JP7399668B2 - Operation planning system for mechanical parking device, operation planning method thereof, operation planning program, mechanical parking device - Google Patents

Description

The present disclosure relates to a driving planning system for a mechanical parking device, a driving planning method thereof, a driving planning program, and a mechanical parking device.

In a mechanical parking device, a vehicle is transported and stored in a storage area. For example, in the berth type (IPS), a vehicle parked at a warehousing berth is transported to a storage area and stored by a transport means such as a warehousing lift or a conveyor, and the vehicle is warehousing (for example, Patent Document 1).

Japanese Patent Application Publication No. 2018-115501

Mechanical parking devices are used in a variety of locations, including office buildings, residential facilities, commercial facilities, hospitals, and under roads. For this reason, the usage status of mechanical parking devices changes depending on various factors.

If the operational settings of the mechanical parking device are made based on the usage situation predicted in advance, if the usage situation changes thereafter, there is a possibility that the operation will continue with settings that are not optimal. For example, the usage status of mechanical parking devices is thought to change due to the construction of large commercial facilities nearby. If the vehicle is not operated with appropriate operating settings, the mechanical parking device may not be able to fully demonstrate its capabilities.

The present disclosure has been made in view of these circumstances, and provides an operation planning system for a mechanical parking device that can be operated according to usage conditions, an operation planning method thereof, an operation planning program, and a mechanical parking system. The purpose is to provide a type parking device.

A first aspect of the present disclosure is a driving planning system that is applied to a mechanical parking device that allows vehicles to enter and exit the parking lot, and which uses a plurality of driving planning systems related to vehicle transportation based on performance data of vehicles that entered and exited the parking lot during a predetermined period in the past. Corresponding to each of the driving modes of Operation of a mechanical parking device, comprising: an estimating unit that outputs an estimated value for one index; and a determining unit that determines the driving mode in the predetermined time period based on the estimated value of the index for each driving mode. It is a planning system.

A second aspect of the present disclosure is a driving planning method that is applied to a mechanical parking device that allows vehicles to enter and exit the parking lot, and the method is based on performance data of vehicles that entered and exited the parking lot during a predetermined period in the past. Corresponding to each of the driving modes of A driving planning method for a mechanical parking device, comprising: outputting an estimated value for one index; and determining the driving mode in the predetermined time period based on the estimated value of the index for each driving mode. It is.

A third aspect of the present disclosure is a driving planning program that is applied to a mechanical parking device that allows vehicles to enter and exit the parking lot, and which performs multiple operations related to vehicle transportation based on performance data of vehicles that entered and exited the parking lot in a predetermined period in the past. Corresponding to each of the driving modes of Mechanical parking for causing a computer to execute a process of outputting an estimated value for one index, and a process of determining the driving mode in the predetermined time period based on the estimated value of the index for each driving mode. This is a device operation planning program.

According to the present disclosure, it is possible to drive according to usage conditions.

1 is a diagram showing a schematic configuration of a mechanical parking device according to an embodiment of the present disclosure. FIG. 1 is a schematic configuration diagram of a parking area of a mechanical parking device according to an embodiment of the present disclosure. FIG. 1 is a schematic configuration diagram of a parking area of a mechanical parking device according to an embodiment of the present disclosure. FIG. 1 is a hardware configuration diagram of a control device according to an embodiment of the present disclosure. FIG. 2 is a functional block diagram showing functions included in a control device according to an embodiment of the present disclosure. 1 is a diagram illustrating an example of a schematic cross-sectional configuration diagram of a mechanical parking device according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of scheduling according to an embodiment of the present disclosure. FIG. 2 is a schematic diagram showing an example of a neuron model. FIG. 1 is a schematic diagram showing an example of a neural network. FIG. 2 is a diagram illustrating an example of a radar chart according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of a radar chart according to an embodiment of the present disclosure. FIG. 3 is a diagram showing a flowchart of scheduling processing in a control device according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of the arrangement of traveling carts according to an embodiment of the present disclosure. FIG. 2 is an image diagram showing an example of performance data according to an embodiment of the present disclosure.

DESCRIPTION OF EMBODIMENTS Below, an embodiment of a driving planning system for a mechanical parking device, a driving planning method thereof, a driving planning program, and a mechanical parking device according to the present disclosure will be described with reference to the drawings.
FIG. 1 is a diagram showing a schematic configuration of a mechanical parking device 60 according to an embodiment of the present disclosure. The mechanical parking device 60 in this embodiment is a berth type (planar reciprocating conveyor type), and allows vehicles to enter and leave the parking lot from, for example, an entry level provided on the ground. In this embodiment, a berth-type mechanical parking device 60 will be described as an example, but the driving planning system can be applied to other types of mechanical parking devices as long as they are mechanical parking devices that allow vehicles to enter and exit the parking lot. The same can be applied to.

As shown in FIG. 1, the boarding floor is provided with a boarding and alighting area (boarding and alighting area) 70, and the boarding and alighting room 70 is provided with a warehousing area (warehousing berth) 70a and a warehousing area (outgoing berth) 70b. It is being A vehicle parked at the warehousing berth 70a (entered by a driver or the like) is stored in a storage area on the storage floor 66. The vehicles stored in the storage area are transported to the unloading berth 70b and unloaded. The storage area is composed of a plurality of storage shelves (for example, a first stage, a second stage, and a third stage). Each storage shelf is provided with a traveling trolley 63 for transporting vehicles within the storage shelf.

FIG. 2 is a schematic diagram of the warehousing berth 70a. As shown in FIG. 2, the warehousing lift 62 is connectable to the warehousing berth 70a. When the warehousing lift 62 is placed on the entrance floor (when the warehousing lift 62 is placed at approximately the same height as the floor surface of the warehousing berth 70a), the vehicles are moved from the warehousing berth 70a by the lateral conveyors 61a to 61d. It is transported laterally to the lift 62. An operation panel 22 is provided outside the warehousing berth 70a. The operation panel 22 is an input device for executing various operations in the mechanical parking device 60. For example, the operation panel 22 is operated when starting transport of a vehicle parked in the warehousing berth 70a.

When warehousing a vehicle, the vehicle parked at the warehousing berth 70a for warehousing is transferred to the warehousing lift 62 by the transverse conveyors 61a to 61d. Then, the vehicle is transported (lowered) to a predetermined storage shelf on the storage floor 66 by the warehousing lift 62, and the vehicle is moved from the warehousing lift 62 to the traveling carriage 63 by the transverse conveyors 61c to 61f. Then, the traveling trolley 63 on which the incoming vehicle is placed travels along the trolley traveling path 64 to a predetermined storage section. Thereafter, the warehousing vehicles on the traveling trolley 63 are stored in a predetermined storage compartment by the transverse conveyors 61e to 61h.

