CN111091226B - Transport capacity prediction method based on actual shipping service and data mining - Google Patents
Transport capacity prediction method based on actual shipping service and data mining Download PDFInfo
- Publication number
- CN111091226B CN111091226B CN201911093827.0A CN201911093827A CN111091226B CN 111091226 B CN111091226 B CN 111091226B CN 201911093827 A CN201911093827 A CN 201911093827A CN 111091226 B CN111091226 B CN 111091226B
- Authority
- CN
- China
- Prior art keywords
- ship
- data
- voyage
- port
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000007418 data mining Methods 0.000 title claims abstract description 15
- 230000002159 abnormal effect Effects 0.000 claims abstract description 7
- 238000001914 filtration Methods 0.000 claims abstract description 7
- 238000001514 detection method Methods 0.000 claims abstract description 4
- 230000007306 turnover Effects 0.000 claims description 37
- 238000004364 calculation method Methods 0.000 claims description 24
- 238000013507 mapping Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 9
- 230000001788 irregular Effects 0.000 claims description 5
- 230000001174 ascending effect Effects 0.000 claims description 4
- 238000002474 experimental method Methods 0.000 claims 1
- 230000000694 effects Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 230000009467 reduction Effects 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 238000007667 floating Methods 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000003795 chemical substances by application Substances 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000010006 flight Effects 0.000 description 2
- 239000004576 sand Substances 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013501 data transformation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000004579 marble Substances 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000005180 public health Effects 0.000 description 1
- 230000001172 regenerating effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G06Q50/40—
Abstract
The invention discloses a capacity prediction method based on actual shipping business and data mining, which comprises the steps of firstly extracting ship experimental data from actual shipping industry data, filtering and clearing abnormal data contained in the ship experimental data based on time consistency detection and a preset ship navigation rule, and then combining adjacent voyages into an effective ship historical voyage; then, carrying out feature selection on the basic data and the processed route data to obtain an effective experimental data set; then calculating the voyage turnaround periods of different ship types and air lines in the effective experimental data set according to the air lines and the ship types, and calculating the ship turnaround periods of different ships in the effective experimental data set according to the ship identifications; and finally, predicting the available transport capacity according to the effective experimental data set, the voyage turnaround periods of different ship types and air routes and the ship turnaround periods of different ships. The method can obtain high-quality experimental data, and predict the duration of the voyage times in real time, thereby predicting the available capacity of the market.
Description
Technical Field
The invention relates to the technical field of data mining, in particular to a capacity prediction method based on actual shipping service and data mining.
Background
The big data has the characteristics of large data volume, multiple types, high processing speed, strong data authenticity, low value density and the like. The big data are analyzed, the value of the big data is mined, and the application prospect is wide. At present, the big data technology is integrated into various industries and plays an important role in evaluation and prediction. The big data can help the government to realize market economic regulation, public health safety prevention, disaster early warning and social public opinion supervision; the system can help the airline companies to save operation cost; the system helps enterprises to improve the pertinence of marketing, and reduces the cost of logistics and inventory.
The marine industry is a traditional industry that has lagged behind relatively combining internet and big data technologies. However, the ship industry is absolutely a global and large-scale industry, which is related to design, manufacture and goods transportation, and related to trade market and life of people, and has wide data sources, contains many knowledge and information with potential values, and needs to be mined and discovered. Therefore, the application of a plurality of big data in the ship industry is started. The "munin (united Navigation through intelligent network)" project has been published in europe, aiming at developing a new generation of control system and communication technology to display and control ships at and out of port, which undoubtedly promotes the development of information-based ships and information-based shipping. In 7 months 2014, the japan ship technical research association sets forth a ship "big data road sign" work, collects the voyages of a plurality of ships and related data thereof to form big data, and plans to be used for projects such as energy-saving voyage of ships, ship type development, equipment remote maintenance, and the like. The trend of future shipping informatization development is provided by the mountains and the Marble cloud, the urgent needs of the shipping industry on big data technology are discussed, and meanwhile, revelations are provided for big data application of the ship industry. In addition, people are also actively exploring the positive influence of the big data age on the transformation development of the shipping industry in the low ebb period of the shipping industry.
The inventor of the present application finds that the method of the prior art has at least the following technical problems in the process of implementing the present invention:
by collecting and carrying out statistical analysis and value mining on data obtained by monitoring the operating ship, a lot of valuable data information can be obtained. However, according to the actual situation of the current coastal shipping in China, some companies provide real-time inquiry of ship dynamics, which can provide certain traversals for shipowners, owners, ship agents, freight agents, crews and their families, but cannot comprehensively and accurately acquire available capacity data, so that the capacity cannot be effectively predicted, and at present, only the information of the available capacity in the market can be acquired from each message network side for ship supply and demand prediction.
