Device and method
Patent Information
- Application Number
- PCT/JP2025/008832
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-03-10
- Publication Date
- 2026-09-17
Smart Images
Figure JP2025008832_17092026_PF_FP_ABST
Abstract
Description
Apparatus and Method
[0001] The present disclosure relates to an apparatus and a method for determining a measure.
[0002] Patent Literature 1 discloses a technique for optimizing the overall situation of behaviors of a large number of users. This technique comprises: content providing means for transmitting content that prompts behavior change to terminal devices of some users; confirmation means for confirming whether a user who has received the content actually changed their behavior in accordance with the content; and content effect determination means for determining characteristics of users who are likely to accept the content based on a confirmation result obtained by the confirmation means and user characteristics stored in a database, and recording a determination result in the database.
[0003] Japanese Unexamined Patent Publication No. 2017-59099
[0004] Conventionally, techniques for promoting user behavior change and optimizing pedestrian flow have been proposed. In the above technique, since measures that are easily accepted by users are presented, there is a possibility that bias may occur in the content of behavior change, which may make it difficult to optimize pedestrian flow.
[0005] An object of the present disclosure is to provide a technique for determining a measure for optimizing pedestrian flow.
[0006] An apparatus according to one embodiment comprises: a first behavior prediction unit that acquires first behavior data predicting respective behaviors of a plurality of users based on position information; a second behavior prediction unit that acquires second behavior data predicting macro pedestrian flow including the plurality of users based on position information; and a measure determination unit that sets a measure for each of the plurality of users, acquires a sum of travel times of the macro pedestrian flow based on correction data which is the second behavior data corrected using reflection data in which the measure is reflected in the first behavior data of each of the plurality of users, and determines the measure for each of the plurality of users such that an evaluation function including the sum of travel times is minimized.
[0007] In the above-described device, both the first behavioral data of individual users and the second behavioral data of macro-level pedestrian flow are predicted based on the user's location information. By using the first and second behavioral data, which are highly correlated with each other, the accuracy of the corrected data for macro-level pedestrian flow that reflects the implemented measures can be improved. Since the evaluation function includes the sum of travel times obtained based on such corrected data, it is possible to accurately predict pedestrian flow after behavioral changes. Therefore, it is possible to determine measures to optimize pedestrian flow.
[0008] According to this disclosure, we can provide technology that determines measures to optimize pedestrian flow.
[0009] This figure shows the configuration of the policy optimization system according to the embodiment of this disclosure. This figure shows an example of micro-behavior prediction data. This figure shows an example of macro-behavior prediction data. This flowchart shows the processing performed by the policy optimization system according to the embodiment of this disclosure. This figure shows the hardware configuration of the policy optimization system according to the embodiment of this disclosure.
[0010] The embodiments of the policy optimization system related to this disclosure will be described in detail below, along with the drawings. In the description of the drawings, the same elements will be denoted by the same reference numeral, and redundant explanations will be omitted.
[0011] Figure 1 shows the configuration of an example policy optimization system. The policy optimization system 1 includes a policy optimization device 10 and user terminals 20 that are connected to the policy optimization device 10 via a communication network or the like. In Figure 1, only one user terminal 20 is shown, but in reality, multiple user terminals 20 are connected to the policy optimization device 10. Note that the policy optimization device 10 and the user terminals 20 may communicate via a database 30 such as an application server. That is, information from the user terminals 20 may be transmitted to the policy optimization device 10 via the database 30, or information acquired by the policy optimization device 10 may be stored in the database 30 and output to the user terminals 20 upon request.
[0012] The user terminal 20 is an information and communication terminal equipped with input / output devices and capable of acquiring location information via GPS (Global Positioning System) or the like. While various information and communication devices (smartphones, mobile phones, smartwatches, wearable devices, laptops, etc.) can be used as hardware, the user terminal 20 in this embodiment is assumed to be a mobile device carried by the user (smartphone, mobile phone, game console, smartwatch, wearable device, glasses-type device (e.g., smart glasses such as so-called MR glasses)). In one example, the user terminal 20 may have an application installed for accessing the database 30 according to this embodiment.
