Service optimization system, service optimization method, and program

Abstract

To improve the accuracy of predicting service demand.SOLUTION: A system of the present disclosure predicts service demand based on real data that includes at least one of a behavior, a behavior history, and a behavior schedule of a customer related to a service, through simulation using a human model that models a service use behavior of the customer. Next, the system of the present disclosure determines service parameter setting values at which a service level is maintained based on the predicted demand, through simulations using a service model that models a relation between service parameters for determining a service provision content, the service demand, and the service level. Then, the system of the present disclosure acquires the real data after the service is provided using the determined service parameter setting values, and feeds back the acquired real data to the input of the simulation using the human model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a system, method, and program for optimizing services using simulation. [Background technology]

[0002] The demand forecasting device disclosed in Japanese Patent Application Laid-Open No. 2005-352728 comprises a model creation device and a forecasting device. The model creation device models customer behavior as customer awareness of a service, including awareness of the existence of the service and awareness of the service's functions. The forecasting device predicts which service a customer will select using the customer behavior model created by the model creation device. This conventional technology aims to improve the accuracy of the customer behavior model. However, even if the accuracy of the behavior model is improved, it still faces difficulties in forecasting actual demand for the service.

[0003] In addition to Japanese Patent Application Laid-Open No. 2005-352728, Japanese Patent Application Laid-Open No. 2013-106507 can also be cited as examples of documents that demonstrate the state of the art in the technical field related to the present disclosure. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-352728 [Patent Document 2] Japanese Patent Application Laid-Open No. 2013-106507 Summary of the Invention [Problem to be solved by the invention]

[0005] The present disclosure has been made in view of the above-mentioned problems, and an object of the present disclosure is to improve the accuracy of predicting demand for services. [Means for solving the problem]

[0006] The present disclosure provides a service optimization system. The service optimization system of the present disclosure includes at least one processor and a program memory coupled to the at least one processor and storing a plurality of instructions. The plurality of instructions are configured to cause the at least one processor to execute the following first to third processes. The first process is a process of predicting demand for a service from real data including at least one of customer behavior, behavior history, and behavior schedule related to the service, through a simulation using a human model that models customer service usage behavior. The second process is a process of determining setting values for service parameters that maintain the service level based on the predicted demand, through a simulation using a service model that models the relationship between service parameters that determine the content of the service to be provided and the demand and service level of the service. The third process is a process of acquiring real data after the service is provided using the determined setting values of the service parameters, and feeding the acquired real data back into the input of the simulation using the human model.

[0007] The present disclosure also provides a program that can be stored in a computer-readable storage medium, the program being configured to cause a computer to execute the first to third processes described above.

[0008] The present disclosure further provides a service optimization method. The service optimization method of the present disclosure includes the following first to third steps. The first step is a step of predicting demand for a service from real data including at least one of customer behavior, behavior history, and behavior schedule related to the service, through a simulation using a human model that models customer service usage behavior. The second step is a step of determining setting values for service parameters that maintain the service level based on the predicted demand, through a simulation using a service model that models the relationship between service parameters that determine the content of the service to be provided and the demand for the service and the service level. The third step is a step of acquiring real data after the service is provided using the determined setting values for the service parameters, and feeding the acquired real data back into the input of the simulation using the human model.

[0009] The above-described technology provided by the present disclosure, i.e., the service optimization system, service optimization method, and program, can improve the accuracy of demand predicted by simulation by feeding back real data from the world in which services are actually provided to the input of a simulation using a human model. By improving the accuracy of demand prediction, it becomes possible to optimize the setting values of service parameters and maintain service levels.

[0010] In the above technology provided by the present disclosure, the human model may be trained based on the error between the actual demand for a service obtained from real data and the demand predicted by a simulation using the human model, thereby improving the accuracy of the human model through training using real data.

[0011] In the technology provided by the present disclosure, the customer may include a customer group, and a human model may be prepared for each attribute of the customer group. By preparing a human model for each attribute of the customer group, it is possible to improve the accuracy of demand forecasting using the human model when forecasting demand for a customer group.

