Operation plan generation system, method, and program

Abstract

To provide an operation plan generation system capable of accurately generating an operation plan of pre-cooling or pre-heating, and a method and program thereof.SOLUTION: An operation plan generation system in an analysis server 24 includes a generation unit which generates, for an object refrigerator being one of each of refrigerators for transportation of a plurality of vehicles A1, A2, and A3, an operation plan of pre-cooling or pre-heating in the object refrigerator on the basis of operation information acquired from each of the refrigerator for transportation corresponding to the object refrigerator.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an operation plan generation system, a method, and a program therefor. [Background technology]

[0002] Vehicles such as refrigerated trucks and trailers equipped with transport refrigeration units transport cargo at a constant temperature to a destination (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2004-251508 Summary of the Invention [Problem to be solved by the invention]

[0004] In order for vehicles to transport goods at constant temperatures, the van body (cargo bed) on which the goods are carried must be pre-cooled (or pre-heated) to a set temperature. For example, low-temperature items such as ice cream and frozen foods must be pre-cooled to a set temperature of -10°C or below. This pre-cooling process takes time, so the driver of the vehicle must turn on the freezer in advance to begin pre-cooling. Using an on-timer for the freezer could be considered, but this would require manual setting according to the situation, which can be time-consuming.

[0005] The present disclosure has been made in consideration of the above circumstances, and aims to provide an operation plan generation system, method, and program that can generate a pre-cooling or pre-heating operation plan with high accuracy. [Means for solving the problem]

[0006] A first aspect of the present disclosure is a generation unit that generates a pre-cooling or pre-heating operation plan for a target refrigerator using a trained model that has been machine-learned using operation information acquired from each of the transport refrigerators corresponding to the target refrigerator, with one of a plurality of transport refrigerators being set as the target refrigerator. and an input unit into which input data is input, wherein the generation unit generates, as the operation plan, an operation start time or a required time for bringing the target refrigerator to a predetermined set temperature at a predetermined completion time, and the generation unit selects the trained model that has been machine-learned corresponding to the target refrigerator to generate the operation plan, and the trained model is machine-learned using teacher data in which a set temperature, a completion time at which pre-cooling or pre-heating to the set temperature is completed, and the operation start time at which pre-cooling or pre-heating to the set temperature is started or the required time required for pre-cooling or pre-heating to the set temperature are associated as the operation information, and the predetermined completion time and the predetermined set temperature input by the input unit are used as input data, and the operation start time or the required time corresponding to the input data is output. This is an operation plan generation system.

[0007] In a second aspect of the present disclosure, one of a plurality of transport refrigeration units is designated as a target refrigeration unit, and a trained model is machine-learned using operation information acquired from each of the transport refrigeration units corresponding to the target refrigeration unit to generate an operation plan for pre-cooling or pre-heating the target refrigeration unit. Generate process and an input process for inputting input data, wherein the generation process generates, as the operation plan, an operation start time or a required time for bringing the target refrigerator to a predetermined set temperature at a predetermined completion time, and the generation process selects the trained model that has been machine-learned corresponding to the target refrigerator to generate the operation plan, and the trained model is machine-learned using teacher data in which, as the operation information, a set temperature, a completion time at which pre-cooling or pre-heating to the set temperature is completed, and the operation start time at which pre-cooling or pre-heating to the set temperature is started or the required time required for pre-cooling or pre-heating to the set temperature are associated with each other, and the predetermined completion time and the predetermined set temperature input in the input process are used as input data, and the operation start time or the required time corresponding to the input data is output. A computer-implemented method for generating an operation plan.

[0008] A third aspect of the present disclosure is a method for generating a pre-cooling or pre-heating operation plan for a target refrigerator using a trained model that is machine-learned using operation information acquired from each of the transport refrigerators corresponding to the target refrigerator, with one of a plurality of transport refrigerators being a target refrigerator. Generate process and an input process for inputting input data, wherein the generation process generates, as the operation plan, an operation start time or a required time for bringing the target refrigerator to a predetermined set temperature at a predetermined completion time, and the generation process selects the trained model that has been machine-learned corresponding to the target refrigerator to generate the operation plan, and the trained model is machine-learned using teacher data in which a set temperature, a completion time at which pre-cooling or pre-heating to the set temperature is completed, and the operation start time at which pre-cooling or pre-heating to the set temperature is started or the required time required for pre-cooling or pre-heating to the set temperature are associated as the operation information, and the predetermined completion time and the predetermined set temperature input by the input process are used as input data, and the operation start time or the required time corresponding to the input data is output. This is an operation plan generation program. [Effects of the Invention]

