Methods for providing transportation locations

Abstract

A method performed by an electronic device is disclosed. The method includes obtaining shipping data associated with the shipment of a container. The method includes: predicting one or more shipping locations of the container based on the shipping data by applying a location prediction model to the shipping data. Optionally, the method includes: providing a supplier delay output indicating an estimate of the delay of the container at the supplier based on one or more predicted shipping locations.

Description

[0001] This disclosure relates to the field of transportation and freight. This disclosure relates to a method and related electronic device for providing transportation location. Background Technology

[0002] Goods shipped may be stored in storage facilities such as warehouses before being delivered directly to the consignee. The distance between the storage facilities and the shipping hub (such as a port or airport) to which the goods were originally delivered can vary greatly.

[0003] For example, contracts can be entered into with suppliers to transport goods from shipping hubs to storage facilities. However, difficulties may arise due to the wide range of distances involved in these journeys.

[0004] For example, the availability of supplier vehicles, operating trucks, and / or drivers can vary significantly. Availability can differ based on the distance traveled to and from delivery hubs and destinations (such as storage facilities and / or consignee locations). Therefore, it may be difficult to find suppliers with available delivery vehicles, potentially leading to logistical inefficiencies. Summary of the Invention

[0005] Therefore, there is a need for an electronic device and a method for predicting shipping locations that mitigates, alleviates, or resolves existing shortcomings and allows for more accurate, robust, and time-saving prediction of container shipping locations and more accurate prediction of supplier delays.

[0006] A method performed by an electronic device is disclosed. The method includes obtaining shipping data associated with the shipment of a container. The method includes: predicting one or more shipping locations of the container based on the shipping data by applying a location prediction model to the shipping data. Optionally, the method includes: providing a supplier delay output indicating an estimate of the delay of the container at the supplier based on one or more predicted shipping locations.

[0007] An electronic device is disclosed, the electronic device including memory circuitry, processor circuitry, and an interface, wherein the electronic device is configured to perform any of the methods disclosed herein.

[0008] A computer-readable storage medium is disclosed for storing one or more programs, the programs including instructions that, when executed by an electronic device having a display and a touch-sensitive surface, cause the electronic device to perform any of the methods disclosed herein.

[0009] One advantage of this disclosure is that the disclosed electronic devices and methods provide predictions of the shipping location of containers, for example, when booking a container shipment. The disclosed shipping location predictions can improve the visibility and / or transparency of container shipments to users (e.g., consignees and / or booking parties), and thereby lead to improved control over shipments from origin to destination, such as when booking a shipment. Furthermore, the predicted shipping location and the resulting supplier delay outputs can make shipping metrics such as search rates and booking search rates more accurate, reliable, and robust. Due to the supplier delay outputs determined based on the predicted shipping location, this disclosure can lead to a reduction in unexpected supplier delays.

[0010] The predicted transport locations in this disclosure can advantageously improve the management and / or organization of inventory (such as fleets, equipment, and drivers). In other words, suppliers can improve the logistical efficiency of their fleets based on the determined transport locations provided by the disclosed electronic devices and methods. In other words, this disclosure can enable improvements in supplier inventory availability, thereby enabling increased availability of the supplier's fleet to transport containers to, for example, storage facilities.

[0011] Furthermore, the transportation locations predicted by the disclosed electronic devices and methods can improve the relevance of supplier services recommended to users (such as those booking freight). For example, supplier services can be recommended to users based on the predicted transportation locations. Attached Figure Description

[0012] The above and other features and advantages of this disclosure will be readily apparent to those skilled in the art from the following detailed description of exemplary embodiments with reference to the accompanying drawings, in which: Figure 1 This is a schematic diagram illustrating an example system in which the disclosed technology is implemented according to this disclosure. Figures 2A to 2C This is a flowchart illustrating an exemplary method for providing a transportation location, performed by an electronic device according to the present disclosure. Figure 3 This is a schematic diagram illustrating an example process in which an example electronic device performs the disclosed technology according to this disclosure, and Figure 4 This is a block diagram illustrating an exemplary electronic device according to the present disclosure. Detailed Implementation

[0013] Various exemplary embodiments and details are described below with reference to the accompanying drawings, where applicable. It should be noted that the drawings may be drawn to scale or not, and elements with similar structure or function are indicated by the same reference numerals in all the drawings. It should also be noted that the drawings are intended only to facilitate the description of embodiments. The drawings are not intended as an exhaustive description of this disclosure or a limitation on the scope of this disclosure. Furthermore, the illustrated embodiments need not possess all the aspects or advantages shown. Aspects or advantages described in connection with a particular embodiment are not necessarily limited to that embodiment and can be practiced in any other embodiment, even if not so shown or explicitly described.

[0014] For clarity, the accompanying drawings are schematic and simplified, and only details that aid in understanding this disclosure are shown, while other details are omitted. Throughout, the same reference numerals are used for the same or corresponding parts.

