JP7033515B2 - Traffic condition prediction device - Google Patents

Description

The present invention relates to a traffic condition prediction device.

As used herein, "traffic capacity" is the structural capacity of an individual road, which is how many vehicles can pass through it within a given period of time. As used herein, "traffic volume" is the number of vehicles that have actually passed a certain road within a certain period of time, as measured by a road sensor such as a camera or a detector installed on the road. As used herein, "traffic demand" is the number of vehicles attempting to pass a road within a certain period of time.

There is known a technology that acquires traffic data such as traffic volume and required time and predicts future traffic conditions (traffic demand, transit time, occurrence of traffic congestion, etc.) at a desired point based on the acquired traffic data. There is. As a method of predicting traffic conditions, there is a method of predicting future traffic conditions by statistical processing methods such as pattern matching and machine learning from accumulated past traffic data and current traffic data. In the technique described in Patent Document 1, the characteristics of fluctuations in traffic volume at a specific date and time and place are statistically analyzed from the accumulated past traffic data, and the characteristics obtained and the current traffic volume are used as self-regressive models. The regression model predicts future traffic volume. Further, in the technique described in Patent Document 2, a traffic congestion prediction model is created based on past traffic congestion data, and the current traffic data is input to the traffic congestion prediction model to predict traffic congestion.

Japanese Unexamined Patent Publication No. 11-160454 Japanese Unexamined Patent Publication No. 2004-272408

If traffic demand exceeds the traffic capacity on a certain road at a certain time, vehicles will stay and traffic congestion will occur, and the traffic volume and traffic demand will not match (traffic volume will be less than traffic demand). It is not possible to grasp the traffic demand from the traffic volume. For example, if a road with a traffic capacity of 2,000 [vehicles / hour] (2,000 vehicles can pass in one hour) has a traffic demand of 3,000 [vehicles / hour], the traffic volume will be It is 2,000 [units / hour] and does not match the traffic demand. Traffic data such as traffic volume measured by road sensors etc. is data about vehicles that actually passed through the road, so if the traffic demand exceeds the traffic capacity and the upper limit of the traffic volume is limited by the traffic capacity, which I can't figure out how much traffic demand was in the time zone. Vehicles that have stagnated due to traffic demand exceeding the traffic capacity will pass through the road after the time when they originally intended to pass, which will affect future traffic volume, required time, and continuation of traffic congestion. Furthermore, if traffic demand that exceeds the traffic capacity continues during that period, the impact will continue thereafter. For this reason, when predicting future traffic volume, required time, and traffic congestion only from traffic data, there is a risk that the prediction error will be large, especially on roads where there is a lot of traffic and traffic demand that exceeds the traffic capacity frequently occurs.

For example, in the method described in Patent Document 1, the future traffic volume is predicted based on the traffic volume measured by the road sensor, but it is limited to the prediction of the traffic volume, and when the traffic demand exceeds the traffic capacity. Traffic demand has not been predicted. In addition, since there are many factors that affect traffic demand, required time, and traffic congestion, there is a concern that the forecast results may differ from the actual ones due to unexpected factors in the conventional forecasting method. For example, in the method described in Patent Document 2, a traffic jam prediction model is created using data of a vehicle that is already in a running state as past traffic data, and the vehicle is not in the running state at that time but is in the running state thereafter. Vehicles that affect traffic on the forecasted road are not considered. For this reason, when predicting traffic demand, required time, and traffic congestion only from past traffic data, the longer the time from the current time to the predicted time, the more the vehicle is not in the current running state but will be in the running state at the predicted time. As a result, the prediction error may increase.

On the other hand, if these prediction models are constructed in consideration of various factors affecting traffic demand, required time, and traffic congestion, prediction errors can be reduced and prediction accuracy can be improved. However, in this case, the amount of computation required to build the prediction model may increase dramatically.

The present invention has been made in view of the above circumstances, and an object of the present invention is to improve the prediction accuracy of future traffic demand and to suppress an increase in the amount of calculation required for constructing a prediction model.

The traffic condition prediction device according to one aspect of the present invention is a population data acquisition unit that acquires population data of a plurality of groups classified according to the position and attribute of a person, and the number of vehicles that have passed the prediction target road. The traffic volume data acquisition unit for each time zone that acquires the traffic volume data for each time zone that shows a certain traffic volume for each time zone, and the traffic volume for each time zone included in the traffic volume data for each time zone are in the corresponding time zone. Use to determine whether the traffic demand matches the traffic demand indicating the number of vehicles trying to pass the forecast target road, and to extract only the traffic volume in the matching time zone as the traffic volume data for learning when constructing the forecast model. Based on the time zone extraction unit, learning traffic volume data, and population data for each group, the traffic demand forecast model construction unit that builds a traffic demand forecast model that predicts traffic demand using population data as explanatory variables, and the population for each group. It is equipped with a traffic demand prediction unit that predicts the traffic demand of the road to be predicted based on the data and the traffic demand prediction model.

In the traffic condition forecasting device according to one aspect of the present invention, a traffic demand forecasting model is constructed by, for example, machine learning, based on traffic volume data and population data. By using population data, it is possible to construct a traffic demand forecast model in consideration of various factors of increase / decrease in traffic demand. For example, population data correlates with the number of vehicles that are not currently in the driving state but will be in the driving state at the predicted time. By acquiring the factors of increase / decrease in traffic demand that are difficult to obtain from traffic volume data from population data and constructing a traffic demand forecast model, the traffic demand forecast model can be used with high accuracy. You can make predictions. Here, for example, when a traffic demand exceeding the traffic capacity occurs, the vehicle stays and traffic congestion occurs, and the traffic volume becomes less than the traffic demand. Therefore, it is necessary to accurately grasp the traffic demand from the traffic volume. I can't. Therefore, if a traffic demand forecast model is simply constructed from traffic volume data, it may not be possible to forecast traffic demand with high accuracy. In this respect, in the traffic condition prediction device according to one aspect of the present invention, the traffic volume and the traffic demand are compared, and only the traffic volume in the time zone in which both are the same is used for constructing the prediction model. That is, for example, if there is a time zone in which the traffic volume and the traffic demand do not match due to a traffic demand exceeding the traffic requirement on the forecast target road, the traffic volume data in the non-matching time zone is excluded. Therefore, a traffic demand forecasting model that more accurately reflects the relationship between population and traffic demand is constructed. By using the traffic demand forecasting model, it becomes possible to predict the traffic demand exceeding the traffic capacity with high accuracy by the extrapolation method based on the population. Further, since the traffic demand forecast model is constructed by excluding some traffic volume data, the amount of calculation required for constructing the traffic demand forecast model can be suppressed. From the above, according to the traffic condition forecasting device according to one aspect of the present invention, it is possible to improve the forecasting accuracy of traffic demand and suppress the increase in the amount of calculation required for constructing the traffic demand forecasting model.

