CA3126555A1 - Optimization method, device, computer equipment and storage medium of logistics transportation network - Google Patents
Optimization method, device, computer equipment and storage medium of logistics transportation networkInfo
- Publication number
- CA3126555A1 CA3126555A1 CA3126555A CA3126555A CA3126555A1 CA 3126555 A1 CA3126555 A1 CA 3126555A1 CA 3126555 A CA3126555 A CA 3126555A CA 3126555 A CA3126555 A CA 3126555A CA 3126555 A1 CA3126555 A1 CA 3126555A1
- Authority
- CA
- Canada
- Prior art keywords
- network
- warehouses
- transportation
- predicted
- models
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000005457 optimization Methods 0.000 title claims abstract description 28
- 238000003860 storage Methods 0.000 title claims abstract description 11
- 238000004590 computer program Methods 0.000 claims description 15
- 238000007637 random forest analysis Methods 0.000 claims description 6
- 230000004048 modification Effects 0.000 description 16
- 238000012986 modification Methods 0.000 description 16
- 230000003068 static effect Effects 0.000 description 8
- 238000010276 construction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 101100288962 Mus musculus Lefty2 gene Proteins 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
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"
-
- 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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Disclosed in the present invention are a method, a device, a computer apparatus, and a storage medium for logistic transportation network optimization. The mentioned method comprises: creating a prediction model to export predicted freight transportation volumes; adjusting warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes; building a basic network, wherein the basic network comprises a portion of warehouses;
allocating collecting and distributing centers for the rest of warehouses; and exporting a logistic transportation network. With comprehensive considerations of current network structures, based on center levels and OD demand information amongst centers, the present invention adopts a multi-network strategy to build the overall transportation network with sequential buildup and integral optimization process, to allow an optimization strategy for adjusting current networks.
allocating collecting and distributing centers for the rest of warehouses; and exporting a logistic transportation network. With comprehensive considerations of current network structures, based on center levels and OD demand information amongst centers, the present invention adopts a multi-network strategy to build the overall transportation network with sequential buildup and integral optimization process, to allow an optimization strategy for adjusting current networks.
Description
OPTIMIZATION METHOD, DEVICE, COMPUTER EQUIPMENT AND STORAGE MEDIUM
OF LOGISTICS TRANSPORTATION NETWORK
Technical Field The present invention relates to the field of logistic transportation, in particular, to a method, a device, a computer apparatus, and a storage medium for logistic transportation network optimization.
Background The cm-rent transportation network constructions for the majority of companies are mature, with relatively comprehensive fundamental constructions. The currently standardized optimization methods investigate an optimized network plan with lowered costs, wherein the network plan satisfies constraint conditions by currently available transportation networks. In the meanwhile, the relative stability of the networks should be considered to control the discrepancy between the optimized network and the current network within a certain range, so as to limit impacts to the current networks. At the same time, multiple network modes are used to eliminate or modify improper paths.
Transportation network plans are featured in static routing plans and dynamic routing plans. Based on the properties of warehouses, the static routing plans determines warehouse levels and relationships amongst warehouses. The dynamic routing plans modifies details of vehicle transportation based on freight volumes. The current methods barely modify or optimize the static routing plans, while the dynamic routing plans are modified and optimized according to history data of the transportation networks. By taking distribution demands between two branches, all time-space curves satisfying requirements are determined, wherein the time-space curve with the lowest costs is identifies as the path plan curve. By identifying the time-space curve with the lowest costs as the path plan curve, the costs are reduced with ensuring the timeliness.
However, the current methods involve several drawbacks and limitations.
1. With a large network scale, the current method requires to search all time-space curves satisfying timeliness requirements, leading to a huge data volume and consequently causing difficult computations and great resource costs;
OF LOGISTICS TRANSPORTATION NETWORK
Technical Field The present invention relates to the field of logistic transportation, in particular, to a method, a device, a computer apparatus, and a storage medium for logistic transportation network optimization.
Background The cm-rent transportation network constructions for the majority of companies are mature, with relatively comprehensive fundamental constructions. The currently standardized optimization methods investigate an optimized network plan with lowered costs, wherein the network plan satisfies constraint conditions by currently available transportation networks. In the meanwhile, the relative stability of the networks should be considered to control the discrepancy between the optimized network and the current network within a certain range, so as to limit impacts to the current networks. At the same time, multiple network modes are used to eliminate or modify improper paths.