[0015] The container disclosed herein refers to a shell in which goods to be shipped are encapsulated for transport. For example, a container may be shipped as part of a shipment. For example, a shipment may involve transporting one or more containers from a first location (such as a location of origin) to a second location (such as a location of destination). For example, the goods disclosed herein may be considered as goods that can be packaged into, for example, rectangular stackable cartons having variable weight and volume and transported in one or more dry containers.

[0016] For example, shipments can be carried out using transport vehicles such as trucks, ships, and / or freight vehicles. Shipments can be carried out using land vehicles, ships, and / or aircraft.

[0017] This disclosure specifically applies machine learning (ML) techniques and / or statistical modeling techniques to predict the transport location to which containers will be delivered and their corresponding turn times. In some examples, the transport location is an inland transport location, such as a location far from a port. In other words, the transport location can be considered as the container's transport destination, such as the final location and / or intermediate location. This allows for transparency and predictability of turn times, such as at the time of booking, which in turn reduces the risk of transport service providers being unable to locate the transport location. For example, contracts can be entered into with transport providers, thus serving users in a more transparent manner and allowing for advantageous selection of transport services between external transport providers and carrier transport services.

[0018] For example, in the import-side journey of a container, users can use their own transportation or a carrier's inland transportation services to transport the container to an inland location and then return empty. To provide inland transportation services to users, carriers enter into contracts with transportation providers (such as trucking providers). The difficulty lies in the transportation providers' ability to provide trucks when delivery to the required location is needed. The visibility of container turnaround time is not considered during the supplier contract period, leading to an unfavorable process affecting delivery and costs.

[0019] This disclosure enables carriers to use machine learning (ML) to predict the transport locations (e.g., inland transport locations) where containers will need to be delivered, thereby controlling supplier contracts and allowing advance control of the truck driver capacity required at a given port. This can further allow for risk mitigation when containers are delayed beyond a predetermined threshold due to reasons attributable to the transport supplier, for example, by generating and including penalty values ​​in contracts with the transport supplier.

[0020] Figure 1 This is a schematic diagram illustrating an example system 1 in which the disclosed technology is implemented according to this disclosure.

[0021] Example system 1 can be considered as a container booking and control system. System 1 includes, for example, the example electronic device 300 disclosed herein.

[0022] Electronic device 300 obtains, for example, shipping data 11 associated with the shipment of a container from data storage 18. Data storage 18 may be a database and / or data repository that stores one or more data structures associated with goods, bookings, shipments, and / or delivery modes. Data storage 18 may access goods or booking data structure 12, goods data structure 14, and / or data structure 16 for delivery modes. In one or more example methods, shipping data includes one or more of the following: transaction data associated with one or more transactions, container movement data associated with the movement of one or more containers, goods data associated with one or more goods, consignee data associated with one or more consignees, container data associated with one or more containers, booking data associated with one or more shipment bookings, port data associated with one or more shipping ports, turnaround time data, time extension data associated with container time extensions, and historical shipping data.

[0023] In some examples, electronic device 300 includes a recommendation system 25 configured to predict one or more shipping locations of a container based on shipping data 11 by applying a location prediction model to the freight data 11. Recommendation system 25 is configured to provide a supplier delay output 27, indicating an estimate of the container's delay at the supplier, based on one or more predicted shipping locations.

[0024] In some examples, shipment data 11 is preprocessed in 24 to generate risk factors, seasonality factors, and / or one or more consignee distribution parameters for at least one consignee.

[0025] In some examples, electronic device 300 includes a turnaround time prediction system 29, which is configured to predict turnaround time parameters 31 based on shipping data 11 or preprocessed data.

[0026] Optionally, the electronic device 300 includes a supplier contract signing system 33, configured to obtain supplier delay output 27 and optionally obtain predicted turnaround time parameters 31. The predicted turnaround time parameters can be used to determine the penalty value for a supplier's delayed return of a container exceeding a threshold.

[0027] In some examples, the supplier contract signing system 33 can generate documented data of the supplier's contract based on the supplier's delayed output 27 and optionally on the penalty value based on the predicted turnaround time parameter 31.

[0028] This disclosure allows for the prediction of how much capacity is needed for any given location, and what the threshold for turnaround time parameters should be, in order to, for example, adjust penalty values ​​in supplier contracts.

[0029] Figure 2 illustrates a flowchart of an exemplary method 100 for predicting a shipping location, performed by an electronic device according to the present disclosure. In some examples, method 100 is a method for providing delayed supplier output (e.g., for providing document data related to supplier shipping). Method 100 is performed by an electronic device, such as the electronic devices disclosed herein, such as... Figure 1 , Figure 3 and Figure 4 Electronic device 300.

[0030] Method 100 includes obtaining (e.g., receiving and / or retrieving) shipment data associated with the shipment of containers in S102. In one or more example methods, the shipment data includes one or more of the following: transaction data associated with one or more transactions, container movement data associated with the movement of one or more containers, commodity data associated with one or more goods, consignee data associated with one or more consignees, container data associated with one or more containers, booking data associated with one or more shipment bookings, port data associated with one or more shipment ports, turnaround time data, time extension data associated with container time extensions, and historical shipment data. In some examples, information may be extracted from the shipment data. In some examples, the shipment may be transformed. In some examples, the shipment data may be obtained from a transaction data source that stores booking and container movement data.

