WO2021085874A1

WO2021085874A1 - Method for providing cargo vehicle dispatch service using o2o-based big data and artificial intelligence

Info

Publication number: WO2021085874A1
Application number: PCT/KR2020/013341
Authority: WO
Inventors: 김민규
Original assignee: 김민규
Priority date: 2019-10-29
Filing date: 2020-09-29
Publication date: 2021-05-06
Also published as: KR102101708B1

Abstract

Provided is a method for providing a cargo vehicle dispatch service using O2O-based big data and artificial intelligence, comprising the steps of: receiving, from at least one vehicle owner terminal and at least one shipper terminal, vehicle data and cargo data; by using previously established case-based artificial intelligence big data, generating answer data for queries of a cargo search event of the at least one vehicle owner terminal and a vehicle search event of the at least one shipper terminal; and transmitting a list included in each piece of the answer data to the at least one shipper terminal and the at least one vehicle owner terminal.

Description

Method of providing cargo dispatch service using O2O-based big data and artificial intelligence

The present invention relates to a method for providing a freight dispatch service using O2O-based big data and artificial intelligence, and provides a method for linking each party of a freight transport contract using big data modeled and learned based on a case.

The knowledge and information revolution based on computers and the Internet is progressing through the innovation of the industrial structure system that superconnects people, objects and spaces, and superintelligifies people, objects and spaces as the trend of the 4th industrial revolution. Even digital operation is difficult. The communication between the borrower, the transport agent, the call center and the freighter is mainly made up of the telephone, and it remains at the level of writing various data manually or sharing a single Excel file with multiple people. The freight transport market is mainly made up of cash transactions, and it is very difficult to collect various data.

At this time, a platform that connects borrowers and shippers using artificial intelligence and smart environment has been researched and developed. In this regard, Korean Patent Registration No. 10-1981475 (announced May 23, 2019) and Korean Patent Publication No. 2013-0082877 (published on July 22, 2013), grouping by location, industry, and grade to form a freight dispatch matching network, shipper's transport condition information, dispatch request information, freight information network, freight transport cost, After collecting the vehicle owner's operation record data and storing it as big data for freight dispatch, the configuration provides real-time big data for freight dispatch, and includes information on the departure date, arrival date, departure place, destination, quantity, weight and volume of the cargo. It manages cargo information and collects vehicle information including vehicle location, loadable weight, and volume information, so that cargo and vehicles are dispatched according to the user's request for dispatch, and the carrier, shipper, and vehicle owner are interconnected. Each configuration is disclosed.

However, even with the above configuration, if the shipper cannot find a borrower who trusts and can leave the cargo due to the nature of mass transportation, the possibility of risk is increased, and the borrower does not pay the price in a timely manner or You still run the risk of meeting a shipper who doesn't pay for it properly. In addition, even if the above-described brokerage is used, anxiety about whether the vehicle owner properly transports from the origin to the destination remains. Furthermore, the current freight vehicle transport business law requires some transport companies to report freight rates and charges, but the current law does not have a means to compel compliance with the declared freight rates other than the reporting obligation, so the volume of cargo is limited compared to many freight forwarders. In the existing freight transport market, the shipper has no choice but to have a dominant position, and based on this dominant position, the shipper actually decides the core of the market factors, such as freight rates. In the real world of transactions, the multi-level transport structure is prevalent due to the smallness of the company, the closedness of transaction information, and the oversupply of the transport service provider and the transport agent, and the freight rate is determined due to dumping force by the shipper with market power. The reality is that the damage is being passed on to the final carrier.

An embodiment of the present invention includes transport, storage, and unloading information so as to reduce the risk burden of borrowers and shippers, and to end issues between major market parties regarding the controversial freight rate system in the freight car transport market. Improve the freight transport structure that provides unrealistic freight rates while providing transparent services for the entire logistics process based on the openness of transaction information, and mediating the optimal parties based on big data and case-based neural network artificial intelligence learning. And it can provide a method of providing freight dispatch service using O2O-based big data and artificial intelligence that enables the transportation service provider's profit structure and quality transportation service to be provided. However, the technical problem to be achieved by the present embodiment is not limited to the technical problem as described above, and other technical problems may exist.

As a technical means for achieving the above-described technical problem, an embodiment of the present invention provides the step of receiving vehicle data and cargo data from at least one vehicle owner terminal and at least one shipper terminal. Generating response data to a query of at least one cargo search event of at least one owner's terminal and a vehicle search event of at least one shipper's terminal using the data, and at least one shipper of a list each included in the answer data And transmitting to the terminal and at least one vehicle owner terminal.

According to any one of the above-described problem solving means of the present invention, transportation and storage to reduce the risk burden of the borrower and shipper, and to end the issue between the major parties in the market regarding the controversial freight rate system in the freight vehicle transportation market. , An unrealistic freight structure is produced while intermediating the optimal party based on the openness of transaction information and transparently proceeding with the integrated service for the entire logistics process, which includes loading and unloading information, and mediating the optimal party based on big data and case-based neural network artificial intelligence learning. It can improve the structure of freight transport, provide the profit structure and quality transport service of the transport service provider, improve the profit structure of freight owners, improve the structure of the freight transport market and reinforce the public interest functions of freight transport.

1 is a view for explaining a cargo dispatch service providing system using O2O-based big data and artificial intelligence according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a cargo dispatch service providing server included in the system of FIG. 1.

3 to 6 are views for explaining an embodiment in which a freight dispatch service using O2O-based big data and artificial intelligence according to an embodiment of the present invention is implemented.