[0009] According to the present disclosure, an effect is achieved in that a pre-cooling or pre-heating operation plan can be generated with high accuracy. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram showing a schematic configuration of an entire system according to a first embodiment of the present disclosure. [Figure 2] FIG. 2 is a schematic configuration diagram showing an example of a hardware configuration of an analysis server according to the first embodiment of the present disclosure. [Figure 3] 1 is a functional block diagram showing functions of an operation plan generation system according to a first embodiment of the present disclosure. [Figure 4] FIG. 1 is a schematic diagram illustrating an example of a neuron model. [Figure 5]FIG. 1 is a schematic diagram illustrating an example of a neural network. [Figure 6] 5 is a flowchart illustrating an example of a procedure of an operation plan generation process according to the first embodiment of the present disclosure. [Figure 7] FIG. 4 is a functional block diagram showing functions of an operation plan generation system according to a second embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] [First embodiment] A first embodiment of an operation plan generation system, method, and program according to the present disclosure will be described below with reference to the drawings.

[0012] FIG. 1 is a diagram showing a schematic configuration of an entire system to which a driving plan generation system 10 according to a first embodiment of the present disclosure is applied. In this embodiment, the driving plan generation system 10 is provided in an analysis server 24 in a server group 22. The server group 22 is capable of communicating with the vehicle side and the user side. The communication method is not limited to, but may be, for example, an Internet line. For example, cellular communication or LPWA can be used. In this embodiment, a case will be described in which the driving plan generation system 10 is provided as the analysis server 24, but it may also be installed on the vehicle side, and the device on which the functions of the driving plan generation system 10 are installed is not limited.

[0013] In this embodiment, the operation plan generation system 10 will be described as generating an operation plan for pre-cooling a transport refrigeration unit, but it may also be configured to generate an operation plan for pre-heating. That is, the operation plan generation system 10 can generate an operation plan for pre-cooling or pre-heating a transport refrigeration unit. When generating an operation plan for pre-heating, the same processing as when generating an operation plan for pre-cooling, which will be described below, can be performed.

[0014] In FIG. 1, the vehicle side shows vehicle A1, vehicle A2, and vehicle A3. In the following description, when it is necessary to distinguish between the vehicles, they will be referred to as vehicle A1, vehicle A2, and vehicle A3, and when it is not necessary to distinguish between the vehicles, they will be referred to as vehicle 1. Vehicle 1 is a truck, and a box-shaped loading platform (hereinafter referred to as "van body") for carrying luggage is provided behind the cab. Vehicle 1 is also equipped with a transport refrigeration unit (land transport refrigeration unit). That is, in vehicle 1, the interior (interior) of the van body can be frozen using the transport refrigeration unit.

[0015] The transport refrigeration unit is controlled by a controller (B1, B2, B3) provided in, for example, the cab. For example, the controllers (B1, B2, B3) are configured with settings such as operation start and set temperature, and the refrigeration cycle of the transport refrigeration unit is controlled based on these settings.

[0016] Vehicle 1 is capable of communicating with the server group 22. For example, a controller of vehicle 1 is capable of communicating with the server group 22 via an Internet line. Vehicle A1 is connected to the server group 22 via a communication device C1 such as a smartphone. Vehicle A2 is connected to the server group 22 via an in-vehicle device C2. Note that if the in-vehicle device C2 is an in-vehicle device of another company, it may be connected to the server group 22 via the other company's server 21. Vehicle A3 is connected to the server group 22 via an in-vehicle communication device C3. The above is an example of a communication method between vehicle 1 and server group 22, and other communication methods may also be used.

[0017] The user side is, for example, an information processing terminal used by the user. The information processing terminal is, for example, a PC 27, a smartphone 28, a tablet, etc. Each information processing terminal is capable of exchanging information with the server group 22.

[0018] The server group 22 is made up of multiple servers. In this embodiment, an analysis server 24, a server 23, and a mail server 26 are provided. The server 23 exchanges information with other servers (24, 26), the vehicle, and the user. The server 23 may exchange information with the user via a website 25. The mail server 26 is a server capable of sending mail and provides information to the user via mail.

[0019] The analysis server 24 is equipped with the functions of the operation plan generation system 10 and is capable of generating an operation plan for a transportation refrigeration unit.

[0020] Fig. 2 is a schematic diagram showing an example of a hardware configuration of an analysis server 24 according to an embodiment of the present disclosure. As shown in Fig. 2, the analysis server 24 is a so-called computer, and includes, for example, a CPU (Central Processing Unit) 11, a main memory 12, a storage unit 13, an external interface 14, a communication interface 15, an input unit 16, and a display unit 17. These units are connected to each other directly or indirectly via a bus, and work together to execute various processes.