[0031] Method 100 includes: predicting one or more transport locations of container S114 based on shipping data by applying a location prediction model to the shipping data. Method 100 includes, for example, predicting one or more transport locations of containers based on shipping data, such as by applying a location prediction model to the shipping data, for example by applying the location prediction model to one or more of the following: transaction data associated with one or more transactions, container movement data associated with the movement of one or more containers, commodity data associated with one or more goods, consignee data associated with one or more consignees, container data associated with one or more containers, booking data associated with one or more shipping bookings, port data associated with one or more shipping ports, turnaround time data, time extension data associated with container time extensions, and historical shipping data. In some examples, one or more transport locations are determined at the time of booking the container for shipment to predict and mitigate any risk that the container return will be delayed beyond a certain threshold. In one or more example methods, the location prediction model includes one or more of the following: label encoding and multi-class classification machine learning models. For example, label encoding and / or multi-class classification machine learning models are performed on the shipping data. For example, the port of Copenhagen could be labeled P1. For example, each port, each commodity, each shipping location, each shipping mode, and each consignee has a unique label, assigned by a label encoding. For example, this label is a numbered label. For example, the ML classification model takes labeled assembly operations as input and provides, in particular, the probability of each shipping location chosen by the user for a given shipment during the booking period as output. In other words, for example, the predicted shipping location includes the probability that the user will choose a shipping location for a given shipment during the booking period.

[0032] Method 100 includes: providing a supplier delay output, based on one or more predicted transport locations, to indicate, S130, an estimate of the delay of the container at the supplier. In some examples, the supplier delay output may be in the form of a value (such as a penalty value, a threshold for the delay of returning the container), which is passed to the supplier contract signing system, such as... Figure 1 The supplier contract signing system 33. In some examples, the output may be a user interface object representing one or more transportation locations. In some examples, delayed supplier output results in the display of a user interface object representing an inland transportation location.

[0033] In some examples, method 100 includes grouping consignees S104 into a first consignee group and a second consignee group, the first consignee group including consignees existing in a system (such as a booking shipment system and / or a transportation location prediction system) on an electronic device, and the second consignee group including consignees not existing in such a system (such as a booking shipment system and / or a transportation location prediction system). For example, the first consignee group includes consignees who have booked shipments and / or used carrier transportation services. For example, the second consignee group includes consignees who have never booked shipments and / or never used carrier transportation services.

[0034] In one or more example methods, the method includes grouping shipments of the shipment data based on one or more of the following: month, destination port, goods, transport location, and transport mode. In one or more example methods, label encoding is performed on the grouped shipments. For example, the shipments are grouped based on the corresponding month, destination port, goods, inland location, and transport mode used by the user. For example, label encoding is performed on the grouped shipments and fitted to a multi-class classification-based machine learning algorithm that predicts the probability of each transport location to be selected by the user.

[0035] In one or more example methods, predicting one or more shipping locations based on shipment data in S114 includes: predicting one or more shipping locations associated with at least one consignee in the second consignee group in S114B based on the shipment data. In some examples, steps S112, S114, and S114B are performed on shipments to consignees in the first consignee group. The shipment data can be obtained by extracting and / or transforming master facility data, which includes data on all facilities (such as warehouses, manufacturing units, etc.) registered by the user (e.g., the consignee) and their geographic codes.

[0036] In one or more example methods, predicting one or more shipping locations associated with at least one consignee in the second consignee group based on freight data S114 includes: determining whether at least one of the one or more shipping locations predicted for the first consignee group in S114C meets a criterion. For example, the shipment data and predictions for the first group of consignees can be reused to fit an additional multi-class classification-based machine learning algorithm that uses month, destination port, commodity, inland location, and shipping mode to check whether there is a shipping location whose predicted probability of selection is higher than a first threshold (such as >75%). For example, at least one shipping location meets the criterion when the predicted probability of selection for at least one of the one or more shipping locations predicted for the first consignee group is higher than the first threshold.

[0037] In one or more example methods, predicting one or more shipping locations based on shipping data S114 includes: when determining that at least one of the one or more shipping locations meets a criterion, providing S114D that at least one shipping location as part of one or more shipping locations associated with at least one consignee in a second group of consignees. In other words, when the probability of a shipping location meets a criterion (e.g., above a first threshold), the shipping location predicted for an existing consignee can be reused for a non-existent consignee.

[0038] In one or more example methods, predicting one or more shipping locations based on shipping data S114 includes: when it is determined that at least one of the one or more shipping locations does not meet a criterion, providing the nearest shipping location S114E as part of one or more shipping locations associated with at least one consignee in the second consignee group. For example, when there is no inland location predicted to meet the criterion (such as > a first threshold, e.g., 75%) with a probability of selection, the shipping location of the non-existent consignee will be selected as the location closest to the destination port.