Therefore, the technical problems of low experimental data quality and poor prediction effect in the transportation capacity prediction exist in the prior art.
Disclosure of Invention
In view of the above, the present invention provides a capacity prediction method based on actual shipping service and data mining, so as to solve or at least partially solve the technical problems of low experimental data quality and poor prediction effect in capacity prediction in the prior art.
In order to solve the technical problem, the invention provides a capacity prediction method based on actual shipping service and data mining, which comprises the following steps:
step S1: extracting ship experimental data from actual shipping industry data, filtering and clearing abnormal data contained in the ship experimental data based on time consistency detection and a preset ship navigation rule, and obtaining processed ship experimental data, wherein the processed ship experimental data comprises basic data and historical course data, the basic data comprises a ship type and a ship identification, and the historical course data comprises a voyage number;
step S2: combining adjacent voyages into an effective historical voyage of the ship according to the time continuity and preset rules of the voyage in the voyage data to obtain processed voyage data;
step S3: carrying out characteristic selection on the basic data and the processed route data to obtain an effective experimental data set;
step S4: calculating the voyage turnaround periods of different ship types and ship routes in the effective experimental data set according to the ship types, and calculating the ship turnaround periods of different ships in the effective experimental data set according to the ship identifications;
step S5: and predicting the available transport capacity according to the effective experimental data set, the voyage turnaround periods of different ship types and air routes and the ship turnaround periods of different ships.
In one embodiment, the basic data includes ship basic attribute data and port basic attribute data, the port basic attribute data includes a port name, and the step S1 specifically includes:
and performing similarity mapping on the irregular port names contained in the basic data and the collected regular port data sets, filling missing values of voyages through the traceability of the ship voyages and based on the adjacent time, and clearing inconsistent data according to the uniqueness rule, the continuity rule and the null value rule.
In one embodiment, step S2 specifically includes:
step S2.1: grouping historical route data according to ship identifications, and performing ascending arrangement according to time;
step S2.2: judging the validity of the airline data according to whether the time of the number of the airline is consistent with the numerical value of the adjacent number of the airline and the reasonability of the parking time;
step S2.3: and if the current voyage number is valid, judging whether the loading and unloading time length of the adjacent voyage number is within a given threshold value, if so, combining the current voyage number and the adjacent voyage number, and repeatedly executing the steps S2.2-S2.3 until the course data of all ships are processed.
In one embodiment, the basic data and the processed route data include continuity data, text attribute data, and data related to determining the time of the route, and step S3 specifically includes:
discretizing continuous data, performing attribute numerical convention on text attribute data, and generalizing data of decision flight time to obtain an effective experimental data set.
In one embodiment, the continuous type data includes a ship type, and the discretization processing of the continuous type data includes: and carrying out data supervision discretization operation according to the intervals, and classifying and discretizing according to a general grouping mode of shipping.
In one embodiment, the text attribute data includes a load port and an unload port, and performing an attribute value reduction on the text attribute data includes:
and establishing a mapping relation of a combination of the loading port and the unloading port, wherein the mapping relation is defined as effective route data, and when the loading port or the unloading port is absent, filling is carried out by adopting a standard loading port and a standard unloading port.
In one embodiment, step S5 specifically includes:
setting two transportation capacity calculation modes according to an effective experimental data set, voyage turnaround periods of different ship types and air routes and ship turnaround periods of different ships, wherein the first transportation capacity calculation mode is used for screening available ships according to set time period calculation, and the second transportation capacity calculation mode is used for selecting a certain ship and calculating future available ship;
and predicting the transport capacity according to the set transport capacity calculation mode.
In one embodiment, the predicting the transport capacity according to the set transport capacity predicting mode comprises the following steps:
screening all ships with current time, the number of days required for reaching a loading port from the current position, the number of voyages from the current time to a specified future time, a turnover period (a specified future time) and ship types (specified ship types +/-10%) as available ships, wherein if the current voyage unloading port is missing, the ship turnover period is taken, otherwise, the ship turnover period is taken;
calculating the number of times n of voyage of the specified ship from the current position to the next loading port + the number of future voyages + the ship turnover period < k >, and listing the current time + the number of days required from the current position to the next loading port + {0,1, …, n }. the corresponding date sequence of the turnover period as the available future ship period, wherein k is the specified number of future voyage, the current shipping port is absent, the turnover period is taken as the ship turnover period, and otherwise, the ship turnover period is taken.