[0013] One example of a policy optimization device 10 is a device that optimizes policies to encourage behavioral changes for individual users in order to optimize traffic. Policies may be specific instructions, suggestions, etc., to change the user's behavior. In this embodiment, the change in a user's behavior in response to a policy is called behavioral change. One example of a policy optimization device 10 has a data acquisition unit 11, a micro-behavior prediction unit 12, a macro-behavior prediction unit 13, a policy decision unit 15, and a congestion verification unit 16.
[0014] The data acquisition unit 11 is a functional unit for acquiring various data necessary for the policy optimization device 10. One example of the data acquisition unit 11 acquires location information from each user's user terminal 20. The location information can be any information that can identify the location of the user terminal 20, such as GPS information, mobile base station connection information, or Wi-Fi (Wireless Fidelity) information. The acquired location information may be time-stamped and managed in a way that is linked to each user. The location information may be acquired at predetermined time intervals.
[0015] The data acquisition unit 11 acquires meteorological data. For example, meteorological data may be acquired from an external database, such as a database of a government agency or company that provides weather forecasts. Alternatively, meteorological data may be stored in database 30. The meteorological data includes past actual weather data and future predicted weather data (weather forecast data). This meteorological data may include date and time information, location information, and weather information. Weather information may include, for example, precipitation. Weather information may also include temperature, wind speed, atmospheric pressure, etc. Location information may be, for example, area information defined as an arbitrary area demarcated by a virtual boundary. Area information may be, for example, based on administrative divisions, or it may be a regional mesh in which an area on a map is divided into a grid.
[0016] The data acquisition unit 11 acquires the policy acceptance rate. The policy acceptance rate may be the percentage of users who are expected to actually implement the policies presented to them. The policy acceptance rate may also be calculated based on the user's past policy acceptance record. That is, the data acquisition unit 11 stores information on the policies presented to the user and whether or not those policies were implemented, and may calculate the policy acceptance rate based on this information. The policy acceptance rate for new users may be a predetermined value for each policy, or it may be the average for all users.
[0017] The data acquisition unit 11 may acquire information about the notification timing. The notification timing is information indicating the timing at which the policy optimization device 10 will notify the user of information regarding the policy. For example, the user can specify the notification timing through settings on an application installed on the user terminal 20. The notification timing specified on the user terminal 20 is acquired by the data acquisition unit 11. The user may set a specific time as the notification timing, or they may choose to be notified at the optimal timing determined by the policy optimization device 10.
[0018] The micro-behavior prediction unit 12 is a functional unit that acquires micro-behavior prediction data predicting the behavior of each user. Figure 2 shows an example of micro-behavior prediction data. One example of the micro-behavior prediction unit 12 predicts the user's future behavior based on the user's location information and weather data. For example, the micro-behavior prediction unit 12 may predict the user's behavior using a machine learning model that has been trained with each user's location information and weather data as explanatory variables and each user's behavior as the dependent variable. As a prediction method, machine learning models effective for time series data, such as LSTM (Long Short Term Memory) and Convolution LSTM, can be used. Information indicating the user's behavior includes information such as the means of transportation, departure date and time, departure point (starting point), and arrival point (ending point). The means of transportation may be any of the following selected from: car, train, bus, airplane, walking, etc. The means of transportation as training data may be estimated, for example, based on the user's travel speed and travel route derived from location information. The departure date and time, departure point, and arrival point may be determined based on location information. For example, micro-behavior prediction data is managed for each user, linked to a user ID that identifies the user.
[0019] The macro-behavior prediction unit 13 is a functional unit that acquires macro-behavior prediction data that predicts macro-level human flow, including users. In other words, the macro-behavior prediction unit 13 predicts the human flow of all people, including those other than users of the application related to the policy optimization device 10. Figure 3 is a diagram showing an example of macro-behavior prediction data. In one example, the macro-behavior prediction unit 13 predicts macro-level human flow based on user location information and weather data. For example, the macro-behavior prediction unit 13 may predict macro-level human flow (macro-behavior prediction data) using micro-behavior prediction data predicted based on user location information and weather data. In other words, the macro-behavior prediction unit 13 may acquire macro-level human flow by aggregating the micro-behavior prediction data for each user and performing an expanded estimation of the aggregated data. Information indicating macro-level human flow includes information such as means of transportation, departure date and time, departure place, arrival place, and number of people traveling. The information on means of transportation, departure date and time, departure place, and arrival place may have the same configuration as the micro-behavior prediction data.
