A method for generating shipment operational data and related electronic device

Abstract

Disclosed is a method, performed by an electronic device. The method comprises obtaining, from a plurality of sources, shipment data indicative of shipment operations. The method comprises predicting, based on the shipment data, a booking ratio by applying a machine-learning optimization model to the shipment data. The method comprises generating, based on the predicted booking ratio, shipment operational data to be used for control of an upcoming shipment.

Description

[0001] A METHOD FOR GENERATING SHIPMENT OPERATIONAL DATA AND RELATED ELECTRONIC DEVICE

[0002] The present disclosure pertains to the field of transport and freight. The present disclosure relates to a method for generating shipment operational data, and related electronic device.

[0003] BACKGROUND

[0004] In an increasingly connected world, there is an ever greater number of shipments transporting commodities to various ports and terminals across the globe. For each of these shipments, there is a shipping operator in charge of the shipment. However, due the variety, complexity, and sheer volume of shipments and services associated with these shipments, it has become increasingly difficult for shipping operators to control shipment operations, e.g. in a more predictable manner.

[0005] SUMMARY

[0006] There is a need for an electronic device and a method that may improve control of shipment operations and / or logistical operations associated with a shipment, e.g. in a more predictable manner.

[0007] Accordingly, there is a need for an electronic device and a method for generating shipment operational data, which mitigate, alleviate or address the shortcomings existing and may allow for provision of more accurate, robust, and / or reliable control of an upcoming shipment (such as operationally).

[0008] Disclosed is a method, performed by an electronic device. The method comprises obtaining, from a plurality of sources, shipment data indicative of shipment operations. The method comprises predicting, based on the shipment data, a booking ratio e.g., by applying a machine-learning optimization model to the shipment data. The method comprises generating, based on the predicted booking ratio, shipment operational data to be used for control of an upcoming shipment.

[0009] Disclosed is an electronic device comprising memory circuitry, processor circuitry, and an interface, wherein the electronic device is configured to perform any of the methods disclosed herein.

[0010] P24-061 PCT1Disclosed is a computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by an electronic device (optionally with a display and a touch-sensitive surface) cause the electronic device to perform any of the methods disclosed herein.

[0011] It is an advantage of the present disclosure that the disclosed electronic device and method provide improved control of the shipping operation, e.g. in a more predictable manner. This may lead to operational efficiency of shipping capacity, e.g. by prediction of the booking ratio for upcoming shipment. Generating the shipment operational data (e.g. for control of the upcoming shipment) may enable optimization of shipping capacity and may lead to a reduction in unused potential freight capacity, and in carbon footprint.

[0012] Further, the disclosed methods and electronic device may enable control of an upcoming shipment to reduce a container turnaround time. It may be appreciated that the method advantageously utilizes generative Al-driven (Artificial Intelligence-driven) data extraction which may result in a reduction error rate and increased accuracy. In other words, obtaining the shipment data (such as Demurrage & Detention data, and / or other suitable data) using generative Al may advantageously enable improved efficiency and / or accuracy of data extraction. Reduction of error rate may lead to faster data processing processes. The model relies on comprehensive data integration which may provide information allowing for data visualization (e.g. dashboards and / or User interfaces, Ul) that may enhance comparison and decision-making in controlling shipment operations based on the disclosed shipment operational data.

[0013] The present disclosure allows consolidating diverse data sources of shipment data, providing a holistic view of the overall system and context, and allowing an improved prediction of booking ratio, and thereby an improved control of the upcoming shipment via the generated shipment operational data disclosed. This may also enable a cross-domain optimization of the shipment operations.

[0014] The present disclosure provides for example an improved scalability and adaptability of the shipment operational data for upcoming shipments. The disclosure allows evolution of the machine-learning optimization model with more data and scaling to support expanding operations. The disclosed technique is scalable for global operations and adapts over time, continuously improving as new data is introduced.

[0015] P24-061 PCT1I BRI EF DESCRIPTION OF THE DRAWINGS

[0016] The above and other features and advantages of the present disclosure will become readily apparent to those skilled in the art by the following detailed description of example embodiments thereof with reference to the attached drawings, in which:

[0017] Fig. 1 is a flow-chart illustrating an example method, performed by an electronic device, for generating shipment operational data according to this disclosure,

[0018] Fig. 2 is a table representative of example shipment data obtained from a publicly available source according to this disclosure,

[0019] Fig. 3 is a table representative of example shipment data obtained from a source of an operator associated with the shipment operations,

[0020] Fig. 4 is a block diagram illustrating an example electronic device according to this disclosure, and

[0021] Fig. 5 is a diagram illustrating schematically an example representation of shipment data obtained via a source from an operator associated with the shipment according to this disclosure, and

[0022] Figs. 6A-6B show diagrams illustrating schematically an example representation of shipment data obtained from a publicly available source according to this disclosure.

[0023] DETAILED DESCRIPTION

[0024] Various example embodiments and details are described hereinafter, with reference to the figures when relevant. It should be noted that the figures may or may not be drawn to scale and that elements of similar structures or functions are represented by like reference numerals throughout the figures. It should also be noted that the figures are only intended to facilitate the description of the embodiments. They are not intended as an exhaustive description of the disclosure or as a limitation on the scope of the disclosure. In addition, an illustrated embodiment needs not have all the aspects or advantages shown. An aspect or an advantage described in conjunction with a particular embodiment is not necessarily limited to that embodiment and can be practiced in any other embodiments even if not so illustrated, or if not so explicitly described.

[0025] The figures are schematic and simplified for clarity, and they merely show details which aid understanding the disclosure, while other details have been left out. Throughout, the same reference numerals are used for identical or corresponding parts.

[0026] P24-061 PCT1A shipment may for example be seen as transportation of an item from a first location (e.g., origin) to a second location (e.g., destination). For example, the shipment may be seen as shipment of a container comprising one or more commodities. A shipment may involve one or more modalities of transportation such as ocean carrier, land carrier, and / or air carrier. For example, the shipment may be carried out using one or more of: land vehicle, aircraft and seacraft (e.g., a cargo ship).

[0027] In the shipping and logistics industry, for example, time allowances and Demurrage & Detention (D&D) impact both customers and shipping companies and play a crucial role in optimizing efficiency of operations. When using shipping services, these factors can influence the booking of shipments.

[0028] The inventors have realized that advanced Al models can provide deep insights from various sours of data (e.g. publicly available shipment data, and private shipment data). The machine learning (ML) models disclosed herein have been found to be judicious for determining how difference in shipment data affects shipment operations. The present disclosure proposes, inter alia, to forecast and / or predict a booking ratio, e.g. based on booking patterns across various shipment data sets by accounting for historical shipment data and evolving demand trends.