7 is a flowchart illustrating a method of providing a freight dispatch service using O2O-based big data and artificial intelligence according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is said to be "connected" with another part, this includes not only "directly connected" but also "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, it means that other components may be further included, and one or more other features, not excluding other components, unless specifically stated to the contrary. It is to be understood that it does not preclude the presence or addition of any number, step, action, component, part, or combination thereof.

The terms "about", "substantially", and the like, as used throughout the specification, are used in or close to the numerical value when manufacturing and material tolerances specific to the stated meaning are presented, and are used to provide an understanding of the present invention. To assist, accurate or absolute numerical values are used to prevent unreasonable use of the stated disclosure by unscrupulous infringers. As used throughout the specification of the present invention, the term "step (to)" or "step of" does not mean "step for".

In the present specification, the term "unit" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Further, one unit may be realized by using two or more hardware, or two or more units may be realized by one piece of hardware.

In the present specification, some of the operations or functions described as being performed by a terminal, device, or device may be performed instead in a server connected to the terminal, device, or device. Likewise, some of the operations or functions described as being performed by the server may also be performed by a terminal, device, or device connected to the server.

In this specification, some of the operations or functions described as mapping or matching with the terminal means mapping or matching the unique number of the terminal or the identification information of the individual, which is the identification information of the terminal. Can be interpreted as.

In the present specification, cargo means luggage or cargo, which is transported by means of transportation such as automobiles, motorcycles, trains, and ships. In the present specification, vehicles, motorcycles, cargo drones, and autonomous vehicles for cargo driving may be used as means of transport for cargo, and the owners of vehicles and motorcycles as well as cargo drone owners and cargo driving autonomous vehicle owners Can be.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a view for explaining a cargo dispatch service providing system using O2O-based big data and artificial intelligence according to an embodiment of the present invention. Referring to FIG. 1, a cargo dispatch service providing system 1 using O2O-based big data and artificial intelligence includes at least one shipper terminal 100, a cargo dispatch service providing server 300, and at least one vehicle owner terminal 400. ), and at least one host terminal 500. However, since the cargo dispatch service providing system 1 using O2O-based big data and artificial intelligence of FIG. 1 is only an embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1.

In this case, each component of FIG. 1 is generally connected through a network 200. For example, as shown in FIG. 1, at least one shipper terminal 100 may be connected to a cargo dispatch service providing server 300 through a network 200. In addition, the freight dispatch service providing server 300 may be connected to at least one shipper terminal 100, at least one vehicle owner terminal 400, and at least one carrier terminal 500 through the network 200. In addition, at least one vehicle owner terminal 400 may be connected to at least one shipper terminal 100, a cargo dispatch service providing server 300, and at least one carrier terminal 500 through the network 200. . In addition, the at least one carrier terminal 500 may be connected to at least one shipper terminal 100, a cargo dispatch service providing server 300, and at least one vehicle owner terminal 400 through the network 200.

Here, the network refers to a connection structure in which information exchange is possible between each node, such as a plurality of terminals and servers, and examples of such networks include RF, 3rd Generation Partnership Project (3GPP) network, and Long Term (LTE). Evolution) network, 5GPP (5th Generation Partnership Project) network, WIMAX (World Interoperability for Microwave Access) network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network) , Personal Area Network (PAN), Bluetooth (Bluetooth) network, NFC network, satellite broadcasting network, analog broadcasting network, Digital Multimedia Broadcasting (DMB) network, and the like, but are not limited thereto.

In the following, the term'at least one' is defined as a term including the singular number and the plural number, and even if the term'at least one' does not exist, each component may exist in the singular or plural, and may mean the singular or plural. It will be self-evident. In addition, it will be possible to change according to the embodiment that each component is provided in the singular or plural.

The at least one shipper terminal 100 may be a shipper's terminal requesting freight transportation using a web page, an app page, a program, or an application related to a freight dispatch service using O2O-based big data and artificial intelligence. At this time, the at least one shipper terminal 100 may be a terminal that registers basic information of the size, volume, weight, origin, and destination of the cargo. In addition, at least one shipper terminal 100, based on a list having at least one level and a result of analyzing the propensity through the contract cumulative history of the shipper terminal 100, from the cargo dispatch service providing server 300 at least one It may be a terminal for which the borrower terminal 100 of is recommended.

Here, the at least one shipper terminal 100 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like. At this time, the at least one shipper terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one shipper terminal 100, for example, as a wireless communication device that is guaranteed portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) ) All types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs may be included.

The freight dispatch service providing server 300 may be a server that provides a freight dispatch service web page, an app page, a program, or an application using O2O-based big data and artificial intelligence. In addition, the cargo dispatch service providing server 300 may be a server that receives data on cargo and vehicles from the shipper terminal 100 and the vehicle owner terminal 400, respectively, and converts the data into a database. In addition, the cargo dispatch service providing server 300 is a server that searches for and recommends a shipper through data about a vehicle input from the vehicle owner terminal 400 or a transport history log when searching for cargo transport from the vehicle owner terminal 400 Can be And, the cargo dispatch service providing server 300, when a contract between the owner and the shipper is established, monitors the driver and the cargo from departure to arrival, and when the monitoring result matches the alarm condition, at least one report center It could be a server that shares the results. In addition, the freight dispatch service providing server 300 collects evaluation data on the shipper and the borrower from the shipper terminal 400 and the shipper terminal 100 after the transport is completed, and It may be a server used as input data for analysis of propensity. In addition, the freight dispatch service providing server 300 may be a server that builds big data using raw data including accumulated history logs or contract details, and then a search event occurs from the borrower or shipper. In this case, it may be a server that generates a recommendation list as a result by inputting a search event to the learned big data as a query. In addition, the freight dispatch service providing server 300 may be a server that mediates at least one owner terminal 400 from the at least one carrier terminal 500 with the at least one shipper terminal 100. And, the cargo dispatch service providing server 300, among at least one of the vehicle owner terminals 400, post a registration examination for the vehicle owner terminal 400 requesting the registration of the crew, and, if the registration is determined, a notice for recruiting crew members, It may be a server that clusters crew members and crew members, analyzes the propensity of crew members and crew members based on the recruited crew members and ongoing contracts and results, and converts them into a database. Likewise, the freight dispatch service providing server 300 may be a server that performs interconnection and matching based on the propensity of the crew chief and the crew chief when the recommendation of the shipper or the crew chief is performed after constructing big data.