[0012] In addition, in the above technology provided by the present disclosure, a human model may be prepared for each customer and personalized using real data acquired for each customer. Personalizing the human model for each customer can improve the accuracy of demand forecasting using the human model when forecasting demand for each customer.

[0013] In the technology provided by the present disclosure, the service may include multiple services, a service model may be prepared corresponding to each of the multiple services, and the human model may model the customer's usage behavior for the multiple services. In this way, when a customer uses multiple services, it is possible to predict demand for each service and maintain the service level for each of the multiple services.

[0014] Furthermore, in the technology provided by the present disclosure, determining the setting values of the service parameters may include acquiring a service provision policy and determining the setting values of the service parameters that maintain the service level based on the predicted demand using the service provision policy as a model. In this way, when a service provision policy exists, the service level can be maintained in accordance with the service policy by optimizing the setting values of the service parameters using the service provision policy as a model. [Effects of the Invention]

[0015] As described above, according to the service optimization system, service optimization method, and program disclosed herein, the accuracy of demand predicted by the simulation can be improved by feeding back real data from around the world where services are actually provided into the input of a simulation using a human model. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a diagram illustrating a configuration of a service optimization system according to a first embodiment of the present disclosure. [Figure 2] FIG. 2 is a diagram illustrating the configuration of the human model shown in FIG. [Figure 3] FIG. 2 is a diagram illustrating a process executed by the human simulator shown in FIG. [Figure 4] FIG. 10 is a diagram illustrating a process executed by a service simulator. [Figure 5] 10 is a flowchart showing a process executed by a parameter setting value determiner. [Figure 6] FIG. 10 is a diagram illustrating a configuration of a service optimization system according to a second embodiment of the present disclosure. [Figure 7] FIG. 7 is a diagram illustrating a process executed by the human simulator shown in FIG. 6. [Figure 8] FIG. 1 is a diagram illustrating an example of a hardware configuration of a service optimization system according to each embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0017] 1. Overview A service optimization system according to an embodiment of the present disclosure is a system for optimizing the relationship between the demand and supply of a service. The service optimization system according to an embodiment of the present disclosure is configured to predict demand for a service and change service parameters that determine the content of the service to be provided based on the predicted demand. Note that a service optimization method according to an embodiment of the present disclosure is realized by computer processing performed in the service optimization system according to an embodiment of the present disclosure.

[0018] The service optimization system according to the embodiment of the present disclosure can be applied to various types of services provided in the real world. A first type of service can be a service in which the service optimization system directly controls service parameters via a network. A second type of service can be a service in which a service provider adjusts service parameters according to information supplied from the service optimization system.

[0019] Specific examples of services belonging to the first type include on-demand bus services and ride-sharing services provided using autonomous vehicles. In these services, service parameters include the number of vehicles in use per time period, the timing of charging each vehicle at a charging station, and the area in which the vehicles patrol. Another specific example of a service belonging to the first type is a logistics service provided using delivery robots. In this service, service parameters include the number of delivery robots in use per time period, the timing of charging each delivery robot at a charging station, and the area in which packages are collected. Another specific example of a service belonging to the first type is a product sales service through an online store. The types of products listed on the web and the order in which they are listed are examples of service parameters for this service.

[0020] A specific example of a service that falls into the second category is a product sales service in a physical store. In this service, the types of products sold, the quantity of each type, and the display location of the products in the store are used as service parameters. In addition, manned taxi services operated by a driver and manned home delivery services in which delivery personnel deliver and collect packages are also services that fall into the second category. Service parameters for a manned taxi service are, for example, the number of taxis to be dispatched and the taxi patrol area. Service parameters for a manned home delivery service are, for example, the number of vehicles and delivery personnel that collect packages, or the collection area for packages.

[0021] In addition, in any of the services, the price of the service provided and the price of the product provided in the service are examples of service parameters. Therefore, dynamic pricing is included in the means for optimizing the relationship between supply and demand of the service that can be adopted by the service optimization system according to the embodiment of the present disclosure.