Transportation network plans are featured in static routing plans and dynamic routing plans. Based on the properties of warehouses, the static routing plans determines warehouse levels and relationships amongst warehouses. The dynamic routing plans modifies details of vehicle transportation based on freight volumes. The current methods barely modify or optimize the static routing plans, while the dynamic routing plans are modified and optimized according to history data of the transportation networks. By taking distribution demands between two branches, all time-space curves satisfying requirements are determined, wherein the time-space curve with the lowest costs is identifies as the path plan curve. By identifying the time-space curve with the lowest costs as the path plan curve, the costs are reduced with ensuring the timeliness.
However, the current methods involve several drawbacks and limitations.
1. With a large network scale, the current method requires to search all time-space curves satisfying timeliness requirements, leading to a huge data volume and consequently causing difficult computations and great resource costs;
2. The dynamic routing plans based on the history data yield a lower feasibility, wherein the history data can only represent the network freight conditions in one period in the past and cannot present the freight demands in the future. As a result, there exist great discrepancy for practical executions.
3. The current method uses exact algorithm to calculate the optimized solution, wherein the discrepancy between the optimized new network and the current network remains undetermined. Where if a new network involves a large discrepancy when the current network has already existed, difficulties for modifications results in little feasibility in practice.
Date Recue/Date Received 2021-09-30
Date Recue/Date Received 2021-09-30
4. The current method is not able to constrain path numbers and ratios requiring modifications in the optimized network.
5. The cm-rent algorithm does not consider the application scenarios for different network modes.
6. The cm-rent method does not consider the costs for network modifications and feasibility for applying the optimization. When the networks mature with fixed branch levels, the practical operations are greatly affected by the level allocations in the current network. Where if the optimized results conflict with the current network in terms of the level allocations, the current network level allocations need to be modified.
However, the network level modifications involves the changes of locations and installments, with a long working period and a high cost, consequently leading to little feasibility of the optimized solution.
Summary In order to solve the forementioned problems, the present invention proposes the following strategies:
A logistic transportation network optimization method, comprises:
creating a prediction model to export predicted freight transportation volumes;
adjusting warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes;
building a basic network, wherein the basic network comprises a portion of warehouses;
allocating collecting and distributing centers for the rest of warehouses; and exporting a logistic transportation network.
Furthermore, the predicting process of the described prediction model comprises:
collecting target features, wherein the described target features comprise at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure.
selecting data volume, wherein the data volume is the overall number of the described target features;
setting a predicted target, wherein the described predicted target is workload; and determining models, wherein predicted values from multiple selected models are combined.
Furthermore, three models including xgboost, random forest, and lightgbm are selected and the final exporting value is obtained by combining the predicted values of the three models.
Furthermore, the described basic network is constrained by network ratio limits and freight transportation volumes to yield the maximum coverage range of the described network.
Furthermore, the collecting and distributing centers with the lowest costs are allocated, by determining costs of collection and distributing from the rest of warehouses individually.
A logistic transportation network optimization device, comprises:
Date Recue/Date Received 2021-09-30 a prediction module, configured to create a prediction model for exporting predicted freight transportation volumes;
an adjustment module, configured to adjust warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes;
a building module, configured to build a basic network, wherein the basic network comprises a portion of warehouses;
an allocation module, configured to allocate collecting and distributing centers for the rest of warehouses; and an export module, configured to export a logistic transportation network.
In particular, the described prediction model also includes:
a feature collecting unit, configured to collect target features, wherein the described target features comprises at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure;
a volume selection unit, configured to select a data volume, wherein the data volume is the overall number of the described target features;
a target setting unit, configured to set a predicted target, wherein the described predicted target is workload; and a combination unit, configured to determine models, wherein predicted values from multiple selected models are combined.
A computer apparatus is provided in the present invention, comprising a memory unit, a processor, and computer programs stored in the memory unit executable on the processor, wherein any of procedures of the described methods are performed when the described processor executes the described computer programs.
A readable computer storage medium with computer programs stored is provided in the present invention, wherein any of procedures of the described methods are performed when the described computer programs are executed on the described processor.
