JPWO2016120918A1 - Prediction system, prediction method, and computer-readable recording medium - Google Patents

Abstract

Provide a prediction system that predicts demand for goods with high accuracy. The prediction system includes at least a reception unit that receives information on a first target that is a target of demand prediction, information on demand prediction of the first target, and demand of a second target that can be substituted for the first target. A prediction unit configured to predict the demand of the first target based on the information related to the prediction and the prediction result of the demand related to the third target including the first and second targets.

Description

  The present invention relates to a prediction system, a prediction method, and a computer-readable recording medium.

  In various industries, demand for various products and services (hereinafter referred to as “products”) is predicted. When forecasting demand for products, etc., a loss of forecast of demand for products, etc. may lead to lost sales opportunities due to sold out or missing items, etc., and losses such as disposal or price cuts due to unsold products. There is. Therefore, it is necessary to predict demand with high accuracy.

  Patent Document 1 describes a demand prediction device. The demand prediction apparatus described in Patent Literature 1 calculates a sales prediction value of a demand prediction target product based on the overall product type demand value of all existing products and the sales trend of existing similar products.

JP 2008-186413 A

  In the prediction of demand for products and the like, there may be products and the like that are similar in nature and characteristics to the products and the like to be predicted. A product or the like to be predicted and a similar product or the like can be substituted for each other, and one of them may be selected and purchased. In such a case, the prediction accuracy of demand for each product or the like may be low. In other words, the technique described in Patent Document 1 has a problem in demand prediction accuracy for each of similar products.

  The present invention has been made to solve the above-described problems, and has as its main object to provide a prediction system that predicts demand for goods and the like with high accuracy.

  The prediction system in one aspect of the present invention has a receiving unit that receives at least information related to a first target that is a demand prediction target, information related to prediction of demand of the first target, and a relationship that can be substituted for the first target. Predicting means for predicting the demand of the first target based on the information on the forecast of the demand of the second target and the prediction result of the demand on the third target including the first and second targets.

  The prediction method in one aspect of the present invention receives at least information related to a first target that is a demand prediction target, and information related to the demand prediction of the first target and a second target that can be substituted for the first target. The demand of the first target is predicted based on the information related to the prediction of the demand and the prediction result of the demand related to the third target including the first and second targets.

  The computer-readable recording medium according to one aspect of the present invention is a computer that can at least receive information on a first target that is a demand prediction target, information on demand prediction of the first target, and the first target. A process of predicting the demand of the first target based on the information on the forecast of the demand of the second target in a possible relationship and the prediction result of the demand on the third target including the first and second targets Store the program to be executed non-temporarily.

  ADVANTAGE OF THE INVENTION According to this invention, the prediction system etc. which estimate the demand for goods etc. with high precision can be provided.

It is a figure which shows the structure of the prediction system in the 1st Embodiment of this invention. It is a figure which shows the relationship of the prediction object of the demand by the prediction system in the 1st Embodiment of this invention. It is another figure which shows the relationship of the prediction object of the demand by the prediction system in the 1st Embodiment of this invention. It is another figure which shows the relationship of the prediction object of the demand by the prediction system in the 1st Embodiment of this invention. It is a flowchart which shows operation | movement of the prediction system in the 1st Embodiment of this invention. It is a figure which shows the structure of the prediction system in the 2nd Embodiment of this invention. It is a figure which shows the structure of the prediction system in the 3rd Embodiment of this invention. It is a figure which shows the structure of the modification of the prediction system in the 3rd Embodiment of this invention. It is a figure which shows the structure of another modification of the prediction system in the 3rd Embodiment of this invention. It is a figure which shows the structure of another modification of the prediction system in the 3rd Embodiment of this invention. It is a figure which shows an example of the information processing apparatus which implement | achieves the prediction system etc. in each embodiment of this invention.

  Embodiments of the present invention will be described with reference to the accompanying drawings. In each embodiment of the present invention, each component of each device represents a functional unit block. Each component of each device can be realized by any combination of an information processing device 500 and software as shown in FIG. 10, for example. The information processing apparatus 500 includes the following configuration as an example.

CPU (Central Processing Unit) 501
ROM (Read Only Memory) 502
-RAM (Random Access Memory) 503
A program 504 loaded into the RAM 503
A storage device 505 for storing the program 504
A drive device 507 for reading / writing the recording medium 506
Communication interface 508 connected to the network 509
An input / output interface 510 for inputting / outputting data
-Bus 511 connecting each component
In addition, there are various modifications to the method for realizing each device. For example, each device can be realized as a dedicated device. Each device can be realized by a combination of a plurality of devices.

(First embodiment)
First, a first embodiment of the present invention will be described. FIG. 1 is a diagram showing a prediction system in the first embodiment of the present invention. FIG. 2 is a diagram illustrating a relationship between demand prediction targets by the prediction system according to the first embodiment of the present invention. FIG. 3A and FIG. 3B are other diagrams showing the relationship of demand prediction targets by the prediction system in the first embodiment of the present invention. FIG. 4 is a flowchart showing the operation of the prediction system in the first embodiment of the present invention.

  As shown in FIG. 1, the prediction system according to the first embodiment of the present invention includes a reception unit 110 and a prediction unit 120. The reception unit 110 receives at least information related to the first target that is a target of demand prediction. The prediction unit 120 predicts the demand of the first target based on the information related to the prediction of the demand related to the first and second targets and the prediction result of the demand related to the third target. The second object is an object that can be substituted for the first object. The third object is an object including the first and second objects.

  First, in this embodiment, an object or the like on which demand is predicted will be described. The first object is an object for which the demand is predicted. The first target is, for example, a product or service that is a target of demand prediction.

  The demand for the first target is a quantity that requires the product or service that is the first target. As an example, when the first target is a product such as rice balls or bread sold at a store, the demand of the first target is, for example, the quantity at which the product is sold at the store. As another example, when the first object is a passenger of a certain transportation facility, the demand of the first object is, for example, the number of passengers of the transportation facility.

  The second object is an object that can be substituted for the first object. In addition, the number of the 2nd object with respect to a 1st object is arbitrary, and several 2nd object may exist. In this case, the fact that the first object and the second object can be substituted for each other means that at least one of the first and second objects is selectively adopted. As an example, when the first and second objects are some products, it is considered that they are in a relationship such that a person who needs the product selects and purchases at least one of them. In other words, the fact that the first object and the second object can be substituted for each other is considered to be in a relationship in which the demand for one of them decreases as the demand for one of them increases. In another aspect, the first object and the second object are also considered to be products that can cause each other a phenomenon called cannibalization in the field of marketing.

  The third object is an object including the first and second objects. As an example, a set of first and second objects becomes the third object. Further, a feature common to the first and second objects, some superordinate concept or category to which the first and second objects belong, or the entire object included in them is the third object.

  As described above, the first object is in a substitutable relationship with the second object. From this relationship, the person who needs the first or second object is considered to select either the first or second object.

  On the other hand, it is considered that the demand for the entire first and second objects may be substantially constant. That is, even when the demand for each of the first and second objects changes, the change in the demand for the third object including them may be smaller than the change in the demand for each of the first and second objects. It is believed that there is. In other words, it is considered that the demand for each of the first and second objects may change within the range of demand for the third object.

  Therefore, the demand for the third target is related to the upper limit of the demand for each of the first and second targets, and is considered to be one of the elements that define the upper limit of the demand for the first target.

  An example of the first to third objects will be described with reference to FIG. This example is an example relating to a product in a store such as a convenience store. It is assumed that a purchaser of a product that is a customer of a store purchases a rice ball or bread. As a product purchased by the purchaser, rice balls include at least tuna mayo rice balls, kelp rice balls, and plum rice balls. Similarly, bread includes at least curry bread, bread, and melon bread.

  In this case, the first object is, for example, either the above-described rice ball or bread. And according to a 1st object, another rice ball or bread is made into a 2nd object. That is, when any of the three types of rice balls is the first target, the remaining two types of rice balls are set as the second target. When any of the three types of bread is the first target, the remaining two types of bread are set as the second target. In addition, when other types of rice balls and bread (not shown) are sold as products, those products may be included in the second object. Finally, depending on the first object, the entire rice ball or the entire bread is set as the third object. That is, when any of the three types of rice balls is the first object, the entire rice ball is the third object.

  In this example, the entire onigiri or the entire bread may be the first target. In this case, the other of the entire rice ball or the entire bread that is not the first target is the second target. And the 3rd object in this case is the whole onigiri and the whole bread which are the whole goods which are purchase objects by a purchaser.

  Moreover, as shown in FIG. 2, the 3rd object with respect to a 1st and 2nd object may have a some hierarchy. That is, there may be a third object with the entire purchase target that includes them, with respect to the individual onigiri or the entire rice ball for the bread and the third object for the entire bread. Details regarding this point will be described later.