FIG. 3 is a schematic configuration diagram of the shipping berth 70b. As shown in FIG. 3, the delivery berth 70b is configured to be connectable with the delivery lift 65. When the unloading lift 65 is placed on the entry floor (when the unloading lift 65 is placed at approximately the same height as the floor of the unloading berth 70b), the vehicles are moved out of the unloading berth 70b by the lateral conveyors 61i to 61l. It is transported laterally to the lift 65. An operation panel 22 is provided outside the shipping berth 70b. The operation panel 22 is an input device for executing various operations in the mechanical parking device 60.

When leaving a vehicle, the vehicle is moved from the storage compartment to the carriage 63 by the transverse conveyors 61e to 61h, and from the carriage 63 to the delivery lift 65 by the transverse conveyors 61e, 61f, 61i, and 61j. . Then, the outgoing vehicle is transported (raised) to the outgoing berth 70b by the outgoing lift 65, and then transferred to the outgoing berth 70b by the lateral conveyors 61i to 61l.

Note that the warehousing lift 62 and the warehousing lift 65 are transport means for transporting the vehicle between the entry floor (warehousing area) and the storage floor (storage area) 66. Note that the transverse conveyors 61a to 61l and the traveling carriage 63 are also conveyance means for conveying vehicles.

The control device (operation planning system) 30 schedules the operation mode of the mechanical parking device 60. Specifically, the control device 30 determines a driving mode suited to the usage situation from among the plurality of driving modes of the mechanical parking device 60, and performs driving in the determined driving mode. Therefore, it is possible to drive in an appropriate driving mode in response to changes in usage conditions caused by environmental changes in the vicinity of the mechanical parking device 60.

FIG. 4 is a diagram showing an example of the hardware configuration of the control device 30 according to this embodiment.
As shown in FIG. 4, the control device 30 is a computer system, and includes, for example, a CPU 11, a ROM (Read Only Memory) 12 for storing programs executed by the CPU 11, and a ROM (Read Only Memory) 12 for storing programs executed by the CPU 11. A RAM (Random Access Memory) 13 that functions as a work area, a hard disk drive (HDD) 14 as a mass storage device, and a communication section 15 for connecting to a network or the like. These parts are connected via a bus 18.

Further, the control device 30 may include an input unit such as a keyboard and a mouse, and a display unit such as a liquid crystal display device that displays data.

Note that the storage medium for storing programs and the like executed by the CPU 11 is not limited to the ROM 12. For example, other auxiliary storage devices such as a magnetic disk, a magneto-optical disk, and a semiconductor memory may be used.

A series of processing steps for realizing various functions described below are recorded in the form of a program in the hard disk drive 14, etc., and the CPU 11 reads this program into the RAM 13 etc. to process information and perform arithmetic processing. As a result, various functions described below are realized. Note that the program may be pre-installed in the ROM 12 or other storage medium, provided as being stored in a computer-readable storage medium, or distributed via wired or wireless communication means. etc. may also be applied. Computer-readable storage media include magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, semiconductor memories, and the like.

FIG. 5 is a functional block diagram showing the functions included in the control device 30. As shown in FIG. 5, the control device 30 mainly includes an estimation section 31, a determination section 32, and a modification section 33. Note that the functions of each part in FIG. 5 are functions as an operation planning system, and may have other functions.

The estimating unit 31 calculates an average entry waiting time and an average exit waiting time in a future predetermined time period based on actual data of vehicles entering and exiting the warehouse in a predetermined period in the past, corresponding to each of a plurality of driving modes related to transportation of vehicles. , the number of times a plurality of predetermined devices in the mechanical parking device 60 are used (hereinafter referred to as the "number of times of use"), and the vehicle occupancy rate in the storage area (hereinafter referred to as the "vehicle occupancy rate"). Output the estimated value for .

The track record data is data related to the track record of vehicles entering and leaving the warehouse during a predetermined period in the past. The past predetermined period is, for example, a past predetermined period (for example, several years) from the present (such as the time of estimation). FIG. 14 is an image diagram showing an example of performance data. Note that the performance data is not limited to the configuration shown in FIG. 14.

The performance data is stored in association with vehicle information, date and time, entry time, exit time, driving mode, and storage location. As will be described later, when applying machine learning and using performance data as training data, the performance data includes vehicle information, date and time, storage time, exit time, driving mode, and storage. It includes information in which positions and indicators are associated with each other. Each parameter of vehicle information, date and time, parking time, parking time, driving mode, and storage position has a value that reflects the usage status of the mechanical parking device 60. Therefore, each parameter has a correlation with each index (average entry waiting time, average exit waiting time, number of times of use, and vehicle inventory rate, which are parameters that change depending on the usage situation).

The vehicle information is information about the stored vehicle, and is vehicle type information (light car, regular car, high roof car, etc.). The vehicle type information must include at least one of type information (sedan, wagon, or sports), manufacturer country information (domestic car or foreign car), and power source information (gasoline car, hybrid car, or electric car). You can also use it as Note that it may also include the ticket issue number at the time of entering the warehouse and the details of the contract (whether it is a fixed-term rental or an hourly rental). Since the vehicle information reflects the usage status of the mechanical parking device 60, it has a correlation with each index (average entry waiting time, average exit waiting time, number of times of use, and vehicle availability rate). .

The actual data includes each of the above parameters (vehicle information, date and time, storage time, exit time, driving mode, and storage location), but other parameters (complementary parameters) may also be added and associated. . Estimation accuracy can be improved by using complementary parameters that reflect the usage status.

For example, the performance data may include information regarding the driver of the vehicle. The information related to the driver of the vehicle includes, for example, gender, age, physical characteristics, number of passengers, and presence or absence of children. The information related to the driver of the vehicle can include at least one of each parameter. Since the information related to the vehicle driver reflects the usage status of the mechanical parking device 60, the correlation with each index (average entry waiting time, average exit waiting time, number of times of use, and vehicle availability rate) is calculated. have.

Furthermore, the performance data may include setting information of a car navigation system installed in the vehicle. The setting information of the car navigation system installed in the vehicle includes, for example, the address of the driver, the destination, and whether or not the position of the mechanical parking device 60 is set in the navigation system. Whether or not the location of the mechanical parking device 60 is set in the navigation means whether or not the location information of the mechanical parking device 60 is set in the navigation, and if it is set, the user is a regular user and cannot use it. It is presumed that you are used to it. Setting information of a car navigation system installed in a vehicle can include at least one of each parameter. Since the setting information of the car navigation system installed in the vehicle reflects the usage status of the mechanical parking device 60, each index (average entry waiting time, average exit waiting time, number of times of use, and inventory rate) There is a correlation with

Furthermore, the performance data may include at least one of the stored parking time of the vehicle and information on the number of times of use. The parking time is the time the vehicle is stored from the time the vehicle enters the garage to the time the vehicle leaves the garage (storage time). The usage frequency information is the total number of times the user has used the mechanical parking device 60 so far. Note that statistical information may also be included. Since the above parameters reflect the usage status of the mechanical parking device 60, they have a correlation with each index (average entry waiting time, average exit waiting time, number of times of use, and inventory rate). .