[0022] A service optimization system according to an embodiment of the present disclosure has a function of predicting demand for a service. Various methods for predicting demand for a service are known. For example, a method for predicting demand using a statistical technique that uses past performance data and a method for predicting demand by analyzing big data using machine learning are known. However, rather than using such conventional methods as they are, a service optimization system according to an embodiment of the present disclosure uses a method for predicting demand based on the behavior of customers to whom the service is to be provided.

[0023] Specifically, a service optimization system according to an embodiment of the present disclosure predicts demand for a service through simulation using a human model. The human model models the behavior of customers using a service. Demand for a service arises as a result of customer behavior, and customer behavior arises as a result of the provision of a service, so customer behavior that leads to demand for a service can be modeled. However, the modeled behavior is routine behavior, and out-of-routine behavior that leads to impulse buying is excluded from modeling. A machine learning model can be used as the human model. There is no limitation on the type of machine learning model. For example, the human model may be a linear regression model, a neural network model, or a deep learning model.

[0024] Simulations using human models use real data obtained in the real world. However, the real data used is data related to the current behavior, past behavior history, and future behavior schedule of customers who are the target of the service. However, only that real data related to the service being predicted is input into the human model. Using such data makes it possible to accurately predict customer demand for the service.

[0025] Furthermore, the service optimization system according to an embodiment of the present disclosure considers the service level when determining the value of the service parameter from the predicted demand. This is because, although there are multiple set values of the service parameter that can satisfy the demand, the service level changes depending on the set value of the service parameter. Taking a ride-sharing service as an example, if usage by six customers is predicted, the service level will differ depending on whether one six-seater vehicle or two four-seater vehicles is provided. In the former case, customers will have to wait longer for the arrival of the vehicle than in the latter case, and the vehicle will also be more crowded.

[0026] In reality, multiple service parameters often exist for one service. In such cases, the service level is determined by a combination of the setting values of the multiple service parameters. Therefore, a service optimization system according to an embodiment of the present disclosure determines the setting values of service parameters that maintain the service level at or above a predetermined threshold level based on predicted demand through a simulation using a service model. The service model models the relationship between the service parameters, the service demand, and the service level. A machine learning model can be used for the service model. There is no limitation on the type of machine learning model.

[0027] Hereinafter, details of a service optimization system according to an embodiment of the present disclosure will be described with reference to the drawings.

[0028] 2. First embodiment The service optimization system according to the first embodiment of the present disclosure is a system for a single service. Fig. 1 is a diagram showing the configuration of the service optimization system 11 according to the first embodiment of the present disclosure.

[0029] 1, the service optimization system 11 comprises a virtual part 11V established in a virtual world and a real part 11R established in the real world. The virtual world in which the virtual part 11V is established is a world realized by a computer that physically constitutes the service optimization system 11. The virtual part 11V may be established on a cloud and connected to the real part 11R via a network including the Internet.

[0030] The virtual part 11V includes a human simulator 21, a service simulator 41, a parameter setting value determiner 61, and a policy determiner 71. These devices 21, 41, 61, and 71 are virtual devices, and are realized by programs executable by a computer. There are no limitations on the hardware configuration of the computer system on which the program realizing the virtual part 11V is executed.

[0031] The human simulator 21 is configured to predict the demand for a service using a human model 31. Details of the human simulator 21 will be described below with reference to FIGS. 2A, 2B, and 3.

[0032] The human model 31 models the behavior of customers when using the services targeted by the service optimization system 11. The configuration of the human model 31 is explained using FIG. 2A. In the configuration shown in FIG. 2A, customers are grouped by attributes, and a human model is prepared for each group. In other words, the human model 31 is represented as a collection of human models prepared for each customer group. Examples of customer attributes include gender and age. According to the configuration shown in FIG. 2A, for example, a 45-year-old woman and a 55-year-old man belong to different groups, and their respective demands are predicted using different human models. In contrast, a 32-year-old woman and a 38-year-old woman belong to the same group, and their respective demands are predicted using the same human model. Other examples of customer attributes include occupation and living area.

[0033] Another configuration of the human model 31 is shown in FIG. 2B. In the configuration shown in FIG. 2B, a human model is prepared for each customer. For example, if there are N customers of a service targeted by the service optimization system 11, a total of N human models are prepared. In other words, the human model 31 is expressed as a collection of human models prepared for each customer. The human model prepared for each customer is personalized using real data acquired for each customer.