[0039] In one or more example methods, method 100 includes determining a turnaround time parameter, S110, indicating the turnaround time of a container based on shipping data. In other words, the turnaround time parameter is generated based on one or more of the following: transaction data associated with one or more transactions, container movement data associated with the movement of one or more containers, commodity data associated with one or more goods, consignee data associated with one or more consignees, container data associated with one or more containers, booking data associated with one or more shipping bookings, port data associated with one or more shipping ports, time extension data associated with container time extensions, and historical shipping data.

[0040] In one or more example methods, determining the S110 turnaround time parameter based on shipping data includes applying a turnaround time prediction model (S110A) to the shipping data. For example, the turnaround time prediction model may be applied to one or more of the following: transaction data associated with one or more transactions, container movement data associated with the movement of one or more containers, commodity data associated with one or more goods, consignee data associated with one or more consignees, container data associated with one or more containers, booking data associated with one or more shipping bookings, port data associated with one or more shipping ports, turnaround time data, time extension data associated with container time extensions, and historical shipping data. In one or more example methods, the turnaround time prediction model is one or more of the following: a linear regression model, a random forest model, a decision tree-based model, an ensemble model, and time series forecasting techniques.

[0041] In one or more example methods, method 100 includes generating a consignee risk factor S106 for at least one consignee (e.g., each of a plurality of consignees) based on shipping data. S106 may be for consignees in the first consignee group.

[0042] In one or more example methods, generating the S106 consignee risk factor based on shipping data includes: calculating one or more consignee distribution parameters for at least one consignee under S106A based on the shipping data. In one or more example methods, the one or more consignee distribution parameters for at least one commodity (e.g., each commodity) indicate, for at least one commodity (e.g., for each commodity), the distribution of at least one of the following: the proportion of delayed containers for at least one consignee, the ratio of the elapsed time extension to the granted time extension for shipment for at least one consignee, and the ratio of the elapsed time extension to the available time extension (e.g., the purchase time extension) for shipment for at least one consignee. For example, for each consignee, one or more consignee distribution parameters are extracted by transforming the shipping data. For example, the one or more consignee distribution parameters for each consignee are used to generate a consignee risk factor for each consignee. In some examples, one or more consignee distribution parameters include one or more of the following: distribution parameters (e.g., percentiles, percentages, etc.) of the distribution of the proportion of delayed containers for at least one consignee (e.g., the proportion of customer-delayed shipments / containers for each consignee); the ratio of the time extension elapsed to the time extension granted for shipments for at least one consignee (e.g., the ratio of days consumed by the consignee to days granted for each consignee); and the ratio of the time extension elapsed to the available time extension for shipments for at least one consignee (e.g., the ratio of days consumed by the customer to days purchased for each consignee).

[0043] In one or more example methods, generating the S106 consignee risk factor based on shipping data includes generating the S106B consignee risk factor based on one or more consignee distribution parameters. In one or more example methods, generating the S106 consignee risk factor includes applying a normalization function (S106C) to one or more consignee distribution parameters. For example, based on one or more distribution parameters, a consignee risk factor is generated for each consignee using normalization and / or a sigmoid function.

[0044] In one or more example methods, the method includes obtaining a seasonality factor for the seasonality of turnaround time parameters at the port (e.g., for combinations of container type / size and goods), as indicated by S108. For example, consignees are classified as large or small consignees based on the number of transactions at each port in the past and any combination of container type / size. For example, consignees with more than 50 transactions are classified as large, while those with fewer than 50 transactions are classified as small. For large customers, the seasonality factor for the turnaround time at the port for a specific commodity and any combination of container type / size can be obtained, for example, from a database. For example, the seasonality factor is associated with the month the container is unloaded. For example, seasonality is not checked for small customers due to inaccuracies.

[0045] In one or more example methods, determining the S110 turnaround time parameter based on shipping data includes aggregating the time extension data and turnaround time data into a dataset. For example, aggregating the time extension data given the granted time extension with the corresponding turnaround time data at each port.

[0046] In one or more example methods, determining the S110 turnaround time parameter based on shipping data includes fitting an aggregated dataset to the S110C location prediction model. For example, fitting the aggregated dataset to various container types or sizes using an ML-based regression model, such as single-book 20-foot, single-book 40-foot, contract-book 20-foot, and contract-book 40-foot data, to predict shipping turnaround times.

[0047] In one or more example methods, determining the S110 turnaround time parameters based on shipment data includes determining the turnaround time parameters for at least one consignee and at least one commodity under S110D based on goodness of fit and seasonality factors. For example, turnaround time parameters can be specifically determined for the consignee and the commodity being shipped, using a turnaround time prediction model and seasonality factors, as well as a derived consignee risk factor.

[0048] In one or more example methods, determining the S110 turnaround time parameter based on shipping data includes: determining the consignee turnaround time parameter associated with S110E and at least one consignee in the second consignee group based on the shipping data. In one or more example methods, determining the consignee turnaround time parameter associated with S110E and at least one consignee in the second consignee group includes: identifying the earliest transaction of at least one consignee of S110EA.