Here, the freight dispatch service providing server 300 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like.

The at least one vehicle owner terminal 400 may be a vehicle owner's terminal using a cargo dispatch service related web page, app page, program, or application using O2O-based big data and artificial intelligence. In this case, when requesting a Crew head among at least one vehicle owner terminal 400, it is defined as a Crew head terminal, but the same reference numerals are used. Likewise, a crew-one terminal that is supported as a crew-one among at least one vehicle-owner terminal 400 and is bundled in one cluster is also the vehicle-owner terminal 400, so the same reference numerals are used. At this time, the at least one vehicle owner terminal 400 may be a terminal of a vehicle owner that registers basic information such as a type, a loading volume, a loading weight, and a destination of a freight vehicle possessed by the freight dispatch service providing server 300. In addition, the at least one borrower terminal 400 may be a terminal that searches for a freight forwarding case to the freight dispatch service providing server 300, and a borrower who enters into a contract with any one shipper terminal 100 of the searched recommended list It can be a terminal. In addition, the at least one vehicle owner terminal 400 may be a terminal connected to the shipper terminal 100 by the carrier terminal 500 or the crew terminal 400. In addition, the at least one vehicle owner terminal 400 may be a terminal serving as a hub connecting at least one monitoring device (not shown) and the cargo dispatch service providing server 300 through a wireless network, but a hub or an access point If is installed separately, it may not play a hub role.

Here, the at least one vehicle owner terminal 400 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like. In this case, the at least one vehicle owner terminal 400 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one vehicle owner terminal 400, for example, as a wireless communication device that is guaranteed portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) ) All types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs may be included.

At least one carrier terminal 500 is a cargo dispatch service providing server using at least one vehicle owner terminal 400 using a web page, app page, program or application related to cargo dispatch service using O2O-based big data and artificial intelligence. It may be a terminal of the host company provided as (300).

Here, the at least one host terminal 500 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like. At this time, the at least one host terminal 500 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one carrier terminal 500, for example, as a wireless communication device that is guaranteed portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular) , PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband) Internet) terminal, smart phone (smartphone), smart pad (smartpad), it may include all kinds of handheld (Tablet PC) based wireless communication devices such as.

2 is a block diagram illustrating a freight dispatch service providing server included in the system of FIG. 1, and FIGS. 3 to 6 are freight dispatch services using O2O-based big data and artificial intelligence according to an embodiment of the present invention. A diagram for explaining an embodiment in which is implemented.

Referring to Figure 2, the cargo dispatch service providing server 300, an input unit 310, a generation unit 320, a transmission unit 330, an estimate unit 340, a transport assistance unit 350, a crew management unit 360 , And a risk management unit 370.

The cargo dispatch service providing server 300 according to an embodiment of the present invention or another server (not shown) operating in conjunction with at least one shipper terminal 100, at least one vehicle owner terminal 400, and at least one When transmitting a cargo dispatch service application, program, app page, web page, etc. using O2O-based big data and artificial intelligence to the carrier terminal 500, at least one shipper terminal 100, at least one owner terminal 400 ), and at least one carrier terminal 500 may install or open a cargo dispatch service application, program, app page, web page, etc. using O2O-based big data and artificial intelligence. In addition, a service program may be driven in at least one shipper terminal 100, at least one vehicle owner terminal 400, and at least one carrier terminal 500 by using a script executed in a web browser. Here, the web browser is a program that enables you to use the web (WWW: world wide web) service, which means a program that receives and displays hypertext described in HTML (hyper text mark-up language). For example, Netscape , Explorer, chrome, etc. In addition, the application refers to an application on the terminal, and includes, for example, an app running on a mobile terminal (smart phone).

Referring to FIG. 2, the input unit 310 may receive vehicle data and cargo data from at least one vehicle owner terminal 400 and at least one shipper terminal 100. At this time, since the types, types, volumes, and weights of cargo are diverse, and the waiting time, loading and unloading difficulty, cargo turnover, and volume scale are different, receiving basic data from the shipper and borrower determines the quality of matching or recommendation afterwards. It is an important factor. Of course, in the case of the shipper, the delivery of cargo of similar size, type, shape, etc. is consigned or the borrower is mostly small, so there is a high possibility that there is only one freight vehicle, so the initially entered data can be used as a default for future recommendations. . In addition, the fundamental reason for the distortion of freight rates is in the structure of the freight transport market. In the freight transport market, information related to freight transport is not delivered fairly to transport operators, transport brokers, and transport affiliates. The information asymmetry that actually monopolizes information on this can be resolved.