[0021] The CPU 11 performs overall control using, for example, an OS (Operating System) stored in a storage unit 13 connected via a bus, and executes various programs stored in the storage unit 13 to perform various processes.

[0022] The main memory 12 is composed of writable memory such as cache memory or RAM (Random Access Memory), and is used as a working area for reading out programs executed by the CPU 11 and writing data processed by the programs.

[0023] The storage unit 13 is a non-transitory computer readable storage medium, such as a read only memory (ROM), a hard disk drive (HDD), or a flash memory. The storage unit 13 stores, for example, an OS for controlling the entire device, such as Windows (registered trademark), iOS (registered trademark), or Android (registered trademark), a basic input / output system (BIOS), various device drivers for operating peripheral devices as hardware, various application software, and various data and files. The storage unit 13 also stores programs for implementing various processes and various data required for implementing the various processes.

[0024] The external interface 14 is an interface for connecting to an external device. Examples of external devices include an external monitor, a USB memory, an external HDD, etc. Although only one external interface is shown in the example shown in FIG. 1, multiple external interfaces may be provided.

[0025] The communication interface 15 functions as an interface for connecting to a network to communicate with other devices and sending and receiving information. For example, the communication interface 15 communicates with other devices via wired or wireless communication. Examples of wireless communication include Bluetooth (registered trademark), Wi-Fi, and communication using a dedicated communication protocol. An example of wired communication is a wired LAN (Local Area Network).

[0026] The input unit 16 is a user interface for giving instructions, such as a keyboard, a mouse, or a touchpad.

[0027] Display unit 17 is, for example, a liquid crystal display, an organic EL (Electroluminescence) display, etc. Display unit 17 may also be a touch panel display on which a touch panel is superimposed.

[0028] 3 is a functional block diagram showing functions of the driving plan generation system 10 in the analysis server 24. As shown in FIG. 3, the driving plan generation system 10 includes a data collection unit 31, a generation unit 32, and a transmission unit 33.

[0029] The functions realized by these units are realized, for example, by processing circuitry. For example, a series of processes for realizing the functions shown below are stored in the storage unit 13 in the form of a program (for example, a route teaching data creation program), and the CPU 11 reads this program into the main memory 12 and executes information processing and calculation processing to realize various functions.

[0030] The program may be pre-installed in the storage unit 13, provided in a state stored in another computer-readable storage medium, or distributed via wired or wireless communication means, etc. Examples of computer-readable storage media include magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, and semiconductor memories.

[0031] The data collection unit 31 collects driving information from each vehicle 1. That is, in the data collection unit 31, driving information corresponding to each vehicle 1 is stored.

[0032] The operating information is information that indicates the operating status of the transport refrigeration unit. In particular, the operating information includes information that indicates the operating status of the transport refrigeration unit for pre-cooling. The operating information may also include information that indicates the refrigeration capacity of the transport refrigeration unit. Specifically, the operating information includes the temperature inside the van body and / or the outside air temperature. In this embodiment, an example will be described in which the temperature inside the van body is included. The operating information also includes information that associates the set temperature for pre-cooling, the completion time when pre-cooling to the set temperature is completed, and the operation start time when pre-cooling to the set temperature begins. Note that the operation start time may be the time required for pre-cooling to the set temperature.

[0033] The data collection unit 31 periodically acquires driving information from each vehicle 1 and stores it as performance data. This allows pre-cooling performance data for each vehicle 1 to be collected.

[0034] The generation unit 32 generates a pre-cooling operation plan. Specifically, the generation unit 32 selects one of the multiple transport refrigeration units as the target refrigeration unit. Then, the generation unit 32 generates a pre-cooling operation plan for the target refrigeration unit based on operation information acquired from each transport refrigeration unit that corresponds to (is similar to) the target refrigeration unit. The pre-cooling operation plan is an operation plan based on predetermined conditions. Specifically, the operation plan is the pre-cooling operation start time or required time for the target refrigeration unit to reach a predetermined set temperature at a predetermined completion time. In this embodiment, a case will be described in which an operation start time is generated as an operation plan. For example, if it is desired to pre-cool the inside temperature to -10°C (predetermined set temperature) at 9:00 a.m. the next morning (predetermined completion time), the time to start pre-cooling in the target refrigeration unit (for example, 7:00 a.m.) is estimated.

[0035] For this reason, the generation unit 32 generates an operation plan using a trained model that has been machine-learned. The trained model will be described later.

[0036] The generation unit 32 generates an operation plan based on the operation information acquired from each transport refrigeration unit corresponding to the target refrigeration unit. In other words, the generation unit 32 uses the operation information of each transport refrigeration unit (including itself) that is similar to the target refrigeration unit.