[0034] Note that the amount of real data required to personalize a human model for each customer is enormous. Therefore, it is not easy to prepare a human model for each new customer. Therefore, it is possible to first use a human model prepared for each attribute of a customer group as shown in FIG. 2A, and then use a human model for each customer as shown in FIG. 2B for customers for whom sufficient real data has been collected for personalization.

[0035] FIG. 3 is a diagram showing the processing executed by the human simulator 21. The human simulator 21 inputs real data relating to a customer's current behavior, past behavior history, and future behavior schedule into the human model 31. The human model 31 outputs service demand predicted from the input real data. By gathering real data relating to past, present, and future behavior, it is possible to accurately predict service demand arising from customers. However, it is not necessary to have all of these three types of real data; it is sufficient that at least some of them are input into the human model 31.

[0036] The human simulator 21 stores the predicted demand for the service output from the human model 31 in storage. The human simulator 21 also acquires data related to the actual demand for the service in the real world. The higher the accuracy of the human model 31, the smaller the error between the predicted demand and the actual demand. The human simulator 21 feeds back the error between the predicted demand and the actual demand to the human model 31, and learns the parameters of the human model 31. Note that learning of the human model 31 may be performed by batch processing or in parallel with the simulation.

[0037] 1, the virtual part 11V will be further described. The human simulator 21 transmits the predicted demand for the service output from the human model 31 to the service simulator 41.

[0038] The service simulator 41 is configured to calculate the service level that can be achieved by the currently set service parameters in response to the predicted demand for the service. The service simulator 41 uses a service model 51 to calculate the service level.

[0039] FIG. 4 is a diagram showing the processing executed by the service simulator 41. The service simulator 41 inputs the predicted demand for the service calculated by the human model 31 into the service model 51. The service simulator 41 also inputs the current setting values of the service parameters into the service model 51. There are multiple service parameters that need to be set to provide a single service. The service simulator 41 inputs the current setting values of all the service parameters into the service model 51. For example, if the service targeted by the service optimization system 11 is a ride-sharing service, the setting values for the number of vehicles in use, the timing of charging each vehicle at a charging facility, the route area for the vehicles, etc. are input into the service model 51. The service model 51 outputs a service level calculated from the input predicted demand for the service and the current setting values of each service parameter.

[0040] 1, the virtual part 11V will be further described. The service simulator 41 transmits the service level output from the service model 51 to the parameter setting value determiner 61.

[0041] The parameter set value determiner 61 is configured to refer to the service level input from the service simulator 41 and determine the set values of the service parameters that can maintain the level of service provided to customers at or above a predetermined threshold level. Specifically, if the input service level does not reach or exceed the predetermined threshold level, the parameter set value determiner 61 changes the set values of the service parameters. When the changed set values of the service parameters are input to the service model 51, the service level output from the service model 51 also changes.

[0042] Additionally, the parameter setting value determiner 61 receives input of setting value candidates for the service parameters from the policy determiner 71. The policy determiner 71 accepts input of service provision policies from the service provider. If the service provider has service provision policies, the values that each service parameter can take are restricted by the service provision policies. The policy determiner 71 reflects the service provision policies determined by the service provider in the setting value candidates for the service parameters. For example, if the service parameters are the number of vehicles used and operating hours in a ride-sharing service, the setting value candidates for each service parameter may be determined so that the operating hours per vehicle decrease as the number of vehicles used increases.

[0043] The parameter set value determiner 61 determines the set value of a service parameter from among the set value candidates of the service parameter input from the policy determiner 71. When the set value candidates of a service parameter are indicated as a range, the set value of the service parameter is determined within that range. Furthermore, when the set value candidates define a relationship between the set values of a plurality of service parameters, the set value of the service parameter is determined in accordance with the defined relationship.

[0044] The above processing executed by the parameter setting value determiner 61 can be represented by a flowchart as shown in Fig. 5. First, in step S1, the service level under the current setting values of the service parameters is acquired from the service simulator 41. In step S2, it is determined whether the service level acquired in step S1 is equal to or greater than a predetermined threshold level.