[0020] The macro behavior prediction unit 13 may also predict macro-level pedestrian flow based on user location information and weather data. For example, the macro behavior prediction unit 13 may use a machine learning model trained with macro-level pedestrian flow as the target variable, using aggregated data (aggregated location information of each user, which is micro-level behavioral data) and weather data as explanatory variables. As a prediction method, machine learning models effective for time-series data, such as LSTM (Long Short Term Memory) and Convolution LSTM, can be used.
[0021] The congestion verification unit 16 is a functional unit that predicts travel time based on macro-behavior prediction data and predicts the occurrence of congestion from the predicted travel time. For example, the congestion verification unit 16 may have a machine learning model that uses macro-behavior prediction data as an explanatory variable and the average travel time from the origin to the destination (OD (Origin-Destination)) as the dependent variable. The congestion verification unit 16 may use such a machine learning model to obtain the average travel time for each OD and predict whether or not congestion will occur at each OD based on the length of the obtained average travel time. As described later, the congestion verification unit 16 may output the obtained average travel time for each OD to the policy decision unit 15.
[0022] The policy decision unit 15 is a functional unit that determines policies for each of multiple users. For example, the policy decision unit 15 proposes a policy to a user when the congestion verification unit 16 predicts that congestion will occur on the user's travel route, as predicted by the micro-behavior prediction unit 12. In addition, if a user has registered their planned activities on the user terminal 20, policies may be proposed to the user according to their planned activities. For example, the data acquisition unit 11 may acquire the user's planned activities, and the congestion verification unit 16 may predict that congestion will occur on the travel route estimated from the planned activities.
[0023] The policy decision unit 15 sets candidate policies for each of the multiple users. For example, the policies may include moving the departure time forward, moving the departure time backward, changing the route (suggesting a detour), or changing the mode of transportation. Changing the mode of transportation would be suggesting that a user who plans to use a car use the train instead. Policies that do not shorten travel time may be excluded from the candidates for each user. For example, if changing the mode of transportation from car to walking results in a longer travel time for the user than before the change, such a change in mode of transportation will be excluded from the candidate policies.
[0024] The policy decision unit 15 acquires corrected data. The corrected data is macro behavior prediction data corrected using reflected data, which is the micro behavior prediction data of multiple users to whom the policy has been applied. In one example, the policy decision unit 15 acquires macro behavior prediction data, micro behavior prediction data, and policy acceptance rate from the macro behavior prediction unit 13, micro behavior prediction unit 12, and data acquisition unit 11, respectively, and generates reflected data. The reflected data is generated according to the candidate policy set for the user. For example, if the candidate behavior change for a user is "shifting departure time forward by one hour," then micro behavior prediction data in which the departure time of the user's micro behavior prediction data has been shifted forward by one hour is generated as reflected data. The policy decision unit 15 corrects the macro behavior prediction data using the reflected data for each user and generates corrected data. For example, the policy decision unit 15 subtracts the micro behavior prediction data from the macro behavior prediction data, adds the reflected data to the macro behavior prediction data after the subtraction, and obtains corrected data. Thus, the corrected data can be said to be macro behavior prediction data in which the user's micro behavior prediction data has been replaced with reflected data.
[0025] In the embodiment of this disclosure, the policy decision unit 15 generates reflection data by referring to the user's policy acceptance rate. That is, the policy decision unit 15 reflects the policy in the micro-behavior prediction data according to the user's policy acceptance rate. For example, when the policy acceptance rate for a certain user is 40%, the policy decision unit 15 may determine whether or not the policy is accepted using random numbers or the like, and generate reflection data if it is determined that the policy is accepted. The policy decision unit 15 may also reflect the policy acceptance rate in the number of users. That is, when the policy acceptance rate for a certain user is 40%, the policy decision unit 15 may count the micro-behavior prediction data in which the policy is reflected as 0.4 people, and the micro-behavior prediction data in which the policy is not reflected as 0.6 people, and obtain these together as reflection data.