One or more technical solutions in the embodiments of the present application have at least one or more of the following technical effects:
the invention provides a capacity prediction method based on actual shipping business and data mining, firstly filtering ship experimental data extracted from actual shipping industry data, filtering and eliminating abnormal data to obtain processed ship experimental data, then merging adjacent voyages according to time continuity and preset rules of the airlines in the airlines data, normalizing a plurality of adjacent airlines into an effective ship historical voyage according to specified rules, then carrying out feature selection on basic data and the processed airlines data to obtain an effective experimental data set, namely obtaining high-quality experimental data, then calculating the transit period of the ship capacity from two dimensions, and finally predicting the available capacity according to the effective experimental data set, the voyage transit periods of different ship types and airlines and the ship transit periods of different ships, the method can convert the ship implicit data into the explicit data available in the actual business, on one hand, high-quality experimental data in the capacity prediction can be obtained, on the other hand, the voyage time length can be predicted in real time, and further the available capacity in the market can be predicted, so that the prediction effect is improved.
Drawings
In order to more clearly illustrate the embodiments or technical solutions of the present invention, the drawings used in the embodiments or technical solutions in the prior art are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is an overall flow chart of a capacity prediction method based on actual shipping service and data mining according to the present invention;
FIG. 2 is a flow chart of a capacity prediction method according to an embodiment of the present invention;
FIG. 3 is a flow chart of the generation of valid voyage data for an embodiment of the present invention;
FIG. 4 is a flow chart of predicting available capacity for a specified time period in accordance with an embodiment of the present invention;
FIG. 5 is a flow chart of specifying a predicted available age for a vessel according to an embodiment of the present invention.
Detailed Description
The invention aims to provide a method for processing and predicting transport capacity data by adopting a big data mining technology aiming at the problem that related personnel cannot comprehensively and accurately acquire available transport capacity data in shipping transaction, thereby achieving the purposes of obtaining high-quality experimental data and improving the prediction effect.
In order to achieve the above object, the main concept of the present invention is as follows:
based on the ship flow rule in the actual shipping service, high-quality effective experimental data are generated through data integration, user-defined interval discretization, attribute stipulation and other operations are carried out on continuous attribute data and text attribute data, ship implicit data are converted into explicit data available in the actual service, the duration of the voyage is predicted in real time, and further the available capacity of the market is predicted. The invention discloses a method for obtaining a ship transport capacity rule through big data analysis and processing, which adopts methods such as interval data discretization and the like to complete data preprocessing operation; and analyzing effective characteristics by combining with shipping business, and predicting available transport capacity aiming at ships and time periods.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
The embodiment provides a capacity prediction method based on actual shipping service and data mining, please refer to fig. 1, and the method includes:
step S1: extracting ship experimental data from actual shipping industry data, filtering and clearing abnormal data contained in the ship experimental data based on time consistency detection and a preset ship navigation rule, and obtaining processed ship experimental data, wherein the processed ship experimental data comprises basic data and historical course data, the basic data comprises ship types and ship identifications, and the historical course data comprises voyage times.
Specifically, experimental data can be continuously acquired by adopting timing scheduling based on data collection and data cleaning technology in data mining, and time consistency and ship navigation rules are detected in the acquisition process to filter and clear abnormal data.
In one embodiment, the basic data includes ship basic attribute data and port basic attribute data, the port basic attribute data includes a port name, and the step S1 specifically includes:
and performing similarity mapping on the irregular port names contained in the basic data and the collected regular port data sets, filling missing values of voyages through the traceability of the ship voyages and based on the adjacent time, and clearing inconsistent data according to the uniqueness rule, the continuity rule and the null value rule.
Specifically, the data include ship basic attributes, historical voyages, current voyage data, current position information, universal port data sets, and the like. The ship navigation rule can be obtained by analyzing historical data, and a universal port name data set can be obtained in advance.