Furthermore, the performance data may include at least one of calendar information, weather information, and disaster information. Calendar information is, for example, information regarding date and time, days of the week, six days of the week, holidays, anniversaries, and the like. Weather information is, for example, information regarding the weather, temperature, humidity, and the like. Disaster information is, for example, information related to accident information, disaster information, transportation information, and the like. Since the above parameters reflect the usage status of the mechanical parking device 60, they have a correlation with each index (average entry waiting time, average exit waiting time, number of times of use, and inventory rate). .

The performance data can include any information other than the above-mentioned parameters as long as it reflects the usage status of the mechanical parking device 60. For example, information about vehicles that could not be stored may be used.

Based on the above performance data, the estimation unit 31 performs estimation corresponding to each driving mode.

The driving mode is a driving mode related to transportation of a vehicle in the mechanical parking device 60, and a plurality of types of modes are prepared in advance. In this embodiment, a case will be described in which the operation mode related to warehousing is used, but it is also possible to apply other operation modes. Note that other operation modes will be described later.

The plurality of driving modes can include at least one of a warehousing priority mode, a warehousing priority mode, a uniform warehousing mode, and a vehicle type priority mode. In this embodiment, a case will be described in which the warehousing priority mode, the exiting priority mode, the uniform warehousing mode, and the vehicle type priority mode are used.

The warehousing priority mode is an operation mode in which warehousing is carried out preferentially. Specifically, this is a mode in which the transport means is operated so as to be able to handle the transport of vehicles parked at the warehousing berth 70a when the vehicles parked at the warehousing berth 70a are transported and stored. For example, the plurality of traveling carts 63 in the storage area are operated with at least one vehicle left open for a vehicle to be stored. Since the other traveling trolley 63 can be used for vehicles leaving the warehouse, it is also possible to leave the vehicle. Further, the time required for warehousing may be shortened by storing the warehousing vehicle near the warehousing lift 62 (shortening the transport distance related to warehousing). In this way, the warehousing priority mode is a driving mode in which the vehicle is driven with priority placed on warehousing the vehicle. Therefore, in the warehousing priority mode, warehousing can be carried out efficiently.

The delivery priority mode is an operation mode in which parking is performed with priority given to delivery. Specifically, this is a mode in which, when a vehicle stored in a storage area is transported and left, the transport means is operated so as to be able to handle the transport of the vehicle leaving the storage area. For example, a plurality of carts in the storage area are used preferentially for vehicles leaving the warehouse. The traveling trolley 63 can be used for warehousing when it is not used for warehousing. Further, the time required for leaving the warehouse may be shortened by storing the leaving vehicle near the leaving lift 65 (shortening the conveyance distance related to leaving the warehouse). In this way, the leaving priority mode is an operation mode in which leaving the vehicle as a priority while also putting the vehicle in the warehouse. Therefore, in the retrieval priority mode, it is possible to carry out warehousing while efficiently retrieving the vehicle.

The uniform warehousing mode is an operation mode in which warehousing is performed so that vehicles are stored spatially uniformly in the storage area. Specifically, when there are multiple stages in the storage area, vehicles are stored in each stage so that the number of vehicles in each stage is equal. Note that each stage may also be stored spatially uniformly. In this way, the uniform warehousing mode is an operation mode that suppresses uneven distribution of vehicle positions in the storage area. In the uniform warehousing mode, uneven distribution of vehicles can be suppressed, and, for example, it is possible to equalize the frequency of use of each device.

The vehicle type priority mode is a driving mode in which a specific vehicle type is preferentially stored in a predetermined area in the storage area. Specifically, if a predetermined area in the storage area is designed to store a specific type of vehicle, the specific type of vehicle is preferentially stored in the predetermined area. For example, if the specifications are designed so that high-roof vehicles can be stored (regular cars can also be stored), as in the third stage of FIG. 6, high-roof vehicles are preferentially stored in the third stage. In this manner, the vehicle type priority mode is a driving mode in which a corresponding vehicle type is preferentially stored in a corresponding area according to the specifications of the storage area. Therefore, in the vehicle type priority mode, it is possible to prevent a situation in which a specific vehicle type cannot be stored even though there is space in the storage area.

In addition to the above-mentioned operation mode, a cycle storage mode may be used. The cycle storage mode is an operation mode in which vehicles are stored in order in each stage (each layer) in the storage area. For example, after storing in the first stage, the next vehicle is stored in the second stage, after storing in the second stage, the next vehicle is stored in the third stage, and after storing in the third stage, the next vehicle is stored in the second stage. This is a mode in which storage is performed in a cycle of storing to the first stage.

Then, the estimating unit 31 outputs estimated values for the indicators in a predetermined time period in the future, corresponding to each driving mode as described above.

The future predetermined time slots are time slots set at equal intervals within a day (for example, every two hours), as shown in FIG. Note that since the time period is the unit in which the driving mode is set, it is not limited to the case where it is set as shown in FIG. 7. Further, the predetermined time period may be set depending on the usage status of the mechanical parking device 60. Specifically, the predetermined time period may be set based on temporal fluctuations in the number of vehicles in stock. For example, by switching the time period at a time when the time fluctuation is large (when the fluctuation value is larger than the threshold value) (in other words, within the time period, the fluctuation is smaller than the threshold value), you can change the time period when the number of vehicles in stock does not change significantly. , the driving mode can be determined. Alternatively, the predetermined time period may be set based on performance data using machine learning, which will be described later.

In this embodiment, a case will be described in which the average parking waiting time, the average exit waiting time, the number of times a plurality of predetermined devices in the mechanical parking device 60 are used, and the occupancy rate in the storage area are used as indicators for estimation. As the index, at least one of the average parking waiting time, the average exit waiting time, the number of times a plurality of predetermined devices in the mechanical parking device 60 are used, and the occupancy rate in the storage area may be used.

The average warehousing waiting time is the average value of the time required for warehousing. Specifically, the average warehousing waiting time is the average value of the time from when entry of a vehicle into the warehousing area is started (permitted) until the warehousing (transportation) of the vehicle is completed. The average value is calculated as the average value of the warehousing waiting time for a predetermined time period. Note that the average waiting time for warehousing is not limited to the above specific example as long as it is a time related to warehousing.

The average delivery waiting time is the average value of the time required for delivery. Specifically, the average exit waiting time is the average value of the time from the start of transport of the stored vehicles until the start (permission) of the vehicles transported to the exit area. The average value is calculated as the average value of the exit waiting time for a predetermined time period. Note that the average exit waiting time is not limited to the above specific example as long as it is a time related to exiting the warehouse.