The method, the device, the computer apparatus, and the storage medium for logistic transportation network optimization in the present invention provides a prediction for freight volume in the next period of time to yield a more accurate plan; appropriate modifications of static routing plans and dynamic routing plans based on the future freight volume, to lower the overall transportation costs; dynamic modifications on warehouse levels based on the future freight volume to improve feasibility of the functionalities of warehouses in the networks; construction of basic networks based on the future freight volume and optimizing models, to maximize basic network utility; and a transportation model based on the future freight Date Recue/Date Received 2021-09-30 volume, to yield more appropriate network transportation structures and cost structures and improve feasibility of the modified network.
Brief descriptions of the drawings Fig. 1 is a flow diagram of the logistic transportation network optimization method in an embodiment of the present invention.
Fig. 2 is a structure diagram of the logistic transportation network optimization device in an embodiment of the present invention.
Fig. 3 is an internal structure diagram of the computer apparatus in an embodiment of the present invention.
Detailed descriptions In order to make the technical strategies of the present invention clearer, the accompany drawings for the present invention will be briefly introduced below. Obviously, the following drawings in the descriptions are used to explain the present invention and shall not restrict the present invention.
In a transportation network, an origin-destination (OD) demand and a corresponding transportation timeliness requirement exist for transportations in between any two branches.
Based on the current network, a proper modification yielding an equal or better timeliness is required to generate a transportation routing plan with lowered costs and controlled changes.
The transportation network optimization and modifications in the present invention are aiming at the transportation demand applicable to the next period of time. The feasibility of the optimized network is determined by the accuracy of the predicted freight volume in the future.
Therefore, the present application provides a logistic transportation network optimization method, applicable to the transportation demands for the network in the next period of time. In an embodiment, as shown in Fig. 1, the method comprises:
S100, creating a prediction model to export predicted freight transportation volumes.
In the present embodiment, a machine learning prediction model is created in the first place. By inputting currently available history data of the current transportation network into the prediction models to predict the freight volumes in the next period of time, the freight volume prediction results are exported.
In particular, the predicting process of the described prediction model in S100 comprises:
collecting target features, wherein the described target features comprise at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure. Preferably, the target features includes: the freight volume demand of any OD pairs in the past period of time, relationships between warehouses, functionality and Date Recue/Date Received 2021-09-30 level of each warehouse, the current transportation network structure, undertaken e-commerce promotion events (if at promoting timing such as November 11th, June 18', August 18', etc), holiday information (if on statutory holidays, traditional holidays or other holidays), trends and periodic rules of historic workload (such as the past one week freight volumes, the past two weeks freight volumes, or trends of the past one week freight volumes), historic numbers of all types of orders, timeliness tables for all locations, and/or weather forecast for delivery locations (such as numbers of raining days or sunny days in the next week), etc.
Selecting data volume, wherein the data volume is the overall number of the described target features.
Preferably, the number of features in a period of time, wherein the number of features for the recent half a year can be selected.
Setting a predicted target, wherein the described predicted target is workload. Preferably, based on demands, the prediction in a particular time period can be performed, such as the predicted workload in the next one month.
Determining models, wherein predicted values from multiple selected models are combined. Preferably, models can be selected such as xgboost, random forest, lightgbm, LR, etc. Due to low performance of regression by LR wherein relatively similar performances of regression by the other three models are obtained, xgboost, random forest, lightgbm are selected and the final exporting value is obtained by combining the predicted values of the three models.
S200, adjusting warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes.
In an embodiment of the present invention, according to the prediction exported from S100, data dimensions such as throughput capacity, correlation with other warehouses and geographical locations are considered to dynamically plan the warehouse levels and numbers of warehouses for each level.
S300, building a basic network, wherein the basic network comprises a portion of warehouses.
In an embodiment of the present invention, after adjusting warehouse levels and numbers of warehouses for each level by S200, certain proportions of higher-level warehouses are selected to build a basic network.
Preferably, building the basic network aims at a maximum coverage range by the network.
In particular, the basic network is built according to the following equation, wherein the target function implies the freight volume of the maximized basic network:
MAX (1 Distii * (wgtij + wgtji) *x11) wherein, xi] is a variable of 0 or 1, representing that if a path is added into the basic network; wgtij is the OD demand freight volume from warehouse i to warehouse j; wgt ji is the OD
demand freight volume from warehouse j to warehouse i; and Distii is the distance between warehouse i and warehouse j.