[0045] If the determination result in step S2 is positive, it means that the current set values of the service parameters can be used to provide the customer with a service at or above the threshold level. In this case, the processing flow proceeds to step S3. In step S3, it is determined that the current set values are to be used as the set values of the service parameters. Then, in step S4, the set values of the service parameters determined in step S3 are set in the real service provider 81, which will be described later.

[0046] On the other hand, a negative determination result in step S2 means that the current setting values of the service parameters cannot provide the customer with a service above the threshold level. In this case, the processing flow proceeds to step S5. In step S5, setting value candidates for the service parameters determined in accordance with the service provision policy are acquired from the policy decider 71. In step S6, the setting values of the service parameters are changed from the current setting values among the setting value candidates acquired in step S5. Then, in step S7, the setting values of the service parameters changed in step S6 are set in the service model 51.

[0047] Returning to Fig. 1, the explanation of the virtual part 11V will be continued. As explained above, the virtual part 11V determines the set values of service parameters that maintain the service level at a predetermined level or higher based on the demand predicted from real data, using the service provision measures as a model. The set values of the service parameters determined by the virtual part 11V are transmitted to the real part 11R.

[0048] The real part 11R includes a real service provider 81 and a real data collector 91. The real service provider 81 is a device that provides services to customers in the real world. The configuration and functions of the real service provider 81 vary depending on the content of the service to be provided.

[0049] For example, in the case of a ride-sharing service provided using an autonomous vehicle, the autonomous vehicle can be the real service provider 81. In the case of a logistics service provided using a delivery robot, the delivery robot can be the real service provider 81. In these services belonging to the first type, the setting values of the service parameters can be set directly in the autonomous vehicle or the delivery robot.

[0050] Furthermore, in the case of a product sales service in a physical store, a computer that is placed in the physical store and that displays the set values of service parameters on a display can be the real service provider 81. In the case of a manned delivery service, a computer that is placed in each delivery vehicle and that instructs the delivery person on the set values of service parameters can be the real service provider 81. In these services that belong to the second type, the set values of service parameters are used as instruction values for the service provider and employees.

[0051] The real data collector 91 is configured to collect real data on customer behavior in the real world where services are provided to customers. The real data collected by the real data collector 91 is real data on the customer's current behavior, past behavior history, and future behavior schedule. The real data collector 91 inputs the collected real data into the human model 31. The configuration and function of the real data collector 91 differ depending on the content of the service that affects behavior.

[0052] For example, if the service provided to a customer is a ride-sharing service provided using an autonomous vehicle, the customer's current location from the nearest station may be collected as real data related to their behavior. In this case, the GPS system of the customer's mobile terminal may be used as real data collector 91. Furthermore, the customer's payment record for the transportation service may be collected as real data related to their past behavior history. In this case, the payment system may be used as real data collector 91. Furthermore, the customer's schedule may be collected as real data related to their future behavior schedule. In this case, a web calendar in which the customer's schedule is registered may be used as real data collector 91.

[0053] If the service provided to a customer is a product sales service in a physical store, the opening and closing of the refrigerator in the customer's home may be collected as real data regarding the customer's current behavior. In this case, a sensor attached to the refrigerator can be real data collector 91. Also, receipts issued at the time of payment may be collected as real data regarding past behavioral history. In this case, a POS system can be real data collector 91. Furthermore, downloading a menu from the Internet can be collected as real data regarding a future behavioral schedule. In this case, a server that accepts the download operation can be real data collector 91.

[0054] As explained above, in the real part 11R, a service is provided to the real world in accordance with the set values of the service parameters determined in the virtual part 11V. Then, in the real part 11R, real data reflecting the results of the service provision is collected, and the collected real data is transmitted to the virtual part 11V. By repeating this process, the service optimization system 11 can improve the accuracy of predicting demand for the service, and can optimize the set values of the service parameters to maintain a high service level.