[0029] Fig. 1 shows a flow diagram of an example method 100, performed by an electronic device, for generating shipment operational data according to the disclosure. The electronic device is the electronic device disclosed herein, such as the electronic device 300 of Fig. 4.

[0030] The method 100 comprises obtaining S102, from a plurality of sources, shipment data indicative of shipment operations. Obtaining S102 the shipment data for example comprises receiving and / or retrieving, from a plurality of sources, shipment data indicative of shipment operations. Obtaining S102 shipment data for example comprises generating shipment data indicative of shipment operations. Shipment data can be seen as data indicative of shipment operations, such as associated with shipment operations. For example, shipment data can be seen as data characterizing shipment operations, such as transport operations of an item from a first location to a second location, e.g. over land, sea, and / or air, including storage at terminals. For example, shipment data can comprise that describes details of an item or of a container being transported from one location to P24-061 PCT1another. In one or more example methods, the shipment data comprises one or more of: booking data, freight data, start time data of the shipment, destination country data, trade direction data, port data, container type data, tariff category data, time extension data, transaction data, currency data, and shipping operator data. Shipment operations can be seen as any operation associated with a shipment, transport of commodities, storage of commodities (e.g., at a freight terminal), demurrage & detention, etc. Shipment data may for example be used in the control of a shipment, such as an upcoming shipment. The shipment data is for example associated with one or more shipments at a given port, such as arriving at and / or departing from a given port. Example shipment data can be seen in Tables 10 and 20 of Fig. 2 and Fig. 3 respectively.

[0031] A source as disclosed herein may be seen as a data source, such as a source of shipment data. In some examples, the source may be a database, a data repository, that can be accessed via a device, such as a server device. For example, a source may comprise a data storage medium such as a database, e.g., comprising one or more interfaces capable of communicating (such as directly and / or indirectly) with the electronic device disclosed herein. The plurality of sources may comprise a first source, a second source, optionally a third source, and optionally a fourth source, etc. In one or more example methods, the plurality of sources comprises a publicly available source, a source from an operator associated with the shipment operations, and / or a source of historical shipment data. The operator associated with the shipment operations may for example be a provider of the shipment vessel and / or an operating crew of the shipment vessel. For example, the operator (such as the consignor) may be a shipping company. The publicly available source may comprise one or more sources of publicly available (e.g., publicly accessible) shipment data. The shipment data obtained via a source from an operator associated with the shipment operations may for example be seen as non-publicly available shipment data, such as private shipment data, e.g. data not accessible to the public. In one or more example methods, one or more sources of the plurality of sources may be associated with a different shipping operator. For example, the publicly available source may be associated with a shipping operator different to the operator associated with the shipment operations. In some examples, the plurality of sources may comprise one or more publicly available sources may be associated different shipping operators. In some examples, the source of the historical shipment data may be the same as the publicly available source (as shown in Figs. 6A-6B) or the source from an operator associated with the shipment operations (as shown in Fig. 5). In other words, historical

[0032] P24-061 PCT1shipment data can come from historical shipment data associated with the operator of the shipment operations and / or from a publicly available source.

[0033] The historical shipment data disclosed herein can be seen as previous shipment data, such as shipment data associated with one or more historical shipments (e.g., previously occurred shipments).

[0034] In one or more example methods, the shipment data comprises one or more of: booking data, freight data, start time data of the shipment, destination country data, trade direction data, port data, container type data, tariff category data, time extension data, transaction data, currency data, and shipping operator data.

[0035] Booking data can be seen as data associated with booking of a shipment, such as a past, current and / or upcoming shipment that has been booked. For example, a booking request is provided by a user, possibly via a booking platform. In some examples, the user may be a booking agent. In some examples, the user is a consignee. The user is for example a user of an electronic device, such as an electronic device communicatively coupled with the electronic device disclosed herein (e.g., electronic device 300 of Fig. 4). The booking data for example comprises information associated with the booking, such as one or more of: booking identifier, user identifier, operator identifier (e.g. identifier of the shipping company), items to ship, origin, destination, container category, service category, time extension options, and D&D options. For example, the booking data may comprise information indicative of how many persons visited a website for booking of the shipment and / or searched for the booking of a given shipment. The booking data may be based on user input provided by the user, such as provided in a booking request. In some examples, the booking data comprises one or more of: shipping operator data, start time data of the shipment, destination country data, trade direction data, port data, container type data, tariff category data, time extension data, transaction data, and currency data. The freight data for example comprises data indicative of freight, such as data indicating one or more operations associated with the freight (e.g., the commodities) being transported in the shipment. In some examples, the freight data comprises data indicative of a charge for a freight. In some examples, the freight data comprises a Basic Ocean Freight (BAS) data associated with a given shipment. The freight data (e.g., the BAS) may for example comprise data associated with one or more charges for the service of transportation of freight from the start of the shipment to the end of the shipment (e.g., from the port of departure to the port of arrival). In some examples, the freight data, e.g., P24-061 PCT1the BAS data, may depend on the container and / or commodity being shipped. In one or more example methods, the BAS may be seen as a minimum charge for a given shipment. A commodity may be seen as an object and / or an item that is to be placed in a container for transport. For example, a commodity may be seen as an item of freight, e.g., an object to be shipped. For example, the commodities may be seen as items that can be packed into containers. In some examples, the freight data one or more of: shipping operator data, start time data of the shipment, destination country data, trade direction data, port data, container type data, tariff category data, time extension data, transaction data, and currency data.

[0036] Table 10 and table 20 shown in Figs. 2-3 show example shipment data obtained in S102. The columns of Table 10 of Fig. 2 and Table 20 of Fig. 3 show (from left to right) the shipping operator data, start time data of the shipment, destination country data, trade direction data, port data, container type data, tariff category data, time extension data, transaction data, and currency data. The start time data of the shipment for example indicates the time of departure (e.g. tentative, planned, and / or scheduled) from the port at the start of the shipment, and / or or at the start of a given trip of the overall shipment. The destination country data is for example indicative of the destination country of the shipment. The trade direction data is for example indicative of the direction of trade of the shipment. For example, the trade direction data may be indicative of export and / or import of commodities. The port data for example indicates the port of arrival and / or departure of a given shipment. For example, the port data may indicate that a shipment may arrive and / or depart from one or more ports in a given area, e.g., in a given country. The container type data for example indicates a container type transported as a part of a shipment. For example, the container type data may indicate one or more properties of the container. For example, the container type data may indicate: a container dimensions, a container conditions (such as dry container or reefer container), open top or closed top, etc. The container type data for example indicates whether the container is a special type of container, such as a flat-rack container, insulated container, ventilated container, etc.