The traffic demand forecast model construction unit may construct a traffic demand forecast model by selecting population data of a group having a high contribution to traffic demand on the forecasted road. As a result, it is possible to secure the accuracy of the traffic demand forecast, limit the population data used for constructing the traffic volume demand forecast model, and suppress an increase in the amount of calculation required for constructing the traffic volume demand forecast model.

The above-mentioned traffic condition forecasting device has a required time data acquisition unit that acquires required time data for each time zone that indicates the required time of a vehicle that has passed the forecast target road for each time zone, and a traffic demand forecast result by the traffic demand forecasting unit. And, based on the required time data for each time zone, the required time forecast model building unit that builds the required time forecast model that predicts the required time for each time zone using the forecast result of traffic demand as an explanatory variable, and the traffic demand forecasting unit predict Time required prediction unit that predicts the required time for each time zone when passing the forecast target road based on the forecast traffic demand forecast result used to predict the required time for each time zone and the required time forecast model. And may be further provided. For example, a vehicle that has stagnated due to a traffic demand exceeding the traffic capacity in a certain period will pass through the road in a time zone after the time zone in which the vehicle was originally intended to pass. In this case, the required time of the vehicle may not be properly predicted. In this respect, in the traffic condition prediction device according to one aspect of the present invention, the required time prediction model is constructed based on the required time for each time zone and the forecast result of the traffic demand, so that the stagnant vehicle as described above has an effect. The required time can be predicted in consideration of the effect. That is, according to the traffic condition prediction device according to one aspect of the present invention, the accuracy of the required time prediction can be improved.

The above-mentioned traffic condition prediction device may further include a traffic congestion prediction unit that predicts traffic congestion based on the prediction result of the required time for each time zone by the required time prediction unit. By predicting traffic congestion based on the highly accurate required time prediction result described above, the accuracy of traffic congestion prediction can be improved.

The above-mentioned traffic condition prediction device has a speed data acquisition unit that acquires speed data for each time zone that indicates the speed of a vehicle that has passed the prediction target road for each time zone, and a time for the status of traffic accidents and traffic regulations on the prediction target road. The traffic abnormality data acquisition unit that acquires the traffic abnormality data for each time zone shown for each zone is further provided, and the usage time zone extraction unit considers at least one of the speed data for each time zone and the traffic abnormality data for each time zone. Then, it may be determined whether the traffic volume of each time zone included in the traffic volume data for each time zone matches the traffic demand of the corresponding time zone. When traffic demand exceeds the traffic capacity, it is considered that the speed of the vehicle will decrease due to congestion. In addition, when a traffic accident or traffic regulation occurs, it is considered that the traffic volume changes irregularly. In this way, the speed data and traffic abnormality data for each time zone are closely related to the traffic volume and the situation of the traffic demand. Therefore, when determining whether or not the traffic volume matches the traffic demand, the time zone is used. By considering the speed data and the traffic abnormality data, it is possible to determine the coincidence between the traffic volume and the traffic demand with high accuracy.

According to the present invention, it is possible to improve the prediction accuracy of future traffic demand and suppress an increase in the amount of calculation required for constructing a prediction model.

It is a figure for demonstrating the function of the traffic condition prediction apparatus which concerns on one Embodiment. It is a figure which shows the composition example of the population data. It is a figure which shows the example of the traffic congestion prediction result. It is a flowchart which shows the process executed by the traffic condition forecasting apparatus which concerns on one Embodiment, and more specifically, is a flowchart which shows the construction process of the traffic demand forecast model. It is a flowchart which shows the process executed by the traffic condition forecasting apparatus which concerns on one Embodiment, and more specifically, is a flowchart which shows the traffic demand forecast process. It is a flowchart which shows the process executed by the traffic condition prediction apparatus which concerns on one Embodiment, and more specifically, is a flowchart which shows the construction process of the required time prediction model. It is a flowchart which shows the process executed by the traffic condition prediction apparatus which concerns on one Embodiment, and more specifically, is a flowchart which shows the required time prediction process and the traffic congestion prediction process. It is a figure which shows the hardware configuration of the traffic condition prediction apparatus which concerns on one Embodiment.

Hereinafter, embodiments of the traffic condition prediction device according to the present invention will be described in detail together with the drawings. In the description of the drawings, the same elements are designated by the same reference numerals, and duplicate description will be omitted.

FIG. 1 is a diagram for explaining the function of the traffic condition prediction device 100 according to the present embodiment. The traffic condition forecasting device 100 creates (constructs) a traffic demand forecasting model used for forecasting traffic demand based on traffic volume data and population data for learning. The traffic condition forecasting device 100 uses the created traffic demand forecasting model to predict traffic demand by inputting population data.

Further, the traffic condition prediction device 100 creates a required time prediction model used for predicting the required time based on the data accumulated for learning the predicted traffic demand and the required time data. The traffic condition prediction device 100 predicts the required time based on the created required time prediction model and the traffic demand predicted by inputting the population data. Then, the traffic condition prediction device 100 predicts the traffic congestion from the required time further predicted from the traffic demand predicted by inputting the population data.

As shown in FIG. 1, the traffic condition prediction device 100 includes a learning population data acquisition unit 111, a learning population data storage unit 112, a traffic volume data acquisition unit 113, a traffic volume data storage unit 114, and a speed. Data acquisition unit 128, speed data storage unit 129, traffic abnormality data acquisition unit 130, traffic abnormality data storage unit 131, usage time zone extraction unit 115, learning traffic volume data storage unit 116, and traffic demand forecast. The model construction unit 117, the traffic demand forecast model storage unit 118, the forecasting population data acquisition unit 110, the traffic demand prediction unit 119, the learning traffic demand prediction result storage unit 120, the required time data acquisition unit 121, and the required time data acquisition unit 121. The required time data storage unit 122, the required time prediction model construction unit 123, the required time prediction model storage unit 124, the traffic demand prediction result acquisition unit 125 for prediction, the required time prediction unit 126, and the traffic congestion prediction unit 127. , Equipped with.

The learning population data acquisition unit 111 acquires population data of a plurality of groups classified according to the position and attribute of a person. The learning population data acquisition unit 111 acquires learning population data (learning population data) used for constructing a traffic demand forecast model from outside the traffic condition prediction device 100. The learning population data acquisition unit 111 acquires the learning population data sampled within a predetermined sampling period as the original data used for the congestion prediction. The learning population data is, for example, the past population data of the whole country recorded in an external database, and is divided into a plurality of groups classified according to the sampling time, the position of the person, and the attribute of the person. In the present embodiment, the learning population data acquisition unit 111 acquires only the learning population data sampled at noon every day. The learning population data acquisition unit 111 inputs learning population data to the learning population data storage unit 112. The learning population data storage unit 112 stores (stores) the learning population data.