[0055] Although the service optimization system 11 according to this embodiment is composed of a virtual part 11V and a real part 11R, only the virtual part 11V may be defined as the service optimization system. In other words, the real part 11R does not necessarily have to constitute a part of the service optimization system. Furthermore, the service optimization system may be composed of the virtual part 11V and either the real service provider 81 or the real data collector 91.

[0056] 3. Second embodiment The service optimization system according to the second embodiment of the present disclosure is a system that targets multiple services. For simplicity, it is assumed that the services targeted for optimization by the service optimization system are two services, service A and service B. Figure 6 is a diagram showing the configuration of the service optimization system 12 according to the second embodiment of the present disclosure.

[0057] 6, the service optimization system 12 comprises a virtual part 12V provided in the virtual world and a real part 12R provided in the real world. However, as described in the first embodiment, only the virtual part 12V may be defined as the service optimization system. In other words, the real part 12R does not necessarily constitute a part of the service optimization system.

[0058] The virtual part 12V includes a human simulator 22, service simulators 42A and 42B, parameter setting value determiners 62A and 62B, and a policy determiner 72. These devices 22, 42A, 42B, 62A, 62B, and 72 are virtual devices, and are realized by programs executable by a computer. There are no limitations on the hardware configuration of the computer system on which the program realizing the virtual part 12V is executed.

[0059] The human simulator 22 is configured to predict demand for services using a human model 32. As in the first embodiment, the human model 32 may be a collection of human models prepared for each customer group grouped according to customer attributes. Alternatively, the human model 32 may be a collection of human models for each customer personalized using real data acquired for each customer. Details of the human simulator 22 will be described below with reference to FIG. 7.

[0060] 7 is a diagram showing the processing executed by the human simulator 22. The human simulator 22 inputs real data relating to a customer's current behavior, past behavior history, and future behavior schedule into the human model 32. The human model 32 models the customer's usage behavior for service A and service B. The relationship between service A and service B regarding the customer's usage behavior is reflected in the human model 32. The human model 32 outputs the demand for service A (predicted demand A) and the demand for service B (predicted demand B) predicted from the input real data.

[0061] The human simulator 22 stores the predicted demand A and the predicted demand B output from the human model 32 in storage. The human simulator 22 also acquires data relating to the actual demand for service A (actual demand A) and the actual demand for service B (actual demand B) in the real world. The higher the accuracy of the human model 32, the smaller the error between the predicted demand A and the actual demand A, and the smaller the error between the predicted demand B and the actual demand B. The human simulator 22 feeds back the error between the predicted demand A and the actual demand A and the error between the predicted demand B and the actual demand B to the human model 32, and learns the parameters of the human model 32. The learning of the human model 32 may be performed by batch processing or in parallel with the simulation.

[0062] Returning to Fig. 6, the description of the virtual part 12V will be continued. The human simulator 22 transmits the predicted demand for the service output from the human model 32 to the service simulators 42A and 42B. The service simulators 42A and 42B are provided independently for each service.

[0063] The service simulator 42A is configured to calculate the service level of service A that can be achieved with the currently set service parameters for the predicted demand A. The service simulator 42A uses a service model 52A to calculate the service level. The service model 52A models the relationship between the service parameters of service A, the demand for service A, and the service level of service A. The service simulator 42A transmits the service level output from the service model 52A to the parameter setting value determiner 62A.

[0064] Furthermore, the service simulator 42B is configured to calculate the service level of service B that can be achieved with the currently set service parameters for the predicted demand B. The service simulator 42B uses a service model 52B to calculate the service level. The service model 52B models the relationship between the service parameters of service B, the demand for service B, and the service level of service B. The service simulator 42B transmits the service level output from the service model 52B to the parameter setting value determiner 62B.

[0065] Parameter setting value determiner 62A changes the setting values of the service parameters of service A when the service level input from service simulator 42A does not reach or exceed a predetermined threshold level. The changed setting values of the service parameters are input to service model 52A, and the service level output from service model 52A also changes. Similarly, parameter setting value determiner 62B changes the setting values of the service parameters of service B when the service level input from service simulator 42B does not reach or exceed a predetermined threshold level. The changed setting values of the service parameters are input to service model 52B, and the service level output from service model 52B also changes.