The implementation of the data processing in step S1 is explained as follows:
extracting and collating the basic data set of the dry bulk cargo ship in China coastal region, and naming the basic data set as ShipData, and recording the data set as S ═ S1,s2,…,stAnd acquiring a basic data set from a ship communication network and a shipping company, wherein the basic data set comprises basic information of the ship, including ship name, mms (unique identifier in the ship industry, namely ship identifier), ship length, ship width, ship type, no-load draft, full-load draft, average speed and the like. The size of the data set ShipData is t × s, i.e. the data set contains t ship records, t ═ 1680, siRepresents the ith record, i ∈ [1, t ]](ii) a Each data ship record siContaining s attributes, s-14, denoted si＝{a1,a2,…,as},aijThe jth attribute representing the ith ship data record, j ∈ [1, s]。
Through acquisition of collected data sets of dry bulk carriersNearly four monthly history route data of all ships are named as DataInit and are recorded as U ═ x1,x2,…,xnAnd arranging according to the unique identifier mmi of the ship and the time of the arrival of the air line at the destination port in an ascending order. The size scale of the data set DataInit is n × m, that is, the data set contains n route history records, where n is 32637, and x isiRepresents the ith record, i ∈ [1, n ]](ii) a Each data historical route record xiContaining m attributes, m being 18, denoted xi＝{a1,a2,…,am},aijThe jth attribute representing the ith route data history, j ∈ [1, m]. Aiming at the abnormal data problem of the original data (namely historical route data), filtering according to the uniqueness of the route of the same time; and respectively existing the partial attributes (arrival time and departure time of the starting port and the destination port) of each historical record in adjacent routes, and filling the data items according to the time line continuity. Aiming at irregular port names which do not accord with the universal names in the original data, a universal port data set PA (p) is collected and sorted1,p2,…,pnAnd establishing a similarity mapping relation between the irregular port name in the original data and the universal port data set PA, and filling a standard port representation attribute value.
Step S2: and combining the adjacent voyages into an effective historical voyage of the ship according to the time continuity and the preset rule of the voyage in the voyage data to obtain the processed voyage data.
Specifically, according to the general concept of the voyage times in the current shipping business and in the cargo transaction, the historical voyage times of the effective ship can be designated as follows: the number of flights from loading to unloading and back to loading, where loading is the departure port and may be designated as the north port, unloading is the destination port and may be designated as the south and south ports. And (5) normalizing a plurality of adjacent routes into an effective historical voyage number of the ship according to a specified rule by adopting the experimental data obtained in the step (S1) and adopting data conversion and data integration in data mining according to the time line continuity of the ship routes. And circularly executing the rules until all data are processed, and generating effective experimental data.
In one embodiment, step S2 specifically includes:
step S2.1: grouping historical route data according to ship identification, and performing ascending arrangement according to time;
step S2.2: judging the validity of the airline data according to whether the time of the number of the airline is consistent with the numerical value of the adjacent number of the airline and the reasonability of the parking time;
step S2.3: and if the current voyage number is valid, judging whether the loading and unloading time length of the adjacent voyage number is within a given threshold value, if so, combining the current voyage number and the adjacent voyage number, and repeatedly executing the steps S2.2-S2.3 until the course data of all ships are processed.
Specifically, referring to fig. 3, in order to generate the effective voyage data, the loading port set PL is given as { p { (p) } based on the ship experimental data processed in step S11,p2,…,pnGiving a time threshold T required for loading and unloading goodsmin,TmaxAnd processing initial course data according to ship grouping and time sequencing, and traversing all course data.
The traversal rules are as follows:
initial harbor psDestination Port peAnd a harbor stopping time LT:
wherein if the route line fA1 (i.e., the origin port belongs to the loading port, the destination port does not belong to the loading port, and the number of times the loading and unloading duration is within a given threshold); then p issFor a port with a valid number of voyages, the subsequent routes are checked in sequence:
wherein if the route fB1 (i.e. the origin port does not belong to the loading port, the duration of the voyage is within the given threshold), then the route psThe port of discharge for the active voyage. Integrating two lane data into oneValid voyage data. And circularly executing the rules until all data are processed, and generating effective experimental data.
Step S3: and (4) carrying out characteristic selection on the basic data and the processed route data to obtain an effective experimental data set.
Specifically, feature selection is carried out on the basic data and the processed route data, namely different attribute data in the basic data and the processed route data are subjected to normalized processing, so that effective high-quality ship data are obtained.
In one embodiment, the basic data and the processed route data include continuity data, text attribute data, and data related to decision time of flight, and step S3 specifically includes:
discretizing continuous data, performing attribute numerical convention on text attribute data, and generalizing data of decision flight time to obtain an effective experimental data set.