[0037] The time extension data for example comprises information indicative of a period (e.g., number of days) for which no charge or no tariff is incurred for container usage. For example, the time extension data is indicative of the demurrage and / or detention cut-off time, such as the last day of demurrage and / or detention before incurring a tariff. Stated differently, the time extension period can for example be seen as demurrage and / or detention free time days (e.g., a free time allowance). The time extension may differ P24-061 PCT1depending on the commodity of a container, the consignee associated with the container, the properties of the container (e.g., type and / or size of the container), the port location, the booking type, the container type etc. In some examples, the time extension may be granted by an authority associated with the port. The time extension data (such as free time allowance) may for example play a role with regard to container management and availability. To enhance retention and increase bookings, it may be appreciated to optimize time allowances and D&D costs, achieving a balance between optimized pricing and revenue generation.

[0038] Demurrage can be seen as the time from which a container arrives at a terminal and / or a port to the time the container is collected from the terminal and / or port to be transported to a recipient and / or destination (e.g., for unloading of at least a part of the container).

[0039] Detention can be seen as the time from which the container is collected from the terminal and / or port (e.g., for unloading) until the container is returned to the terminal and / or port (e.g., after unloading). For example, the returned container may comprise fewer items than prior to transportation from the terminal and / or port towards the recipient. In some examples, the returned container may be empty. Demurrage and / or detention may be referred to herein as D&D or DND.

[0040] The tariff category data is for example indicative of the tariff type associated with the shipment, such as for a given commodity of the shipment. The tariff category data for example indicates whether the tariff associated with the shipment is a demurrage tariff, a detention tariff, or a merged tariff (both demurrage and detention).

[0041] The transaction data is for example indicative of a transaction associated with the shipment. For example, the transaction data is indicative of a rate applied to a transaction associated with the shipment. The transaction is for example a transaction between a shipping operator (e.g., consignor) and a user (e.g., consignee). The transaction data for example comprises one or more tariffs associated with D&D (such as D&D tariffs) and / time extension. For example, transaction data comprise one or more D&D tariff rates indicative of a cost incurred for each day where the free time period is exceeded.

[0042] Each of tables 10, 20 shown in Figs. 2-3 comprise time extension data (shown in columns 9-11, 12-13) and transaction data (shown in columns 12 and 14). The transaction data shown in tables 10, 20 of Figs. 2-3 comprises data associated with a time extensions called “Slab 1” and “Slab 2”. “Slab 1” and “Slab 2” are time periods (e.g., a range of time), such as a number of days. The start of the time period “Slab 1” is denoted by “Slab 1 start P24-061 PCT1day” in column 10 and the end of the time period “Slab 1” is denoted by “Slab 1 end day” in column 11. Transaction data in form of a rate (e.g., in currency USD) for each day of “Slab 1” is indicated in column 12 as “Slab 1 rates”. For example, in row 2 of Table 10 shown in Fig. 2, when the 21 days of free time are exceeded in Slab 1, a D&D tariff of 37 USD is incurred for each day beyond the 21 days.

[0043] The start of the time period extension for e.g. “Slab 2” is denoted by “Slab 2 start day” in column 13 and the end of the time period “Slab 2” is denoted by “Slab 2 end day” in column 14. The rate time period for “Slab 2” has no designated end as illustrated by the term “onwards” used in column 14. A rate (e.g., in currency USD) for each day of “Slab 2” is indicated in column 15 as “Slab 2 rates”.

[0044] The currency data is for example indicative of the currency associated with the shipment data. For example, when the currency data is indicative of “USD” any cost provided in the shipment data is in the currency “USD”. In other words, any cost shown in the shipment data may be configured to correspond with that of the currency data.

[0045] The shipping operator data is for example indicative of the operator of the shipment, such as indicative of the company operating (e.g., in charge of) the shipment, such as operating a vessel associated with the shipment. For example, the shipping operator data comprises a shipping operator identifier.

[0046] The historical shipment data as disclosed for example comprises historical booking data, historical freight data, historical start time data of the shipment, historical destination country data, historical trade direction data, historical port data, historical container type data, historical tariff category data, historical time extension data, historical transaction data, historical currency data, historical shipping operator data, historical booking ratios, import demand data and / or export demand data.

[0047] In some examples, the shipment data, such as the historical shipment data, comprises information indicative of a demand for a given shipment.

[0048] Table 1 shows example historical shipment data, including historical booking data inter alia. The historical shipment data shown in Table 1 was obtained via a source from an operator associated with the shipment operations. Table 1 shows an example month that corresponds with a row of the historical shipment data. Table 1 shows an example historical booking data (e.g., number of bookings, an example historical import and / or export demand data, e.g., number of searches), an example historical booking ratio, the P24-061 PCT1number of Forty Foot Equivalent (FFE) booked, an example historical freight data (e.g., base freight, such as BAS, total freight and / or average base freight per FFE). For example, the electronic device may be configured to predict a booking ratio for a given month (e.g., a given row of Table 1). FFE is for example a standard unit of measurement in container shipping, representing a 40-foot long container.

[0049]

[0050] Table 1

[0051] Table 1 of historical shipment data is for example associated with a given location (such as country and / or port) and / or a given shipping operator. For the historical shipment data, the number of searches shown in Table 1 is a number of visitors on a booking website of the shipping operator. The booking ratio (e.g. historical booking ratio) shown in Table 1 is determined based on the number of bookings and the number of searches of Table 1 , e.g

[0052] P24-061PCT1from the historical shipment data (e.g. historical booking ratio is calculated as (historical Number of Bookings / historical Number of Searches)* 100). The example historical shipment data shown in Table 1 is historical shipment data after the pre-processing S102A of the historical shipment data. The historical shipment data can be used for training the machine-learning optimization model.

[0053] The method 100 comprises predicting S104, based on the shipment data, a booking ratio by applying S104A a machine-learning optimization model to the shipment data. The booking ratio can be seen as a ratio indicative of the likelihood of a shipment to be booked by a user. For example, the booking ratio may be indicative of a number of persons that have searched for a booking, or viewed a booking but have not proceeded to finalize and complete the booking with a booking transaction. For example, when only historical shipment data is used as input into statistical functions: when there has been 100 searches resulting in 50 booked transactions, the past booking ratio is 50%. However, in the present disclosure, a future booking ratio requires more advanced modelling as disclosed herein.

[0054] In some examples, predicting S104 the booking ratio can be seen as determining a future booking ratio for an upcoming shipment. This may be seen in some instances, as predicting the demand for a shipment, e.g., a shipment having one or more given tariffs, such as D&D tariffs. The booking ratio can for example indicate the user context or user conditions for booking, where the user context or user conditions may be seen as circumstances (e.g., increased demand for shipments) that can affect the likelihood of user booking a shipment.