Population data is data showing how many people existed in a certain area at a certain time. In this embodiment, the population data is of a plurality of groups classified according to the sampling time, the position of the person (eg, regional mesh, etc.), and the attributes of the person (eg, gender, age, place of residence, etc.). It shows the population belonging to each. The sampling time is the time when the data was sampled. The position and attribute of a person are the position and attribute of a person to be sampled (sampling target person) of data. For example, the population data is generated from the location information that is periodically sampled in advance of the mobile terminal and the registration information about the attributes of the user of the mobile terminal, which are stored in an external database or the like. In the present embodiment, as shown in FIG. 2, the population data is composed of the items of "time", "location identification number", "gender", "age", "residential place", and "population". ..

"Time" indicates the date and time when the data was sampled. The “position identification number” indicates a number that identifies the position of the sampling target person. For example, "JIS X 0410 area mesh code" is used as the "position identification number". For the "position identification number", a regional mesh code such as a secondary mesh, a tertiary mesh, or a half regional mesh may be used, or a prefecture code or a city / ward / town / village code of the place of residence may be used. You may. In this embodiment, since the half area mesh is used as the "position identification number", the position identification numbers are classified into about 1.5 million types.

“Gender” indicates the gender of the sampling target. "Age" indicates the age group of the person to be sampled. In the present embodiment, the "age" is 20 years old or younger, 21-25 years old, 26-30 years old, 31-35 years old, 36-40 years old, 41-45 years old, 46-50 years old, 51-55 years old, 56 years old. It is classified into 13 types: 60 years old, 61-65 years old, 66-70 years old, 71-75 years old, and 76 years old or older. "Residence" indicates the residence of the person to be sampled. For the "place of residence", a regional mesh code such as a secondary mesh, a tertiary mesh, or a half regional mesh may be used, or a prefecture code or a city / ward / town / village code of the place of residence may be used. May be good. In this embodiment, as the "residential place", a city / ward / town / village code classified into about 2000 pieces is used.

In the present embodiment, the population data is divided into a plurality of groups according to the "position identification number" corresponding to the position of the person and the "gender", "age", and "place of residence" corresponding to the attribute of the person. being classified. Therefore, the population data can be classified into groups of about 1.5 million x 2 x 13 x about 2000. That is, the space in which the combination of the position and the attribute of a person is a variable theoretically extends to about 80 billion dimensions.

The traffic volume data acquisition unit 113 acquires traffic volume data for each time zone, which indicates the traffic volume, which is the number of vehicles passing through the forecast target road, for each time zone. The traffic volume data acquisition unit 113 acquires traffic volume data from the outside of the traffic condition prediction device 100. The traffic volume data acquisition unit 113 acquires the traffic volume data at the traffic demand forecast position sampled within a predetermined sampling period as the original data used for the traffic demand forecast. The traffic volume data indicates, for example, the traffic volume according to the traffic demand forecast position for predicting the traffic demand among the past traffic volume data recorded in the external database. The traffic volume data acquisition unit 113 inputs traffic volume data to the traffic volume data storage unit 114. The traffic volume data storage unit 114 stores (stores) traffic volume data.

The traffic volume data is data showing the number of vehicles passing by (traffic volume) according to the position. For example, using a vehicle detector installed on the road, the traffic volume information is measured at regular sampling periods. It is stored in advance in an external database to be managed. For example, the traffic volume data is composed of items of "time" and "traffic volume" for each section (so-called link) in which a road (for example, a major road in the whole country) is divided in advance. In the present embodiment, the above-mentioned pre-divided section (position) is the target of prediction (road to be predicted). "Time" indicates the date and time at the start of the sampling period. "Traffic volume" indicates the traffic volume at the target position during the sampling period. In this embodiment, the traffic volume data is sampled at 10 minute intervals.

The speed data acquisition unit 128 acquires speed data for each time zone, which indicates the speed of the vehicle passing through the prediction target road for each time zone. The speed data acquisition unit 128 acquires speed data used for extracting learning traffic volume data from the outside of the traffic condition prediction device 100. The speed data acquisition unit 128 acquires speed data at a traffic demand forecast position sampled within a predetermined sampling period. The speed data indicates, for example, the speed according to the traffic demand forecast position for predicting the traffic demand among the past speed data recorded in the external database. The speed data acquisition unit 128 inputs speed data to the speed data storage unit 129. The speed data storage unit 129 stores (stores) speed data.

The speed data is data showing the speed of the vehicle according to the position. For example, it is measured at regular sampling periods using a vehicle detector installed on the road and stored in advance in an external database that manages the speed information. Has been done. For example, the speed data uses the average speed of vehicles that have passed the target position within the sampling period for each sampling period. For example, velocity data is composed of items of "time" and "velocity". "Time" indicates the date and time of the start of the sampling period. "Speed" displays the speed of the vehicle during the sampling period. The speed does not have to be the average speed of the vehicle within the sampling period, and the maximum speed, the minimum speed, the median speed, and the like may be used. In this embodiment, the velocity data is sampled at 5 minute intervals.

The traffic abnormality data acquisition unit 130 acquires traffic abnormality data for each time zone, which indicates the status of traffic accidents and traffic restrictions on the forecasted road for each time zone. The traffic abnormality data acquisition unit 130 acquires traffic abnormality data used for extracting learning traffic volume data from the outside of the traffic condition prediction device 100. The traffic abnormality data acquisition unit 130 acquires traffic abnormality data at a traffic demand forecast position sampled within a predetermined sampling period. The traffic abnormality data indicates, for example, the presence or absence of a traffic abnormality according to the traffic demand forecast position for predicting the traffic demand among the past traffic abnormality data recorded in the external database. The traffic abnormality data acquisition unit 130 inputs traffic abnormality data to the traffic abnormality data storage unit 131. The traffic abnormality data storage unit 131 stores (stores) the traffic abnormality data.

The traffic abnormality data is data indicating a traffic abnormality according to a position, and is stored in advance in an external database that manages the traffic abnormality information. A traffic abnormality is an event that can affect the traffic conditions at a target location due to a traffic accident, road construction, traffic regulation, or the like. For example, the traffic abnormality data is composed of the items of "time" and "presence or absence of traffic abnormality". "Time" indicates the start date and time of the period separated by a fixed time interval. "Traffic abnormality" indicates the presence or absence of traffic abnormality at the target position within the period. The format of the traffic abnormality data is not limited to the above format, and may be data that exists only at the time when the traffic abnormality occurs, for example.