Date Recue/Date Received 2021-09-30 ratio * linenum 0 xij = 0 wgtij < loadration * minioad 0 The two equations above are two constraint conditions, wherein the constraint condition 0 implies the basic network satisfying the network proportion constraint, and the constraint condition 0 implies the limited freight volume for adding a path y into the basic network.
S400, allocating collecting and distributing centers for the rest of warehouses.
In an embodiment of the present invention, besides the warehouses selected to build the basic network in S300, the rest of warehouses are allocated with collecting and distributing centers. Preferably, collecting and distributing centers are allocated strategically to minimize collecting and distributing costs.
In particular, collecting and distributing centers are allocated according to the following equation, wherein the target function is the minimized costs for collecting and distributing centers.
Xin_ab XOUt ab DiStOut b * ) minload - minload a b wherein a implies the at OD demand pair, and b implies the selected collecting or distributing warehouse b.
= out Xi nab ab 0 Xi nab leftb 0 Xou tab lefty 0 a E
b Xinab = transW gta 0 ELI Xmitab = transW gta 0 The five equations above are five constraint conditions, wherein the constraint condition 0 implies the main freight volume of the OD demand pair a to transfer in and leave from the warehouse b; the constraint condition 0 and the constraint condition 0 imply the freight volumes passing the warehouse b being equal or less than the remaining freight capacity in the warehouse b;
and the constraint condition 0 and the constraint condition 0 indicate that the overall freight volume for the OD demand a to pass each warehouses is equal to the overall freight volume required by the OD
demand a.
S500, exporting a logistic transportation network.
In an embodiment of the present invention, after the forementioned procedures, an optimized logistic transportation network is exported.
Date Recue/Date Received 2021-09-30 The forementioned logistic transportation network optimization method predicts the transportation network conditions in the next period of time based on history order routing data, history transportation data, routing plan data and so on. According to the predicted data, the warehouse levels are adjusted and a basic network is built based on an optimized model, wherein the network modifications are controlled in line with constraint conditions, the final optimized network is computed and exported.
In an embodiment of the present invention, as shown in Fig. 2, a logistic transportation network optimization device adopting the forementioned method is provided, comprising:
a prediction module 100, an adjustment module 200, a building module 300, an allocation module 400, and an export module 500.
In particular, the prediction module 100, configured to create a prediction model for exporting predicted freight transportation volumes.
In an embodiment of the present invention, the described prediction model 100 also includes:
a feature collecting unit, configured to collect target features, wherein the described target features comprises at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure;
a volume selection unit, configured to select a data volume, wherein the data volume is the overall number of the described target features;
a target setting unit, configured to set a predicted target, wherein the described predicted target is workload; and a combination unit, configured to determine models, wherein predicted values from multiple selected models are combined.
In particular, three models including xgboost, random forest, and lightgbm are selected and the final exporting value is obtained by combining the predicted values of the three models.
The adjustment module 200, configured to adjust warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes.
The building module 300, configured to build a basic network, wherein the basic network comprises a portion of warehouses.
The allocation module 400, configured to allocate collecting and distributing centers for the rest of warehouses.
The export module 500, configured to export a logistic transportation network.
The detailed specifications of the logistic transportation network optimization device can refer to the forementioned specifications for the logistic transportation network optimization method, which is not explained in detail hereby. Each the described module in the logistic transportation network optimization device, in whole or in part, may be accomplished by software, hardware, and their combinations. Each the
However, the network level modifications involves the changes of locations and installments, with a long working period and a high cost, consequently leading to little feasibility of the optimized solution.
Summary In order to solve the forementioned problems, the present invention proposes the following strategies:
A logistic transportation network optimization method, comprises:
creating a prediction model to export predicted freight transportation volumes;
adjusting warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes;
building a basic network, wherein the basic network comprises a portion of warehouses;
allocating collecting and distributing centers for the rest of warehouses; and exporting a logistic transportation network.
Furthermore, the predicting process of the described prediction model comprises:
collecting target features, wherein the described target features comprise at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure.
selecting data volume, wherein the data volume is the overall number of the described target features;
setting a predicted target, wherein the described predicted target is workload; and determining models, wherein predicted values from multiple selected models are combined.
Furthermore, three models including xgboost, random forest, and lightgbm are selected and the final exporting value is obtained by combining the predicted values of the three models.
Furthermore, the described basic network is constrained by network ratio limits and freight transportation volumes to yield the maximum coverage range of the described network.