The generation unit 320 uses pre-built case-based artificial intelligence big data to query the cargo search event of the at least one owner terminal 400 and the vehicle search event of the at least one shipper terminal 100. Answer data can be generated. In this case, the case-based artificial intelligence big data that has been previously constructed may be big data that has been learned using Long Short-Term Memory models (LSTM) among Recurrent Neural Networks (RNNs). RNN is one of the models of artificial neural networks. It is a neural network model that considers the current input along with the previous input as input, and is an algorithm suitable for time series data learning. The existing general neural network model can be said to be independent of the input order because it only processed one current input as an input, but since the RNN considers both the current input and the previous input, it exhibits a property dependent on the input order. Therefore, it may be suitable for the problem of inferring the propensity of the current contract from the previous contract in matching in consideration of the propensity of the shipper or the borrower, as in an embodiment of the present invention. Recurrent, which is the word representing R of the RNN, means that the same operation is applied to all sequence elements, and the output of the previous sequence is considered together as the input of the current sequence. RNNs can be subdivided into several categories depending on how they are implemented. One of them is Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU). In one embodiment of the present invention, the LSTM is limited, but the use of the GRU is not excluded. Through this, it is possible to resolve the constant supply and demand imbalance between the quantity of cargo that is demanded and the number of trucks that are supplied, and eliminate the structure in which those who have cargo transport information, including the shipper, are relatively superior, and fundamentally eliminate the distorted freight rate decision structure. can do.

The transmission unit 330 may transmit a list each included in the answer data to at least one shipper terminal 100 and at least one borrower terminal 400. At this time, when a transport contract between the shipper and the borrower is established due to agreement between the parties, a contract may be prepared in a preset contract format and transmitted to the shipper terminal 100 and the borrower terminal 400. Even after the contract is completed, whether or not the cargo has been fulfilled according to the contract is followed up by the transport assistant 350, which will be described in detail in the transport assistant 350.

The estimating unit 340 receives vehicle data and cargo data from at least one vehicle owner terminal 400 and at least one shipper terminal 100 in the input unit 310, and then is pre-mapped to a route included in the cargo data. Cargo data may be delivered by extracting at least one stored carrier terminal. In addition, the estimating unit 340 may filter the received estimate data based on pre-built case-based artificial intelligence big data when receiving the estimate data from at least one carrier terminal. In addition, the estimating unit 340 may transmit the filtered quotation data to at least one shipper terminal 100, and when any one of the quotation data is selected from the at least one shipper terminal 100, the selected quotation data is provided. Transport approval events can be delivered to the terminal of one carrier. At this time, the above-described big data is also used when filtering the estimate data. Of course, it will be apparent that modeling and learning for filtering the estimate data can be performed separately from the above-described big data.

In addition, the estimating unit 340 may generate an alarm by detecting a normal estimate and an abnormal estimate with an abnormal behavior detection algorithm. At this time, the estimating unit 340 may be used not only to match the two parties with big data in consideration of the sequence as described above, but also to actually verify the quotation submitted by one party. First of all, in order to detect abnormal behavior, it is necessary to first learn the normal behavior. Until the data of the normal estimate classified as normal behavior are accumulated, the intervention of the person performing the evaluation is required. Initially, the normal behavior pattern of the estimate data classified as a normal estimate is extracted and learned, and the day containing the behavior pattern that deviates from this can be determined and detected as the day when the abnormal behavior occurred. Upon completion of the learning, you can obtain an Autoencoder that compresses and contains information on normal behavior for each quotation, and loads the quotation sequence to detect abnormal behavior, extracts the pattern of the sequence, and autoencoder. The loss value is calculated by comparing the output obtained by putting it as an input of and using this to detect a malicious anomaly estimate. Here, the auto-encoder is an unsupervised learning method in which artificial intelligence learns itself without providing a separate answer as one of the machine learning algorithms. The auto-encoder can be defined as a model composed by connecting an encoder and a decoder. The input value is used as an invisible output value through the encoder, and the final output value is output by using the output value as the input of the decoder. The operation of the autoencoder is to make the final output value and the input value outputted through this process have the same value.The autoencoder's operation is to create a model containing a compressed representation of the input value by repeating the encoding and decoding process in the learning process. It is the purpose. Using these features, it can be used for abnormality detection or data generation model training. Of course, in addition to the above-described autoencoder, there may be various methods for verifying the quotation, so it will be said that it is not limited to the above-described embodiment.

The transport assistance unit 350 transmits the list included in the response data from the transmission unit 330 to at least one shipper terminal 100 and at least one vehicle owner terminal 400, and then at least one shipper terminal 100 ) And at least one vehicle owner terminal 400, when mutual selection and approval are made, location information, weight information, image and image information may be collected from at least one monitoring device installed in the vehicle of the at least one vehicle owner terminal 400. I can. At this time, the transport assistant 350 may divide the collected weight, image, and video information into a driving area and a cargo area, and check whether a condition of a preset warning event is satisfied by analyzing the driver and cargo status information. For example, it may be a case that the driver is drowsy, the weight of the cargo is reduced, or the cargo fixing string recognized by the camera is broken or loosened, but is not limited to the listed ones. Accordingly, when the state information of the driver or the cargo satisfies the condition of a preset warning event, the transport assistant 350 may transmit the state information of the driver or the cargo to at least one preset report center.

In this case, the at least one monitoring device may include at least one camera and at least one weight sensor, but is not limited to those listed as described above. In addition, the transport assistant 350 divides the collected weight, image, and video information into a driving area and a cargo area, and analyzes the status information of the driver and cargo to check whether the condition of a preset warning event is satisfied, the image and Analysis of image information may be based on Convolutional Neural Networks (CNN), and analysis of collected text may be based on Recurrent Neural Networks (RNN). In this case, the CNN is a network structure using a convolutional layer, which is suitable for image processing, and can classify images based on features in the image by inputting image data.