Does the usage time zone extraction unit 115 match the traffic volume of each time zone included in the traffic volume data for each time zone with the traffic demand indicating the number of vehicles trying to pass the forecast target road in the corresponding time zone? Is determined, and only the traffic volume in the matching time zone is extracted as the traffic volume data for learning at the time of constructing the prediction model. The usage time zone extraction unit 115 refers to the traffic volume data storage unit 114, the speed data storage unit 129, and the traffic abnormality data storage unit 131, and determines that the traffic volume and the traffic demand match for each position. Extract traffic data. In this way, the usage time zone extraction unit 115 considers the speed data for each time zone and the traffic abnormality data for each time zone, and the traffic volume for each time zone included in the traffic volume data for each time zone corresponds to the corresponding time. Determine if it matches the traffic demand of the zone.

Hereinafter, an example will be described in which the usage time zone extraction unit 115 determines whether or not the traffic volume and the traffic demand match with respect to the traffic volume data at a certain traffic demand forecast position by using the speed data and the traffic abnormality data. do. When traffic demand exceeds the traffic capacity of the target location, the speed is reduced due to congestion at the target location. In addition, when lane reduction or speed regulation occurs due to traffic abnormalities, the traffic volume changes irregularly, so it cannot be determined whether it matches the traffic demand. The use time zone extraction unit 115 uses the speed data to determine whether the speed at a certain time of the predicted position is below a predetermined speed threshold. As a result of the determination, if the speed is below the threshold value, it is determined that congestion has occurred at the target position and the traffic volume and the traffic demand do not match, and the traffic volume data at the target time is excluded. The speed threshold is a parameter and can be changed by the user. The speed threshold is set to 70 [km / hour], for example, on a highway. Further, the usage time zone extraction unit 115 uses the traffic abnormality data to exclude the traffic volume data at the time when the traffic abnormality occurred. The usage time zone extraction unit 115 inputs the traffic volume data at the time not excluded to the learning traffic volume data storage unit 116. The learning traffic volume data storage unit 116 stores (stores) the traffic volume data at the time not excluded. The exclusion of traffic volume data performed by the usage time zone extraction unit 115 is not limited to the method using the speed data and the traffic abnormality data described above. For example, only one of the speed data and the traffic abnormality data may be used, or the traffic jam data may be used.

The traffic demand forecast model construction unit 117 constructs a traffic demand forecast model that predicts traffic demand using population data as an explanatory variable based on learning traffic volume data and learning population data (population data for each group). The traffic demand forecast model construction unit 117 refers to the learning population data storage unit 112 and the learning traffic volume data storage unit 116, and uses a forecast regression equation (traffic demand forecast model) for forecasting traffic demand for each position. To build. The traffic demand forecast model construction unit 117 constructs a traffic demand forecast model by Elastic Net regression using the population data for learning as an explanatory variable and the traffic volume data for learning as an objective variable. The traffic demand forecast model construction unit 117 constructs a traffic demand forecast model using a plurality of sets of sampling periods, with the learning population data and the learning traffic volume data showing information on the same day as one set. The regression used in the traffic demand forecast model construction unit 117 is not limited to the Elastic Net regression. Instead of Elastic Net regression, for example, Lasso regression, Ridge regression, etc. may be used. Further, variable selection by logistic regression or the like, dimension compression by PLS regression, PCA or the like, linear regression, neural network or the like may be combined. In this embodiment, a model is constructed in which the daily population at noon and the traffic volume data in units of 5 minutes in the afternoon of the same day are set as one set, and the traffic demand in units of 5 minutes in the afternoon of the same day is predicted from the population at noon. The traffic demand forecast model construction unit 117 inputs the traffic demand forecast model to the traffic demand forecast model storage unit 118. The traffic demand forecast model storage unit 118 stores (stores) the traffic demand forecast model.

The traffic demand forecast model construction unit 117 may construct the traffic demand forecast model by preferentially selecting the population data of the group having a high contribution to the traffic demand on the forecast target road. Whether or not the contribution is high is determined by, for example, a regression coefficient, a correlation coefficient, a weight in a neural network, or the like.

The forecasting population data acquisition unit 110 acquires forecasting population data used for forecasting traffic demand from outside the traffic condition forecasting device 100. The forecasting population data is population data at a predetermined date and time, and is, for example, the latest population data among the available population data, population data at a predetermined time on the day, and the like. In the present embodiment, the forecasting population data acquisition unit 110 acquires population data sampled at noon on the day of traffic demand forecasting.

The traffic demand forecasting unit 119 predicts the traffic demand of the forecast target road based on the forecasting population data (population data for each group) and the traffic demand forecasting model. The traffic demand forecasting unit 119 predicts traffic demand using the traffic demand forecasting model stored in the traffic demand forecasting model storage unit 118 and the forecasting population data acquired by the forecasting population data acquisition unit 110. In the present embodiment, the traffic demand forecasting unit 119 inputs the forecasting population data, and uses the traffic demand forecasting model acquired from the traffic demand forecasting model storage unit 118 to determine the traffic demand forecasting position in the traffic demand forecasting time zone. The traffic demand forecast result is output to the outside of the traffic condition forecasting device 100. Further, the traffic demand forecasting unit 119 inputs the traffic demand forecasting result to the learning traffic demand forecasting result storage unit 120 and the forecasting traffic demand forecasting result acquisition unit 125. The learning traffic demand forecast result storage unit 120 stores (stores) the traffic demand forecast result as learning data for constructing a required time prediction model.

The required time data acquisition unit 121 acquires the required time data for each time zone, which indicates the required time of the vehicle passing through the prediction target road for each time zone. The required time data acquisition unit 121 acquires the required time data from the outside of the traffic condition prediction device 100. The required time data acquisition unit 121 acquires the required time data at the required time predicted position sampled within a predetermined sampling period as the original data used for the required time prediction. The required time data indicates, for example, the required time according to the required time predicted position for predicting the required time among the past required time data recorded in the external database. The required time data acquisition unit 121 inputs the required time data to the required time data storage unit 122. The required time data storage unit 122 stores (stores) the required time data.

The required time data is data showing the required time to pass from the entrance to the exit of the target position, is measured at a fixed sampling period, and is stored in advance in an external database that manages the required time information. .. For example, the required time data is composed of the items of "time" and "required time" for each position. "Time" indicates the date and time when the vehicle arrived at the entrance of the target position. In "Time required", the length of time it takes for the vehicle that has reached the entrance of the target position to reach the exit of the target position is displayed. In this embodiment, the required time data is sampled at 30-minute intervals.

The required time prediction model construction unit 123 predicts the required time for each time zone using the traffic demand forecast result as an explanatory variable based on the traffic demand forecast result by the traffic demand forecasting unit 119 and the required time data for each time zone. Build a time prediction model. The required time prediction model construction unit 123 refers to the learning traffic demand forecast result storage unit 120 and the required time data storage unit 122, and uses a prediction regression equation (required time prediction model) for predicting the required time for each position. To build. The required time prediction model construction unit 123 constructs a required time prediction model by Elastic Net regression using the traffic demand forecast result for learning as an explanatory variable and the required time data as an objective variable. The required time prediction model building unit 123 constructs a required time prediction model using a plurality of sets of sampling periods, with the learning traffic demand forecast result and the required time data showing information on the same day as one set. The regression used in the time required prediction model construction unit 123 is not limited to the Elastic Net regression. Instead of Elastic Net regression, for example, Lasso regression, Ridge regression, etc. may be used. Further, variable selection by logistic regression or the like, dimension compression by PLS regression, PCA or the like, linear regression, neural network or the like may be combined. In this embodiment, the traffic demand forecast result in 5-minute units in the afternoon of each day and the required time data in 30-minute units in the afternoon of the same day are set as one set, and the required time in 30-minute units is calculated from the traffic demand forecast result in 5-minute units. Build a forecasting model. The required time prediction model construction unit 123 inputs the required time prediction model to the required time prediction model storage unit 124. The required time prediction model storage unit 124 stores (stores) the required time prediction model.

The forecasting traffic demand forecast result acquisition unit 125 acquires the forecasting traffic demand forecast result used for forecasting the required time from the traffic demand forecasting unit 119. The forecast traffic demand forecast result is a traffic demand forecast result of a predetermined date and time. For example, the latest population data among the available population data, the population data at a predetermined time of the day, etc. are used for the traffic demand. It is a traffic demand forecast result predicted by the forecasting unit 119. In the present embodiment, the traffic demand forecast result acquisition unit 125 for forecasting uses the population at noon on the day when the required time is predicted, and the traffic demand forecast is predicted by the traffic demand forecast unit 119 in 5-minute units on the afternoon of the day. Get the result.

The required time forecasting unit 126 passes through the forecast target road based on the forecasting traffic demand forecast result used for predicting the required time for each time zone predicted by the traffic demand forecasting unit 119 and the required time prediction model. Predict the required time for each time zone. The required time forecasting unit 126 uses the required time forecasting model stored in the required time forecasting model storage unit 124 and the forecasting traffic demand forecasting result acquired by the forecasting traffic demand forecasting result acquisition unit 125. Predict. In the present embodiment, the required time forecasting unit 126 inputs the forecasting traffic demand forecasting result, and uses the required time forecasting model acquired from the required time forecasting model storage unit 124 to predict the required time in the required time forecasting time zone. The time required prediction result of the position is output.

The traffic congestion prediction unit 127 predicts traffic congestion based on the prediction result of the required time for each time zone by the required time prediction unit 126. In the present embodiment, "presence or absence of traffic jam", "start time of traffic jam", and "end time of traffic jam" in the afternoon of the same day are obtained from the prediction result of the required time in units of 30 minutes in the afternoon of the same day at the position of the traffic jam prediction target for each day. , And predict the "peak time of traffic jam". The traffic jam required time threshold Tj for determining that a traffic jam is occurring at the position of the traffic jam prediction target is the traffic jam required time threshold value: using the speed v for determining the traffic jam and the length m of the position of the traffic jam prediction target. It is calculated as Tj = m / v. The speed v determined to be congested is a parameter and can be arbitrarily set by the user. For example, on a highway, 40 [km / hour] is set. When there is at least one required time prediction result that exceeds Tj among the required time prediction results in 30-minute units in the afternoon of the traffic jam prediction target day at the position of the traffic jam prediction target predicted by the required time prediction unit 126. It is judged that there is a traffic jam, and if there is no traffic jam, it is judged that there is no traffic jam.

When it is determined that there is a traffic jam, the traffic jam prediction unit 127 predicts the start time of the traffic jam, the end time of the traffic jam, and the peak time of the traffic jam. In the present embodiment, the time when the required time of the required time prediction result in units of 30 minutes in the afternoon of the congestion prediction target position on the congestion prediction target day first exceeds the congestion required time threshold Tj is set as the congestion start time. Next, the time of the last transition from the state in which the required time exceeds the congestion required time threshold Tj to the state in which the required time falls below the congestion required time threshold Tj is set as the end time of the traffic jam. Next, the time with the longest required time is defined as the peak time of traffic congestion. FIG. 3A shows an example of output of the prediction result from the traffic congestion prediction unit 127 when it is determined that there is a traffic jam. For example, as shown in FIG. 3A, a value indicating "congestion is present" is output for the item "presence or absence of congestion", and "start time", "end time", and "congestion peak time" are output. The derived results are output for each item. The above is an example of the embodiment, and the determination of the start time of the traffic jam, the end time of the traffic jam, and the peak time of the traffic jam is not limited to the above. For example, on a road where traffic jams occur and end repeatedly in the afternoon, it is assumed that the traffic jams occur many times instead of once, not only at the time when the traffic jam required time threshold Tj is first exceeded. , It is conceivable that the start time will be set as another traffic jam each time it exceeds.

The traffic congestion prediction unit 127 does not predict the "start time", "end time", and "congestion peak time" when it is determined that there is no congestion. FIG. 3B shows an output example of the prediction result from the traffic congestion prediction unit 127 when it is determined that there is no congestion. For example, as shown in FIG. 3 (b), for the item of "presence or absence of traffic jam", a value indicating "no traffic jam" is output, and "start time", "end time", and "end time", which are not predicted, are output. And the item of "traffic jam peak time" is NULL.

Next, the processing executed by the traffic condition prediction device 100 will be described with reference to FIGS. 4 to 7. FIG. 4 is a flowchart showing the construction process of the traffic demand forecast model. FIG. 5 is a flowchart showing a traffic demand forecast process. FIG. 6 is a flowchart showing the construction process of the required time prediction model. FIG. 7 is a flowchart showing a required time prediction process and a traffic congestion prediction process.

As shown in FIG. 4, in the traffic demand forecast model construction process, various data are first acquired from the outside of the traffic condition forecasting device 100 (step S1). Specifically, the learning population data acquisition unit 111 acquires the learning population data, the traffic volume data acquisition unit 113 acquires the traffic volume data for each time zone, and the speed data acquisition unit 128 acquires the speed data for each time zone. The traffic abnormality data acquisition unit 130 acquires the traffic abnormality data for each time zone.