[0026] The policy decision unit 15 obtains the sum of travel times for macro-level human traffic based on the correction data. One example of the policy decision unit 15 determines policies for each of multiple users such that the evaluation function, which includes the sum of travel times, is minimized. For example, when the policy decision unit 15 obtains the sum of travel times for all ODs where congestion is predicted, it may use the evaluation function shown in the following formula: od ｉ The number of people is the number of people in the corrected macro behavior prediction data (corrected data) that uses the i-th OD where congestion is predicted, and od ｉ The average travel time is the average travel time predicted based on corrected macro-behavioral prediction data.
[0027] The policy decision unit 15 passes the correction data to the congestion verification unit 16, receives the average travel time corresponding to the correction data from the congestion verification unit 16, and obtains the total travel time of macro-level pedestrian flow corresponding to the correction data using the above formula. The policy decision unit 15 searches for the combination of user and policy that minimizes the total travel time by repeatedly changing the candidate policy for each user and obtaining the corresponding total travel time of the correction data. In other words, the policy decision unit 15 optimizes the combination of user and policy. Algorithms such as grid search and Bayesian optimization may be used for the search.
[0028] The policy decision unit 15, based on the search results, determines the combination of user and policy that minimizes the total travel time. The content of the policy determined for each user is output to each user terminal 20 according to the notification timing specified by the user.
[0029] In addition, in other embodiments of this disclosure, the evaluation function may include the total travel time and the sum of the incentive amounts given for each type of measure, and the measure determination unit 15 may determine the incentive amounts for each type of measure so as to minimize the evaluation function. The incentive amounts may be points with monetary value, points usable within the service related to the measure optimization device 10, etc.
[0030] In this example, the policy decision unit 15 generates reflected data by referring to the user's policy acceptance rate and the incentive amount for each policy. That is, the policy decision unit 15 reflects the policies in the micro-behavior prediction data according to the user's policy acceptance rate and incentive amount. For example, compared to a case where the incentive amount for a certain policy is 100 points, a policy with an incentive amount of 500 points is considered to be more likely to be accepted by users. In other words, the incentive amount can be considered a factor that changes the user's policy acceptance rate. In this case, the incentive amount can be replaced with an added value that is added to the policy acceptance rate. In one example, the relationship between the incentive amount and the added value may be predetermined or may be obtained for each user.
[0031] For example, suppose the incentive amounts are 100 points, 300 points, and 500 points, and the added values are 10%, 15%, and 20%. Also, suppose a user's policy acceptance rate is 40% and the incentive amount is 300 points. In this case, the policy decision unit 15 generates reflected data assuming that the policy acceptance rate for that user is 55%. Then, the policy decision unit 15 calculates the value of an evaluation function that includes the sum of travel time and the sum of incentive amounts based on the correction data generated using this reflected data. The policy decision unit 15 searches for the combination of policy and incentive amount that minimizes the evaluation function by repeatedly changing the policy and incentive amount for each user and obtaining the corresponding total travel time and total incentive amount from the correction data. Note that there is an upper limit on the sum of incentive amounts, which is a constraint condition in the search.
[0032] The policy decision unit 15, based on the search results, determines the combination of policy and incentive amount that minimizes the evaluation function. The determined combination of policy and incentive amount is output to the user terminal 20 according to the notification timing specified by the user. In this case, each user terminal may be presented with the same information in the form of a list of multiple policies and their corresponding incentive amounts. The user will then consider which of the presented policies to implement or not implement, taking the incentive amount into consideration.
[0033] Figure 4 is a flowchart showing an example of the operation of the policy optimization device. As shown in Figure 4, in the policy optimization device 10, first, the data acquisition unit 11 acquires various data such as location information, weather data, policy acceptance rate, and notification timing from the user terminal 20 (step S1). Next, the micro-behavior prediction unit 12 generates micro-behavior prediction data for each user based on the acquired location information and weather data (step S2). The micro-behavior prediction data includes information such as the predicted means of transportation, departure date and time, departure point, and arrival point for each user. Next, the macro-behavior prediction unit 13 aggregates the micro-behavior prediction data and predicts macro-level human flow (step S3). This generates macro-behavior prediction data. The macro-behavior prediction data includes information such as the means of transportation, departure date and time, departure point, arrival point, and number of travelers for each OD. Next, the congestion verification unit 16 predicts travel time based on the macro-behavior prediction data and predicts the occurrence of congestion (step S4). The congestion verification unit 16 uses a machine learning model to acquire the average travel time for each OD and predicts whether or not congestion will occur. Next, the policy decision unit 15 determines policies for multiple users (step S5). The policy decision unit 15 reflects the user's policies in the micro-behavior prediction data and obtains reflected data, and uses the obtained reflected data to obtain correction data from the macro-behavior prediction data. Then, based on the correction data, the policy decision unit obtains the sum of the travel times for each OD and determines the policies for each user so that the value of the evaluation function including the total travel time is minimized. The policy decision unit 15 notifies the user terminal 20 of the optimized policies.