In addition, Faster R-CNN can also be used, which has a structure that can simultaneously detect and classify objects through the learning process. The structure of Faster R-CNN can be divided into a convolutional layer for extracting and classifying features of an image, and a region proposal network for extracting candidate regions. In the input image, features of the image are extracted through the convolutional layer, the extracted features are extracted as candidate regions through the Region proposal network, and finally detected as an object through RoI pooling, Classifier, and Bounding box regressor. The result of monitoring the driver and cargo can be used as a learning data set, and the location where the image was recorded, the type of vehicle, and the driver's identifier (ID) are recorded for each frame, and the state of the driver and the cargo by searching for the same object in previous frames Change can be tracked.

For example, to change the driver's state, a gaze recognizer using CNN can be used, which receives the target's eye image or face image as an input, and outputs the gaze direction as a coordinate value according to the labeling type of the database. , Or, supervised learning technology, which has a form of outputting the class number of the gazing area, can be used. The gaze recognizer to be used in the vehicle, for example, pre-designates the gaze area from the front based on the driver, assigns a class number suitable for it to each image in the database, and learns it using a network, and then the driver from the input image image It can be configured to output the result class number for which area you are gazing. At this time, in the case of ADAS (Advanced Driver Assistance System), which receives images in front of the vehicle for detecting signs and traffic lights, RGB cameras are mainly used because of the importance of color information of objects and the presence of light sources such as headlights that illuminate the front. In the case of an image, information on shades is more important than information on colors. In addition, in the case of driving a vehicle at night when there is insufficient light, it is difficult to properly recognize the driver's face because it is difficult to properly recognize the driver's face with an RGB camera. Even in a night environment, an infrared (IR) camera that can take a relatively intact image of the driver's face can be used.

At this time, rather than inserting the face image as it is, dividing the face image vertically and using the upper-half face can intensively refer only the data necessary for gaze recognition. After extracting only the face part, the area corresponding to the upper half of the face may be cropped and used as input data for the network. This is because the data of the image is lost to some extent in the process of convolution in the CNN, so the data that performed less convolution and the data that performed more convolution were concatenated to make a full connection. When used by projecting together in a fully-connected layer, the information of the input image can be referenced to some degree of preserved data, and high accuracy can be recorded even if only a part is used, while the computational amount and throughput are low. Therefore, it can be performed smoothly even in a low specification server.

For changes in cargo conditions, cargo safety monitoring technology (CSM) may be used. In addition, before monitoring, it is also possible to use Cargo Balancing Control (CBC), which can perform loading modeling based on the size and type of cargo from the time of loading. In the case of the former, there is an open/closed and abnormal condition of the vehicle door, operation and abnormality of transport facilities, proper weight distribution and abnormality of cargo load, CCTV screen and status to manage abnormality in the vehicle, abnormality of balance control device, etc. A device that can display information related to the loaded cargo can be used. In monitoring the safety status of a vehicle carrying a loaded cargo, an unmanned monitoring function for the status of the cargo in real time, a function to check the fastening status of the loaded cargo, a function to monitor a fire in the cargo or a fire in the vehicle, among the cargoes If there is a cargo that will generate harmful gases, it can additionally have a monitoring function to see if a problem such as leakage occurs. In addition, a function to check the temperature and humidity status of the vehicle, a function to check the weight of the loaded cargo to prevent it from becoming overweight, organize the cargo loading status by video control when loading and unloading cargo, monitor the fastening status of the loaded cargo (image control to monitor cargo shaking), and cargo when loading and unloading cargo. The configuration of monitoring the arrangement status, guiding the balance control status of the loaded cargo, warning and screen display to the driver in case of an abnormality in the loaded cargo, checking the weight when loading and unloading the cargo vehicle, generating a warning sound when overweight loading occurs, and providing information to the driver through the vehicle control device. Can be added. In the latter case, modeling on the mass distribution of ULD and the driving safety of a freight train can be used assuming that a unit load device (ULD) of a freight vehicle is loaded in the vehicle. Of course, it will be apparent that various methods other than the above-described methods may be used and may be changed according to embodiments.

The crew management unit 360, after receiving vehicle data and cargo data from at least one vehicle owner terminal 400 and at least one shipper terminal 100 from the input unit 310, the at least one vehicle owner terminal 400 When a crew (Crew) registration request event occurs, it is possible to examine whether or not the crew jang terminal 400 that generated the crew registration request event has been registered. Judging may be mandatory or optional. Here, the crew management unit 360, when the registration request of the crew terminal 400 is approved as a result of the examination, may post a crew member recruitment announcement, and at least one crew one terminal 400 that generated a crew member support event. ) Information is transmitted to the crew jang terminal 400, and using the information of at least one crew-one terminal 400 selected by the crew jang terminal 400, it is possible to learn the crew-one information and analyze the propensity of the crew jang. have. In addition, the crew management unit 360 transmits a crew-one subscription approval to at least one crew-one terminal 400 selected by the crew-one terminal 400, and at least one crew approved based on the crew-jang terminal 400 The original terminal 400 may be clustered. In addition, the crew management unit 360 transmits a crew-one subscription approval to at least one crew-one terminal 400 selected by the crew-one terminal 400, and at least one crew approved based on the crew-jang terminal 400 After clustering the original terminal 400, the freight transport contracts of the clustered crew may be collected, accumulated as a history log, and stored. In addition, the crew management unit 360 builds big data by learning contract propensity as input data by learning at least one condition included in the stored freight forwarding contract, and when the freight forwarding request is input as query data, the learned contract propensity It is possible to determine whether to recommend or not by calculating the degree of similarity with.