Specifically, feature selection may be performed on the basic data in step S1 and the route data processed in step 2 according to the attribute importance based on a data reduction and a data transformation technique in data mining, where the feature selection specifically includes: and performing data supervision discretization operation on ship type and other continuous data according to regions, and classifying the ship type and other continuous data into k types according to a general grouping mode of shipping and discretizing. Performing attribute numerical specification on text attribute data (loading port, unloading port and the like), establishing a mapping relation of a combination of the loading port and the unloading port, wherein the specification is processable data, and when the unloading port is absent, taking a standard route (loading port-Jiangyin, loading port-south) according to the ship type; and generalizing the data of the decision-making voyage time to obtain a final decision attribute of the voyage, wherein the decision attribute refers to an attribute directly influencing a predicted capacity result, one part of the decision attribute is obtained from original data, and the other part of the decision attribute is obtained by processing data integration of multiple dimensions for multiple times. Through the processing, high-quality effective experimental data are finally obtained.
Wherein, continuous type data includes the ship type, carries out the discretization to continuous type data and handles, includes: and carrying out data supervision discretization operation according to the intervals, and classifying and discretizing according to a general grouping mode of shipping.
The text attribute data comprises a loading port and an unloading port, and the attribute numerical reduction of the text attribute data comprises the following steps:
and establishing a mapping relation of a combination of the loading port and the unloading port, wherein the mapping relation is defined as effective route data, and when the loading port or the unloading port is absent, filling is carried out by adopting a standard loading port and a standard unloading port.
In the specific implementation process, according to a general processing method of ship and cargo transaction in shipping, feature selection is carried out on basic data and processed ship route data according to attribute importance, data supervision discretization operation is carried out on a self-defined interval of ship type and other continuous data, and a self-defined interval separation data set KD is recorded as [ k ═ k [1,k2,…,kn]In units of ten thousand tons, the interval discretization is defined as follows:
where i ∈ [1, n ]]，kiNumber of i-th partitions. f. ofCAnd representing the discretization result of the attribute value.
And the big data is converted into the processable numerical data after numerical reduction is carried out on the text data which is difficult to process by a calculation method by adopting rule-based conversion. The historical voyage data attribute shipping port p in the processed route data in step S2dAnd port of discharge puRecording the combination as an effective route, and establishing a route mapping data set VD ═ v1,v2,…,vn]When the loading port and the unloading port are missing, the standard loading port and the standard unloading port are used for filling.
Where k is the value after the flight path mapping, vkShipment port p for valid airline, voyage recorddAnd port of discharge puAnd vkCorresponds to, fDAnd expressing the result after numerical value reduction.
Referring to fig. 2, a flowchart of the capacity prediction method according to an embodiment of the present invention is shown, which only shows some, but not all, operations of steps S1-S5, including filling missing values of experimental data in step S1, regenerating effective voyage experimental data in step S2, and selecting features in step S3: discretizing the continuous attributes, stipulating the text-type attributes, calculating the voyage transit time and the ship transit time in step S4, and finally predicting the available capacity in the specified time period and the available ship time in step S5.
Step S4: and calculating the voyage turnaround periods of different ship types and ship routes in the effective experimental data set according to the ship types, and calculating the ship turnaround periods of different ships in the effective experimental data set according to the ship identifications.
Specifically, step S4 is to calculate the turnaround time of the ship' S capacity from two dimensions based on the high quality valid data set obtained in step 3.
In a specific implementation process, by analyzing distribution of the voyage duration, data values are large in difference and have extreme values, so that a mode is adopted as the trend degree in data concentration, voyage duration attribute values are grouped based on a statistical method, the frequency of each group is obtained, the larger the frequency value is, the larger the effect of the group of mark values on the overall level is, and conversely, the smaller the frequency value is, the smaller the effect of the group of mark values on the overall level is. The average of the corresponding set of data sets with the greatest frequency is thus obtained as the most representative voyage turnaround.
Step S5: and predicting the available transport capacity according to the effective experimental data set, the voyage turnaround periods of different ship types and air routes and the ship turnaround periods of different ships.
In one embodiment, step S5 specifically includes:
setting two transportation capacity calculation modes according to an effective experimental data set, voyage turnaround periods of different ship types and air routes and ship turnaround periods of different ships, wherein the first transportation capacity calculation mode is used for screening available ships according to set time period calculation, and the second transportation capacity calculation mode is used for selecting a certain ship and calculating future available ship;
and predicting the transport capacity according to the set transport capacity calculation mode.
Specifically, the invention makes two modes or methods for calculating available capacity from the perspective of shipping practitioners as entities by analyzing the rules of the ship transaction.