[0034] As described above, one example of a policy optimization device 10 includes: a micro-behavior prediction unit 12 (first behavior prediction unit) that acquires micro-behavior prediction data (first behavior data) that predicts the behavior of each of multiple users based on location information; a macro-behavior prediction unit 13 (second behavior prediction unit) that acquires macro-behavior prediction data (second behavior data) that predicts macro-level human flow including multiple users based on location information; and a policy decision unit 15 that sets policies for each of multiple users, acquires the sum of travel times of macro-level human flow based on corrected data which is macro-level behavior prediction data corrected using reflected data in which the behavior change policies are reflected in the micro-behavior prediction data of each of the multiple users, and determines behavior change policies for each of the multiple users so as to minimize the evaluation function which includes the sum of travel times.
[0035] In the above-described device, both micro-behavioral prediction data for individual users and macro-behavioral prediction data for macro-level pedestrian flow are predicted based on the user's location information. By using highly correlated micro-behavioral prediction data and macro-behavioral prediction data, the accuracy of the corrected macro-level pedestrian flow data, which reflects behavioral change measures, can be improved. Since the evaluation function includes the sum of travel times obtained based on such corrected data, it is possible to accurately predict pedestrian flow after behavioral change. Therefore, it is possible to determine behavioral change measures that optimize pedestrian flow.
[0036] The policy decision unit 15 outputs the content of the decided behavioral change measures to the user terminals 20 (terminal devices) of multiple users. By outputting the content of the behavioral change measures to the user terminals 20, the behavioral change measures presented to the users can be implemented by some users. This can optimize the flow of people so that travel time is reduced.
[0037] The policy decision unit 15 may calculate the sum of travel times by referring to the acceptance rate of behavioral change measures for each of the multiple users. By referring to the acceptance rate of behavioral change measures, the sum of travel times can be calculated with greater accuracy. Since the acceptance rate is referred to in the process of optimizing the measures, it can contribute sufficiently to optimizing transportation even if the implementation of the measures is left to the free will of the users.
[0038] The micro-behavior prediction unit 12 may construct a learning model that predicts behavior based on the past behavioral data and past weather data of multiple users, and acquire micro-behavior prediction data using the learning model. By using such a learning model, changes in human flow due to weather effects can be efficiently reflected in behavior predictions.
[0039] The policy decision unit 15 may subtract micro-behavior prediction data from macro-behavior prediction data, add reflection data to the macro-behavior prediction data from which the micro-behavior prediction data has been subtracted, and obtain corrected data. Because highly related micro-behavior prediction data and macro-behavior prediction data are used, appropriate corrected data is obtained by replacing the behavior prediction data before and after the user's behavior change within the macro-behavior prediction data.
[0040] The policy optimization device 10 may include a congestion verification unit 16 (verification unit), which is a learning model that takes macro behavior prediction data as input data and outputs travel time for macro-level human flow. The policy decision unit 15 outputs correction data to the congestion verification unit 16, and the congestion verification unit 16 inputs the correction data into the learning model and obtains the sum of the travel times obtained. Since the travel time is output by the learning model, highly accurate output results can be obtained in a short time.
[0041] The evaluation function may include the sum of travel times and the sum of the incentive amounts given for each type of behavioral change measure. The measure decision unit 15 may determine the incentive amount for each type of behavioral change measure so as to minimize the evaluation function. By setting the optimal incentive amount, users can be encouraged to take behavioral change measures, and pedestrian flow can be predicted with higher accuracy.