The risk management unit 370 transmits the list included in the response data from the transmission unit 330 to at least one shipper terminal 100 and at least one borrower terminal 400, and then at least one shipper terminal 100 ) And at least one owner's terminal 400 to mutually select and approve, and when the transport of the cargo is completed, at least one shipper's terminal 100 and at least one owner's terminal 400 give points of the owner and the shipper. I can. And, the risk management unit 370 may generate a list having at least one level based on the assigned score, and add the list having at least one level to the pre-built case-based artificial intelligence big data for learning. You can proceed. For example, level 1 may be a group most favored by contracting parties, and level N may be a group most avoided.

In matching each party throughout an embodiment of the present invention, existing contracts are analyzed and contracted parties are matched based on transaction propensity, and the results of big data, artificial intelligence, and modeling are used. For example, by analyzing the positive and negative rates of each contract, the type of contract with the most positive and negative tendencies among the contracts, the type of contract, and the relationship between the basic data of both parties, it is possible to know which parties to match in the future. . For example, Naive Bayes can be used to extract Korean words, frequencies, and statistics for determining positive and negative propensity, and the Decision Tree is extracted from Naive Bayes. It can be used for intuitive data classification for each word, and the distance between classified data groups can be extracted using K-nearest neighbors. Based on KNN, it is possible to analyze the relationship between key keywords.

Naive Bayes is used to record document classification and number of times for a specific word, and is a method of selecting an item with a high probability between two categories as part of the base theorem. In Bayes'rule, p(c|x)=p(x|c)p(c)/p(x) is used to classify data by conditional probability. When x is defined as the input data and c is the classification item in the data classification, p(c|x) can be known while knowing p(x|c). At this time, when the Bayesian rule is applied, p(ci|x,y)=p(x,y|ci)p(ci)/p(x,y), and two rules, p(c1|x) By comparing ,y) and p(c2|x,y), the probability of belonging to the category c1 can be calculated based on the data identified as x,y. This Bayesian rule can be used to calculate the unknown from known data.

The decision tree is used to classify a data set, and has a form of repeating a recursive classification procedure until the data set is classified from the first parent node. Among the methods of determining data division, a quantization method is applied, and information gain is calculated and the properties of each node are checked and divided. At this time, the formula L=log2p(xi) is used to calculate the probability that the classification item will be selected. Here, p(xi) is the probability that the classification item of xi is selected, and the method of measuring information about the information gain is the entropy of H=-Σ(xi)log2p(xi) (i is from 0 to n). Use a calculation formula. Entropy H is calculated using the probability (xi) of the classified data item L and added together. Finally, the index of the best attribute is returned by comparing the changes before and after the comparison, and the data can be classified for the attribute with the highest information gain (entropy) by repeating the previous process.

K-nearest neighbor is the most well-known algorithm in the field of pattern recognition, comparing all existing data (requires classified data items) and new data to classify data, and measures the distance to the top k most similar data. This is how to classify. Euclidean distance can be used to measure the distance. The distance can be calculated from the attribute values between the two data groups, and to obtain the distance of each attribute value, an integer numeric value is used as the attribute value, and data can be normalized for this purpose. Normalization can be user-defined, such as 0 to 1 or 1 to 1. Normalization proceeds with normalization using the smallest and largest values in the set, and the maximum and minimum values used in the normalization process can be used to predict the distance for the unclassified type by determining the close distance.

Apriori is an algorithm that finds rules that can best explain the major relationships observed in data variables such as association rules, preferences, and information filtration, and can be used to find important associations between data with high multidimensional or complex relationships. In the Apriori algorithm, relationships are expressed in two forms: Frequent item sets or Association rules. The frequent item set is a collection of items that occur frequently together, and the association rule suggests that there is strength in the relationship between items. In a specific data set, the above two methods can be used to determine whether there is a relationship to a set of items. Support is a type of clustering in which data with a lot of correlation is grouped by defining the ratio and reliability of a data set containing specific data in a data group as an association rule, and can be used to find data that is likely to satisfy the purpose at the same time. Of course, it is obvious that in addition to the above-described methods, it is possible to derive the propensity from each contract relationship and model big data based on the case using various algorithms.

Hereinafter, the operation process according to the configuration of the cargo dispatch service providing server of FIG. 2 will be described in detail with reference to FIGS. However, it will be apparent that the embodiment is only any one of various embodiments of the present invention, and is not limited thereto.

3 to 6 are diagrams illustrating a process in which data is transmitted and received between components included in the cargo dispatch service providing system using O2O-based big data and artificial intelligence of FIG. 1 according to an embodiment of the present invention. Hereinafter, an example of a process in which data is transmitted/received between each of the components will be described through FIGS. 3 to 6, but the present application is not limited to such an embodiment, and FIG. 3 according to various embodiments described above. It is apparent to those skilled in the art that the process of transmitting and receiving data shown in FIG. 6 may be changed.

3 shows a general dispatch service connecting the borrower and the shipper. Referring to (a) of FIG. 3A, the cargo dispatch service providing server 300 before the artificial intelligence dispatch proceeds registers basic information from the vehicle owner terminal 400, authenticates the vehicle owner, and records a history log of the previous cargo transport. For analysis, data is collected from borrowers and public institutions, and based on this, a configuration for providing a list of recommended cargo or urgent cargo requests is shown. At this time, it is possible to further use data from the Ministry of National Highway and Transportation to collect not only vehicle owners and shippers, but also distance data for distance measurement from public institutions, meteorological data for checking regional weather, traffic volume, accidents, rest area information, and the like. 3A(b) shows data transmission/reception between the vehicle owner terminal 400 and each component after the vehicle is dispatched. The freight dispatch service providing server 300 may share information from departure to arrival and transmit an arrival notification to the shipper terminal 100 to share the delivery progress.