Wherein, according to the fortune prediction mode that sets up, predict the fortune, include:
screening all ships with current time, the number of days required for reaching a loading port from the current position, the number of voyages from the current time to a specified future time, a turnover period (a specified future time) and ship types (specified ship types +/-10%) as available ships, wherein if the current voyage unloading port is missing, the ship turnover period is taken, otherwise, the ship turnover period is taken;
calculating the number of times n of voyage of the specified ship from the current position to the next loading port + the number of future voyages + the ship turnover period < k >, and listing the current time + the number of days required from the current position to the next loading port + {0,1, …, n }. the corresponding date sequence of the turnover period as the available future ship period, wherein k is the specified number of future voyage, the current shipping port is absent, the turnover period is taken as the ship turnover period, and otherwise, the ship turnover period is taken.
Specifically, in the first calculation method, a time period is set to indicate that a day in the future is designated as a start time and an end time (optional). The number of flights from the current time to the specified future time is (specified future time-current time)/T, and an integer is taken down. In actual calculation, the floating range of the tonnage of the target ship is +/-10%, and the actual ship can float backwards within 7 days. By performing loop calculation on different ships, the ships available in a future period of time can be calculated. And when the ending time is missing or equal to the starting time, returning to the available ship on the day of the starting time.
In the second calculation method, a certain ship is selected, a future available ship age (for example, an available ship age of k days in the future) is calculated, and the future available ship ages of a batch of ships can be calculated through loop calculation. The calculation method of case 2.
In a specific implementation, a future day is designated as a start time and an end time (optional) for the case (1). And calculating and screening available ships. Referring to fig. 4, the specific algorithm is as follows:
appointing a certain future day as a starting time to be recorded as TsAnd the end time (optional) is marked as TeAnd if the designated ship type is marked as ST, the floating range of the tonnage of the target ship is [0.9ST,1.1 ST%]And screening a target ship data set with the tonnage st within the tonnage floating range of the target ship from the basic data set ShipData, and integrating the data of the ship turnover period and the ship turnover period obtained in the step S4 and the current ship data according to the unique ship identifier mmsi.
And for each target ship, taking the ship turnover period in the current turnover period T if the current shipping port is missing, and taking the shipping turnover period in the reverse way. Recording the time of arrival of the current voyage at the loading port as TdUsing TrIndicating the next arrival time at the port, i.e. the time of release of the transport. Then:
Tr＝Td+T
recording the current time as TnowThen the current time is up to the specified future time TsThe method for calculating the flight times comprises the following steps:
according to the screening condition that "the current time + the number of days required to arrive at the loading port from the current position + the current time to the specified future time x the ship turnover T is the specified future time" and "the ship type is the specified ship type ± 10%", the calculation method for specifying whether the ship belongs to the result ship is defined as follows:
fE＝Tr+Nv×T
fEfor the ship to pass through NvThe time of releasing the transport capacity again after each voyage, and the value T when the Te value is lostSAnd finishing at 24 points on the day. If:
fE∈[Ts,Te]
the vessel will release capacity as available capacity for a specified period of time. And traversing all target ships to obtain an available transport capacity data set.
And (3) specifying a certain ship according to the case (2), and calculating the available ship stage of the ship within k days in the future. Referring to fig. 5, the specific algorithm is as follows:
and integrating the ship turnover and voyage turnover data obtained in the step S4 and the current voyage data of the ship according to the unique ship identifier mmisi. And if the current voyage unloading port is missing, taking the current turnover period T as the ship turnover period, and otherwise, taking the voyage turnover period. The method for calculating the number of voyages of the ship from the current time to k days in the future is defined as follows:
recording the time of arrival of the current voyage at the loading port as TdUsing TrIndicating the next arrival time at the loading port of the vessel, i.e. the time of release of the transport capacity. Then:
Tr＝Td+T
the calculation mode of the available ship stage in k days in the future of the ship is defined as follows:
f(n)＝Td+nT,n∈[1,Nv]
f(n)and calculating the result of the time of the capacity release of the ship in k days.