[0066] Service parameter setting value candidates are input to each of the parameter setting value determiners 62A and 62B from the policy determiner 72. The policy determiner 72 accepts input of provision policies for each of services A and B from the service provider, and reflects the provision policies for each of services A and B determined by the service provider in the setting value candidates for the service parameters of each of services A and B. The parameter setting value determiner 62A determines the setting value of the service parameter of service A from the setting value candidates for the service parameter of service A. The parameter setting value determiner 62B determines the setting value of the service parameter of service B from the setting value candidates for the service parameter of service B.

[0067] Furthermore, the parameter setting value determiners 62A and 62B have a function to arbitrate the setting values of the service parameters between them. This function is effective, for example, when there is overlap in the resources used by service A and service B and the resources are finite. As a specific example, assume that service A is an on-demand bus service using autonomous vehicles, and service B is a logistics service using delivery robots, and that the autonomous vehicles and delivery robots share a charging station. Because the autonomous vehicles and delivery robots cannot use the same charging station at the same time, the optimal charging timing for service A and the optimal charging timing for service B are not necessarily compatible. In such a case, the parameter setting value determiners 62A and 62B use their arbitration function to shift the setting values of each service parameter from the optimal value. In this way, optimization of the service as a whole is achieved.

[0068] The virtual part 12V determines the set values of the service parameters for each of the services A and B, which maintain the service level at or above a predetermined level based on the demand predicted from real data, using the service provision measures as a model. The set values of the service parameters for each of the services A and B determined in the virtual part 12V are transmitted to the real part 12R.

[0069] The real part 12R comprises real service providers 82A and 82B and a real data collector 92. The real service provider 82A is a device that provides service A to customers in the real world. The real service provider 82B is a device that provides service B to customers in the real world. The specific example of the real service provider 81 introduced in the first embodiment also applies to the real service providers 82A and 82B.

[0070] The real data collector 92 is configured to collect real data on customer behavior in the real world where services A and B are provided to the customer. The real data collected by the real data collector 92 is real data on the customer's current behavior, past behavior history, and future behavior schedule. The real data collected by the real data collector 92 is real data related to at least one of service A and service B. The real data collector 92 inputs the collected real data into the human model 32.

[0071] In the real part 12R, services A and B are provided to the real world in accordance with the set values of the service parameters determined in the virtual part 12V. Then, in the real part 11R, real data reflecting the results of providing services A and B is collected, and the collected real data is transmitted to the virtual part 12V. By repeating this process, the service optimization system 12 can improve the accuracy of predicting demand for each of services A and B, and can optimize the set values of the service parameters for each of services A and B to maintain a high service level for each of services A and B.

[0072] 4. Hardware Configuration Finally, the hardware configuration of the service optimization system will be described. Fig. 8 is a diagram showing an example of the hardware configuration of the service optimization system. The service optimization system 10 shown in Fig. 8 corresponds to both the service optimization system 11 according to the first embodiment and the service optimization system 12 according to the second embodiment.

[0073] The service optimization system 10 shown in Fig. 8 is composed of an IoT server 100, a simulation server 110, and a service server 120. The IoT server 100 is a server that receives real data from the real world and corresponds to the input interface of the virtual parts 11V and 12V of the service optimization systems 11 and 12. The service server 120 is a server that transmits setting values of service parameters to various devices (corresponding to real service providers) in the real world and corresponds to the output interface of the virtual parts 11V and 12V. The simulation server 110 corresponds to the virtual parts 11V and 12V themselves.

[0074] Each of the servers 100, 110, 120 includes a processor 101, 111, 121, a program memory 102, 112, 122, and a storage 104, 114, 124. The program memory 102, 112, 122 and the storage 104, 114, 124 are coupled to the processor 101, 111, 121. The program memory 102, 112, 122 stores a plurality of instructions 103, 113, 123. The instructions 103, 113, 123 are executed by the processor 101, 111, 121 to realize each function of the service optimization system 10.