[0055] The machine learning optimization model is for example configured to take as input the shipment data. For example, the machine learning optimization model is configured to take as input one or more of: D&D tariffs, free time days, BAS, import / export demand. The machine-learning optimization model can be seen as providing the booking ratio as an output.

[0056] The method 100 comprises generating S106, based on the predicted booking ratio, shipment operational data to be used for control of an upcoming shipment. For example, the shipment operational data is generated and used for control of an upcoming shipment, such as assigning containers to vessels, triggering booking of Berth in upcoming terminals, and / or of transportation inside and outside the terminal, and / or BAS for upcoming shipment, D&D for upcoming shipment etc. For example, the method 100 P24-061 PCT1comprises controlling, based on the generated shipment operational data, an upcoming shipment. In some examples, the shipment operational data can be seen as operational data (e.g. rate data, time extension data, D&D data, tariff data) as updated based on the predicted booking ratio. In other words, generating S106 the shipment operational data may comprise updating, based on the predicted booking ratio, the rate data, time extension data, D&D data, tariff data. For example, the shipment operational data may be seen as dynamically updated operational data for the upcoming shipment, such as generated dynamically based on booking ratio predicted in S104. In some examples, generating S106 shipment operational data can be seen as optimizing (e.g. adjusting dynamically) the shipment operational data based on the predicted booking ratio, e.g. to maximize the overall results of the booking (e.g. acceptance by the user, and pricing, and resource usage (e.g. shipping space and / container space)). In other words, by generating (e.g. adjusting) shipment operational data (e.g. D&D tariffs) dynamically based on the predicted booking ratio, operational efficiency (and / or revenue) can be maximized by balancing costs with anticipated demand variations. For example, D&D data is adjusted dynamically and the predictions of booking ratios at different BAS rates and Import / export demand is generated. This can allow characterizing a trade-off between demand and D&D data. The predicted booking ratio may for example allow for predictive information for operational planning, such as when to provide reduced D&D tariffs and / or when to adjust free time periods.

[0057] This proactive approach to D&D tariff setting and / or free time period setting can advantageously enable increased revenue for the shipping operator, reduced container turnaround time of a shipment, and / or increased the likelihood that the user selects and completes the booking (which may lead satisfaction of customers booking the shipment). Table 2 shows example shipment data and corresponding predicted booking ratios provided by the disclosed method. The first 3 columns (3 leftmost columns) of Table 2 show example shipment data used to generate the predicted booking ratio (4th column). In other words, the shipment data of Table 2 is taken as input into the disclosed machinelearning optimization model which outputs the predicted booking ration illustrated in Table 2. The booking ratio is provided, e.g. by the machine-learning optimization model. Column 4 of Table 2 shows the input of the machine-learning optimization model. The average base freight shown (e.g., average, avg, BAS) in Table 2 corresponds with the average based freight. Table 2 shows example transaction data (e.g., D&D tariffs), example import and / or export demand data (e.g., import and export demands).

[0058] P24-061 PCT1The example predicted booking ratios shown in Table 2 is predicted in S104 of the method disclosed herein.

[0059] &

[0060]

[0061] Table 2.

[0062] In one or more example methods, the shipment operational data comprises updated rate data (e.g. Freight per FFE rate data, D&D rate data), updated time extension period data, and / or updated tariff data (e.g. D&D tariff data, Freight per FFE tariff data). Shipment operational data may for example be used for assisting in planning and controlling shipment operations. For example, the shipment operational data may be seen as data associated with an operation of a shipment, such as an upcoming shipment. In some examples, shipment operational data may be used to optimize operation of a shipment, while reducing shipment operational costs and / or carbon footprint. The updated rate data

[0063] P24-061PCT1for example comprises updated D&D rate data (e.g., updated D&D tariff data). In other words, the updated rate data may comprise an updated charge based on the predicted booking ratio that may be incurred for each day demurrage and / or detention used beyond the allocated free time days. The updated time extension period data for example comprises updated time extension data, e.g., an updated number of free time days, such as D&D free time days.

[0064] In one or more example methods, obtaining S102 the shipment data comprises preprocessing S102A the shipment data. In one or more example methods, pre-processing S102A the shipment data comprises applying S102AA one or more: a conversion, an error correction technique, a normalization of the shipment data, a noise reduction technique, removing null data elements, and an extraction technique. In other words, preprocessing S102A the shipment data may comprise applying advanced large language models (LLMs) to extract structured information from unstructured documents, such as PDF files, excel files, and / or tariff documents from public sources, such as from other shipping operators. Obtaining S102 the shipment data for example comprises obtaining (such as extracting) the shipment data, such as time extension data and / or transaction data (such as D&D tariff data), from a source that is from an operator associated with the shipment operations. For example, obtaining the shipment data from an operator associated with the shipment operations may comprise accessing internal data sources to extract the shipment data. This extraction may enable processing accurately shipment data independently from where the shipment data comes from, thereby providing a foundation or basis for accurate comparison with shipment data obtained from a publicly available source (e.g., externally obtained).

[0065] Obtaining S102 the shipment data for example comprises obtaining (such as extracting) historical shipment data, historical booking ratios, and / or historical freight data (such as historical BAS) from one or more of the plurality of sources. The historical shipment data may be applied for identifying patterns and trends in the shipment data, which may be integrated into the machine learning optimization model.

[0066] In some examples, the applying S102AA the conversion may comprise converting one or more data elements of the shipment data (such as D&D data). Shipment data (such as D&D data) obtained in S102 from different sources may comprise different data elements for equivalent data sets. For example, shipment data obtained from a publicly available source may be labelled as “shipment start date” to indicate the start time data of the

[0067] P24-061 PCT1shipment whereas shipment data obtained from an operator associated with the shipment operations may be labelled as “effective start date” (as shown in column 2 of Tables 10 and 20 of Fig. 2 and Fig 3 respectively). This discrepancy in the shipment data may lead to inaccuracy when predicting S104 the booking ratio. For example, applying S102AA the conversion allows consolidating the data elements of the shipment data to the same terminology, such as a converting data elements of the shipment data to a common data label (e.g., to a standard terminology), thereby enabling an improved accuracy of the booking ratio predicted in S104.

[0068] An error correction technique as disclosed herein may be seen as a method or process that identifies and / or corrects errors within the shipment data (such as D&D data), such as prior to predicting S104 the booking ratio. For example, the error correction technique can be seen as a technique configured to correct, rectify and / or adjust erroneous data entries in the shipment data, that have been flagged as erroneous. In some examples, applying S102AA the error correction technique comprises identifying and / or correcting one or more errors in the shipment data by applying a machine learning (ML) model, such as a Large Language Model (LLM) to the shipment data. Applying S102AA the error correction technique may comprise removing, substituting, and / or interpolating (such using an ML model) at least a part of the shipment data. The ML model, such as the LLM, may for example be configured, such as prompted and / or trained, to perform error correction of the shipment data. The error correction technique disclosed herein may advantageously enable a reduction in the redundancies of the shipment data, thereby enabling the predicted booking ratio and / or the generated shipment operational data to have an improved accuracy.

[0069] Normalization of the shipment data (e.g., using one or more normalization techniques) may for example be seen an adjustment that puts the shipment data to a predetermined format, e.g., prior to predicting S104 the booking ratio. For example, normalization of the shipment data may comprise the application one or more formatting and / or scaling techniques. Applying S102AA the normalization for example comprises one or more normalisation techniques such as: min-max normalization, z-score normalization, and decimal scaling to the shipment data.

[0070] The noise reduction technique may for example be seen as a technique for reducing noise in the shipment data. For example, the noise reduction technique comprises filtering the shipment data to reduce noise (e.g., reduce unwanted variations and / or disturbances)

[0071] P24-061 PCT1of the shipment data, e.g., prior to predicting S104 the booking ratio. In other words, the noise reduction technique can be seen as improving the accuracy of the shipment data. Applying S102AA the noise reduction technique may for example comprise applying one or more of: Fourier transform filters, Gaussian smoothing, and median filtering techniques to the shipment data.

[0072] The removal of null elements for example comprises identifying and / or removing null entries in the shipment data, e.g., prior to predicting S104 the booking ratio. In one or more examples, the removal of null elements comprises discarding or imputing values in place of null entries to enrich data completeness. The term "null" may refer to any data point that is absent, undefined, or lacking value within a dataset. In some examples, preprocessing S102A comprises applying S102AA a removal of zero value elements, such as elements having a zero value.

[0073] In one or more example methods, the extraction technique comprises a transformer model configured to extract shipment data, and relation between elements of the shipment data, and context. Context as disclosed herein may for example be seen as information indicative of which column, property, and / or attribute holds which value. In one or more example methods, the context can be seen as the information conveyed by the each data element of the shipment data. For example, the second row of the second column conveys that the effective start data of the shipment is 5thJan 2024.

[0074] The transformer model is for example a machine learning model, e.g., a neural network. For example, the transformer model may be seen as a neural network based on a transformer architecture. In some examples, the transformer model is an LLM, such as a generative pre-trained transformer (GPT) model. In one or more example methods, the transformer model comprises a generative transformer model, such as a multi modal and / or multilingual generative pre-trained transformer model. In one or more example methods, the transformer model comprise the GPT 4o model. The transformer model may for example be trained to extract shipment data. In some examples, the transformer model is configured to extract a relation between discrete data elements of the shipment data (e.g., between rows and / or columns of Table 2 and / or Table 3). The transformer model may for example be seen as a generative artificial intelligence (Al) model. In some examples, applying S102AA the transformer model to the shipment data using an Application Programming Interface (API).

[0075] P24-061 PCT1For example, applying S102AA an extraction technique to the shipment data comprises applying a generative Al model (e.g. a transformer model) to the shipment data. For example, applying S102AA the extraction technique to the shipment data comprises extracting, e.g., using the generative Al model (e.g. transformer model), transaction data (such as D&D tariff data) from obtained shipment data, such as shipment data obtained from a publicly available source. The transformer model may for be configured to, e.g. identify and / or extract key information, including time extension data (e.g., free time days) and / or transaction data (such as D&D tariffs). In some examples, the pre-processing S102A of the shipment data allows integrating shipment data obtained from one or more different sources into the same platform, thereby enabling improved consistency in formatting and a reduction in redundancies. In other words, integrating the shipment data obtained from one or more different sources may enable consolidation of shipment data, (such as time extension data) and transaction data across two or more different shipping operators. In some examples, integrating the shipment data comprises cleaning and / or harmonizing the shipment data, thereby ensuring accuracy and consistency of across the obtained shipment data to facilitate an improved accuracy of the prediction of the booking ratio by the machine learning optimization model.

[0076] In one or more example methods, the machine-learning optimization model is configured to determine how a variation in the shipment data impacts the booking ratio. Stated differently, the machine-learning optimization model may be configured to determine how the variation in shipment data (e.g., one or more of: D&D tariffs, D&D free time days, BAS, import / export demand, and / or past booking ratios) affect a future booking ratio. The machine learning optimization model is for example configured to perform an optimization of one or more shipment operations of the shipment (e.g. booking ratio) based on the shipment data provided as input.

[0077] In one or more example methods, the machine-learning optimization model is configured to determine future patterns (such as future demand) in booking ratios. For example, the machine-learning optimization model may be configured to determine (such as evaluate), e.g., based on the obtained shipment data, one or more of the following: past performance (such as past results), a D&D tariff sensitivity, and market demands. In other words, the machine-learning optimization model may be configured to predict (such as forecast) booking ratios (such as booking patterns) across various D&D tariffs by accounting for historical shipment data on e.g., historical BAS, historical D&D charges (such as tariffs), and / or evolving shipment demand trends. In some examples, the

[0078] P24-061 PCT1machine learning optimization model can be seen as capable of predicting future trends in container bookings and / or future user behaviour (such as future demand). The machine learning optimization model can be seen as configured to determine (such as assess) how the variation in shipment data (such as a variation in one or more of: D&D tariffs, time extension data, freight data (such as BAS), an import / export demand, and / or past booking ratios affect a future Booking ratio. It may be appreciated that the disclosed models enable dynamic tariff adjustments (for optimizing (maximum) results, e.g. revenue). This may also lead to operational efficiency: Faster provision of information leading to e.g. cost savings and reduced D&D penalties through optimized tariff structures.

[0079] In one or more example methods, the machine-learning optimization model comprises one or more regression models. The one or more regression models may for example be configured to characterize the relationship between the booking ratio and at least one element of the shipment data. For example, by applying a regression technique to the obtained shipment data, the relationship between the shipment data and the booking ratio is characterized with coefficients / states etc and can be used to predict the value of the booking ratio based on at least one element of the shipment data. In one or more example methods, the one or more regression models comprise one or more supervised learning regression models.