[0037] The generation unit 32 acquires target refrigerator data. The target refrigerator data may be characteristic information (equipment information) related to the target refrigerator, or may be location information. The generation unit 32 then sets each transport refrigerator corresponding to the target refrigerator based on the characteristic information or location information related to the target refrigerator.

[0038] Specifically, the characteristic information includes manufacturer information for the target refrigeration unit, model information for the target refrigeration unit, specification information for the van body in which the target refrigeration unit is installed, and information about the age of the van body. Specification information for the van body includes the size of the van body (loading capacity, height, depth, etc.). When characteristic information is used, a range of characteristic information is set that is expected to have a pre-cooling situation similar to the characteristic information for the target refrigeration unit. Then, the operating information of other transport refrigeration units that fall within this range is also used.

[0039] The location information is the location information of the target refrigeration unit and the location information of the location where pre-cooling is performed by the target refrigeration unit. When using location information, a range of location information is set that is expected to be close to the location information of the target refrigeration unit. Then, the operating information of other transport refrigeration units included in this range is also used.

[0040] In this embodiment, the generation unit 32 generates an operation plan using a trained model. For this reason, it is preferable that multiple trained models are prepared based on the combination of each target chiller. Then, the generation unit 32 selects a trained model that has been trained using the operating information of each transport chiller with a similar pre-cooling situation based on the target chiller data, and executes generation of an operation plan. If training time for the trained model is available, it is also possible to set each corresponding transport chiller based on the target chiller data, and execute generation of an operation plan after learning using the operating information of each transport chiller.

[0041] Next, we will explain the machine-learned trained model. As described above, the generation unit 32 generates a driving start time (driving plan) using a trained model that has been machine-learned. The trained model is trained using supervised learning. Supervised learning is a learning method that models the characteristics of data using training data (a data set associated with correct answer information).

[0042] In supervised learning, by providing pairs of input and result (label) data, the system learns the features of the dataset and generates a trained model (predictive model) that estimates the result from the input. In other words, the trained model can inductively acquire the relationships in the trained dataset. Machine learning can be achieved using algorithms such as neural networks and SVMs, which will be described later.

[0043] A neural network is configured, for example, by a neural network modeled after a neuron model as shown in FIG. 4. FIG. 4 is a schematic diagram showing an example of a neuron model. As shown in FIG. 4, a neuron outputs an output y in response to a plurality of inputs x (inputs x1, x2, x3). Each of the inputs 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. The output is expressed by an activation function or the like using the inputs x, weights w, and bias θ.

[0044] Combining neurons such as those in Figure 4 results in the neural network shown in Figure 5, for example. Figure 5 shows an example of a neural network with three weighted layers made up of combined neurons. Figure 5 is a schematic diagram showing a neural network with three weighted layers D1 to D3. As shown in Figure 5, multiple 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.

[0045] Specifically, inputs x1 to x3 are multiplied by corresponding weights and input to three neurons N11 to N13. In FIG. 5, the weights multiplied by the inputs are collectively labeled w1. Neurons N11 to N13 output z11 to z13, respectively. z11 to z13 are collectively labeled feature vector z1 and can be considered as a vector extracted from the feature quantities of the input vector. This feature vector z1 is a feature vector between weights w1 and w2.

[0046] z11 to z13 are multiplied by the corresponding weights and input to two neurons N21 and N22. In FIG. 5, the weights multiplied by the feature vectors are collectively labeled w2. Neurons N21 and N22 output z21 and z22, respectively. In FIG. 5, z21 and z22 are collectively labeled as feature vector z2. This feature vector z2 is a feature vector between weights w2 and w3.

[0047] The feature vectors z21 and z22 are multiplied by the corresponding weights and input to three neurons N31 to N33. In Fig. 5, the weights multiplied by the feature vectors are collectively labeled w3. Then, neurons N31 to N33 output results y1 to y3, respectively.

[0048] Neural networks operate in two modes: learning mode and prediction mode. In learning mode, weights w are learned using a learning dataset (teacher data). In prediction mode, the learned weights w are used to output (output a value corresponding to the learned correct answer information) for the input.

[0049] For example, weights w1 to w3 can be learned using backpropagation. Error information enters from the right side (output side) and flows to the left side (input side). Backpropagation is a technique for adjusting (learning) each weight for each neuron so as to reduce the difference between the output y when input x is input and the true output y (teacher). In this way, a trained model is generated by adjusting the weights so that an output (correct value) corresponding to the input is output based on the teacher data. Therefore, by inputting an input to the trained model, an output (an estimated value corresponding to the correct value) can be obtained.