[0075] 8 is merely an example of a hardware configuration applicable to the service optimization systems 11 and 12. For example, the above three servers 100, 110, and 120 may be integrated into one server or may be distributed across more servers. [Explanation of symbols]

[0076] 10, 11, 12 Service Optimization System 11V, 12V Virtual Parts 11R, 12R Real Part 21, 22 Human Simulator 31, 32 Human Model 41, 42A, 42B Service Simulator 51, 52A, 52B service models 61, 62A, 62B Parameter setting value determiner 71, 72 Policy decider 81, 82A, 82B Real service provider 91, 92 Real Data Collector 100 IoT servers 110 Simulation Server 120 Service Server

Claims

1. at least one processor; a program memory coupled to the at least one processor and having a plurality of instructions stored therein; The plurality of instructions may include instructions to the at least one processor: predicting demand for the service from real data including at least one of the customer's behavior, behavior history, and behavior schedule related to the service through a simulation using a human model that models the customer's service usage behavior; calculating a service level that can be achieved with the current setting values of the service parameters in response to the predicted demand through simulation by a service simulator provided independently for each service, using a service model that models the relationship between service parameters that determine the content of the service to be provided and the demand for the service and the level of the service, and determining setting values of the service parameters that will maintain the calculated service level at or above a predetermined level; acquiring real data after the service is provided using the determined setting values of the service parameters, and feeding the acquired real data back as input for a simulation using the human model; When the service includes a plurality of services, the setting values of the service parameters are adjusted among the plurality of services, and the setting values of the service parameters of each of the plurality of services are shifted from the optimum value so as to optimize the plurality of services as a whole. A service optimization system characterized by:

2. The service optimization system according to claim 1, The plurality of instructions are transmitted to the at least one processor. and learning the human model based on an error between an actual demand for the service obtained from the real data and a demand predicted by a simulation using the human model. A service optimization system characterized by:

3. 3. The service optimization system according to claim 1, the customer comprises a customer group; The human model is prepared for each attribute of the customer group. A service optimization system characterized by:

4. 3. The service optimization system according to claim 1, The human model is prepared for each customer and personalized using real data acquired for each customer. A service optimization system characterized by:

5. 3. The service optimization system according to claim 1, the service includes a plurality of services; the service model is prepared corresponding to each of the plurality of services; The human model models the customer's usage behavior for the plurality of services. A service optimization system characterized by:

6. 3. The service optimization system according to claim 1, Determining the setting value of the service parameter includes obtaining a service provision policy, and determining a setting value of the service parameter that maintains the level from the predicted demand based on the service provision policy. A service optimization system characterized by:

7. predicting demand for the service from real data including at least one of the customer's behavior, behavior history, and behavior schedule related to the service through a simulation using a human model that models the customer's service usage behavior; calculating a service level that can be achieved with the current setting values of the service parameters in response to the predicted demand through simulation by a service simulator provided independently for each service, using a service model that models the relationship between service parameters that determine the content of the service to be provided and the demand for the service and the level of the service, and determining setting values of the service parameters that will maintain the calculated service level at or above a predetermined level; acquiring real data after the service is provided using the determined setting values of the service parameters, and feeding the acquired real data back as input for a simulation using the human model; When the service includes a plurality of services, arbitrating the set values of the service parameters among the plurality of services and shifting the set values of the service parameters of each of the plurality of services from the optimum value so as to optimize the plurality of services as a whole. A service optimization method characterized by:

8. predicting demand for the service from real data including at least one of the customer's behavior, behavior history, and behavior schedule related to the service through a simulation using a human model that models the customer's service usage behavior; calculating a service level that can be achieved with the current setting values of the service parameters in response to the predicted demand through simulation by a service simulator provided independently for each service, using a service model that models the relationship between service parameters that determine the content of the service to be provided and the demand for the service and the level of the service, and determining setting values of the service parameters that will maintain the calculated service level at or above a predetermined level; acquiring real data after the service is provided using the determined setting values of the service parameters, and feeding the acquired real data back as input for a simulation using the human model; When the service includes a plurality of services, arbitrating the set values of the service parameters among the plurality of services, and shifting the set values of the service parameters of each of the plurality of services from the optimum value so as to optimize the plurality of services as a whole. A program characterized by:

Images