The number of times a plurality of predetermined devices in the mechanical parking device 60 are used is the number of times each device used for transporting a vehicle is used. The predetermined device is a device designated in advance by an operator, maintenance person, or the like. Although the predetermined device may be set without being limited to any type, it is preferable that the predetermined device be the same type of device. For example, if the number of times of use of a plurality of traveling carts 63 (traveling carts 63 provided at each stage) is uneven, uneven wear will occur. In such a case, it may be necessary to perform maintenance individually or to plan the maintenance period according to the traveling trolley 63 that is the most worn. For this reason, it is preferable that each device wears out equally. For this reason, in the case of a three-stage configuration as shown in FIG. 6, by using the three traveling trolleys 63 as predetermined equipment, the number of times each traveling trolley 63 is used (the bias in the number of times each traveling trolley 63 is used) is evaluated. can do. The number of uses is the number of times each device is used in a predetermined time period. Note that the predetermined device may be a different type of device, and is not limited to the traveling trolley 63 as described above.

The vehicle occupancy rate in the storage area is the ratio of the number of stored vehicles to the number of vehicles that can be stored in the storage area (maximum number of parked vehicles). The occupancy rate may be the maximum occupancy rate in a predetermined time period, the minimum value, or the average value per predetermined time period. The higher the vehicle presence rate, the more efficiently the mechanical parking device 60 can receive vehicles.

The estimation unit 31 outputs estimated values of the above-mentioned indicators based on the performance data. The estimation method can be selected as appropriate as long as it is based on actual data. In this embodiment, we will explain a case where machine learning (supervised learning) is applied to output estimated values for each index, but other machine learning methods (such as reinforcement learning) may also be used. , the value corresponding to the index may be estimated on a rule basis, and the estimation method is not limited. A specific example of performing estimation by applying machine learning will be described below. Supervised learning is a learning method that models the characteristics of data using teacher data (data set to which correct answer information is connected).

Machine learning is performed based on performance data collected in the mechanical parking device 60. Note that performance data collected at other adjacent mechanical parking devices or other mechanical parking devices whose usage status is estimated to be similar may be integrated and used. The performance data may be stored in a storage section provided in the control device 30, or may be collected in another information processing device (server) and used by downloading or the like at the time of estimation. Perform supervised learning based on collected performance data.

In supervised learning, a set of input and result (label) data is given to a machine learning device to learn the features in the data set and generate a predictive model that estimates the result from the input. That is, the predictive model allows relationships in the learned data set to be acquired inductively. Machine learning can be realized using algorithms such as neural networks and SVM, which will be described later. Supervised learning is repeatedly performed by dividing performance data into predetermined units of time and using each divided unit as one input. The predetermined unit time is, for example, a predetermined time period in the future that is an estimation target.

The neural network is configured, for example, by a neural network modeled after a neuron model as shown in FIG. FIG. 8 is a schematic diagram showing an example of a neuron model. As shown in FIG. 8, the neuron outputs an output y in response to a plurality of inputs x (inputs x1, x2, x3). Each input x1 to x3 is multiplied by a weight w (w1, w2, w3) corresponding to each input x. As a result, the neuron outputs an output y expressed by the following equation (1). Note that in equation (1), the input xi (i=1, 2, 3), the output y, and the weight wi (i=1, 2, 3) are all vectors. Further, θ is a bias, and fk is an activation function.

When neurons like those shown in FIG. 8 are combined, the neural network shown in FIG. 9 is obtained, for example. FIG. 9 shows an example of a neural network with three layers of weights combining neurons. FIG. 9 is a schematic diagram showing a neural network having three layers of weights D1 to D3. As shown in FIG. 9, a plurality of inputs x (inputs x1 to x3) are input from the left side of the neural network, and results (y1 to y3) are output from the right side.

Specifically, the inputs x1 to x3 are multiplied by corresponding weights to each of the three neurons N11 to N13 before being input. In FIG. 9, the weights applied to the inputs are collectively labeled w1. Neurons N11 to N13 output z11 to z13, respectively. z11 to z13 are collectively referred to as a feature vector z1, and can be regarded as a vector obtained by extracting the feature amount of the input vector. This feature vector z1 is a feature vector between weight w1 and weight w2.

z11 to z13 are inputted after being multiplied by corresponding weights to each of the two neurons N21 and N22. In FIG. 9, the weights applied to the feature vectors are collectively labeled w2. Neurons N21 and N22 output z21 and z22, respectively. In FIG. 9, z21 and z22 are collectively labeled as a feature vector z2. This feature vector z2 is a feature vector between weight w2 and weight w3.

The feature vectors z21 and z22 are inputted after being multiplied by a weight corresponding to each of the three neurons N31 to N33. In FIG. 9, the respective weights applied to the feature vectors are collectively labeled w3. Then, neurons N31 to N33 output results y1 to y3, respectively.

Neural networks operate in a learning mode and a prediction mode. In the learning mode, weights w are learned using a learning data set (teacher data). In the prediction mode, the learned weight w is used to perform output (output of a value corresponding to the learned correct answer information).

Regarding learning, it is preferable to perform learning (online learning) each time teacher data (performance data) is collected or every time a certain amount of data is accumulated, but it is also possible to perform batch learning.

The weights w1 to w3 can be learned by backpropagation. Error information enters from the right side (output side) and flows to the left side (input side). The error backpropagation method is a method of adjusting (learning) the weights of each neuron so as to reduce the difference between the output y when the input x is input and the true output y (teacher). In this way, the weights are adjusted and a predictive model is generated so that an output (correct value) corresponding to the input is output based on the teacher data. Therefore, by inputting input to the prediction model, an output (estimated value corresponding to the correct value) can be obtained.

Neural networks can also have more than three layers (deep learning). Alternatively, dimension compression may be performed using an autoencoder.

The estimation unit 31 outputs an estimated value of a predetermined index using a prediction model generated by supervised learning. In this embodiment, the teacher data is performance data, and is information in which vehicle information, date and time, storage time, exit time, driving mode, storage position, and index are associated with each other. Among these, the inputs are vehicle information, date and time, storage time, exit time, driving mode, and storage position. Then, the output becomes an index (at least one of the average waiting time for warehousing, the average waiting time for leaving warehousing, the number of times of use, and the vehicle inventory rate).

As mentioned above, the performance data includes information about the driver of the vehicle, setting information of the car navigation system installed in the vehicle, stored parking time of the vehicle, information on the number of times it has been used, calendar information, weather information, and disaster information. When information such as (complementary parameters) is included, at least one of the complementary parameters may be added to the input data.

In this embodiment, a case will be described in which information in which vehicle information, date and time, storage time, exit time, driving mode, storage position, and index are associated is used as the teacher data. In the training data, data corresponding to the input (vehicle information, date and time, storage time, exit time, driving mode, and storage position) and data corresponding to the output (indices, average waiting time for entering, average waiting time for exiting, There is a correlation with at least one of the number of times of use and the vehicle occupancy rate. Specifically, the index is a value that changes according to the usage status of the mechanical parking device 60, and the data corresponding to the input (including the above-mentioned complementary parameters) is data indicating the usage status. In this way, since there is a correlation between the data corresponding to the input and the data corresponding to the output in the training data, by performing supervised learning based on the training data, it is possible to appropriately set the weight of each neuron. It can be adjusted to perform machine learning. The prediction model may be stored in a storage section provided in the control device 30, or may be stored in another information processing device (server).

Then, the estimation unit 31 uses a prediction model machine-learned using the teacher data to input the vehicle information, date, parking time, exit time, driving mode, and storage location into a predetermined time period. Outputs the estimated value of the index corresponding to. Once the predictive model is generated, the output is obtained according to the input. For this reason, input data (vehicle information, date, parking time, exit time, driving mode, and storage location) similar to those used as the data corresponding to the input of the training data are input into the prediction model, and the output is (estimated value of the indicator). By inputting a data set for each driving mode to the prediction model, it is possible to obtain an output corresponding to the driving mode. Specifically, by inputting a dataset corresponding to the warehousing priority mode (input data that includes the warehousing priority mode as an operation mode) to the prediction model, it is possible to output estimated values of indicators corresponding to the warehousing priority mode. can. Similarly, estimated values of corresponding indicators can be output for other driving modes.

Regarding the input data in the prediction mode, in the actual data, it may be each input data (vehicle model information, etc.) from the previous day that corresponds to the predetermined time period of the estimation target, or the corresponding day of the previous month that corresponds to the predetermined time period of the estimation target. Each input data (vehicle type information, etc.) may be used. That is, the input data in the prediction mode can be data whose usage status is estimated to be the same as the predetermined time period to be estimated. For example, if the predetermined time period to be estimated falls on a holiday, it is preferable to use data corresponding to the holiday. Even when input data is extracted and used from past performance data, by using a prediction model that reflects past performance, it is possible to obtain output that better reflects current usage conditions.

In the present embodiment, in the learning mode, a predetermined unit time is used as one input, and the performance data is divided and used as teacher data. Then, the unit time is set as a predetermined time period to be estimated. Therefore, even in the prediction mode, by associating input data with a predetermined time period (unit time), it is possible to obtain an estimated value of an index in a predetermined time period. However, if the unit time in the learning mode and the predetermined time period of the estimation target in the prediction mode are different, the output corresponding to the unit time is obtained by the input corresponding to the unit time, and the predetermined time period is calculated based on the output. It is also possible to calculate an estimated value of the index in the band.

Note that when a complementary parameter is added as input data of teacher data in the learning mode, the complementary parameter may be added to the input in the prediction mode.

By inputting the input to the prediction model in this way, the output is the estimated value of the index (at least one of the following: average waiting time for warehousing, average waiting time for leaving warehousing, number of uses, and inventory rate), and the training data (parameters used in ) can be obtained.

The estimation unit 31 obtains an output for each driving mode (one of the input parameters). The estimated value of the index corresponding to each driving mode, which is the output, is used in the determination unit 32 described later.

The determining unit 32 determines the driving mode in a predetermined time period based on the estimated value of the index for each driving mode. Specifically, the determining unit 32 compares the estimated values of the indexes estimated for each driving mode, and determines the driving mode with the most suitable index. Whether or not it is suitable is determined by a maintenance person or the like depending on the mechanical parking device 60. For example, when it is decided to select the driving mode in which the warehousing waiting time is estimated to be the lowest, the driving mode in which the estimated value of the index (average warehousing waiting time) is estimated to be the lowest is selected.

Furthermore, when using multiple types of indicators, evaluation may be performed using a radar chart with each indicator as an item. In such a case, for example, the estimated value of each index is scored (weighted) and charted. Regarding scoring, weighting may be adjusted so that a predetermined index to be emphasized has a high score. For example, if the average parking waiting time, vehicle availability rate, and bias in the number of times of use are used as indicators, radar charts such as those shown in FIGS. 10 and 11 are created. The bias in the number of uses is the degree of variation (degree of bias) in the estimated number of times each device is used, and is, for example, a standard deviation. It is assumed that FIG. 10 corresponds to the warehousing priority mode, and FIG. 11 corresponds to the warehousing priority mode. A comprehensive evaluation of multiple indicators can be made based on the area of the shape connecting the evaluation points of each item on the radar chart. The area of the chart in FIG. 10 is larger than the area of the chart in FIG. Therefore, it can be estimated that the warehousing priority mode is a more suitable operation mode.

In this way, the determining unit 32 determines the driving mode for each time period. Once each driving mode is determined, scheduling is performed as shown in FIG. 7, for example. The driving mode is appropriately selected according to each time zone and day of the week.

The modification unit 33 modifies the prediction model based on the output estimated value of the index and the value corresponding to the index obtained by actually performing warehousing. Although it is possible to estimate an index using a prediction model, the estimated index (output value) may differ from the value corresponding to the index (correct value) obtained by actually performing warehousing. In such cases, more accurate output values can be obtained by modifying the prediction model.

The correction is performed, for example, by backpropagation (error backpropagation method). In the prediction model, weights are set (learned) as shown in FIG. 9, and output is performed from input based on the weights. In backpropagation, each weight is modified based on the error between the output value and the correct value. Specifically, the error is returned in the opposite direction from the output side to correct the error of each neuron (correct the weight). Note that as long as the model can be modified based on the output value and the correct value, the present invention is not limited to backpropagation and can be applied.

In this way, since the prediction model can be modified to improve the learning accuracy, more accurate output can be obtained.

Next, an example of the scheduling process by the above-mentioned control device 30 will be described with reference to FIG. 12. FIG. 12 is a flowchart illustrating an example of a procedure for scheduling processing according to this embodiment. The flow shown in FIG. 12 is repeatedly executed at a predetermined control cycle.

First, an estimated value of an index is calculated for each driving mode based on performance data (S101). The indicators are the average waiting time for entering the parking lot, the average waiting time for leaving the parking lot, the number of times a plurality of predetermined devices in the mechanical parking device 60 are used, and the occupancy rate in the storage area. In S101, when estimation is performed by applying machine learning, an estimated value of the index is output based on a prediction model machine learned based on performance data.

Next, a driving mode in a predetermined time period is determined based on the estimated value of the index for each driving mode (S102). In S102, for example, as shown in FIG. 7, operation modes are scheduled for days of the week and times.

Then, information on the determined operation mode (scheduling information on the operation mode) is output (S103), and operation is performed. In S103, the already set scheduling information (for example, the scheduling information corresponding to the previous week) is updated with the newly set scheduling information, and operations are performed based on the updated scheduling information. Note that the scheduling information set in S102 may be referred to during operation.

In the flow of FIG. 12, scheduling is performed on a weekly basis (for example, FIG. 14), but the timing of scheduling is not limited to weekly units.

Although the case where a plurality of operation modes such as the warehousing priority mode are used has been described above, it is also possible to use other operation modes. Specifically, an operation mode related to the arrangement position of the conveying means may be used. The operation mode related to the arrangement position of the transport means includes a warehousing arrangement mode in which the transport means for transporting the vehicle is arranged based on the warehousing area into which the incoming vehicle enters, and a warehousing arrangement mode in which the transport means is arranged based on the exit area into which the outgoing vehicle enters. an intermediate placement mode in which the conveying means is placed at an intermediate position between the warehousing area and the leaving area; and an in-vehicle placement mode in which the conveying means is placed based on vehicle positions stored in the storage area. At least one of a warehousing arrangement mode, a warehousing arrangement mode, an intermediate arrangement mode, and a vehicle arrangement mode may be used. Specifically, the conveying means is a traveling cart 63. That is, each operation mode is an operation mode related to the arrangement position (standby position) of the traveling trolley 63.

In the warehousing arrangement mode, the traveling cart 63 is placed on standby near the warehousing lift 62 (for example, at a position connectable to the warehousing lift 62). According to the warehousing arrangement mode, warehousing can be completed more quickly. The delivery arrangement mode is a mode in which the traveling cart 63 is placed on standby near the delivery lift 65 (for example, at a position connectable to the delivery lift 65). According to the delivery arrangement mode, vehicles around the delivery lift 65 can be delivered earlier. As shown in FIG. 13, the intermediate placement mode is a mode in which the traveling cart 63 is placed on standby at an intermediate position between the warehousing lift 62 and the warehousing lift 65 (that is, the intermediate position of the storage shelf (carriage running path 64)). According to the intermediate arrangement mode, it is possible to efficiently deal with both warehousing and warehousing. The vehicle location mode is a mode in which the traveling trolley 63 is placed on standby near the vehicle position (vehicle location) stored in the storage area. In the vehicle location mode, it is preferable to have the traveling trolley 63 wait in the vicinity of an area where more vehicles are stored. According to the vehicle location mode, vehicles around the standby position can be taken out of the warehouse more efficiently.

The warehousing arrangement mode, the warehousing arrangement mode, the intermediate arrangement mode, and the vehicle arrangement mode may be used as driving modes in place of the above-mentioned warehousing priority mode, etc., or may be used in combination with the warehousing priority mode, etc. to set the driving mode. You can also do it. The combination includes, for example, an operation mode that is the warehousing priority mode and the warehousing arrangement mode.

As explained above, according to the driving planning system for a mechanical parking device according to the present embodiment, the driving planning method thereof, the driving planning program, and the mechanical parking device, performance data of vehicles entering and leaving the parking lot in a predetermined period in the past can be used. Since the estimated value of the index corresponding to each driving mode is output based on the above, it is possible to estimate the index corresponding to each driving mode by reflecting the usage status of the mechanical parking device 60. Since the operating mode of the mechanical parking device 60 in a future predetermined time period is determined based on the estimated index, it is possible to reflect the usage status of the mechanical parking device 60 and determine an appropriate operating mode. becomes. That is, since it is possible to determine an appropriate driving mode among the plurality of driving modes in response to changes in usage conditions, it is possible to more effectively demonstrate the capabilities of the mechanical parking device 60.

In addition, as an index to be estimated, at least one of the following: average parking waiting time, average parking waiting time, average parking waiting time in a predetermined time period in the future, the number of times a plurality of predetermined devices in the mechanical parking device 60 are used, and the occupancy rate in the storage area. By using , it becomes possible to conduct evaluations effectively by reflecting changes in the usage situation.

By including vehicle information, date and time, storage time, exit time, driving mode, and storage location in the performance data, it becomes possible to appropriately reflect changes in usage conditions.

A predictive model is machine learned using the information contained in the actual data (information that associates vehicle information, date and time, storage time, exit time, driving mode, storage position, and index) as training data. By using this, it is possible to effectively output estimated values of indicators using machine learning. When performing estimation, the estimated value of the index corresponding to a predetermined time period is output using vehicle information, date, parking time, exit time, driving mode, and storage location as input data, so the teacher You can get output from a predictive model trained on data.

By correcting the prediction model using the estimated value of the index and the value corresponding to the index (that is, the correct value) obtained by actually performing warehousing, it becomes possible to output a more accurate estimated value.

In other words, it is possible to drive according to the usage situation, and it is possible to more effectively demonstrate the capabilities of the mechanical parking device 60.

The present disclosure is not limited to the above-described embodiments, and various modifications can be made without departing from the gist of the invention.

The driving planning system for a mechanical parking device, the driving planning method thereof, the driving planning program, and the mechanical parking device described in each of the embodiments described above are understood as follows, for example.
The operation planning system for a mechanical parking device (60) according to the present disclosure is an operation planning system applied to a mechanical parking device (60) that allows vehicles to enter and exit the parking lot. Based on the performance data of , the average parking waiting time, average exit waiting time, and multiple predetermined devices in the mechanical parking device (60) in a predetermined time period in the future are determined corresponding to each of the plurality of driving modes related to vehicle transportation. an estimation unit (31) that outputs an estimated value for at least one index of the number of times the vehicle has been used and the vehicle occupancy rate in the storage area; A determination unit (32) that determines the driving mode is provided.

According to the driving planning system according to the present disclosure, the mechanical parking device (60) It becomes possible to estimate the index corresponding to each driving mode by reflecting the usage status of the vehicle. Then, in order to determine the driving mode of the mechanical parking device (60) in a future predetermined time period based on the estimated index, an appropriate driving mode is determined by reflecting the usage status of the mechanical parking device (60). It becomes possible to do so. In other words, since it is possible to determine an appropriate driving mode among a plurality of driving modes in response to changes in usage conditions, it is possible to more effectively demonstrate the capabilities of the mechanical parking device (60). Become.

In addition, as an index to be estimated, at least one of the following: average parking waiting time, average exit waiting time, number of times a plurality of predetermined devices in the mechanical parking device (60) are used, and vehicle occupancy rate in the storage area in a predetermined time period in the future. By using one, it becomes possible to reflect changes in the usage situation and perform evaluations effectively.

In the driving planning system according to the present disclosure, the performance data may include vehicle information, date and time, storage time, exit time, driving mode, and storage location.

According to the driving planning system according to the present disclosure, the actual data includes vehicle information, date and time, storage time, exit time, driving mode, and storage location, so that changes in usage conditions can be appropriately handled. It becomes possible to reflect.

In the driving planning system according to the present disclosure, the performance data includes information in which vehicle information, date and time, storage time, exit time, driving mode, storage position, and index are associated with each other. The estimating unit (31) calculates the vehicle information, the date, the parking time, the leaving time, the driving mode, and the storage position using a predictive model machine-learned using the information as training data. As the input data, an estimated value of the index corresponding to the predetermined time period may be output.

According to the driving planning system according to the present disclosure, information included in performance data (information in which vehicle information, date and time, storage time, exit time, driving mode, storage position, and index are associated) By using a predictive model that has been machine learned using the data as training data, it is possible to effectively output estimated values of indicators using machine learning. When performing estimation, the estimated value of the index corresponding to a predetermined time period is output using vehicle information, date, parking time, exit time, driving mode, and storage location as input data, so the teacher You can get output from a predictive model trained on data.

In the driving planning system according to the present disclosure, the performance data may include information regarding a driver of the vehicle.

According to the driving planning system according to the present disclosure, since the performance data includes information regarding the driver of the vehicle, it is possible to appropriately reflect changes in the usage situation. The information regarding the driver of the vehicle may include, for example, at least one of gender, age, physical characteristics, number of passengers, and whether or not the driver has children.

In the driving planning system according to the present disclosure, the performance data may include setting information of a car navigation system installed in the vehicle.

According to the driving planning system according to the present disclosure, since the performance data includes setting information of the car navigation system installed in the vehicle, it becomes possible to appropriately reflect changes in usage conditions. The setting information of the car navigation system installed in the vehicle can include, for example, at least one of the address of the driver, the destination, and whether or not the position of the mechanical parking device (60) is set in the navigation system. .

In the driving planning system according to the present disclosure, the performance data may include at least one of the stored parking time of the vehicle and information on the number of times of use.

According to the driving planning system according to the present disclosure, by including at least one of the stored parking time of the vehicle and the number of usage information in the performance data, it is possible to appropriately reflect changes in the usage status. Become.

In the operation planning system according to the present disclosure, the performance data may include at least one of calendar information, weather information, and disaster information.

According to the operation planning system according to the present disclosure, by including at least one of calendar information, weather information, and disaster information in the performance data, it is possible to appropriately reflect changes in usage conditions.

In the driving planning system according to the present disclosure, the driving modes include a warehousing priority mode in which warehousing is prioritized, a warehousing priority mode in which warehousing is prioritized over exit, and a warehousing priority mode in which vehicles are stored spatially uniformly in a storage area. The system may include at least one of a uniform warehousing mode in which warehousing is performed, and a vehicle type priority mode in which a specific vehicle type is preferentially stored in a predetermined area in the storage area.

According to the operation planning system according to the present disclosure, the operation mode of the mechanical parking device (60) can be appropriately set according to the parking.

In the driving planning system according to the present disclosure, the driving mode includes a warehousing arrangement mode in which a transport means for transporting vehicles is arranged based on a warehousing area into which an incoming vehicle enters, and a warehousing arrangement mode in which transport means for transporting vehicles are arranged based on an exit area into which an outgoing vehicle is driven. a delivery arrangement mode in which the means is arranged, an intermediate arrangement mode in which the transport means is arranged at an intermediate position between the storage area and the delivery area, and a delivery arrangement mode in which the transport means is arranged based on the vehicle position stored in the storage area. It may also be at least one of the vehicle location modes.

According to the driving planning system according to the present disclosure, the driving mode of the mechanical parking device (60) can be appropriately set depending on the arrangement position of the conveyance means.

In the operation planning system according to the present disclosure, the predetermined time period may be set based on temporal fluctuations in the number of vehicles in stock.

According to the operation planning system according to the present disclosure, since the predetermined time period for determining the operation mode is set based on the temporal fluctuation in the number of vehicles in stock, it is possible to set the time period more appropriately. For example, by switching the time period at a time when the time fluctuation is large (when the fluctuation value is larger than the threshold value) (in other words, within the time period, the fluctuation is smaller than the threshold value), you can change the time period when the number of vehicles in stock does not change significantly. , the driving mode can be determined.

In the operation planning system according to the present disclosure, a modification unit (33 ) may be provided.

According to the operation planning system according to the present disclosure, a more accurate prediction model is corrected using the estimated value of the index and the value corresponding to the index obtained by actually executing warehousing (i.e., the correct value). It becomes possible to output estimated values.

A mechanical parking device (60) according to the present disclosure includes a warehousing area, a storage area for storing a vehicle, a conveying means for transporting a vehicle between the warehousing area and the storage area, and the mechanical parking device (60) described above. ) is equipped with an operation planning system.

The operation planning method for a mechanical parking device (60) according to the present disclosure is an operation planning method applied to a mechanical parking device (60) that allows vehicles to enter and exit the parking lot, and includes vehicles that have entered and exited the parking lot during a predetermined period in the past. Based on the performance data of , the average parking waiting time, average exit waiting time, and multiple predetermined devices in the mechanical parking device (60) in a predetermined time period in the future are determined corresponding to each of the plurality of driving modes related to vehicle transportation. a step of outputting an estimated value for at least one index of the number of times of use of the vehicle and a vehicle presence rate in a storage area, and determining the driving mode in the predetermined time period based on the estimated value of the index for each driving mode. and a step of determining.

The operation plan program for the mechanical parking device (60) according to the present disclosure is an operation plan program applied to the mechanical parking device (60) that allows vehicles to enter and exit the parking lot, and includes vehicles that have entered and exited the parking lot during a predetermined period in the past. Based on the performance data of , the average parking waiting time, average exit waiting time, and multiple predetermined devices in the mechanical parking device (60) in a predetermined time period in the future are determined corresponding to each of the plurality of driving modes related to vehicle transportation. The driving mode in the predetermined time period is determined based on the estimated value of at least one of the indicators of the number of times of use and the vehicle occupancy rate in the storage area, and the estimated value of the indicator for each driving mode. The computer executes the process to be determined.

11: CPU
12:ROM
13: RAM
14: Hard disk drive 15: Communication section 18: Bus 22: Operation panel 30: Control device 31: Estimating section 32: Determining section 33: Modifying section 60: Mechanical parking device 61a to 61l: Traverse conveyor 62: Incoming lift 63: Traveling bogie 64: Bogie running path 65: Outgoing lift 66: Storage floor 70: Getting on/off room 70a: Incoming/outgoing berth 70b: Outgoing berth N11 to N13, N21 to N23, N31 to N33: Neuron

Claims (15)