Referring to Figure 3b, (a) shows what kind of data to be collected, registered, and authenticated before registering the cargo in the shipper terminal 100. The shipper of the shipper terminal 100 transmits the requested details and contracts to the freight dispatch service providing server 300, as well as the borrower, and proceeds with a process for receiving recommendations through the freight dispatch service provision server 300. I can. (b) And, when the cargo is registered, the cargo dispatch service providing server 300 may recommend the borrower to the shipper terminal 100, and through a channel formed between the approved shipper terminal 400 and the shipper terminal 100 Call reception/reception and message reception/reception may be possible.

4 illustrates a method of estimating a cargo among cargo dispatch services according to an embodiment of the present invention. Referring to (a) of FIG. 4A, when a quotation is requested from the shipper terminal 100, at least one carrier terminal ( After loading the information of 500) and performing filtering, the cargo information is transmitted to the filtered host carrier terminal 500. At this time, referring to (b), when receiving a quotation request from the agent terminal 500, the quotation can be directly transferred, but in order to refer to how the previous quotation was received from the shipper terminal 100, the freight dispatch service providing server ( 300) may request a previous quotation or a recommended quotation, and as a feedback on this, a recommended quotation may be provided from the freight dispatch service providing server 300 to the carrier terminal 500. The freight dispatch service providing server 300 may transmit a quote to the shipper terminal 100 to select any one or a plurality of carriers when a quote is received. Referring to FIG. 4B, a process after selecting a freight estimate from the shipper terminal 100 is shown, and the host carrier terminal 500 receives transportation approval and connects at least one owner terminal 400 to the freight dispatch service providing server 300. It is selected through and communication and information sharing are performed through a preset channel.

5 shows a driver assistance method in an embodiment of the present invention, in which location information is transmitted to the cargo dispatch service providing server 300 through the vehicle owner terminal 400, and images and image data are transmitted to the vehicle owner terminal 400. When transmitted to the cargo dispatch service providing server 300 from, the cargo dispatch service providing server 300 analyzes text, images, and images to share reports and warnings to the report center, respectively.

6 shows a crew platform in an embodiment of the present invention. Referring to (a) of FIG. 6A, the vehicle owner terminal 400 requests a cargo dispatch service providing server 300 to register and make a payment. In this case, the cargo dispatch service providing server 300 publishes a crew member recruitment announcement, and when requesting a crew member from at least one borrower terminal 400, the cargo dispatch service providing server 300 analyzes the propensity of the crew chief. When crew members are filtered and recruitment is completed, the propensity of the group is analyzed by grouping or clustering of crew members and crew members. And, (b) the freight dispatch service providing server 300 analyzes the propensity of the shipper terminal 100 and the propensity of the crew terminal 400 to perform matching, and when a dispute occurs, the dispute as shown in FIG. 6B The occurrence is coordinated and the appointment of an attorney is brokered.

The matters that are not described with respect to the method of providing a cargo dispatch service using O2O-based big data and artificial intelligence of FIGS. 2 to 6 are described above with respect to the method of providing a cargo dispatch service using O2O-based big data and artificial intelligence through FIG. 1. Since the content is the same as the description or can be easily inferred from the description, the following description will be omitted.

7 is a flowchart illustrating a method of providing a freight dispatch service using O2O-based big data and artificial intelligence according to an embodiment of the present invention. Referring to FIG. 7, the cargo dispatch service providing server receives vehicle data and cargo data from at least one vehicle owner terminal and at least one shipper terminal (S7100).

In addition, the freight dispatch service providing server uses pre-established case-based artificial intelligence big data to receive response data to a query of a freight search event of at least one owner's terminal and a vehicle search event of at least one shipper's terminal. Generate (S7200).

In addition, the cargo dispatch service providing server transmits a list each included in the answer data to at least one shipper terminal and at least one vehicle owner terminal (S7300).

The matters that are not described about the method of providing a freight dispatch service using O2O-based big data and artificial intelligence of FIG. 7 are described above with respect to a method of providing a freight dispatch service using O2O-based big data and artificial intelligence through FIGS. 1 to 6. Since the content is the same as the description or can be easily inferred from the description, the following description will be omitted.

A method of providing a cargo dispatch service using O2O-based big data and artificial intelligence according to an embodiment described with reference to FIG. 7 is a method of providing a cargo dispatch service using an application executed by a computer or a recording medium including a computer-executable instruction such as a program module. It can also be implemented in a form. Computer-readable media can be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Further, the computer-readable medium may include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

The above-described method for providing a cargo dispatch service using O2O-based big data and artificial intelligence according to an embodiment of the present invention includes an application basically installed in a terminal (this includes a program included in a platform or operating system basically installed in the terminal). It may be executed by the application store server, an application, or an application (that is, a program) directly installed on the master terminal through an application providing server such as a web server related to the service. In this sense, the method for providing a cargo dispatch service using O2O-based big data and artificial intelligence according to an embodiment of the present invention described above is implemented as an application (ie, program) that is basically installed in the terminal or directly installed by the user, and It can be recorded on a computer-readable recording medium such as E.

The above description of the present invention is for illustrative purposes only, and those of ordinary skill in the art to which the present invention pertains will be able to understand that other specific forms can be easily modified without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative and non-limiting in all respects. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. do.