[0049] In one or more example methods, determining the consignee turnaround time parameter associated with S110E and at least one consignee in the second consignee group includes: for a commodity, determining the consignee turnaround time parameter associated with S110EB and at least one consignee in the second consignee group based on the consignee turnaround time parameters of consignees in the first and second consignee groups. For example, the consignee's transactions are arranged chronologically to identify the consignee's first transaction and its corresponding commodity. For example, the average turnaround time for all consignees of a particular commodity is determined as the turnaround time parameter for shipments from new consignees (e.g., consignees in the second consignee group). For example, when there is no historical shipment data for that commodity, the average turnaround time for all commodities is presented as the turnaround time parameter for shipments from consignees in the second consignee group (e.g., new consignees).

[0050] In one or more example methods, the method includes obtaining the historical booking-to-quote ratio (S116). For example, the historical booking-to-quote ratio is based on historical data for each container arriving at a given port. For instance, selecting the top two shipping locations with the highest probability of being selected by the user.

[0051] In one or more example methods, the booking-to-quote ratio predicted by S118 is determined by applying the estimate to the historical booking-to-quote ratio, for example, by applying a time series estimate, such as a monthly, quarterly, or yearly estimate.

[0052] In one or more example methods, the method includes determining, through simulation, the number of containers to be picked up at each predicted shipping location within each time period of S120, using a predicted booking-to-quote ratio and one or more predicted shipping locations. For example, the simulation is a Monte Carlo simulation that takes the predicted booking-to-quote ratio and one or more predicted shipping locations as input and provides the number of containers to be picked up at each predicted shipping location within each time period (e.g., by week, month, quarter, or year). For example, for shipping location A, 100 containers are predicted, but the predicted booking-to-quote ratio is 25%, so only 25 containers are estimated to be booked. For example, the simulation could provide the following example table, where frequency M represents monthly and Y represents yearly: Table 1.

[0053] In one or more example methods, the method includes estimating the percentage of containers expected to return within the turnaround time parameter by applying a Poisson distribution to the aggregated turnaround time parameter for each transport location. For example, the Poisson distribution is used to estimate the percentage of containers that will return within the turnaround time indicated in the predicted turnaround time parameter.

[0054] In one or more example methods, the method includes comparing a turnaround time parameter for each transport location with a threshold (such as a second threshold) S124. This threshold may be based on supply and demand at each transport location.

[0055] Table 2.

[0056] In one or more example methods, based on the comparison and aggregation value parameters, a penalty value is determined for the S126 supplier to cause container delays exceeding a certain time threshold. For example, the aggregation value parameter is obtained for each port. In some examples, the aggregation value parameter includes an aggregation contribution yield (CY) value, such as the contribution / revenue share to the gross profit per shipment / container at each port.

[0057] In one or more example methods, providing the S128 Supplier Delay Output includes generating S128A document data based on penalty values. For example, the document data is associated with documents (such as contracts) and / or contractual terms with the transportation supplier. For example, the Supplier Delay Output includes penalty values ​​that are part of a contract with the supplier. For example: “The above is the threshold time for the supplier to accept the returned container, otherwise the supplier will pay $10 per day.” Figure 3 The diagram schematically illustrates Example System 2 according to this disclosure, and Example system 2 can be considered as a container booking and control system. System 2 includes, for example, the example electronic device 300 disclosed herein.

[0058] Electronic device 300 obtains, for example, shipping data 11 associated with the shipment of a container from data storage 18. Data storage 18 may be a database and / or data repository that stores one or more data structures associated with goods, bookings, shipments, and / or delivery modes. Data storage 18 may access goods or booking data structure 12, goods data structure 14, and / or data structure 16 for delivery modes. In one or more example methods, shipping data includes one or more of the following: transaction data associated with one or more transactions, container movement data associated with the movement of one or more containers, goods data associated with one or more goods, consignee data associated with one or more consignees, container data associated with one or more containers, booking data associated with one or more shipment bookings, port data associated with one or more shipping ports, turnaround time data, time extension data associated with container time extensions, and historical shipping data.

[0059] In some examples, electronic device 300 includes a recommendation system 25 configured to predict one or more shipping locations of a container based on shipping data 11 by applying a location prediction model to the freight data 11. Recommendation system 25 is configured to provide a supplier delay output 27, indicating an estimate of the container's delay at the supplier, based on one or more predicted shipping locations. For example, recommendation system 25 is configured to provide the supplier delay output to a master database 27 (such as a master table), which is coupled to a subordinate database 26 (such as a subordinate table).

[0060] In some examples, shipment data 11 is preprocessed in 24 to generate risk factors, seasonality factors, and / or one or more consignee distribution parameters for at least one consignee.

[0061] Prediction and preprocessing can occur once a day.

[0062] In some examples, electronic device 300 includes a turnaround time prediction system 29, which is configured to predict turnaround time parameters 31 based on shipping data 11 or preprocessed data.

[0063] Optionally, the electronic device 300 includes a supplier contract signing system 33 configured to acquire supplier delay output 27 and optionally acquire predicted turnaround time parameters 31. The predicted turnaround time parameters can be used to determine penalty values ​​for supplier delays exceeding a threshold. The master database 27 provides data to the supplier contract signing system 33 upon request, and the subordinate database 26 provides data during data refresh.

[0064] In some examples, the supplier contract signing system 33 can generate documented data for supplier contracts based on supplier delay output 27 and, optionally, penalty values ​​based on predicted turnaround time parameter 31. For example, the supplier contract signing system 33 includes a recommendation application programming interface that receives a request 34 and provides a response 35 based on the supplier delay output.

[0065] Figure 4 A block diagram of an exemplary electronic device 300 according to the present disclosure is shown. The electronic device 300 includes memory circuitry 301, processor circuitry 302, and interface 303. The electronic device 300 is configured to perform... Figures 2A to 2C Any of the methods disclosed herein. In other words, the electronic device 300 is configured to predict the transportation location.

[0066] Electronic device 300 is configured (such as via interface 303 and / or processor circuitry 302) to obtain shipping data associated with the shipment of the container.

[0067] Electronic device 300 is configured (such as via processor circuitry 302) to predict one or more shipping locations of a container based on shipping data, for example by applying a location prediction model to the shipping data.

[0068] Optionally, the electronic device 300 is configured (such as via interface 303 and / or processor circuitry 302) to provide a supplier delay output indicating an estimate of the delay of the container at the supplier based on one or more predicted shipping locations.

[0069] Processor circuit 302 is optionally configured to execute Figure 2A Any of the operations disclosed in C (such as any one or more of the following: S102, S104, S106, S106A, S106B, S106C, S108, S110, S110A, S110B, S110C, S110D, S110E, S110EA, S110EB, S112, S114, S114A, S114B, S114C, S114D, S114E, S116, S118, S118A, S120, S122, S122A, S124, S126, S128, S128A). The operation of the electronic device 300 can be manifested in the form of executable logic routines (e.g., lines of code, software programs, etc.) stored on a non-transitory computer-readable medium (e.g., memory circuitry 301) and executed by processor circuitry 302.

[0070] Furthermore, the operation of electronic device 300 can be considered as a method configured to be performed by electronic device 300. Additionally, while the described functions and operations can be implemented in software, such functions can also be implemented via dedicated hardware or firmware, or some combination of hardware, firmware, and / or software.

[0071] The memory circuit 301 may be one or more of a buffer, flash memory, hard disk drive, removable media, volatile memory, non-volatile memory, random access memory (RAM), or other suitable devices. In a typical arrangement, the memory circuit 301 may include non-volatile memory for long-term data storage and volatile memory used as system memory for the processor circuit 302. The memory circuit 301 may exchange data with the processor circuit 302 via a data bus. Control lines and an address bus may also exist between the memory circuit 301 and the processor circuit 302. Figure 4 (Not shown in the image). Memory circuit 301 is considered a non-transitory computer-readable medium.

[0072] The storage circuit 301 can be configured to store shipping data, predicted shipping locations, supplier delay outputs, location prediction models, and turnaround time prediction models in a portion of the memory.

[0073] The following clauses set forth an implementation scheme for the methods and products (electronic devices) according to this disclosure: 1. A method performed by an electronic device, the method comprising: Obtain shipping data associated with the shipment of containers; Based on the shipping data, one or more shipping locations of the container are predicted by applying a location prediction model to the shipping data; and A supplier delay output is provided based on one or more predicted shipping locations, indicating an estimate of the delay of the container at the supplier.

[0074] 2. The method according to Clause 1, wherein the method includes grouping the shipments of the shipment data based on one or more of the following: month, destination port, commodity, transport location, and transport mode.

[0075] 3. The method according to any one of the preceding clauses, wherein the location prediction model comprises one or more of the following: label encoding and multi-class classification machine learning models.

[0076] 4. The method according to Clause 3, wherein the label coding is performed on the grouped shipments.

[0077] 5. The method according to any one of the preceding clauses, wherein the method comprises grouping consignees into: a first group of consignees, the first group of consignees including consignees present in the system of the electronic device; and a second group of consignees, the second group of consignees including consignees not present in the system.

[0078] 6. The method according to Clause 5, wherein predicting the one or more transport locations based on the shipment data comprises: predicting one or more transport locations associated with at least one consignee in the second group of consignees based on the shipment data.

[0079] 7. The method according to Clause 6, wherein predicting one or more transport locations associated with at least one consignee in the second group of consignees based on the shipment data comprises: - Determine whether at least one of the one or more transport locations predicted for the first consignee group meets the criteria; - When it is determined that at least one of the one or more transport locations meets the criteria, the at least one transport location is provided as part of the one or more transport locations associated with at least one consignee in the second consignee group; and - When it is determined that at least one of the one or more transport locations does not meet the criteria, the nearest transport location is provided as part of the one or more transport locations associated with at least one consignee in the second consignee group.

[0080] 8. The method according to any one of the preceding clauses, the method comprising determining a turnaround time parameter indicating the turnaround time of a container based on the shipping data.

[0081] 9. The method according to Clause 8, wherein determining the turnaround time parameter based on the shipment data comprises: applying a turnaround time prediction model to the shipment data.

[0082] 10. The method according to Clause 9, wherein the turnaround time prediction model is one or more of the following: a linear regression model, a random forest model, a decision tree-based model, an ensemble model, and a time series forecasting technique.

[0083] 11. The method according to any one of the preceding clauses, wherein the shipping data includes one or more of the following: transaction data associated with one or more transactions, container movement data associated with the movement of one or more containers, commodity data associated with one or more commodities, consignee data associated with one or more consignees, container data associated with the one or more containers, booking data associated with one or more shipping bookings, port data associated with one or more shipping ports, turnaround time data, time extension data associated with the time extension of the containers, and historical shipping data.

[0084] 12. The method according to any one of the preceding clauses, wherein the method includes generating a consignee risk factor for at least one consignee based on the shipment data.

[0085] 13. The method according to Clause 12, wherein generating the consignee risk factor based on the shipment data includes: Based on the shipping data, one or more consignee distribution parameters are calculated for the at least one consignee, wherein the one or more consignee distribution parameters for the at least one commodity indicate a distribution of at least one of the following for the at least one commodity: - For the proportion of delayed containers of the at least one consignee, - The ratio of the time extension already used to the time extension granted for shipment to the at least one consignee, and - The ratio of the time elapsed to the available time elapsed for shipments to the at least one consignee; and The consignee risk factor is generated based on one or more consignee distribution parameters.

[0086] 14. The method according to any one of Clauses 12 to 13, wherein generating the consignee risk factor comprises applying a standardized function to one or more consignee distribution parameters.

[0087] 15. The method according to any one of the preceding clauses, wherein the method includes obtaining a seasonality factor indicating the seasonality of the turnaround time parameter at the port.

[0088] 16. The method according to any one of Clauses 8 and 15 and Clauses 7 to 14, wherein determining the turnaround time parameter based on the shipment data comprises: Aggregate the time extension data and the turnaround time data into a dataset; and The aggregated dataset is fitted into the location prediction model; and - Based on the goodness of fit and seasonality factors, determine the turnaround time parameters for at least one consignee and at least one commodity.

[0089] 17. The method according to any one of Clauses 5 and 8 and Clauses 6 to 7 and 9 to 16, wherein determining the turnaround time parameter based on the shipment data comprises: determining a consignee turnaround time parameter associated with at least one consignee in the second group of consignees based on the shipment data.

[0090] 18. The method according to Clause 17, wherein determining the consignee turnaround time parameter associated with the at least one consignee in the second consignee group comprises: - The earliest transaction identifying the at least one consignee; and - For a certain commodity, based on the recipient turnaround time parameters of the recipients in the first recipient group and the second recipient group, determine the recipient turnaround time parameters associated with at least one recipient in the second group.

[0091] 19. The method according to any one of the preceding clauses, the method comprising: - Obtain historical booking to quote ratios; and - Determine the predicted booking-to-offer ratio by estimating the historical booking-to-offer ratio.

[0092] 20. The method according to Clause 19, wherein the method comprises: - Through simulation, using the predicted booking-to-quote ratio and one or more predicted shipping locations, determine the number of containers to be picked up at each predicted shipping location within each time period.

[0093] 21. The method according to any one of the preceding clauses, the method comprising estimating the percentage of containers expected to return within the turnaround time parameter by applying a Poisson distribution to an aggregated turnaround time parameter for each transport location.

[0094] 22. The method according to any one of the preceding clauses, the method comprising: - Compare the turnaround time parameter of each of the one or more predicted transport locations with a threshold; and - Based on the comparison and aggregation value parameters, determine the penalty value for the supplier to cause container delays exceeding a certain time threshold for each of the one or more predicted shipping locations.

[0095] 23. The method according to Clause 22, wherein providing the supplier delay output based on the one or more predicted shipping locations comprises: generating file data for each of the one or more predicted shipping locations based on the penalty value associated with the supplier.

[0096] 24. An electronic device comprising memory circuitry, processor circuitry, and an interface, wherein the electronic device is configured to perform any of the methods according to any one of claims 1 to 23.

[0097] 25. A computer-readable storage medium storing one or more programs, said one or more programs including instructions that, when executed by an electronic device, cause the electronic device to perform any of the methods described in claims 1 to 23.

[0098] The use of terms such as "first," "second," "third," and "fourth," "primary," "secondary," and "auxiliary," etc., does not imply any particular order, but is included to identify individual elements. Furthermore, the use of terms such as "first," "second," "third," and "fourth," "primary," "secondary," and "auxiliary," etc., does not indicate any order or importance, but is used to distinguish one element from another. Note that the use of the terms "first," "second," "third," and "fourth," "primary," "secondary," and "auxiliary," etc., here and elsewhere, is solely for labelling purposes and is not intended to indicate any particular spatial or temporal order. Moreover, the labeling of the first element does not imply the existence of a second element, and vice versa.