[0050] Neural networks can also have more than three layers (deep learning).

[0051] The generation unit 32 outputs the operation start time using a trained model generated by supervised learning. The training data uses data collected by the data collection unit 31. Specifically, the training data associates the set temperature (target temperature) for pre-cooling, the completion time when pre-cooling to the set temperature is completed, and the operation start time when pre-cooling to the set temperature begins (or the required time required for pre-cooling to the set temperature). In other words, past pre-cooling operation records are used as training data. Appropriately set training data may also be used instead of operating records. The input is the set temperature and completion time. The output is the operation start time (or required time).

[0052] Since machine learning is performed using past pre-cooling operation records as training data, the weights of each neuron can be appropriately adjusted to perform machine learning.

[0053] Then, using a trained model machine-learned using teacher data, the generation unit 32 takes the desired set temperature and completion time (i.e., the scheduled set temperature and completion time) as input data and outputs the operation start time corresponding to this input data. For example, if you want to pre-cool the inside temperature to -10°C (set temperature) at 9:00 a.m. the next day (completion time), by inputting 9:00 a.m. (completion time) and -10°C (set temperature), the generation unit 32 outputs the operation start time (e.g., 7:00 a.m.) that will complete pre-cooling at the completion time.

[0054] Other operating information may be used as input data for the training data. For example, the refrigerator temperature (and / or outside air temperature) may be used. In this case, the set temperature, the completion time, the operation start time, and the refrigerator temperature at this operation start time may be associated with each other to form the training data. Machine learning is performed using this training data to generate a trained model. Then, in prediction mode, the desired set temperature, the completion time, and the predicted value of the refrigerator temperature are used as input data, and the operation start time corresponding to this input data is output.

[0055] The transmission unit 33 transmits the driving plan to a predetermined terminal. The predetermined terminal may be a terminal on the user's side or a terminal in each vehicle.

[0056] For example, when an operation start time (operation plan) is generated as described above, the user is notified of the operation start time, and pre-cooling operation begins automatically at the operation start time. The user may also set the pre-cooling start timer of the transport refrigeration unit based on the operation start time. In this way, generating the operation start time makes it easier for the user to organize their work.

[0057] Next, an example of an operation plan generation process by the above-described operation plan generation system 10 will be described with reference to Fig. 6. Fig. 6 is a flowchart showing an example of the procedure of an operation plan generation process according to this embodiment.

[0058] First, a target refrigeration unit is set (S101). Next, a trained model that has been machine-learned based on the operating information of each transport refrigeration unit corresponding to the target refrigeration unit is selected (S102).

[0059] Next, input data is input to the trained model (S103). The input data may be, for example, a desired set temperature and a completion time.

[0060] Next, the operation start time corresponding to the input data is output from the trained model (S104).

[0061] In this way, the trained model created through machine learning is used to generate the driving start time as part of the driving plan.

[0062] In the above example, a case where an operation plan is generated using a trained model that has been machine-learned has been described, but the method is not limited to using a trained model as long as a future pre-cooling operation plan is generated based on collected operation information. Specifically, an operation plan may be generated using a rule-based method, simulation method, or other method based on operation information (not limited to information related to pre-cooling) obtained from each transport refrigeration unit that corresponds to the target refrigeration unit (having similar pre-cooling conditions).

[0063] As described above, the operation plan generation system, method, and program according to this embodiment can generate a pre-cooling or pre-heating operation plan for a target chiller unit based on operation information acquired from each transport chiller unit corresponding to the target chiller unit. By using operation information for multiple transport chillers, the accuracy of operation plan generation can be improved. Furthermore, even for multiple transport chillers, by using operation information for each transport chiller unit corresponding to the target chiller unit, the accuracy of the operation plan for the target chiller unit can be improved.

[0064] By generating an operation plan, the user's work efficiency is improved.

[0065] Second Embodiment Next, an operation plan generation system, a method, and a program according to a second embodiment of the present disclosure will be described. The operation plan generation system, method, and program according to this embodiment will be described below, focusing mainly on the differences from the first embodiment.

[0066] As shown in FIG. 7, the operation plan generation system 10 according to this embodiment includes an estimation unit .

[0067] The estimation unit 34 estimates the planned completion time or the planned remaining time for completing pre-cooling to the set temperature as the completion schedule based on the operating state of the target refrigerator. In this embodiment, a case where the planned remaining time for completion is estimated as the completion schedule will be described, but the same can be applied to the planned completion time.