The embodiments for the implementation of the invention have been described together in the best mode for the implementation of the above invention.

The present invention relates to a method of providing a freight dispatch service using O2O-based big data and artificial intelligence, while lowering the risk burden of borrowers and shippers, and ending issues between major market parties regarding the controversial freight rate system in the freight car transport market. In order to be able to do so, we transparently provide an integrated service for the entire logistics process, including transportation, storage, and unloading information, based on the openness of transaction information, and mediate the optimal party based on big data and case-based neural network artificial intelligence learning. At the same time, it is possible to improve the freight transport structure that mass-produces an unrealistic freight structure, provide the profit structure and high-quality transport service of the transport service provider, improve the profit structure of freight car owners, as well as improve the freight transport market structure and the public interest of freight transport. It has industrial applicability because it can strengthen its function.

Claims

In the freight dispatch service providing method executed in the freight dispatch service providing server,

Receiving vehicle data and cargo data from at least one vehicle owner terminal and at least one shipper terminal;

Generating response data to a query of a cargo search event of the at least one vehicle owner terminal and a vehicle search event of the at least one shipper terminal using pre-established case-based artificial intelligence big data;

Transmitting a list each included in the answer data to the at least one shipper terminal and at least one borrower terminal;

A method of providing a freight dispatch service using big data and artificial intelligence based on O2O (Online to Offline) including a.
The method of claim 1,

The pre-established case-based artificial intelligence big data is big data learned by LSTM (Long Short-Term Memory models) among RNNs (Recurrent Neural Networks), providing cargo dispatch service using O2O-based big data and artificial intelligence. Way.
The method of claim 1

After receiving vehicle data and cargo data from the at least one vehicle owner terminal and at least one shipper terminal,

Extracting at least one host carrier terminal previously mapped and stored on a route included in the cargo data and transmitting the cargo data;

Filtering the received estimate data based on the pre-established case-based artificial intelligence big data when receiving the estimate data from the at least one host terminal;

Transmitting the filtered estimate data to the at least one shipper terminal;

When the at least one shipper terminal selects any one of the quotation data, transmitting a transport approval event to the carrier terminal that provided the selected quotation data;

The method of providing a freight dispatch service using O2O-based big data and artificial intelligence that further comprises.
The method of claim 1,

After the step of transmitting the list each included in the answer data to the at least one shipper terminal and at least one borrower terminal,

When the at least one shipper terminal and at least one vehicle owner terminal select and approve each other, collecting location information, weight information, image and image information from at least one monitoring device installed in the vehicle of the at least one vehicle owner terminal step;

Dividing the collected weight, image, and video information into a driving area and a cargo area, and analyzing status information of a driver and cargo to check whether a condition of a preset warning event is satisfied;

When the condition information of the driver or the cargo satisfies a condition of a preset warning event, transmitting the condition information of the driver or cargo to at least one preset report center;

The method of providing a freight dispatch service using O2O-based big data and artificial intelligence that further comprises.
The method of claim 4,

The at least one monitoring device includes at least one camera and at least one weight sensor, a method for providing a cargo dispatch service using O2O-based big data and artificial intelligence.
The method of claim 4,

In the step of dividing the collected weight, image, and video information into a driving area and a cargo area, and checking whether the condition of a preset warning event is satisfied by analyzing the driver and cargo status information,

The analysis of the image and video information is based on CNN (Convolutional Neural Networks), and the analysis of the collected text is based on RNN (Recurrent Neural Networks). .
The method of claim 1,

After receiving vehicle data and cargo data from the at least one vehicle owner terminal and at least one shipper terminal,

If a Crew chapter registration request event occurs among the at least one vehicle owner terminal, examining whether or not the Crewjang terminal that generated the Crew chapter registration request event has been registered;

Posting a crew member recruitment notice when the request for registration of the crew terminal terminal is approved as a result of the examination;

Transmit information of at least one crew-one terminal that generated the crew-one support event to the crew-one terminal, and learn information of crew-one using information of at least one crew-one terminal selected by the crew-one terminal. Analyzing intestinal tendencies;

Transmitting a crew-one subscription approval to at least one crew-one terminal selected by the crew-one terminal, and clustering the approved at least one crew-one terminal based on the crew-one terminal;

A freight dispatch service providing method using O2O-based big data and artificial intelligence further comprising a.
The method of claim 7,

After transmitting a crew-one subscription approval to at least one crew-one terminal selected by the crew-one terminal, and clustering the approved at least one crew-one terminal based on the crew-one terminal,

Collecting freight transport contracts of the clustered crew, accumulating and storing them as a history log;

Building big data by learning contract propensity as input data on at least one condition included in the stored freight transport contract;

When a cargo transport request is input as query data, determining whether to recommend or not by calculating a degree of similarity with the learned contract propensity;

The method of providing a freight dispatch service using O2O-based big data and artificial intelligence that further comprises.
The method of claim 1,

After the step of transmitting the list each included in the answer data to the at least one shipper terminal and at least one borrower terminal,

Selecting and approving each other at the at least one shipper terminal and at least one owner terminal, and assigning scores from the at least one shipper terminal and at least one owner terminal to the owner and the shipper from the at least one shipper terminal and at least one owner terminal;

Generating a list having at least one level based on the assigned score;

Adding the list having the at least one level to the pre-built case-based artificial intelligence big data to proceed with learning;

The method of providing a freight dispatch service using O2O-based big data and artificial intelligence that further comprises.
A computer-readable recording medium on which a program for executing the method of any one of claims 1 to 9 is recorded.