[0042] The apparatus of the present disclosure has the following configuration: [1] an apparatus comprising: a first behavior prediction unit that acquires first behavior data predicting the behavior of each of a plurality of users based on location information; a second behavior prediction unit that acquires second behavior data predicting macro-level human flow including the plurality of users based on the location information; and a policy determination unit that sets a policy for each of the plurality of users, acquires the sum of travel times of the macro-level human flow based on corrected data which is the second behavior data corrected using reflected data in which the policy is reflected in the first behavior data of each of the plurality of users, and determines the policy for each of the plurality of users such that an evaluation function including the sum of travel times is minimized. [2] the apparatus according to [1], wherein the policy determination unit outputs the content of the determined policy to each of the plurality of users' terminal devices. [3] the apparatus according to [1] or [2], wherein the policy determination unit calculates the sum of travel times by referring to the acceptance rate of the policy for each of the plurality of users. [4] The apparatus according to any one of [1] to [3], wherein the first behavior prediction unit constructs a learning model that predicts the behavior based on the past behavior data and past weather data of each of the plurality of users, and acquires the first behavior data using the learning model. [5] The apparatus according to any one of [1] to [4], wherein the policy decision unit subtracts the first behavior data from the second behavior data, adds the reflection data to the second behavior data from which the first behavior data has been subtracted, and acquires the correction data. [6] The apparatus according to any one of [1] to [5], further comprising a verification unit which is a learning model that outputs the travel time of the macro human flow using the second behavior data as input data, wherein the policy decision unit outputs the correction data to the verification unit, and the verification unit inputs the correction data to the learning model and acquires the sum of the travel times obtained. [7] The apparatus according to any one of [1] to [6], wherein the evaluation function includes the sum of the travel times and the sum of the incentive amounts given for each type of policy, and the policy decision unit determines the incentive amount for each type of policy so that the evaluation function is minimized.[8] A method comprising: a step of acquiring first behavior data obtained by predicting each behavior of a plurality of users based on position information; a step of acquiring second behavior data obtained by predicting macro pedestrian flow including the plurality of users based on the position information; a step of setting a policy for each of the plurality of users; a step of acquiring a total travel time of the macro pedestrian flow based on correction data that is the second behavior data corrected using the first behavior data of each of the plurality of users in which the policy is reflected; and a step of determining the policy for each of the plurality of users such that an evaluation function including the total travel time is minimized.
[0043] Note that, the block diagrams used in the description of the above embodiments show blocks in units of functions. These functional blocks (components) are realized by any combination of at least one of hardware and software. In addition, the method for implementing each functional block is not particularly limited. That is, each functional block may be implemented using one physically or logically coupled device, or may be implemented using two or more physically or logically separated devices connected directly or indirectly (for example, using a wired connection, a wireless connection, or the like) and using the plurality of devices. Functional blocks may be implemented by combining software with the one device or the plurality of devices.
[0044] Functions include, but are not limited to, judgment, determination, decision, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, picking, establishment, comparison, assumption, expectation, assumption, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assigning. For example, a functional block (component) that enables transmission is referred to as a transmitting unit or a transmitter. In any case, as described above, the implementation method is not particularly limited.
[0045] For example, the measure optimization apparatus 10 according to an embodiment of the present disclosure may function as a computer that executes information processing according to the present disclosure. FIG. 5 is a diagram illustrating an example of a hardware configuration of the measure optimization apparatus 10 according to an embodiment of the present disclosure. The above-described measure optimization apparatus 10 may be physically configured as a computer apparatus including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like.
[0046] Note that in the following description, the term "apparatus" can be read as a circuit, a device, a unit, or the like. The hardware configuration of the measure optimization apparatus 10 may be configured to include one or more of each apparatus illustrated in the drawing, or may be configured to exclude some apparatuses.
[0047] Each function in the measure optimization apparatus 10 is implemented by loading predetermined software (program) onto hardware such as the processor 1001 and the memory 1002, causing the processor 1001 to perform arithmetic operations, controlling communication by the communication device 1004, and controlling at least one of reading and writing of data in the memory 1002 and the storage 1003.
[0048] For example, the processor 1001 operates an operating system to control the entire computer. The processor 1001 may be configured by a central processing unit (CPU) including an interface with peripheral devices, a control unit, an arithmetic unit, a register, and the like. For example, each function in the above-described measure optimization apparatus 10 may be implemented by the processor 1001.