  Then, each component of the prediction system 100 in this embodiment is demonstrated.

  The accepting unit 110 accepts information related to the first object as described above. The information relating to the first object includes at least information relating to the object that is the first object. Moreover, the reception part 110 may receive the information regarding the time when the demand of the 1st object used as a prediction object generate | occur | produces as information regarding a 1st object, for example.

  In addition to this, the reception unit 110 receives information on the prediction of demand related to the first to third objects, the prediction result of the demand related to the first to third objects, and the like according to the operation of the prediction unit 120 described later. May be.

  As described above, the prediction unit 120 predicts the demand of the first target based on the information related to the prediction of demand related to the first and second targets and the prediction result of the demand related to the third target.

  As described above, each of the first object and the second object has a replaceable relationship. Therefore, the demand for the first target may affect the demand for each of the second targets for various reasons. That is, when forecasting demand, the demand forecast accuracy for the first and second objects may be lower than the demand forecast accuracy for the third target, such as the demand for the whole rice ball. In other words, the demand prediction result for the third target is less likely to be affected by various causes than the demand prediction results for the first and second targets, and the variance is often small.

  The relationship between the first and second objects and the third object will be further described using the above-described examples of rice balls and bread. It is likely that the purchaser will first decide whether to buy bread or rice balls, and then decide whether to purchase individual types of rice balls or bread. In addition, the purchaser is influenced by various factors including the purchaser's own mood and physical condition, the weather and temperature at the time of purchase, events, sales, campaigns, and other environmental factors, and the individual rice ball or bread he / she wishes to purchase. Is considered to be determined. That is, the fluctuation in demand for individual rice balls and bread is considered to be relatively large compared to the fluctuation in demand for whole rice balls and bread.

  That is, it is considered that the predicted result of demand for each type of rice ball or bread is relatively more blurred than the actual demand (that is, the quantity actually purchased or the like). In addition, it is considered that the predicted results of demand for whole rice balls and whole bread are relatively less blurred than the predicted results of demand for each individual.

  Therefore, in the prediction system in the present embodiment, when the prediction unit 120 predicts the demand of the first target, as described above, in addition to the information related to the prediction of the demand related to the first and second targets, the prediction unit 120 relates to the third target. Use the demand forecast results. By doing in this way, the prediction part 120 can predict the demand regarding a 1st object accurately compared with the case where the individual demand of only a 1st object is estimated directly.

  As an example when the demand is predicted, the prediction unit 120 predicts the demand based on the prediction result of the demand related to the third target and the ratio of the first target in the target including the second target. For example, when the ratio of the first target in the object including the second target is 0.2 and the predicted value of the demand for the third target is 630, the prediction unit 120 determines the demand for the first target. The predicted value is 630 × 0.2 = 126. That is, the prediction unit 120 calculates that the result of predicting the demand of the first target is 126.

  In the present embodiment, the prediction unit 120 can use values obtained by various methods as the prediction result of the demand of the third target and the ratio of the first target. As an example of the prediction result of the demand of the third target, the prediction unit 120 can use the prediction result of the demand for the third target determined as shown in FIGS. 3A and 3B.

  That is, the prediction unit 120 can use the prediction result of the demand with the superordinate concept including the first or second target as the third target. FIG. 3A is a diagram illustrating an example of the first to third objects in this case.

  In the example shown in FIG. 3A, for example, when the first object is a Tuna Mayo rice ball, the alternative object is a kelp rice ball or a plum rice ball that can be substituted. And "the whole rice ball" including these various rice balls becomes a 3rd object. Similarly, the second or third target is determined for other types of rice balls or bread. The prediction unit 120 predicts the demand of the first target based on the prediction result of the demand related to the third target thus determined.

  As in this example, the predicted result of the demand for the third target, which is a superordinate concept including the first or second target, relates to the predicted result of the demand when compared with the third target defined in another format. The variance is considered small. Therefore, the prediction unit 120 can improve the prediction accuracy of the demand of the first target by using the prediction result related to the demand of the third target determined in this way.

  Moreover, the prediction part 120 can use the prediction result of the demand with respect to the whole for the whole object regarding the 1st or 2nd object as a 3rd object. FIG. 3B is a diagram illustrating an example of the first to third objects in this case.

  In the example shown in FIG. 3B, for example, when the first object is a Tuna Mayo rice ball, the second object is a kelp rice ball or a plum rice ball that can be substituted as in the example shown in FIG. 3A. And the whole of all types of rice balls that are the first or second target are set as the third target. Similarly, the second or third target is determined for other types of rice balls or bread. The prediction unit 120 may predict the demand of the first target based on the demand prediction result related to the third target thus determined.

  Note that the demand for the third target is obtained by using multiple regression analysis, ARMA (Autogressive moving average), SVR (Support Vector Regression), and the like. That is, the demand for the third target is obtained using a prediction formula or the like obtained in advance based on these methods.

  In addition, as the ratio of the first object, the prediction unit 120 can use the ratio of the first object obtained by various methods. For example, the prediction unit 120 can use the ratio of the first target obtained based on the prediction results of the individual demands regarding each of the first and second targets.

  In this case, the demand of the first or second target is predicted in consideration of the weather and temperature at the time when the demand is predicted for the first target, the presence of an event, a sale, a campaign, or the like. That is, the ratio of the first target in consideration of factors that may affect the demand of the first target described above is obtained. Therefore, by using the ratio of the first target obtained in this way, the prediction unit 120 can predict the demand of the first target with higher accuracy.

  Note that the demand for the first or second target is obtained using multiple regression analysis, ARMA, SVR, or the like, for example, similarly to the demand for the third target.

  Moreover, you may obtain | require the ratio of a 1st object based on the track record of the past demand regarding the 1st or 2nd object. The ratio of the first target can be easily obtained based on the past demand record related to the first or second target. Further, in this case, the ratio of the first target may be obtained based on the past demand actual result at the time when the demand is predicted for the first target and the above-described factor is similar.

  Next, the operation of the prediction system 100 according to the first embodiment of the present invention will be described using the flowchart of FIG.

  First, the reception part 110 receives the information regarding a 1st object (step S101). As described above, the reception unit 110 may receive information necessary for the prediction unit 120 in addition to the information related to the first target. In addition, the reception unit 110 may store the received information about the first target in a storage unit (not shown).

  Subsequently, the prediction unit 120 calculates the demand of the first target based on the prediction result of the demand related to the third target including the first target and the second target that can be substituted for the first target. Prediction is performed (step S102). The prediction result of the demand of the first target obtained by the prediction unit 120 is output from an arbitrary output unit (not shown) including a display device and a communication network, for example. Moreover, the prediction result of the demand of the 1st object calculated | required in the prediction part 120 may be memorize | stored in arbitrary memory | storage means so that it may be referred when needed.

  As described above, the prediction system 100 according to the first embodiment of the present invention is a demand prediction result regarding the third target including the first target and the second target that can be substituted for the first target. Based on this, the demand for the first target is predicted. When the prediction result of the demand for the third target including the first target that is the prediction target is smaller than the prediction result of the demand for the first or second target that is the individual object, Conceivable. In the prediction system 100 according to the present embodiment, the prediction unit 120 predicts the demand of the first target based on the prediction result of the demand for the third target described above. Therefore, the prediction system 100 in the present embodiment can predict the demand for goods and the like with high accuracy.

(Modification of the first embodiment)
Various modifications can be considered for the prediction system 100 in the present embodiment.

  As an example, as described above with reference to FIG. 2, the third object may have a plurality of hierarchies. That is, a plurality of third objects may exist, and at least one of them may be in a relationship including the other.

  In this case, the number of hierarchies is arbitrary. In the example illustrated in FIG. 2, there is a third target for the entire target of purchase including the third target for the entire rice ball and the entire bread. That is, in the example shown in FIG. 2, there are two layers as the third target. However, the third object may further have a plurality of hierarchies. The number of hierarchies is determined as appropriate according to the characteristics of the object to be predicted for demand.

  In this case, the prediction unit 120 predicts the demand of the first target using the plurality of third targets. That is, the prediction unit 120, for example, based on the prediction result of the demand for the third target in the upper hierarchy and the information on the demand of the plurality of third targets in the lower hierarchy (for example, the ratio of each demand, etc.) Predict demands of a plurality of third objects in the lower hierarchy. Then, the prediction unit 120 calculates the demand of the first target based on the prediction result of the demand for each of the third targets in the lower hierarchy predicted as described above and information on the demands of the first and second targets. Predict.

  In this case, the demand prediction result for the third target in the lower hierarchy may be output from any output means together with the demand prediction result for the first target, or may be stored in any storage means. May be.

  In the present embodiment, the range of the target object that is the second target is appropriately determined according to the characteristics of the first target and the like. That is, when there are a plurality of objects that can be candidates for the second object, all of them may be the second object, or some of them may be the second object. In the example of FIG. 2, when there are other types of rice balls and bread different from the illustrated rice balls and bread, those rice balls and bread may be further included in the second object. The range of the target object to be the second target is appropriately determined according to the relationship with the first target, the required prediction accuracy, and the like.

  Further, information regarding the relationship between the first object and each of the objects that can be the second object may be used. For example, in the example of bread described above, for a certain type of bread, it is considered that there is a bread that is more likely to be purchased instead, or a bread that is less likely to be purchased instead. That is, when there are a plurality of second objects, the degree of possibility of substitution with each of the first objects may be different.

  Therefore, the range of the target object that is the second target may be determined based on information obtained by weighting the possibility of substitution between the first target and each of the second target by any method. The prediction unit 120 may further predict the demand of the first target by further using information on the weighting. In the prediction unit 120 or the like, when the ratio of the first target described above is obtained, information regarding this weighting may be considered.

(Second Embodiment)
Subsequently, a second embodiment of the present invention will be described. FIG. 5 is a diagram illustrating a configuration of the prediction system 200 according to the second embodiment of the present invention.

  As shown in FIG. 5, the prediction system 200 in the second embodiment further includes a first prediction unit 121, a second prediction unit 122, and a third prediction unit 123. The prediction system 200 according to the second embodiment of the present invention is different from the prediction system 100 according to the first embodiment of the present invention in these points. About a point other than this, the structure and operation | movement of the prediction system 200 in the 2nd Embodiment of this invention are the same as that of the prediction system 100 in the 1st Embodiment of this invention.

  In the present embodiment, the first prediction unit 121 predicts the demand for the first target based on the information related to the demand prediction for the first target. Similarly, the 2nd prediction part 122 predicts the demand of the 2nd object based on information about prediction of the demand of the 2nd object. The 3rd prediction part 123 predicts the demand of the 3rd object based on the information about prediction of the demand of the 3rd object.

  In the present embodiment, the first target demand predicted by the first prediction unit 121 is a provisional prediction used when the prediction unit 120 predicts the first target demand.

  As described above, the prediction unit 120, as an example, predicts the demand of the first target based on the ratio of demand to the first target and the demand prediction result regarding the third target. Moreover, if the ratio of the demand of the first target is obtained based on the demand prediction result for each of the first and second objects, a highly accurate ratio is obtained. That is, the prediction unit 120 directly determines the demand of the first target by using the demand prediction result regarding the third target that has relatively small fluctuations in demand compared to each of the first target and the second target. Compared with the case where it calculates | requires, it can estimate with sufficient precision.

  And the prediction system 200 in this embodiment is further provided with the 1st prediction part 121, the 2nd prediction part 122, and the 3rd prediction part 123, According to the situation of the factor etc. which influence demands, such as a 1st object. Thus, it becomes possible to obtain a prediction result of each demand. Therefore, in the present embodiment, the prediction unit 120 can predict the demand for the first target with higher accuracy in accordance with changes in demand related to the first target and factors that affect the demand.

  Each of the first prediction unit 121, the second prediction unit 122, and the third prediction unit 123 predicts each demand by an arbitrary method. When forecasting demand, techniques such as multiple regression analysis, ARMA, and SVR are used. That is, each of the 1st prediction part 121, the 2nd prediction part 122, and the 3rd prediction part 123 predicts each demand, for example using the prediction formula etc. which were beforehand obtained based on these methods.

  Moreover, as information about the prediction of the demand regarding the 1st-3rd object, the past demand results about each object, its moving average, etc. are mentioned as an example. When the first to third objects are products as shown in FIG. 2, the information related to demand prediction includes, for example, information on the presence or absence of sales and campaigns related to the products, etc., and their details. It is. Moreover, the presence or absence of the event in the vicinity of the store etc. where the said goods etc. are sold, the detail, or weather data, such as the weather in the vicinity of the said store, and temperature, etc. may be contained in the information regarding demand prediction.

  As described above, the prediction system 200 according to the second embodiment of the present invention further includes the first prediction unit 121, the second prediction unit 122, and the third prediction unit 123 with respect to the prediction system 100 according to the first embodiment. Prepare. With such a configuration, the prediction system 200 in the present embodiment predicts the demand of the first target according to the demand of the first target, the second target, etc., changes in factors that affect the demand, and the like. It becomes possible. Therefore, the prediction system 200 in this embodiment can predict the demand of the first target with high accuracy.

  Note that the prediction system 200 in the present embodiment may be configured to include at least a part of the first prediction unit 121, the second prediction unit 122, and the third prediction unit 123. The prediction system 200 in the present embodiment may be configured such that any one of the first prediction unit 121, the second prediction unit 122, and the third prediction unit 123 also serves as other functions.

(Third embodiment)
Subsequently, a second embodiment of the present invention will be described. FIG. 6 is a diagram illustrating a configuration of a prediction system 300 according to the third embodiment of the present invention. FIG. 7 is a diagram illustrating a configuration of a modified example of the prediction system 300 according to the third embodiment of the present invention. FIG. 8 is a diagram illustrating a configuration of another modified example of the prediction system 300 according to the third embodiment of the present invention. FIG. 9 is a diagram illustrating a configuration of another modified example of the prediction system 300 according to the third embodiment of the present invention.

  As shown in FIG. 6, the prediction system 300 according to the third embodiment of the present invention further includes a relationship storage unit 130. The prediction system 300 according to the third embodiment of the present invention is different from the prediction system 100 according to the first embodiment of the present invention in these points. About a point other than this, the structure and operation | movement of the prediction system 300 in the 3rd Embodiment of this invention are the same as that of the prediction system 100 in the 1st Embodiment of this invention.

  In the present embodiment, the relationship storage unit 130 stores information related to the relationship among the first object, the second object, and the third object. This relationship is, for example, the relationship shown in FIG. 2, FIG. 3A or FIG. 3B. In other words, the relationship storage unit 130 stores information on an object that can be a second object with respect to the first object and an object that can be a third object with respect to the first and second objects.

  In forecasting demand, even if the first target is determined, the second target and the third target for the first target may not always be clear. When the first target is determined, the expert regarding demand prediction can determine the second or third target that can affect the demand of the first target. However, it may be difficult for a person with little knowledge or experience regarding demand prediction to determine the second or third target for the first target.

  In contrast, the relationship storage unit 130 stores information related to the relationship among the first object, the second object, and the third object. By doing in this way, it becomes possible for the user of the prediction system 300 in this embodiment to perform the prediction of the demand regarding a 1st object using the information memorize | stored in the relationship memory | storage part 130. FIG.

  The relationship storage unit 130 can store arbitrary information that can be the first, second, or third target. That is, the relationship storage unit 130 may store information on an object that may not be used for prediction in the prediction unit 120 when a specific object is the first object. The prediction unit 120 appropriately selects information related to an object that can be the second or third target with respect to the first target, and predicts demand related to the first target.

  As described above, the prediction system 300 according to the third embodiment of the present invention includes the relationship storage unit 130. By setting it as such a structure, a 2nd and 3rd object is easily determined with respect to the 1st object with respect to the prediction object of a demand. Therefore, the prediction system 300 in this embodiment makes it possible to easily make a demand for the first target.

  In the present embodiment, the prediction system 300 may include a prediction information storage unit 140 as illustrated in FIG. In this case, the prediction information storage unit 140 stores information necessary for predicting each demand, such as an explanatory variable regarding each demand of the first to third objects. The explanatory variable is a variable that represents information that may affect the demand for goods, for example. Or an explanatory variable may be a variable showing the information which may have a correlation with demands, such as goods. In addition, the prediction information storage unit 140 stores a prediction formula used when demand prediction is performed in the prediction unit 120 or the like. In addition to this, the prediction information storage unit 140 may store information required in the prediction unit 120 and the like. When the prediction information storage unit 140 is provided, the prediction unit 120 uses the information stored in the prediction information storage unit 140 to predict demand related to the first target.

  Moreover, the prediction system 300 in this embodiment may be used in combination with the prediction system 200 in the second embodiment of the present invention. That is, the prediction system 300 in the present embodiment may further include a first prediction unit 121, a second prediction unit 122, and a third prediction unit 123, as shown in FIG. By setting it as such a structure, even if it is a case where it is a case where it is a case where it is used by the person who is not familiar with the prediction of a demand, the prediction system 300 can predict the demand of a 1st object accurately.

Further, in this case, the prediction system 300 may include a prediction information storage unit 140 as shown in FIG. In this case, the information stored in the prediction information storage unit 140 is also used when the first prediction unit 121, the second prediction unit 122, or the third prediction unit 123 performs prediction.
(Application example of prediction system)
The prediction system in each embodiment of the present invention is applied to prediction of demand for various objects in addition to the above-described demand prediction of foods and various products including rice balls and bread.

  As an example, the prediction system in each embodiment of the present invention is used for prediction of demand related to products and the like different from the above-described example in a store or the like.

  That is, the prediction system in each embodiment of the present invention is capable of predicting demand relating to products and the like different from the rice balls and bread described with reference to FIG. Moreover, the prediction system in each embodiment of this invention can also forecast the demand as the whole of the goods etc. which belong to a certain category by making the whole rice ball or the whole bread into the 1st or 2nd object.

  Moreover, the prediction system in each embodiment of this invention can estimate the sales in a certain store etc. as a demand. In this case, for example, if a store or the like is located in the vicinity of the first target and in the vicinity of the store, the store that is in a competitive relationship with the store is in a competitive relationship with the second target and the store. It is possible to consider the entire possible store as the third target.

  As another example, the prediction system in each embodiment of the present invention is used for prediction of demand of transportation facilities such as Shinkansen.

  When moving between points, a plurality of transportation facilities may be available. In this case, some transportation may be replaceable with other transportation. In addition, a train, a flight, or the like in a certain time zone of a certain transportation may be replaced with a train, a flight, or the like in another time zone in the transportation. On the other hand, the number of passengers moving between the points may be generally constant regardless of the type of transportation selected.

  In other words, in this case, it is considered that a certain transportation facility can be regarded as a first target, another transportation facility as a second target, and the entire transportation facility that transports between the points as a third target. It is done. Then, it is considered that passengers of these transportation facilities can be regarded as demands of the first to third objects. Similarly, a train etc. in a specific time zone of a certain transportation is a first subject, a train in another time zone of this transportation is a second subject, and the whole train etc. Can be regarded as a third object. And the passenger for every time slot | zone of the said transportation can be considered as the demand of the 1st or 2nd object, and the whole passenger who moves between the said points in the said transportation can be considered as the demand of the 3rd object.

  And the prediction system in each embodiment of the present invention is capable of predicting the number of passengers of a certain transportation facility and the number of passengers such as trains in a specific time zone of a certain transportation facility based on such information on demand. Is possible. Thus, by predicting demand in transportation using the prediction system in each embodiment of the present invention, it is possible for transportation operators, etc., to determine more appropriate operation plans and charges based on demand. It becomes.

  As another example, the prediction system in each embodiment of the present invention is used for prediction of demand in various facilities such as an amusement park.

  In an amusement park or the like, a visitor selects and uses some of a plurality of attractions prepared therein. In this case, an amusement park attraction may be in a replaceable relationship with another attraction. Similarly, for example, a tourist facility or the like existing in a specific area such as a sightseeing spot may have a replaceable relationship with another facility existing in the area. Alternatively, a commercial facility or the like that exists in a predetermined area such as a downtown area may have a replaceable relationship with another facility that exists in the area. On the other hand, the number of visitors and visitors in the amusement park or in a specific area may be substantially constant depending on factors such as day of the week and weather.

  That is, in this case, it is possible to regard any one of the attractions, etc. prepared in the facility such as the amusement park as the first object, the other attractions as the second object, and the amusement park as the third object. It is. Further, it is possible to regard the users of these attractions as the demand of the first or second target, and regard the visitors to the amusement park as the demand of the third target. Similarly, a facility person existing in a certain area can be regarded as the first object, another facility existing in the area can be regarded as the second object, and the area can be regarded as the third object. Further, it is possible to regard visitors to each facility as the demand of the first or second target, and visitors to the area as the demand of the third target.

  And the prediction system in each embodiment of this invention can estimate the number of users, such as a certain attraction, and the number of visitors to a certain facility based on the information regarding these demands, etc. By predicting the number of visitors by using the prediction system in each embodiment of the present invention, a management company such as an amusement park appropriately performs operations such as attractions according to demand, and promotions that stimulate demand. Can be performed.

  As yet another example, the prediction system in each embodiment of the present invention is used for predicting the audience rating of a television or the like.

  A viewer such as a television selects and views a channel related to a desired broadcasting organization or a program on the channel according to the status of the program being broadcast. In this case, a viewer can replace a certain channel or program with another channel or program. On the other hand, it is considered that the total number of viewers such as televisions is generally constant in accordance with the day of the week or the time zone.

  That is, in this case, a program or the like broadcast by a certain broadcasting organization can be the first target. In this case, it is possible to regard other programs and the like broadcasted by a broadcasting organization or the like in the time zone in which the program is broadcast as the second target, and programs and the like in the time zone as the third target. is there. Further, the audience rating (or the number of viewers) of each program is regarded as the demand of the first or second target, and the total audience rating (or the total number of viewers) of the program or the like in the time zone is regarded as the demand of the third target. Can be considered.

And the prediction system in each embodiment of this invention estimates the audience rating (or the number of viewers) of a certain channel based on the information regarding these demands, etc.
By predicting the audience rating using the prediction system in each embodiment of the present invention, it is possible to appropriately set commercial charges according to demand and organize programs.

  While the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. The configurations in the embodiments can be combined with each other without departing from the scope of the present invention.

  This application claims the priority on the basis of Japanese application Japanese Patent Application No. 2015-013294 for which it applied on January 27, 2015, and takes in those the indications of all here.

100, 200, 300 Prediction system 110 Reception unit 120 Prediction unit 121 First prediction unit 122 Second prediction unit 123 Third prediction unit 130 Relationship storage unit 140 Prediction information storage unit 500 Information processing device 501 CPU
502 ROM
503 RAM
504 Program 505 Storage device 506 Recording medium 507 Drive device 508 Communication interface 509 Network 510 Input / output interface 511 Bus

Claims (12)

Accepting means for accepting at least information on a first object that is a demand forecast object;
Information relating to prediction of demand for the first object, information relating to prediction of demand for the second object that can be substituted for the first object, and demand relating to the third object including the first and second objects A prediction system comprising: a predicting unit that predicts the demand of the first target based on the prediction result. The prediction means includes
The prediction system according to claim 1, wherein the demand of the first target is predicted based on a ratio of the first target in the demand of the first and second targets and a prediction result of the demand related to the third target. . First prediction means for predicting demand of the first object;
Second prediction means for predicting demand of the second target;
Third prediction means for predicting the demand of the third target,
The prediction means includes a provisional prediction result of demand related to the first target predicted by the first prediction means, and a prediction result regarding demand related to the second target predicted by the second prediction means; The prediction system according to claim 1 or 2, wherein the demand of the first target is predicted based on a prediction result of the demand related to the third target predicted by the third prediction unit. There are a plurality of third objects, and at least one of the third objects includes the other third objects;
The prediction means predicts the demand of the other third object based on the information on the demand of each of the other third objects and the prediction result of the demand on the at least one third object,
And the information regarding the prediction of the demand of the first object, the information regarding the prediction of the demand of the second object, and the prediction result of the other third object including the first and other second objects The prediction system according to claim 1, wherein the demand of the first target is predicted based on the prediction target.   The prediction system according to any one of claims 1 to 4, further comprising a relationship storage unit that stores information regarding a relationship between the first object, the second object, and the third object.   The prediction system according to any one of claims 1 to 5, further comprising a prediction information storage unit that stores at least information required for prediction of demand in the prediction unit. The first and second objects are foods belonging to the same category,
The third object is the entire category,
The prediction system according to any one of claims 1 to 6, wherein the prediction unit predicts the sales quantity of the food that is the first target as demand. Each of the first and second objects is a transportation facility in a predetermined section,
The third object is the entire transportation that transports the predetermined section,
The prediction system according to any one of claims 1 to 6, wherein the prediction means predicts the number of passengers of the transportation facility that is the first target as demand. Each of the first and second objects is a facility in a predetermined area,
The third object is the predetermined area,
The prediction system according to any one of claims 1 to 6, wherein the prediction means predicts the number of visitors to the facility as the first target as demand. The first and second objects are programs that are broadcast by different broadcasting organizations,
The third target is the entire program broadcast by the broadcasting organization,
The prediction system according to any one of claims 1 to 6, wherein the prediction unit predicts the audience rating of the first target program as demand. Accept at least information about the first target for which the demand is forecast,
Information relating to prediction of demand for the first object, information relating to prediction of demand for the second object that can be substituted for the first object, and demand relating to the third object including the first and second objects A prediction method for predicting the demand of the first target based on the prediction result. On the computer,
A process of accepting at least information related to a first target that is a target of demand prediction;
Information relating to prediction of demand for the first object, information relating to prediction of demand for the second object that can be substituted for the first object, and demand relating to the third object including the first and second objects The computer-readable recording medium which stored the program which performs the process which estimates the demand of said 1st object based on the prediction result of this.

Images