It should be understood that the above description of the preferred embodiments is given for clarity and not for any purpose of limitation, and that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (6)
1. A capacity prediction method based on actual shipping service and data mining is characterized by comprising the following steps:
step S1: extracting ship experimental data from actual shipping industry data, filtering and clearing abnormal data contained in the ship experimental data based on time consistency detection and a preset ship navigation rule, and obtaining processed ship experimental data, wherein the processed ship experimental data comprises basic data and historical course data, the basic data comprises a ship type and a ship identification, and the historical course data comprises a voyage number;
step S2: combining adjacent voyages into an effective historical voyage of the ship according to the time continuity and preset rules of the voyage in the voyage data to obtain processed voyage data;
step S3: carrying out characteristic selection on the basic data and the processed route data to obtain an effective experimental data set;
step S4: calculating the voyage turnaround periods of different ship types and the air routes in the effective experimental data set according to the air routes and the ship types, and calculating the ship turnaround periods of different ships in the effective experimental data set according to the ship identifications;
step S5: predicting the available transport capacity according to the effective experimental data set, the voyage turnaround periods of different ship types and air routes and the ship turnaround periods of different ships;
wherein, step S5 specifically includes:
setting two transportation capacity calculation modes according to an effective experiment data set, voyage turnover periods of different ship types and air routes and ship turnover periods of different ships, wherein the first transportation capacity calculation mode is used for screening available ships according to set time period calculation, and the second transportation capacity calculation mode is used for selecting a certain ship and calculating future available ship;
predicting the transport capacity according to the set transport capacity calculation mode;
wherein, according to the fortune prediction mode that sets up, predict the fortune, include:
screening all ships with current time, the number of days required for reaching a loading port from the current position, the number of voyages from the current time to a specified future time, a turnover period (a specified future time) and ship types (specified ship types +/-10%) as available ships, wherein if the current voyage unloading port is missing, the ship turnover period is taken, otherwise, the ship turnover period is taken;
calculating the number of times n of voyage of the specified ship from the current position to the next loading port + the number of future voyages + the ship turnover period < k >, and listing the current time + the number of days required from the current position to the next loading port + {0,1, …, n }. the corresponding date sequence of the turnover period as the available future ship period, wherein k is the specified number of future voyage, the current shipping port is absent, the turnover period is taken as the ship turnover period, and otherwise, the ship turnover period is taken.
2. The method of claim 1, wherein the basic data includes ship basic attribute data and port basic attribute data, the port basic attribute data includes a port name, and the step S1 specifically includes:
and performing similarity mapping on the irregular port names contained in the basic data and the collected regular port data sets, filling missing values of voyages through the traceability of the ship voyages and based on the adjacent time, and clearing inconsistent data according to the uniqueness rule, the continuity rule and the null value rule.
3. The method according to claim 1, wherein step S2 specifically comprises:
step S2.1: grouping historical route data according to ship identifications, and performing ascending arrangement according to time;
step S2.2: judging the validity of the airline data according to whether the time of the number of the airline is consistent with the numerical value of the adjacent number of the airline and the reasonability of the parking time;
step S2.3: and if the current voyage number is valid, judging whether the loading and unloading time of the adjacent voyage number is within a given threshold value, if so, combining the current voyage number and the adjacent voyage number, and repeatedly executing the steps S2.2-S2.3 until the processing of course data of all ships is finished.
4. The method of claim 1, wherein the basic data and the processed airline data include continuity data, text attribute data, and data associated with determining a time of flight, and step S3 specifically includes:
discretizing continuous data, performing attribute numerical convention on text attribute data, and generalizing data of decision flight time to obtain an effective experimental data set.
5. The method of claim 4, wherein the continuous data includes a ship model, and discretizing the continuous data includes: and carrying out data supervision discretization operation according to the intervals, and classifying and discretizing according to a general grouping mode of shipping.
6. The method of claim 4, wherein the textual attribute data includes a load port and an unload port, and wherein numerically specifying attributes of the textual attribute data comprises:
and establishing a mapping relation of a combination of the loading port and the unloading port, wherein the mapping relation is defined as effective route data, and when the loading port or the unloading port is absent, filling is carried out by adopting a standard loading port and a standard unloading port.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911093827.0A CN111091226B (en) | 2019-11-11 | 2019-11-11 | Transport capacity prediction method based on actual shipping service and data mining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911093827.0A CN111091226B (en) | 2019-11-11 | 2019-11-11 | Transport capacity prediction method based on actual shipping service and data mining |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111091226A CN111091226A (en) | 2020-05-01 |
| CN111091226B true CN111091226B (en) | 2022-06-14 |
Family
ID=70394072
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911093827.0A Active CN111091226B (en) | 2019-11-11 | 2019-11-11 | Transport capacity prediction method based on actual shipping service and data mining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111091226B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111625217A (en) * | 2020-05-08 | 2020-09-04 | 江苏新扬子造船有限公司 | Method for counting marine valve members with weight above certain based on AM secondary development |
| CN111966926B (en) * | 2020-07-02 | 2023-12-01 | 海南太美航空股份有限公司 | Target route measuring and calculating method and system |
| CN111966927B (en) * | 2020-07-02 | 2023-12-01 | 海南太美航空股份有限公司 | Target route measuring and calculating method and system |
| CN112699467A (en) * | 2020-12-29 | 2021-04-23 | 香港理工大学深圳研究院 | Method for supervising and checking ships and allocating inspectors in port country |
| CN113610347B (en) * | 2021-07-05 | 2024-04-05 | 上海海事大学 | Ship shipping market capacity assessment method |
| CN113780952B (en) * | 2021-09-16 | 2024-03-15 | 上海船舶运输科学研究所有限公司 | Container airliner line capacity analysis method and system |
| CN113837801A (en) * | 2021-09-24 | 2021-12-24 | 中远海运科技股份有限公司 | Ship identification method |
| CN116095114B (en) * | 2023-01-11 | 2023-11-03 | 上海船舶运输科学研究所有限公司 | Ship-shore data transmission method based on Internet of things mode |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103645705A (en) * | 2013-12-05 | 2014-03-19 | 中国海洋石油总公司 | LNG multi-point transport ship transport-capacity prediction scheduling method |
| CN106529781A (en) * | 2016-11-02 | 2017-03-22 | 合肥飞友网络科技有限公司 | Air line transport power evaluation system and work method thereof |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SG11201600405WA (en) * | 2013-07-19 | 2016-02-26 | Nippon Yusen Kk | Device for allocating ships, method for allocating ships, program and recording media thereof |
-
2019
- 2019-11-11 CN CN201911093827.0A patent/CN111091226B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103645705A (en) * | 2013-12-05 | 2014-03-19 | 中国海洋石油总公司 | LNG multi-point transport ship transport-capacity prediction scheduling method |
| CN106529781A (en) * | 2016-11-02 | 2017-03-22 | 合肥飞友网络科技有限公司 | Air line transport power evaluation system and work method thereof |
Non-Patent Citations (2)
| Title |
|---|
| New inland river channel transit capacity evaluation method based on dynamic quaternion ship domain model;Shaoman Liu 等;《 2015 54th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)》;20151001;全文 * |
| 采用改进模拟植物生长算法的世界石油船队总运力预测;王诺 等;《上海海事大学学报》;20190630;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111091226A (en) | 2020-05-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111091226B (en) | Transport capacity prediction method based on actual shipping service and data mining | |
| Barua et al. | Machine learning for international freight transportation management: A comprehensive review | |
| Yang et al. | Can we trust the AIS destination port information for bulk ships?–Implications for shipping policy and practice | |
| Wang et al. | Quantitative analysis of the impact of COVID-19 on ship visiting behaviors to ports-A framework and a case study | |
| Pani et al. | Prediction of late/early arrivals in container terminals-a qualitative approach | |
| Pani et al. | A data mining approach to forecast late arrivals in a transhipment container terminal | |
| Štepec et al. | Machine learning based system for vessel turnaround time prediction | |
| CN112016730A (en) | Port berth loading and unloading efficiency excavating method, device, equipment and storage medium | |
| CN116843071B (en) | Transportation network operation index prediction method and device for intelligent port | |
| CN109409811A (en) | A kind of Logistic Scheduling method, apparatus, computer equipment and storage medium | |
| Sharma et al. | A Study on Decision-Making of the Indian Railways Reservation System during COVID-19 | |
| CN115456474A (en) | Port operation scheduling method and system based on artificial intelligence | |
| Wang et al. | Fundamental challenge and solution methods in prescriptive analytics for freight transportation | |
| Bai et al. | Identifying port congestion and evaluating its impact on maritime logistics | |
| Asborno et al. | Assigning a commodity dimension to AIS data: Disaggregated freight flow on an inland waterway network | |
| Zhong et al. | A novel oil port risk and efficiency performance measured by using AIS data and maritime open data: the case of Guangzhou, China | |
| Bonham et al. | Analysing port and shipping operations using big data | |
| Chen | Research on Marine Economic Development Information Management System Based on Supply Chain Technology | |
| Zhou et al. | Estimation of shipment size in seaborne iron ore trade | |
| CN114037257B (en) | Special ship competitiveness analysis method and system | |
| Suprata et al. | Analysing the cause of idle time in loading and unloading operation at Indonesian international port container terminal: Port of Tanjung Priok case study | |
| CN115456258A (en) | Method for predicting transportation capacity of competitor ship and computer readable medium | |
| Altan et al. | Discovering gateway ports in maritime using temporal graph neural network port classification | |
| Lirn et al. | Potential for Transport in Asia Pacific Region | |
| Makarova et al. | Improvement opportunities in commodity trucks delivery in globalized markets |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |