated using the observation value acquired by the observation value acquisition means.

5. An information processing device as described in claim 1 or 2, comprising a predicted value acquisition means for acquiring a predicted value of the second parameter from the posterior distribution of the second parameter calculated based on transition examples of the procurement cost of the product, and the measure acquisition means acquires the measure by optimization calculation using the predicted value of the second parameter acquired by the predicted value acquisition means.

6. An information processing device as described in claim 5, further comprising an observation value acquisition means for acquiring an observation value of the procurement cost of the product, wherein the predicted value acquisition means acquires a predicted value of the second parameter from the posterior distribution of the second parameter updated using the observation value acquired by the observation value acquisition means.

7. An information processing device as described in claim 1 or 2, comprising a predicted value acquisition means for acquiring a predicted value of the first parameter using a prediction model of the first parameter generated by machine learning based on examples of the price applied to the product and the demand corresponding to that price, and the measure acquisition means for acquiring the measure by an optimization calculation using the predicted value of the first parameter acquired by the predicted value acquisition means.

8. An information processing device as described in claim 1 or 2, comprising a predicted value acquisition means for acquiring a predicted value of the second parameter using a prediction model of the second parameter generated by machine learning based on transition examples of the procurement cost of the product, and the measure acquisition means acquires the measure by optimization calculation using the predicted value of the second parameter acquired by the predicted value acquisition means.

9. An optimization support method including: a strategy acquisition process in which at least one processor acquires a strategy for calculating the optimal price and procurement quantity of the product, the strategy being generated by optimization calculation using a predicted value of a first parameter representing the probability distribution of demand for the product relative to its price and a predicted value of a second parameter representing the transition probability of the procurement cost of the product; and an optimal value calculation process in which the strategy is used to calculate the price and procurement quantity of the product.

10. An optimization support program that causes a computer to function as: a policy acquisition means that acquires a policy for calculating the optimal price and procurement quantity of the product, generated by optimization calculation using a predicted value of a first parameter that represents the probability distribution of demand for the product's price and a predicted value of a second parameter that represents the transition probability of the procurement cost of the product; and an optimal value calculation means that calculates the price and procurement quantity of the product using the policy.

Images