[0049] Furthermore, the processor 1001 reads programs (program code), software modules, data, etc., from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes accordingly. The program used is one that causes the computer to execute at least a part of the operations described in the above embodiment. For example, each function in the policy optimization device 10 may be implemented by a control program stored in the memory 1002 and operated on the processor 1001. Although the above-described processes have been explained as being executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The program may also be transmitted from a network via a telecommunications line.
[0050] The memory 1002 is a computer-readable recording medium and may consist of at least one of the following: ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), RAM (Random Access Memory), etc. The memory 1002 may also be called a register, cache, main memory, etc. The memory 1002 can store executable programs (program code), software modules, etc., for carrying out information processing according to one embodiment of the present disclosure.
[0051] The storage 1003 is a computer-readable recording medium and may consist of at least one of the following: an optical disc such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital multipurpose disk, a Blu-ray® disc), a smart card, flash memory (e.g., a card, a stick, a key drive), a floppy® disk, a magnetic strip, etc. The storage 1003 may also be called an auxiliary storage device. The storage medium provided by the policy optimization device 10 may be, for example, a database, server, or other suitable medium including at least one of the memory 1002 and the storage 1003.
[0052] The communication device 1004 is hardware (transceiver / receiver device) for communicating between computers via at least one of a wired network and a wireless network, and is also referred to as a network device, network controller, network card, communication module, etc.
[0053] The input device 1005 is an input device that accepts input from an external source (e.g., a keyboard, mouse, microphone, switch, button, sensor, etc.). The output device 1006 is an output device that outputs to an external source (e.g., a display, speaker, LED lamp, etc.). The input device 1005 and the output device 1006 may be configured as an integrated unit (e.g., a touch panel).
[0054] Furthermore, each device, such as the processor 1001 and memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or different buses may be configured for each device.
[0055] Furthermore, the policy optimization device 10 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), and an FPGA (Field Programmable Gate Array), and some or all of each functional block may be realized by such hardware. For example, the processor 1001 may be implemented using at least one of these hardware components.
[0056] The processing procedures, sequences, flowcharts, etc., of each aspect / embodiment described in this disclosure may be reordered, provided they do not contradict each other. For example, the methods described in this disclosure present various step elements using exemplary order and are not limited to the specific order presented.
[0057] Input and output information may be stored in a specific location (e.g., memory) or managed using a management table. Input and output information may be overwritten, updated, or appended to. Output information may be deleted. Input information may be transmitted to other devices.
[0058] The determination may be made by a value represented by one bit (0 or 1), by a boolean value (true or false), or by a numerical comparison (for example, a comparison with a predetermined value).
[0059] Each aspect / embodiment described in this disclosure may be used individually, in combination, or switched between as needed during implementation. Furthermore, notification of specific information (e.g., notification that "X is") is not limited to explicit notification, but may also be implicit (e.g., by not providing such notification).
[0060] Although the present disclosure has been described in detail above, it will be clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the intent and scope of the present disclosure as defined by the claims. Therefore, the descriptions in the present disclosure are illustrative and not intended to be restrictive in any way.
[0061] Software should be broadly interpreted to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, execution threads, procedures, functions, and so on, whether they are called software, firmware, middleware, microcode, hardware description languages, or by any other name.
[0062] Furthermore, software, instructions, information, etc., may be transmitted and received via a transmission medium. For example, if software is transmitted from a website, server, or other remote source using at least one of wired technology (such as coaxial cable, fiber optic cable, twisted pair, or digital subscriber line (DSL)) and wireless technology (such as infrared or microwave), then at least one of these wired and wireless technologies is included in the definition of a transmission medium.
[0063] The terms “system” and “network” as used in this disclosure are interchangeable.
[0064] Furthermore, the information, parameters, etc., described in this disclosure may be expressed using absolute values, relative values from a predetermined value, or corresponding other information.