[0080] In one or more example methods, the one or more regression models comprises a polynomial regression model, a random forest model, and / or a gradient boosted regression model. A polynomial regression model may be seen as a model characterizing the relationship between the booking ratio and at least one element of the shipment data by fitting a polynomial equation to obtained shipment data and can for example be used to predict the value of the booking ratio based on at least one element of the shipment data. A random forest model may be seen as an ensemble learning method for classification, regression and other tasks that operates by constructing a collection of decision trees based on the shipment data. For example, the shipment data may be subject to bagging and feature randomness for building each individual tree to predict the booking ratio. A gradient boosted regression model may be seen as a model configured to carry out an ensemble learning method using pseudo-residuals that operates by constructing a collection of decision trees, e.g., based on the voyage data. The decision trees constructed using the gradient boosted regression model may for example be seen as gradient-boosted trees. In other words, the gradient boosted regression model may focus P24-061 PCT1on minimizing errors by building models sequentially and / or refining performance on difficult-to-predict data points. The gradient boosted regression model may advantageously enable improved capturing of patterns, such as pattern recognition, and may allow for improved accuracy of predictions, e.g., predicting the booking ratio.

[0081] In one or more example methods, the one or more regression models comprise a linear regression model.

[0082] In one or more example methods, the method 100 comprises determining S108 results by applying S108A a maximization technique. For example, determining S108 the results may comprise determining one or more values, such as values determined based on the shipment operational data generated in S106. The maximisation technique may for example be seen as a technique for maximising the results, e.g., for determining a maximum value of the results determined in S108. In some examples, the results may comprise a revenue, such as a revenue associated with the shipment operational data generated in S106. In one or more example methods, the maximization technique comprises a log-likelihood maximisation. A log-likelihood maximisation may be seen as a statistical technique used to estimate the parameters of a probability distribution that best fit observed shipment operational data. The log-likelihood maximisation may comprise maximizing the logarithm of the likelihood function rather than the likelihood function itself. In one or more example methods, the machine-learning optimization model is trained based on historical booking ratios and historical shipment data. In some examples, the machine-learning optimization model is trained based on historical BAS, historical import and / or export demand, historical D&D tariffs, and / or historical booking ratios. In one or more example methods, the machine learning optimization model is retrained periodically, e.g. monthly. The machine learning optimization model may be retrained when new historical shipment data is available from the historical shipment data source. The method 100 for example comprises periodically obtaining (such as extracting) historical shipment from the historical shipment data source. For example, the method 100 may comprise obtaining updated historical shipment data corresponding with the latest time period that has elapsed, such as the most recently elapsed month. In one or more example methods, when the machine-learning optimization model comprises one or more regression models, one or more of the regression models may be trained using supervised learning.

[0083] P24-061 PCT1In one or more example methods, the machine-learning optimization model is trained via feedback loop to continuously or periodically retrain the machine-learning optimization model based on updated historical shipment data and / or historical booking ratios.

[0084] In one or more example methods, the method 100 comprises displaying S110 a first user interface object representative of shipment data from a first source of the plurality of sources, a second user interface object representative of shipment data from a second source of the plurality of sources, and / or a third user interface object representative of shipment data from a third source of the plurality of sources. A user interface object (such as the first, second, and / or third user interface object) may comprise one or more, such as a plurality of, user interface objects. The first source is for example a publicly available source, such as a source of publicly available data. The second source is for example a source from an operator associated with the shipment operations. The third source is for example a source of historical shipment data. For example, the first, second, and third sources may be different sources.

[0085] In one or more example methods, displaying S110 may comprise providing one or more first user interface objects, one or more second user interface objects, and / or one or more third user interface objects (as shown in Figs 6A-B). A user interface object may refer herein to a graphical representation of a shipment data, the predicted booking ratio, the generated shipment operational data, the determined results, and / or any associated information that is displayed on a display interface (such as a screen) of the electronic device. The user interface object may be user-interactive, or selectable by a user input. For example, an image (e.g., icon), a button, and text (e.g., hyperlink) each optionally comprising a user interface object. The user interface object may form part of a widget. A widget may be seen as a mini-application that may be used by the user.

[0086] In some examples, the method 100 comprises displaying S110 the first user interface object, the second user interface object, and / or the third user interface object as a part of a user interface.

[0087] When the shipment data is pre-processed, such as integrated, the shipment data may be provided via one or more user interface objects, e.g., thereby allowing a user to compare shipment data of the operator of the shipment operations with publicly available shipment data. In some examples, the user interface object (such as first, second, and / or third user interface object) may comprise a bar chart, heat map, and / or trend lines, e.g., provide users with clear insights into market positioning, enabling them to make strategic P24-061 PCT1decisions. In some examples, displaying the first, second, and / or third user interface objects may enable a user to observe how D&D tariffs are in various regions or lanes, enable identification of opportunities for tariff adjustments, and prediction of market reactions. The disclosed methods and electronic device may allow for integration of publicly available shipment data (e.g., tariff data) and shipment data obtained via a source from an operator associated with the shipment operations, thus enabling provision of realtime (such as near real-time and / or pseudo real-time, (e.g., updated hourly, daily, weekly, monthly) visually-driven comparisons for immediate insights into tariff positioning.

[0088] Fig. 2 is a table 10 representative of example shipment data obtained from a publicly available source according to this disclosure. For example, the shipment data of table 10 may be obtained from a first source, such as a publicly available source of shipment data. For example, the table 10 comprises shipment data associated with an operator different to the operator of the shipment operations. For example, the shipping operators “ABCD” and ZXYW” of Table 10 may be different operators to the operator “MKS” of the shipment operations of table 20 of Fig. 3. Each row of Table 1 corresponds with a container of a shipment.

[0089] The rates of each slab, e.g., time period, may vary based on the container type which is indicated in the container type data. For example, in rows 2 and 3 of table 20 the only difference is the container type (as indicated in the container type data) and the slab 1 and 2 rates. In row 2, the container type is a 20 footer container, and in row 3 the container is a 40 footer container (larger than the 20 footer container). The slab 1 and slab 2 rates are greater for the 40 footer container than for the 20 footer container. Hence, as shown in Table 2, the transaction data (e.g., the D&D tariff) may vary based on the container type.

[0090] Fig. 3 is a table 20 representative of example shipment data obtained from a source of an operator associated with the shipment operations. For example, the shipment data of table 20 may be obtained from a second source, such as a source from an operator associated with the shipment operations. The table 20 comprises shipment data associated with the operator of the shipment operations.

[0091] Fig. 4 shows a block diagram of an example electronic device 300 according to the disclosure. The electronic device 300 comprises memory circuitry 301, processor circuitry

[0092] P24-061 PCT1302, and an interface 303. The electronic device 300 is configured to perform any of the methods disclosed in Fig. 1. In other words, the electronic device 300 is configured for generating shipment operational data. In some examples, the electronic device 300 can be seen as a booking ratio prediction device. In some examples, the electronic device 300 can be seen as a shipment operational data generation device.