[0068] The operating conditions are parameters that affect pre-cooling. Specifically, the operating conditions are at least one of the following: the internal temperature (temperature inside the van body), the external temperature (temperature outside the van body), the refrigerant temperature of the compressor in the refrigeration cycle of the transport refrigeration unit, the pressure of the compressor, and the rotation speed of the compressor. The refrigerant temperature of the compressor is the temperature of the refrigerant at the suction or discharge part of the compressor. The pressure of the compressor is the pressure at the suction or discharge part of the compressor.

[0069] In this embodiment, the estimation unit 34 generates the scheduled remaining time to completion using a trained model that has been machine-learned. For example, the estimated remaining time to completion until the set temperature (-10°C) is output as 1 hour or the like. The trained model will be described later.

[0070] The estimation process by the estimation unit 34 is performed, for example, after the start of pre-cooling operation. That is, the estimation unit 34 estimates the planned remaining time to completion based on the actual operating state of the target refrigerator after the start of pre-cooling operation. For example, by starting pre-cooling at the operation start time generated by the generation unit 32, it is possible to predict that the predetermined set temperature will be reached at the predetermined completion time, but by estimating the planned remaining time to completion based on the actual operating state of the target refrigerator after the start of pre-cooling operation, it is possible to more accurately estimate the time to complete pre-cooling.

[0071] Next, we will explain the machine-learned trained model. The trained model is the same as in Figures 4 and 5. The estimation unit 34 outputs the planned remaining time to completion using the trained model generated by supervised learning. Specifically, the training data corresponds the operating state, set temperature, and planned remaining time to completion of the target refrigerator. The training data may be the operating history of the planned remaining time to completion corresponding to the operating state and set temperature, or may be the planned remaining time to completion set corresponding to the operating state and set temperature. The training data may be composed of operating information acquired from the target refrigerator, or may be composed of operating information of a transportation refrigerator other than the target refrigerator. The input is the operating state and set temperature. The output is the planned remaining time to completion.

[0072] Then, using a trained model learned by machine learning using teacher data, the estimation unit 34 inputs the current operating state of the target refrigerator and the set temperature that is the pre-cooling target, and outputs the planned remaining time to completion corresponding to this input data. For example, if you want to pre-cool the refrigerator temperature to -10°C (set temperature) at 9:00 AM the next morning (completion time) and pre-cooling starts at 7:00 AM (operation start time), and pre-cooling is likely to be completed quickly based on the actual operating state, the planned remaining time to completion is output as, for example, one hour left until pre-cooling is completed.

[0073] In the above example, a case where a completion schedule is generated using a trained model that has been machine-learned has been described, but the method is not limited to using a trained model as long as the completion schedule is generated based on the operating state. Specifically, the completion schedule may be generated using a rule-based method, a simulation method, or the like based on the operating state of the target chiller.

[0074] As described above, the operation plan generation system, method, and program according to this embodiment can more accurately estimate the scheduled completion time or the scheduled remaining time for completing pre-cooling to the set temperature based on the operating state of the target chiller. By training the trained model using training data in which the operating state and the completion schedule are associated, it is possible to use the operating state as input data and output a completion schedule corresponding to the input data.

[0075] The present disclosure is not limited to the above-described embodiments, and various modifications are possible within the scope of the invention. It is also possible to combine the various embodiments.

[0076] The operation plan generation system, method, and program described in each of the above-described embodiments can be understood, for example, as follows. The operation plan generation system (10) according to the present disclosure includes a generation unit (32) that generates an operation plan for pre-cooling or pre-heating for one of a plurality of transport refrigeration units as a target refrigeration unit based on operation information acquired from each of the transport refrigeration units corresponding to the target refrigeration unit.

[0077] According to the operation plan generation system of the present disclosure, a pre-cooling or pre-heating operation plan for a target refrigerator can be generated based on operation information acquired from each transport refrigerator corresponding to the target refrigerator. By using operation information for multiple transport refrigerators, the accuracy of the generation of the operation plan can be improved. Furthermore, even for multiple transport refrigerators, by using operation information for each transport refrigerator corresponding to the target refrigerator, the accuracy of the operation plan for the target refrigerator can be improved. By generating an operation plan, the user's work efficiency is improved.

[0078] In the operation plan generation system according to the present disclosure, the generation unit may generate the operation plan based on the operation information of each of the transport refrigeration units corresponding to the target refrigeration units, which is set based on characteristic information about the target refrigeration units.

[0079] According to the operation plan generation system according to the present disclosure, it is possible to effectively set each transport refrigeration unit corresponding to the target refrigeration unit based on characteristic information related to the target refrigeration unit.

[0080] In the operation plan generation system according to the present disclosure, the characteristic information may be at least one of manufacturer information of the target refrigerator, model information of the target refrigerator, specification information of the van body in which the target refrigerator is installed, and aging information of the van body.