A driving planning system applied to a mechanical parking device that allows vehicles to enter and exit the parking lot,
Based on the actual data of vehicles entering and exiting the parking lot during a predetermined period in the past, the average parking waiting time, average waiting time for exiting the parking lot, and mechanical parking system are calculated based on the actual data of vehicles that entered and exited the parking lot during a predetermined period in the past, corresponding to each of multiple driving modes related to vehicle transportation. an estimation unit that outputs an estimated value for at least one index of the number of uses of a plurality of predetermined devices provided in the storage area and the vehicle presence rate in the storage area;
a determining unit that determines the driving mode in the predetermined time period based on the estimated value of the index for each driving mode;
Equipped with
The performance data includes vehicle information, date and time, parking time, exit time, driving mode, and storage position of the mechanical parking device. A driving planning system applied to a mechanical parking device that allows vehicles to enter and exit the parking lot,
Based on the actual data of vehicles entering and exiting the parking lot during a predetermined period in the past, the average parking waiting time, average waiting time for exiting the parking lot, and mechanical parking system are calculated based on the actual data of vehicles that entered and exited the parking lot during a predetermined period in the past, corresponding to each of multiple driving modes related to vehicle transportation. an estimation unit that outputs an estimated value for at least one index of the number of uses of a plurality of predetermined devices provided in the storage area and the vehicle presence rate in the storage area;
a determining unit that determines the driving mode in the predetermined time period based on the estimated value of the index for each driving mode;
Equipped with
The performance data includes information in which vehicle information, date and time, storage time, exit time, the driving mode, the storage position, and the index are associated with each other,
The estimator uses a predictive model machine-learned using the information as training data, and uses the vehicle information, date, parking time, exit time, driving mode, and storage location as input data. An operation planning system for a mechanical parking device that outputs an estimated value of the index corresponding to a predetermined time period. The driving planning system for a mechanical parking device according to claim 1 or 2 , wherein the performance data includes information regarding a driver of the vehicle. The driving planning system for a mechanical parking device according to claim 1 , wherein the performance data includes setting information of a car navigation system installed in the vehicle. 5. The operation planning system for a mechanical parking device according to claim 1 , wherein the performance data includes at least one of stored vehicle parking time and usage frequency information. The operation planning system for a mechanical parking device according to claim 1 , wherein the performance data includes at least one of calendar information, weather information, and disaster information. The operation modes include a warehousing priority mode that prioritizes warehousing, a warehousing priority mode that prioritizes unloading, a uniform warehousing mode that stores vehicles spatially uniformly in the storage area, and a predetermined warehousing mode in which vehicles are stored spatially uniformly in the storage area. 7. The driving planning system for a mechanical parking device according to claim 1, further comprising at least one vehicle type priority mode in which a specific vehicle type is preferentially stored in a region. The operation mode includes a warehousing arrangement mode in which the transport means for transporting vehicles is arranged based on a warehousing area into which an incoming vehicle enters, an exit arrangement mode in which the transport means is arranged based on an exit area into which an outgoing vehicle enters; At least one of an intermediate arrangement mode in which the transport means is arranged at an intermediate position between the warehousing area and the outgoing area, and a vehicle arrangement mode in which the transport means is arranged based on the vehicle position stored in the storage area. An operation planning system for a mechanical parking device according to any one of claims 1 to 7 . The operation planning system for a mechanical parking device according to any one of claims 1 to 8 , wherein the predetermined time period is set based on a temporal change in the number of parked vehicles. The mechanical parking according to claim 2 , further comprising a modification unit that modifies the prediction model based on the output estimated value of the index and the value corresponding to the index obtained by actually performing parking. Equipment operation planning system. Incoming goods area,
a storage area for storing vehicles;
a conveyance means for conveying a vehicle between the storage area and the storage area;
The mechanical parking device according to any one of claims 1 to 10 includes a driving planning system;
Mechanical parking device equipped with. A driving planning method applied to a mechanical parking device that allows vehicles to enter and exit the parking lot, the method comprising:
Based on the actual data of vehicles entering and exiting the parking lot during a predetermined period in the past, the average parking waiting time, average waiting time for exiting the parking lot, and mechanical parking system are calculated based on the actual data of vehicles that entered and exited the parking lot during a predetermined period in the past, corresponding to each of multiple driving modes related to vehicle transportation. an estimation step of outputting an estimated value for at least one index of the number of uses of a plurality of predetermined devices provided in the storage area and the vehicle presence rate in the storage area;
a determining step of determining the driving mode in the predetermined time period based on the estimated value of the index for each driving mode;
has
The performance data includes vehicle information, date and time, parking time, exit time, driving mode, and storage position . An operation planning program applied to a mechanical parking device that allows vehicles to enter and exit the parking lot,
Based on the actual data of vehicles that entered and left the parking lot during a predetermined period in the past, the average parking waiting time, average waiting time for leaving the parking lot, and mechanical parking system in a predetermined time period in the future are calculated based on the actual data of vehicles that entered and exited the parking lot during a predetermined period in the past. an estimation process that outputs an estimated value for at least one index of the number of uses of a plurality of predetermined devices provided in the storage area and the vehicle presence rate in the storage area;
A determination process that determines the driving mode in the predetermined time period based on the estimated value of the index for each driving mode;
make the computer run
The performance data is an operation plan program for the mechanical parking device that includes vehicle information, date and time, parking time, exit time, driving mode, and storage position . A driving planning method applied to a mechanical parking device that allows vehicles to enter and exit the parking lot, the method comprising:
Based on the actual data of vehicles that entered and left the parking lot during a predetermined period in the past, the average parking waiting time, average waiting time for leaving the parking lot, and mechanical parking system in a predetermined time period in the future are calculated based on the actual data of vehicles that entered and exited the parking lot during a predetermined period in the past. an estimation step of outputting an estimated value for at least one index of the number of uses of a plurality of predetermined devices provided in the storage area and the vehicle presence rate in the storage area;
a determining step of determining the driving mode in the predetermined time period based on the estimated value of the index for each driving mode;
has
The performance data includes information in which vehicle information, date and time, storage time, exit time, the driving mode, the storage position, and the index are associated with each other,
In the estimation step, a machine learning prediction model is used with the information as training data, and the vehicle information, date, parking time, exit time, driving mode, and storage location are input data. An operation planning method for a mechanical parking device that outputs an estimated value of the index corresponding to a predetermined time period. An operation planning program applied to a mechanical parking device that allows vehicles to enter and exit the parking lot,
Based on the actual data of vehicles entering and exiting the parking lot during a predetermined period in the past, the average parking waiting time, average waiting time for exiting the parking lot, and mechanical parking system are calculated based on the actual data of vehicles that entered and exited the parking lot during a predetermined period in the past, corresponding to each of multiple driving modes related to vehicle transportation. an estimation process that outputs an estimated value for at least one index of the number of times a plurality of predetermined devices installed in the vehicle are used and the vehicle presence rate in the storage area;
A determination process that determines the driving mode in the predetermined time period based on the estimated value of the index for each driving mode;
make the computer run
The performance data includes information in which vehicle information, date and time, storage time, exit time, the driving mode, the storage position, and the index are associated with each other,
In the estimation process, a machine learning prediction model is used with the information as training data, and the vehicle information, date, parking time, exit time, driving mode, and storage location are input data. An operation planning program for a mechanical parking device that outputs an estimated value of the index corresponding to a predetermined time period.

Images