Subsequently, the usage time zone extraction unit 115 extracts the usage time zone (step S2). Specifically, the usage time zone extraction unit 115 indicates the traffic demand indicating the number of vehicles trying to pass the forecast target road in the corresponding time zone for the traffic volume in each time zone included in the traffic volume data for each time zone. It is determined whether or not it matches with, and only the traffic volume in the matching time zone is extracted as the training traffic volume data at the time of constructing the prediction model. The usage time zone extraction unit 115 considers the speed data for each time zone and the traffic abnormality data for each time zone, and the traffic volume for each time zone included in the traffic volume data for each time zone corresponds to the traffic demand for the corresponding time zone. To determine if it matches.

Subsequently, the traffic demand forecast model construction unit 117 constructs a traffic demand forecast model (step S3). Specifically, the traffic demand forecast model construction unit 117 creates a traffic demand forecast model that predicts traffic demand using population data as an explanatory variable based on learning traffic volume data and learning population data (population data for each group). To construct. Then, the constructed traffic demand forecast model is stored by the traffic demand forecast model storage unit 118 (step S4).

As shown in FIG. 5, in the traffic demand forecasting process, first, the forecasting population data acquisition unit 110 acquires the forecasting population data from the outside of the traffic condition forecasting device 100 (step S11). Subsequently, the traffic demand forecasting unit 119 predicts the traffic demand (step S12). Specifically, the traffic demand forecasting unit 119 predicts the traffic demand of the forecast target road based on the forecasting population data (population data for each group) and the traffic demand forecasting model (step S12). Then, the forecast result is output to the outside by the traffic demand forecast unit 119, and the traffic demand forecast result is stored by the learning traffic demand forecast result storage unit 120 (step S13).

As shown in FIG. 6, in the required time prediction model construction process, first, the required time data acquisition unit 121 acquires the required time data for each time zone from the outside of the traffic condition prediction device 100 (step S21). Subsequently, the required time prediction model building unit 123 constructs the required time prediction model (step S22). Specifically, the required time forecast model construction unit 123 is required for each time zone using the traffic demand forecast result as an explanatory variable based on the traffic demand forecast result by the traffic demand forecasting unit 119 and the required time data for each time zone. Build a time prediction model to predict time. Then, the required time prediction model stored in the required time prediction model storage unit 124 stores the constructed required time prediction model (step S23).

As shown in FIG. 7, in the required time prediction process and the traffic congestion prediction process, the required time is first predicted by the required time prediction unit 126 (step S31). Specifically, the required time forecasting unit 126 predicts based on the forecasting traffic demand forecast result used for predicting the required time for each time zone predicted by the traffic demand forecasting unit 119 and the required time prediction model. Predict the time required for each time zone when passing through the target road. Subsequently, the traffic congestion prediction unit 127 predicts the traffic congestion (step S32). Specifically, the traffic congestion prediction unit 127 predicts the traffic congestion based on the prediction result of the required time for each time zone by the required time prediction unit 126. Then, the traffic congestion prediction unit 127 outputs the traffic congestion prediction result to the outside (step S33).

Next, the operation and effect of the traffic condition prediction device 100 will be described.

The traffic condition prediction device 100 is a learning population data acquisition unit 111 that acquires learning population data of a plurality of groups classified according to the position and attribute of a person, and the number of vehicles that have passed the prediction target road. The traffic volume data acquisition unit 113 that acquires traffic volume data for each time zone that indicates the traffic volume for each time zone, and the traffic volume for each time zone included in the traffic volume data for each time zone are predicted roads in the corresponding time zone. It is determined whether it matches the traffic demand indicating the number of vehicles trying to pass through, and only the traffic volume in the matching time zone is extracted as the traffic volume data for learning at the time of building the prediction model. A traffic demand forecast model building unit 117 that builds a traffic demand forecasting model that predicts traffic demand using population data as an explanatory variable based on the traffic volume data for learning and the population data for each group, and the population data for each group. It also has a traffic demand prediction unit 119 that predicts the traffic demand of the road to be predicted based on the traffic demand prediction model.
In the traffic condition forecasting device 100 described above, a traffic demand forecasting model is constructed by, for example, machine learning, based on the traffic volume data and the population data. By using population data, it is possible to construct a traffic demand forecast model in consideration of various factors of increase / decrease in traffic demand. For example, population data correlates with the number of vehicles that are not currently in the driving state but will be in the driving state at the predicted time. By acquiring the factors of increase / decrease in traffic demand that are difficult to obtain from traffic volume data from population data and constructing a traffic demand forecast model, the traffic demand forecast model is used to forecast traffic demand with high accuracy. It can be performed. Here, for example, when a traffic demand exceeding the traffic capacity occurs, the vehicle stays and traffic congestion occurs, and the traffic volume becomes less than the traffic demand. Therefore, it is necessary to accurately grasp the traffic demand from the traffic volume. I can't. Therefore, if a traffic demand forecast model is simply constructed from traffic volume data, it may not be possible to forecast traffic demand with high accuracy. In this respect, in the traffic condition prediction device 100, the traffic volume and the traffic demand are compared, and only the traffic volume in the time zone in which both are the same is used for constructing the prediction model. That is, for example, if there is a time zone in which the traffic volume and the traffic demand do not match due to the traffic demand exceeding the traffic requirement on the forecast target road, the traffic volume data in the non-matching time zone is excluded. Therefore, a traffic demand forecasting model that more accurately reflects the relationship between population and traffic demand is constructed. By using the traffic demand forecasting model, it becomes possible to predict the traffic demand exceeding the traffic capacity with high accuracy by the extrapolation method based on the population. Further, since the traffic demand forecast model is constructed by excluding some traffic volume data, the amount of calculation required for constructing the traffic demand forecast model can be suppressed. From the above, according to the traffic condition prediction device 100, it is possible to improve the prediction accuracy of the traffic demand and suppress the increase in the amount of calculation required for constructing the traffic demand prediction model.

The traffic demand forecast model construction unit 117 may construct a traffic demand forecast model by selecting population data of a group having a high contribution to traffic demand on the forecast target road. As a result, it is possible to secure the accuracy of the traffic demand forecast, limit the population data used for constructing the traffic volume demand forecast model, and suppress an increase in the amount of calculation required for constructing the traffic volume demand forecast model.

In the traffic condition forecasting device 100 described above, the traffic demand by the required time data acquisition unit 121 and the traffic demand forecasting unit 119 for acquiring the required time data for each time zone indicating the required time of the vehicle passing through the forecast target road for each time zone. Based on the forecast result of the above and the required time data for each time zone, the required time forecast model building unit 123 for constructing the required time forecast model for predicting the required time for each time zone using the predicted result of the traffic demand as an explanatory variable, and the traffic demand Predicts the required time for each time zone when passing through the forecast target road based on the forecast traffic demand forecast result used for predicting the required time for each time zone predicted by the forecasting unit 119 and the required time forecast model. The required time forecasting unit 126 is further provided. For example, a vehicle that has stagnated due to a traffic demand exceeding the traffic capacity in a certain period will pass through the road in a time zone after the time zone in which the vehicle was originally intended to pass. In this case, the required time of the vehicle may not be properly predicted. In this regard, since the traffic condition prediction device 100 builds a required time prediction model based on the required time for each time zone and the forecast result of traffic demand, it is required in consideration of the influence of the stagnant vehicle as described above. You can predict the time. That is, according to the traffic condition prediction device 100, the accuracy of the required time prediction can be improved.

The above-mentioned traffic condition prediction device 100 further includes a traffic congestion prediction unit 127 that predicts traffic congestion based on the prediction result of the required time for each time zone by the required time prediction unit 126. By predicting traffic congestion based on the highly accurate required time prediction result described above, the accuracy of traffic congestion prediction can be improved.

The above-mentioned traffic condition prediction device 100 includes a speed data acquisition unit 128 that acquires speed data for each time zone indicating the speed of a vehicle that has passed the prediction target road for each time zone, and a traffic accident or traffic regulation status on the prediction target road. Further includes a traffic abnormality data acquisition unit 130 for acquiring traffic abnormality data for each time zone, and a usage time zone extraction unit 115 for at least one of the speed data for each time zone and the traffic abnormality data for each time zone. In consideration of either one, it is determined whether the traffic volume in each time zone included in the traffic volume data for each time zone matches the traffic demand in the corresponding time zone. When traffic demand exceeds the traffic capacity, it is considered that the speed of the vehicle will decrease due to congestion. In addition, when a traffic accident or traffic regulation occurs, it is considered that the traffic volume changes irregularly. In this way, the speed data and traffic abnormality data for each time zone are closely related to the traffic volume and the situation of the traffic demand. Therefore, when determining whether or not the traffic volume matches the traffic demand, the time zone is used. By considering the speed data and the traffic abnormality data, it is possible to determine the coincidence between the traffic volume and the traffic demand with high accuracy.

Finally, the hardware configuration of the traffic condition prediction device 100 will be described with reference to FIG. The traffic condition prediction device 100 described above may be physically configured as a computer device 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.

In the following description, the word "device" can be read as a circuit, a device, a unit, or the like. The hardware configuration of the traffic condition prediction device 100 may be configured to include one or more of the devices shown in the figure, or may be configured not to include some of the devices.

For each function in the traffic condition prediction device 100, by loading predetermined software (program) on hardware such as the processor 1001 and the memory 1002, the processor 1001 performs an calculation, and communication by the communication device 1004, the memory 1002, and the memory 1002 and the like. It is realized by controlling the reading and / or writing of data in the storage 1003.

Processor 1001 operates, for example, an operating system to control the entire computer. The processor 1001 may be configured by a central processing unit (CPU) including an interface with a peripheral device, a control device, an arithmetic unit, a register, and the like. For example, the control function of the traffic demand forecasting unit 119 of the traffic condition forecasting device 100 may be realized by the processor 1001.

Further, the processor 1001 reads a program (program code), a software module and data from the storage 1003 and / or the communication device 1004 into the memory 1002, and executes various processes according to these. As the program, a program that causes a computer to execute at least a part of the operations described in the above-described embodiment is used. For example, the control function of the traffic demand forecasting unit 119 of the traffic condition forecasting device 100 may be realized by a control program stored in the memory 1002 and operated by the processor 1001, and the other functional blocks may be similarly realized. May be good. Although it has been described that the various processes described above are executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. Processor 1001 may be mounted on one or more chips. The program may be transmitted from the network via a telecommunication line.

The memory 1002 is a computer-readable recording medium, and is composed of at least one such as a ROM (Read Only Memory), an EPROM (Erasable Programmable ROM), an EEPROM (Electrically Erasable Programmable ROM), and a RAM (Random Access Memory). May be done. The memory 1002 may be referred to as a register, a cache, a main memory (main storage device), or the like. The memory 1002 can store a program (program code), a software module, or the like that can be executed to implement the wireless communication method according to the embodiment of the present invention.

The storage 1003 is a computer-readable recording medium, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, an optical magnetic disk (for example, a compact disk, a digital versatile disk, a Blu-ray). It may consist of at least one (registered trademark) disk), smart card, flash memory (eg, card, stick, key drive), floppy (registered trademark) disk, magnetic strip, and the like. The storage 1003 may be referred to as an auxiliary storage device. The storage medium described above may be, for example, a database, server or other suitable medium containing memory 1002 and / or storage 1003.

The communication device 1004 is hardware (transmission / reception device) for communicating between computers via a wired and / or wireless network, and is also referred to as, for example, a network device, a network controller, a network card, a communication module, or the like.

The input device 1005 is an input device (for example, a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives an input from the outside. The output device 1006 is an output device (for example, a display, a speaker, an LED lamp, etc.) that outputs to the outside. The input device 1005 and the output device 1006 may have an integrated configuration (for example, a touch panel).

Further, each device such as the processor 1001 and the memory 1002 is connected by a bus 1007 for communicating information. The bus 1007 may be composed of a single bus or may be composed of different buses between the devices.

Further, the traffic condition prediction device 100 includes hardware such as a microprocessor, a digital signal processor (DSP: Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), and an FPGA (Field Programmable Gate Array). It may be configured to include, and a part or all of each functional block may be realized by the hardware. For example, the processor 1001 may be implemented on at least one of these hardware.

Although the present embodiment has been described in detail above, it is clear to those skilled in the art that the present embodiment is not limited to the embodiment described in the present specification. This embodiment can be implemented as an amendment or modification without departing from the spirit and scope of the present invention as determined by the description of the scope of claims. Therefore, the description herein is for purposes of illustration only and has no limiting implications for this embodiment.

Each aspect / embodiment described herein includes LTE (Long Term Evolution), LTE-A (LTE-Advanced), SUPER 3G, IMT-Advanced, 4G, 5G, FRA (Future Radio Access), W-CDMA. (Registered Trademarks), GSM (Registered Trademarks), CDMA2000, UMB (Ultra Mobile Broad-band), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, UWB (Ultra-Wide) Band), WiMAX®, and other systems that utilize suitable systems and / or extended next-generation systems based on them may be applied.

The processing procedures, sequences, flowcharts, and the like of each aspect / embodiment described in the present specification may be rearranged in order as long as there is no contradiction. For example, the methods described herein present elements of various steps in an exemplary order and are not limited to the particular order presented.