[0081] According to the operation plan generation system of the present disclosure, by setting the characteristic information to at least one of manufacturer information of the target refrigeration unit, model information of the target refrigeration unit, specification information of the van body in which the target refrigeration unit is installed, and aging information of the van body, it is possible to effectively set each transport refrigeration unit corresponding to the target refrigeration unit.

[0082] In the operation plan generation system according to the present disclosure, the generation unit may generate the operation plan based on the operation information of each of the transport refrigeration units corresponding to the target refrigeration units, which is set based on location information regarding the target refrigeration units.

[0083] According to the operation plan generation system according to the present disclosure, each transport refrigeration unit corresponding to the target refrigeration unit can be effectively set based on the location information related to the target refrigeration unit.

[0084] In the operation plan generation system according to the present disclosure, the generation unit may generate, as the operation plan, an operation start time or a required time for achieving a predetermined set temperature at a predetermined completion time.

[0085] According to the operation plan generation system of the present disclosure, the operation plan is set as the operation start time or required time to reach a specified set temperature at a specified completion time, allowing the user to understand the pre-cooling or pre-heating operation plan and improving work efficiency.

[0086] In the operation plan generation system according to the present disclosure, the generation unit generates the operation plan using a trained model that has been machine-learned, and the trained model has been machine-learned using training data that associates, as the operation information, a set temperature, a completion time at which pre-cooling or pre-heating to the set temperature is completed, and the operation start time at which pre-cooling or pre-heating to the set temperature begins or the required time required for pre-cooling or pre-heating to the set temperature, and the completion time and the set temperature are used as input data, and the operation start time or the required time corresponding to the input data is output.

[0087] According to the operation plan generation system of the present disclosure, by training a learned model using training data that corresponds the set temperature, the completion time when pre-cooling or pre-heating to the set temperature is completed, and the operation start time when pre-cooling or pre-heating to the set temperature begins or the required time required for pre-cooling or pre-heating to the set temperature, it is possible to use the completion time and set temperature as input data and output the operation start time or required time corresponding to the input data.

[0088] The operation plan generation system according to the present disclosure may include an estimation unit (34) that estimates, as a completion schedule, a planned completion time or a planned remaining time for completing pre-cooling or pre-heating to the set temperature based on the operating state of the target refrigerator.

[0089] According to the operation plan generation system of the present disclosure, the scheduled completion time or the scheduled remaining time for completing pre-cooling or pre-heating to the set temperature can be more accurately estimated based on the operating state of the target refrigerator.

[0090] In the operation plan generation system according to the present disclosure, the estimation unit generates the completion schedule using a trained model that has been machine-learned, and the trained model has been machine-learned using training data that corresponds the operating state, the set temperature, and the completion schedule, and the operation state and the set temperature are used as input data, and the completion schedule corresponding to the input data is output.

[0091] According to the operation plan generation system of the present disclosure, by training a learned model using teacher data in which an operation state and a completion schedule are associated, it is possible to use the operation state as input data and output a completion schedule corresponding to the input data.

[0092] In the operation plan generation system according to the present disclosure, the operating state may be at least one of an internal temperature, an outside temperature, a refrigerant temperature of a compressor, a pressure of the compressor, and a rotation speed of the compressor.

[0093] According to the operation plan generation system of the present disclosure, the completion schedule can be more accurately estimated by defining the operating state as at least one of the internal temperature, the external temperature, the compressor refrigerant temperature, the compressor pressure, and the compressor rotation speed.

[0094] The driving plan generation system according to the present disclosure may include a transmission unit (31) that transmits the driving plan to a predetermined terminal.

[0095] According to the driving plan generation system according to the present disclosure, by transmitting the driving plan to a predetermined terminal, for example, a user can understand the driving plan.

[0096] The operation plan generation method according to the present disclosure includes a step of generating an operation plan for pre-cooling or pre-heating for one of a plurality of transport refrigeration units as a target refrigeration unit based on operation information acquired from each of the transport refrigeration units corresponding to the target refrigeration unit.