[0093] The electronic device 300 is configured to obtain (e.g. using the processor circuitry 302 and / or interface 303), from a plurality of sources, shipment data indicative of shipment operations.

[0094] The electronic device 300 is configured to predict (e.g. using the processor circuitry 302), based on the shipment data, a booking ratio, e.g. by applying (e.g., using the processor circuitry 302) a machine-learning optimization model to the shipment data.

[0095] The electronic device 300 is configured to generate (e.g. using the processor circuitry 302 and / or interface 303), based on the predicted booking ratio, shipment operational data to be used for control of an upcoming shipment.

[0096] The processor circuitry 302 is optionally configured to perform any of the operations disclosed in Fig. 1 (such as any one or more of: S102, S102A, S102AA, S104, S104A, S106, S108, S108A, S110). The operations of the electronic device 300 may be embodied in the form of executable logic routines (e.g., lines of code, software programs, etc.) that are stored on a non-transitory computer readable medium (e.g., the memory circuitry 301) and are executed by the processor circuitry 302).

[0097] Furthermore, the operations of the electronic device 300 may be considered a method that the electronic device 300 is configured to carry out. Also, while the described functions and operations may be implemented in software, such functionality may as well be carried out via dedicated hardware or firmware, or some combination of hardware, firmware and / or software.

[0098] The memory circuitry 301 may be one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, a random access memory (RAM), or other suitable device. In a typical arrangement, the memory circuitry 301 may include a non-volatile memory for long term data storage and a volatile memory that functions as system memory for the processor circuitry 302. The memory circuitry 301 may exchange data with the processor circuitry 302 over a data bus. Control lines and an

[0099] P24-061 PCT1address bus between the memory circuitry 301 and the processor circuitry 302 also may be present (not shown in Fig. 4). The memory circuitry 301 is considered a non-transitory computer readable medium.

[0100] The memory circuitry 301 may be configured to store shipment data, shipment operational data, the booking ratio, the machine learning optimization model, the transformer model, determined future patterns in booking ratios, results, historical booking ratios, and / or historical shipment data in a part of the memory.

[0101] Fig. 5 is a diagram illustrating schematically an example representation of shipment data obtained via a source from an operator associated with the shipment according to this disclosure. Fig. 5 shows historical shipment data obtained via a source from an operator associated with the shipment operations. The x-axis shown in Fig. 5 includes months in the “YearMonth” format as used in Table 1. For example, 202406 on the x-axis of Fig. 5 indicates June 2024. The y-axis shown in Fig. 5 includes numbers 0-90. Fig. 5 shows example historical booking ratio (indicated by the line with diamond points). Fig. 5 shows example historical transaction data, such as example D&D tariffs associated with slab 1 (e.g., indicated by the line with square points). Fig. 5 shows example historical transaction data, such as example D&D tariffs associated with slab 2 (indicated by the line with triangle points). Fig. 5 shows example historical time extension data, such as example free time days (indicated by the line with cross points). In some examples, the electronic device may be configured to display a user interface object representative of historical shipment data, such as the diagram shown in Fig. 5.

[0102] Figs. 6A-6B show diagrams illustrating schematically an example representation of shipment data obtained from a publicly available source according to this disclosure. Figs. 6A-6B shows historical shipment data obtained from a publicly available source. The x-axis shown in Figs. 6A-6B includes months in the “YearMonth” format as used in Table 1 and in Fig. 5. The y-axis shown in Fig. 6A includes numbers 0-90 and the y-axis shown in Fig. 6B includes numbers 0-100. Figs. 6A-6B shows example historical transaction data, such as example D&D tariffs associated with slab 1 (e.g., indicated by the line with triangle points). Figs. 6A-6B show example historical transaction data, such as example D&D tariffs associated with slab 2 (indicated by the line with cross points). Figs. 6A-6B shows example historical time extension data, such as example free time days (indicated by the line with star points). In some examples, the electronic device may be configured to

[0103] P24-061 PCT1display a user interface object representative of historical shipment data, such as the diagrams shown in Figs. 6A-6B.

[0104] Embodiments of methods and products (electronic device) according to the disclosure are set out in the following items:

[0105] Item 1. A method, performed by an electronic device, the method comprising:

[0106] obtaining, from a plurality of sources, shipment data indicative of shipment operations;

[0107] predicting, based on the shipment data, a booking ratio by applying a machine-learning optimization model to the shipment data; and

[0108] generating, based on the predicted booking ratio, shipment operational data to be used for control of an upcoming shipment.

[0109] Item 2. The method according to item 1, wherein the shipment data comprises one or more of: booking data, freight data, start time data of the shipment, destination country data, trade direction data, port data, container type data, tariff category data, time extension data, transaction data, currency data, and shipping operator data.

[0110] Item 3. The method according to any of the previous items, wherein the plurality of sources comprises a publicly available source, a source from an operator associated with the shipment operations, and / or a source of historical shipment data.

[0111] Item 4. The method according to any of the previous items, wherein obtaining the shipment data comprises pre-processing the shipment data.

[0112] Item 5. The method according to item 4, wherein pre-processing the shipment data comprises applying one or more: a conversion, an error correction technique, a normalization of the shipment data, a noise reduction technique, removing null data elements, and an extraction technique.

[0113] Item 6. The method according to item 5, wherein the extraction technique comprises a transformer model configured to extract shipment data, and relation between elements of the shipment data, and context.

[0114] P24-061 PCT1Item 7. The method according to any of the previous items, wherein the machine-learning optimization model is configured to determine how a variation in the shipment data impacts the booking ratio.

[0115] Item 8. The method according to any of the previous items, wherein the machine-learning optimization model is configured to determine future patterns in booking ratios.

[0116] Item 9. The method according to any of the previous items, wherein the machine-learning optimization model comprises one or more regression models.

[0117] Item 10. The method according to item 9, wherein the one or more regression models comprises a polynomial regression model, a random forest model, and / or a gradient boosted regression model.

[0118] Item 11. The method according to any of the previous items, the method comprising determining results by applying a maximization technique.

[0119] Item 12. The method according to item 10, wherein the maximization technique comprises a log-likelihood maximisation.

[0120] Item 13. The method according to any of the previous items, wherein the shipment operational data comprises updated rate data, updated time extension period data, and / or updated tariff data.

[0121] Item 14. The method according to any of the previous items, wherein the machinelearning optimization model is trained based on historical booking ratios and historical shipment data.