[0099] Understandable. Figures 1 to 4This includes some circuits or operations shown in solid lines and some circuits or operations shown in dashed lines. The circuits or operations included in the solid lines are those included in the most broad exemplary embodiments. The circuits or operations included in the dashed lines are exemplary embodiments that can be included in or part of the circuits or operations of the solid-line example embodiments, or are further circuits or operations that can be taken in addition to the circuits or operations of the solid-line example embodiments. It should be understood that these operations do not need to be performed in the order presented. Furthermore, it should be understood that not all operations need to be performed. The exemplary operations can be performed in any order and in any combination.

[0100] It should be noted that the word "including" does not necessarily exclude the existence of other elements or steps besides those listed.

[0101] It should be noted that the words "one" or "a kind" preceding an element do not preclude the existence of multiple such elements.

[0102] It should be noted that the term "indication" can be considered as "associated with," "related to," "describe," "represent," and / or "define." The terms "indication," "associated with," "related to," "describe," "represent," and "define" are used interchangeably. The term "indication" can be considered as indicating a relationship. For example, weight data indicating weight may include one or more weight parameters.

[0103] It should be noted that the term "based on" can be considered as "as a function of" and / or "derived from". The terms "based on" and "as a function of" are used interchangeably. For example, parameters determined "based on" a dataset can be considered as parameters determined "as a function of". In other words, the parameters can be the output of one or more functions that take the dataset as input.

[0104] Functions can describe the relationship between inputs and outputs, such as mathematical relationships, database relationships, hardware relationships, logical relationships, and / or other suitable relationships.

[0105] It should also be noted that any reference numerals in the drawings do not limit the scope of the claims, exemplary embodiments may be implemented at least in part by both hardware and software, and several “components,” “units,” or “devices” may be represented by the same hardware article.

[0106] The various exemplary methods, apparatuses, nodes, and systems described herein are described in the general context of method steps or processes. In one aspect, these method steps or processes may be implemented by a computer program product embodied in a computer-readable medium, including computer-executable instructions, such as program code, that are executed by a computer in a networked environment. Computer-readable media may include removable and non-removable storage devices, including but not limited to read-only memory (ROM), random access memory (RAM), optical disc (CD), digital versatile disc (DVD), etc. Generally, program circuitry may include routines, programs, objects, components, data structures, etc., that perform a specified task or implement a particular abstract data type. Computer-executable instructions, associated data structures, and program circuitry represent examples of program code for performing steps of the methods disclosed herein. Specific sequences of such executable instructions or associated data structures represent examples of corresponding actions for implementing the functionality described in such steps or processes.

[0107] Although features have been shown and described, it should be understood that they are not intended to limit the claimed disclosure, and it will be apparent to those skilled in the art that various changes and modifications can be made without departing from the scope of the claimed disclosure. Accordingly, this specification and drawings are to be considered illustrative rather than restrictive. The intent of the claimed disclosure is to cover all alternatives, modifications, and equivalents.

Claims

1. A method performed by an electronic device, the method comprising: Obtain shipping data associated with the shipment of containers; Based on the shipping data, one or more shipping locations of the container are predicted by applying a location prediction model to the shipping data; as well as A supplier delay output is provided based on one or more predicted shipping locations, indicating an estimate of the delay of the container at the supplier.

2. The method of claim 1, wherein the method includes grouping the shipments of the shipment data based on one or more of the following: month, destination port, commodity, transport location, and transport mode.

3. The method according to any one of the preceding claims, wherein the location prediction model comprises one or more of the following: label encoding and multi-class classification machine learning model.

4. The method of claim 3, wherein the label encoding is performed on the grouped shipments.

5. The method according to any one of the preceding claims, wherein the method comprises grouping consignees into: a first group of consignees, the first group of consignees including consignees present in the system of the electronic device; and a second group of consignees, the second group of consignees including consignees not present in the system.

6. The method of claim 5, wherein predicting the one or more transport locations based on the shipping data comprises: Based on the shipping data, predict one or more shipping locations associated with at least one consignee in the second consignee group.

7. The method of claim 6, wherein predicting one or more shipping locations associated with at least one consignee in the second group of consignees based on the shipping data comprises: - Determine whether at least one of the one or more transport locations predicted for the first consignee group meets the criteria; - When it is determined that at least one of the one or more transport locations meets the criteria, the at least one transport location is provided as part of the one or more transport locations associated with at least one consignee in the second consignee group; as well as - When it is determined that at least one of the one or more transport locations does not meet the criteria, the nearest transport location is provided as part of the one or more transport locations associated with at least one consignee in the second consignee group.

8. The method according to any one of the preceding claims, the method comprising: Based on the shipping data, a turnaround time parameter indicating the turnaround time of the container is determined; Determining the turnaround time parameter based on the shipping data includes applying a turnaround time prediction model to the shipping data.

9. The method according to any one of the preceding claims, the method comprising: - Compare the turnaround time parameter of each of the one or more predicted transport locations with a threshold; as well as - Based on the comparison and aggregation value parameters, determine the penalty value for the supplier to cause container delays exceeding a certain time threshold for each of the one or more predicted shipping locations.

10. The method of claim 8, wherein providing the supplier delay output based on the one or more predicted transport locations comprises: Based on the penalty value associated with the supplier, document data is generated for each of the one or more predicted shipping locations.

Images