[0097] The operation plan generation program according to the present disclosure causes a computer to execute a process of generating a pre-cooling or pre-heating operation plan for one of a plurality of transport refrigeration units as a target refrigeration unit based on operation information acquired from each of the transport refrigeration units corresponding to the target refrigeration unit. [Explanation of symbols]

[0098] 1, A1, A2, A3: Vehicle 10: Operation plan generation system 11: CPU 12: Main memory 13: Storage section 14: External interface 15: Communication interface 16: Input section 17:Display section 21: Other company's server 22: Server group 23: Server 24: Analysis Server 25:Website 26: Mail server 27: PC 28: Smartphone 31: Data collection section 32: Generation part 33: Transmission unit 34:Estimation part C1: Communication equipment C2: Onboard equipment C3: In-vehicle communication equipment

Claims

1. a generation unit that generates a pre-cooling or pre-heating operation plan for one of the plurality of transport refrigeration units as a target refrigeration unit using a trained model that has been machine-learned using operating information acquired from each of the transport refrigeration units corresponding to the target refrigeration unit; an input unit to which input data is input, the generation unit generates, as the operation plan, an operation start time or a required time for causing the target refrigerator to reach a predetermined set temperature at a predetermined completion time; the generation unit selects the machine-learned trained model corresponding to the target chiller and generates the operation plan; The trained model is machine-learned using training data in which the operation information is associated with a set temperature, a completion time at which pre-cooling or pre-heating to the set temperature is completed, and the operation start time at which pre-cooling or pre-heating to the set temperature is started or the required time required for pre-cooling or pre-heating to the set temperature, and the predetermined completion time and the predetermined set temperature input by the input unit are used as input data, and the trained model outputs the operation start time or the required time corresponding to the input data. Driving plan generation system.

2. 2. The operation plan generation system according to claim 1, wherein the generation unit generates the operation plan based on the operation information of each transport refrigeration unit corresponding to the target refrigeration unit that is set based on characteristic information about the target refrigeration unit.

3. 3. The operation plan generation system according to claim 2, wherein the characteristic information is at least one of manufacturer information of the target chiller, model information of the target chiller, specification information of a van body in which the target chiller is installed, and aging information of the van body.

4. 4. The operation plan generation system according to any one of claims 1 to 3, wherein the generation unit generates the operation plan based on the operation information of each transport refrigeration unit corresponding to the target refrigeration unit that is set based on location information related to the target refrigeration unit.

5. 5. The operation plan generation system according to claim 1, further comprising an estimation unit that estimates, as a completion schedule, a planned completion time or a planned remaining time for completing pre-cooling or pre-heating to the set temperature based on an operating state of the target refrigerator.

6. the estimation unit generates the completion schedule using a second trained model that has been machine-learned; The operation plan generation system of claim 5, wherein the second trained model is machine-learned using training data in which the operating state, the set temperature, and the completion schedule are associated with each other, and the second trained model uses the operating state and the set temperature as input data and outputs the completion schedule corresponding to the input data.

7. 7. The operation plan generation system according to claim 5, wherein the operating state is at least one of an internal temperature, an external temperature, a refrigerant temperature of a compressor, a pressure of the compressor, and a rotation speed of the compressor.

8. The driving plan generation system according to claim 1 , further comprising a transmitting unit that transmits the driving plan to a predetermined terminal.

9. a generation step of generating a pre-cooling or pre-heating operation plan for one of the plurality of transport refrigeration units as a target refrigeration unit using a trained model that has been machine-learned using operating information acquired from each of the transport refrigeration units corresponding to the target refrigeration unit; an input step of inputting input data; the generating step generates, as the operation plan, an operation start time or a required time for the target refrigerator to reach a predetermined set temperature at a predetermined completion time; the generation step selects the trained model that has been machine-learned and corresponds to the target chiller, and generates the operation plan; The trained model is machine-learned using training data in which the operation information is associated with a set temperature, a completion time at which pre-cooling or pre-heating to the set temperature is completed, and the operation start time at which pre-cooling or pre-heating to the set temperature is started or the required time required for pre-cooling or pre-heating to the set temperature, and the predetermined completion time and the predetermined set temperature input in the input step are used as input data, and the trained model outputs the operation start time or the required time corresponding to the input data. A computer-implemented method for generating a driving plan.

10. a generation process of generating a pre-cooling or pre-heating operation plan for one of the plurality of transport refrigeration units as a target refrigeration unit using a trained model that has been machine-learned using operating information acquired from each of the transport refrigeration units corresponding to the target refrigeration unit; and causing the computer to execute an input process in which input data is input. the generation process generates, as the operation plan, an operation start time or a required time for the target refrigerator to reach a predetermined set temperature at a predetermined completion time; the generation process selects the machine-learned trained model corresponding to the target chiller and generates the operation plan; The trained model is machine-learned using training data in which the operation information is associated with a set temperature, a completion time at which pre-cooling or pre-heating to the set temperature is completed, and the operation start time at which pre-cooling or pre-heating to the set temperature is started or the required time required for pre-cooling or pre-heating to the set temperature, and the predetermined completion time and the predetermined set temperature input by the input processing are used as input data, and the trained model outputs the operation start time or the required time corresponding to the input data. Driving plan generation program.

Images