The input / output information and the like may be stored in a specific place (for example, a memory) or may be managed by a management table. Information to be input / output may be overwritten, updated, or added. The output information and the like may be deleted. The input information or the like may be transmitted to another device.

The determination may be made by a value represented by 1 bit (0 or 1), by a boolean value (Boolean: true or false), or by comparing numerical values (for example, a predetermined value). It may be done by comparison with the value).

Each aspect / embodiment described in the present specification may be used alone, in combination, or may be switched and used according to the execution. Further, the notification of predetermined information (for example, the notification of "being X") is not limited to the explicit one, but is performed implicitly (for example, the notification of the predetermined information is not performed). May be good.

Software, whether called software, firmware, middleware, microcode, hardware description language, or other names, instructions, instruction sets, codes, code segments, program codes, programs, subprograms, software modules. , Applications, software applications, software packages, routines, subroutines, objects, executable files, execution threads, procedures, features, etc. should be broadly interpreted.

Further, software, instructions, and the like may be transmitted and received via a transmission medium. For example, the software may use wired technology such as coaxial cable, fiber optic cable, twisted pair and digital subscriber line (DSL) and / or wireless technology such as infrared, wireless and microwave to website, server, or other. When transmitted from a remote source, these wired and / or wireless technologies are included within the definition of transmission medium.

The information, signals, etc. described herein may be represented using any of a variety of different techniques. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description are voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, light fields or photons, or any of these. It may be represented by a combination of.

The terms described herein and / or the terms necessary for understanding the present specification may be replaced with terms having the same or similar meanings.

Further, the information, parameters, etc. described in the present specification may be represented by an absolute value, a relative value from a predetermined value, or another corresponding information. ..

The user terminal may be a mobile communication terminal, a subscriber station, a mobile unit, a subscriber unit, a wireless unit, a remote unit, a mobile device, a wireless device, a wireless communication device, a remote device, a mobile subscriber station, an access terminal, etc. It may also be referred to as a mobile device, wireless device, remote device, handset, user agent, mobile client, client, or some other suitable term.

As used herein, the terms "determining" and "determining" may include a wide variety of actions. "Judgment", "decision" is, for example, calculating, computing, processing, deriving, investigating, looking up (eg, table, database or another). It may include searching in the data structure), considering that the confirmation (ascertaining) is "judgment" and "decision". Also, "judgment" and "decision" are receiving (for example, receiving information), transmitting (for example, transmitting information), input (input), output (output), and access. It may include (for example, accessing data in memory) to be regarded as "judgment" or "decision". In addition, "judgment" and "decision" are considered to be "judgment" and "decision" when the things such as solving, selecting, choosing, establishing, and comparing are regarded as "judgment" and "decision". Can include. That is, "judgment" and "decision" may include considering some action as "judgment" and "decision".

The phrase "based on" as used herein does not mean "based on" unless otherwise stated. In other words, the statement "based on" means both "based only" and "at least based on".

As used herein by designations such as "first", "second", etc., any reference to that element does not generally limit the quantity or order of those elements. These designations can be used herein as a convenient way to distinguish between two or more elements. Thus, references to the first and second elements do not mean that only two elements can be adopted there, or that the first element must somehow precede the second element.

As long as "include", "including", and variations thereof are used herein or within the scope of the claims, these terms are similar to the term "comprising". In addition, it is intended to be inclusive. Moreover, the term "or" as used herein or in the claims is intended to be non-exclusive.

In the present specification, a plurality of devices shall be included unless the device has only one apparently in context or technically.

The entire disclosure is intended to include more than one, unless the context clearly indicates the singular.

100 ... Traffic condition forecasting device, 111 ... Population data acquisition unit for learning, 113 ... Traffic volume data acquisition unit, 114 ... Traffic volume data storage unit, 115 ... Usage time zone extraction unit 117 ... Traffic demand forecast model construction unit 119 ... Traffic demand forecasting unit, 121 ... Required time data acquisition unit, 123 ... Required time forecast model construction unit, 126 ... Required time forecasting unit, 127 ... Traffic congestion forecasting unit, 128 ... Speed data acquisition unit, 130 ... Traffic abnormality data acquisition Department.

Claims (4)

A population data acquisition unit that acquires population data of multiple groups classified according to the position and attributes of people,
A traffic volume data acquisition unit for each time zone that acquires traffic volume data for each time zone that indicates the traffic volume, which is the number of vehicles that have passed the forecast target road, for each time zone.
It is determined whether the traffic volume in each time zone included in the traffic volume data for each time zone matches the traffic demand indicating the number of vehicles trying to pass through the forecast target road in the corresponding time zone, and matches. A usage time zone extraction unit that extracts only the traffic volume in the time zone as learning traffic volume data when constructing a prediction model,
A traffic demand forecasting model building unit that builds a traffic demand forecasting model that predicts traffic demand using population data as an explanatory variable based on the learning traffic volume data and the population data for each group.
Based on the population data and the traffic demand forecast model for each group, the traffic demand forecasting unit that predicts the traffic demand of the forecasted road, and the traffic demand forecasting unit.
A speed data acquisition unit that acquires speed data for each time zone that indicates the speed of a vehicle that has passed the prediction target road, and a time that indicates the status of traffic accidents and traffic restrictions on the prediction target road for each time zone. It is equipped with at least one of the traffic abnormality data acquisition units that acquire traffic abnormality data by band.
The usage time zone extraction unit considers at least one of the time zone speed data and the time zone traffic abnormality data, and determines the traffic volume of each time zone included in the time zone traffic volume data. A traffic condition prediction device that determines whether or not the traffic demand matches the traffic demand in the corresponding time zone . The traffic condition forecasting device according to claim 1, wherein the traffic demand forecast model building unit selects the population data of a group having a high contribution to traffic demand on the forecasted road and builds the traffic demand forecasting model. A required time data acquisition unit that acquires required time data for each time zone that indicates the required time of a vehicle that has passed the predicted road for each time zone, and a required time data acquisition unit.
Based on the traffic demand forecast result by the traffic demand forecasting unit and the required time data for each time zone, a required time prediction model for predicting the required time for each time zone is constructed using the predicted result of the traffic demand as an explanatory variable. Time required prediction model construction department and
The time when passing through the forecast target road based on the forecast traffic demand forecast result used for predicting the required time for each time zone predicted by the traffic demand forecasting unit and the required time forecast model. The traffic condition prediction device according to claim 1 or 2, further comprising a required time prediction unit for predicting the required time for each band. The traffic condition prediction device according to claim 3, further comprising a traffic congestion prediction unit that predicts traffic congestion based on the prediction result of the required time for each time zone by the required time prediction unit.

Images