[0065] As used in this disclosure, the terms “determining” and “determining” may encompass a wide variety of actions. “Determining” may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, or inquiring (e.g., searching in a table, database, or other data structure), or ascertaining. “Determining” may also include receiving (e.g., receiving information), transmitting (e.g., sending information), inputting, outputting, or accessing (e.g., accessing data in memory). Furthermore, "judgment" and "decision" can include considering something as having been "judged" or "decided" after resolving, selecting, choosing, establishing, comparing, etc. In other words, "judgment" and "decision" can include considering something as having been "judged" or "decided" after some action. Also, "judgment (decision)" can be reinterpreted as "assuming," "expecting," or "considering."
[0066] The terms “connected,” “coupled,” or any variation thereof, mean any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are “connected” or “coupled” with each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, “connection” may be reinterpreted as “access.” As used in this disclosure, two elements may be considered to be “connected” or “coupled” with each other using at least one of one or more wires, cables, and printed electrical connections, and, in some non-limiting and non-exclusive examples, electromagnetic energy having wavelengths in the radio frequency domain, microwave domain, and optical (both visible and invisible) domain.
[0067] In this disclosure, the phrase "based on" does not mean "based solely on" unless otherwise specified. In other words, the phrase "based on" means both "based solely on" and "based at least on."
[0068] Any reference to elements using the designations “first,” “second,” etc., as used in this disclosure does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient way to distinguish between two or more elements. Accordingly, references to the first and second elements do not imply that only two elements may be employed, or that the first element must precede the second element in any way.
[0069] Where the terms “include,” “including,” and variations thereof are used in this disclosure, these terms are intended to be inclusive, as is the term “comprising.” Furthermore, the term “or” as used in this disclosure is not intended to mean exclusive OR.
[0070] In this disclosure, if articles are added through translation, such as a, an, and the in English, this disclosure may include the fact that the noun following these articles is plural.
[0071] In this disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "combine" may be interpreted similarly to "different."
[0072] 10...Policy optimization device, 11...Data acquisition unit, 12...Micro-behavior prediction unit, 13...Macro-behavior prediction unit, 15...Policy decision unit, 16...Traffic congestion verification unit, 1001...Processor, 1002...Memory, 1003...Storage, 1004...Communication device, 1005...Input device, 1006...Output device, 1007...Bus.
Claims
1. A device comprising: a first behavior prediction unit that acquires first behavior data predicting the actions of multiple users based on location information; a second behavior prediction unit that acquires second behavior data predicting macro-level human flow including the multiple users based on the location information; and a policy determination unit that sets a policy for each of the multiple users, acquires the sum of travel times of the macro-level human flow based on corrected data which is the second behavior data corrected using reflected data in which the policy is reflected in the first behavior data of each of the multiple users, and determines the policy for each of the multiple users such that an evaluation function including the sum of travel times is minimized.
2. The apparatus according to claim 1, wherein the policy decision unit outputs the content of the decided policy to each of the terminal devices of the multiple users.
3. The apparatus according to claim 1, wherein the policy decision unit calculates the sum of the travel times by referring to the acceptance rate of the policy for each of the plurality of users.
4. The apparatus according to claim 1, wherein the first behavior prediction unit constructs a learning model that predicts the behavior based on the past behavior data and past weather data of each of the plurality of users, and acquires the first behavior data using the learning model.
5. The apparatus according to claim 1, wherein the policy decision unit subtracts the first action data from the second action data, adds the reflection data to the second action data from which the first action data has been subtracted, and obtains the correction data.
6. The apparatus according to claim 1, further comprising a verification unit which is a learning model that outputs the travel time of the macro-level human flow using the second behavioral data as input data, wherein the policy decision unit outputs the correction data to the verification unit, and the verification unit inputs the correction data to the learning model to obtain the sum of the travel times obtained.
7. The apparatus according to claim 1, wherein the evaluation function includes the sum of the travel times and the sum of the incentive amounts given for each type of measure, and the measure determination unit determines the incentive amounts for each type of measure such that the evaluation function is minimized.
8. A method comprising: acquiring first behavioral data predicting the actions of multiple users based on location information; acquiring second behavioral data predicting macro-level human flow including the multiple users based on location information; setting measures for each of the multiple users; acquiring the sum of travel times of the macro-level human flow based on corrected data which is the second behavioral data corrected using the first behavioral data of each of the multiple users to which the measures have been applied; and determining the measures for each of the multiple users such that an evaluation function including the sum of travel times is minimized.