[0122] Item 15. The method according to any of the previous items, the method comprising displaying a first user interface object representative of shipment data from a first source of the plurality of sources, a second user interface object representative of shipment data from a second source of the plurality of sources, and / or a third user interface object representative of shipment data from a third source of the plurality of sources.

[0123] Item 16. 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 of items 1-15.

[0124] P24-061 PCT1Item 17. A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by an electronic device cause the electronic device to perform any of the methods of items 1-16.

[0125] The use of the terms “first”, “second”, “third” and “fourth”, “primary”, “secondary”, “tertiary” etc. does not imply any particular order, but are included to identify individual elements. Moreover, the use of the terms “first”, “second”, “third” and “fourth”, “primary”, “secondary”, “tertiary” etc. does not denote any order or importance, but rather the terms “first”, “second”, “third” and “fourth”, “primary”, “secondary”, “tertiary” etc. are used to distinguish one element from another. Note that the words “first”, “second”, “third” and “fourth”, “primary”, “secondary”, “tertiary” etc. are used here and elsewhere for labelling purposes only and are not intended to denote any specific spatial or temporal ordering. Furthermore, the labelling of a first element does not imply the presence of a second element and vice versa.

[0126] It may be appreciated that the Figures comprise some circuitries or operations which are illustrated with a solid line and some circuitries or operations which are illustrated with a dashed line. The circuitries or operations which are comprised in a solid line are circuitries or operations which are comprised in the broadest example embodiment. The circuitries or operations which are comprised in a dashed line are example embodiments which may be comprised in, or a part of, or are further circuitries or operations which may be taken in addition to the circuitries or operations of the solid line example embodiments. It should be appreciated that these operations need not be performed in order presented.

[0127] Furthermore, it should be appreciated that not all of the operations need to be performed. The example operations may be performed in any order and in any combination.

[0128] It is to be noted that the word "comprising" does not necessarily exclude the presence of other elements or steps than those listed.

[0129] It is to be noted that the words "a" or "an" preceding an element do not exclude the presence of a plurality of such elements.

[0130] It is to be noted that the term "indicative of" may be seen as “associated with”, “related to”, “descriptive of”, “characterizing”, and / or “defining”. The terms “indicative of”, “associated with”, “related to”, “descriptive of”, “characterizing”, and “defining” can be used interchangeably. The term “indicative of” can be seen as indicating a relation. For example, weight data indicative of weight may comprise one or more weight parameters. P24-061 PCT1It is to be noted that the word "based on" may be seen as “as a function of’ and / or “derived from”. The terms “based on” and “as a function of” can be used interchangeably. For example, a parameter determined “based on” a data set can be seen as a parameter determined “as a function of” the data set. In other words, the parameter may be an output of one or more functions with the data set as an input.

[0131] A function may be characterizing a relation between an input and an output, such as mathematical relation, a database relation, a hardware relation, logical relation, and / or other suitable relations.

[0132] It should further be noted that any reference signs do not limit the scope of the claims, that the example embodiments may be implemented at least in part by means of both hardware and software, and that several "means", "units" or "devices" may be represented by the same item of hardware.

[0133] The various example methods, devices, nodes and systems described herein are described in the general context of method steps or processes, which may be implemented in one aspect by a computer program product, embodied in a computer-readable medium, including computer-executable instructions, such as program code, executed by computers in networked environments. A computer-readable medium may include removable and non-removable storage devices including, but not limited to, Read Only Memory (ROM), Random Access Memory (RAM), compact discs (CDs), digital versatile discs (DVD), etc. Generally, program circuitries may include routines, programs, objects, components, data structures, etc. that perform specified tasks or implement specific abstract data types. Computer-executable instructions, associated data structures, and program circuitries represent examples of program code for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represents examples of corresponding acts for implementing the functions described in such steps or processes.

[0134] Although features have been shown and described, it will be understood that they are not intended to limit the claimed disclosure, and it will be made obvious to those skilled in the art that various changes and modifications may be made without departing from the scope of the claimed disclosure. The specification and drawings are, accordingly to be regarded in an illustrative rather than restrictive sense. The claimed disclosure is intended to cover all alternatives, modifications, and equivalents.

[0135] P24-061 PCT1

Claims

CLAIMS1. A method, performed by an electronic device, the method comprising:obtaining, from a plurality of sources, shipment data indicative of shipment operations;predicting, based on the shipment data, a booking ratio by applying a machine-learning optimization model to the shipment data; andgenerating, based on the predicted booking ratio, shipment operational data to be used for control of an upcoming shipment.

2. The method according to claim 1, wherein the shipment data comprises one or more of: booking data, freight data, start time data of the shipment, destination country data, trade direction data, port data, container type data, tariff category data, time extension data, transaction data, currency data, and shipping operator data.

3. The method according to any of the previous claims, wherein the plurality of sources comprises a publicly available source, a source from an operator associated with the shipment operations, and / or a source of historical shipment data.

4. The method according to any of the previous claims, wherein obtaining the shipment data comprises pre-processing the shipment data.

5. The method according to claim 4, wherein pre-processing the shipment data comprises applying one or more: a conversion, an error correction technique, a normalization of the shipment data, a noise reduction technique, removing null data elements, and an extraction technique.P24-061 PCT16. The method according to claim 5, wherein the extraction technique comprises a transformer model configured to extract shipment data, and relation between elements of the shipment data, and context.

7. The method according to any of the previous claims, wherein the machine-learning optimization model is configured to determine how a variation in the shipment data impacts the booking ratio.

8. The method according to any of the previous claims, wherein the machine-learning optimization model is configured to determine future patterns in booking ratios.

9. The method according to any of the previous claims, wherein the machine-learning optimization model comprises one or more regression models.

10. The method according to claim 9, wherein the one or more regression models comprises a polynomial regression model, a random forest model, and / or a gradient boosted regression model.

11. The method according to any of the previous claims, the method comprising determining results by applying a maximization technique.

12. The method according to claim 11, wherein the maximization technique comprises a log-likelihood maximisation.P24-061 PCT113. The method according to any of the previous claims, wherein the shipment operational data comprises updated rate data, updated time extension period data, and / or updated tariff data.

14. The method according to any of the previous claims, wherein the machine-learning optimization model is trained based on historical booking ratios and historical shipment data.

15. The method according to any of the previous claims, the method comprising displaying a first user interface object representative of shipment data from a first source of the plurality of sources, a second user interface object representative of shipment data from a second source of the plurality of sources, and / or a third user interface object representative of shipment data from a third source of the plurality of sources.

16. 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 of claims 1-15.

17. A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by an electronic device cause the electronic device to perform any of the methods of claims 1-15.P24-061 PCT1

Images