Furthermore, the collecting and distributing centers with the lowest costs are allocated, by determining costs of collection and distributing from the rest of warehouses individually.
A logistic transportation network optimization device, comprises:
Date Recue/Date Received 2021-09-30 a prediction module, configured to create a prediction model for exporting predicted freight transportation volumes;
an adjustment module, configured to adjust warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes;
a building module, configured to build a basic network, wherein the basic network comprises a portion of warehouses;
an allocation module, configured to allocate collecting and distributing centers for the rest of warehouses; and an export module, configured to export a logistic transportation network.
In particular, the described prediction model also includes:
a feature collecting unit, configured to collect target features, wherein the described target features comprises at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure;
a volume selection unit, configured to select a data volume, wherein the data volume is the overall number of the described target features;
a target setting unit, configured to set a predicted target, wherein the described predicted target is workload; and a combination unit, configured to determine models, wherein predicted values from multiple selected models are combined.
A computer apparatus is provided in the present invention, comprising a memory unit, a processor, and computer programs stored in the memory unit executable on the processor, wherein any of procedures of the described methods are performed when the described processor executes the described computer programs.
A readable computer storage medium with computer programs stored is provided in the present invention, wherein any of procedures of the described methods are performed when the described computer programs are executed on the described processor.
The method, the device, the computer apparatus, and the storage medium for logistic transportation network optimization in the present invention provides a prediction for freight volume in the next period of time to yield a more accurate plan; appropriate modifications of static routing plans and dynamic routing plans based on the future freight volume, to lower the overall transportation costs; dynamic modifications on warehouse levels based on the future freight volume to improve feasibility of the functionalities of warehouses in the networks; construction of basic networks based on the future freight volume and optimizing models, to maximize basic network utility; and a transportation model based on the future freight Date Recue/Date Received 2021-09-30 volume, to yield more appropriate network transportation structures and cost structures and improve feasibility of the modified network.
Brief descriptions of the drawings Fig. 1 is a flow diagram of the logistic transportation network optimization method in an embodiment of the present invention.
Fig. 2 is a structure diagram of the logistic transportation network optimization device in an embodiment of the present invention.
Fig. 3 is an internal structure diagram of the computer apparatus in an embodiment of the present invention.
Detailed descriptions In order to make the technical strategies of the present invention clearer, the accompany drawings for the present invention will be briefly introduced below. Obviously, the following drawings in the descriptions are used to explain the present invention and shall not restrict the present invention.
In a transportation network, an origin-destination (OD) demand and a corresponding transportation timeliness requirement exist for transportations in between any two branches.
Based on the current network, a proper modification yielding an equal or better timeliness is required to generate a transportation routing plan with lowered costs and controlled changes.
The transportation network optimization and modifications in the present invention are aiming at the transportation demand applicable to the next period of time. The feasibility of the optimized network is determined by the accuracy of the predicted freight volume in the future.
Therefore, the present application provides a logistic transportation network optimization method, applicable to the transportation demands for the network in the next period of time. In an embodiment, as shown in Fig. 1, the method comprises:
S100, creating a prediction model to export predicted freight transportation volumes.
In the present embodiment, a machine learning prediction model is created in the first place. By inputting currently available history data of the current transportation network into the prediction models to predict the freight volumes in the next period of time, the freight volume prediction results are exported.
In particular, the predicting process of the described prediction model in S100 comprises:
collecting target features, wherein the described target features comprise at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure. Preferably, the target features includes: the freight volume demand of any OD pairs in the past period of time, relationships between warehouses, functionality and Date Recue/Date Received 2021-09-30 level of each warehouse, the current transportation network structure, undertaken e-commerce promotion events (if at promoting timing such as November 11th, June 18', August 18', etc), holiday information (if on statutory holidays, traditional holidays or other holidays), trends and periodic rules of historic workload (such as the past one week freight volumes, the past two weeks freight volumes, or trends of the past one week freight volumes), historic numbers of all types of orders, timeliness tables for all locations, and/or weather forecast for delivery locations (such as numbers of raining days or sunny days in the next week), etc.
Selecting data volume, wherein the data volume is the overall number of the described target features.
Preferably, the number of features in a period of time, wherein the number of features for the recent half a year can be selected.
Setting a predicted target, wherein the described predicted target is workload. Preferably, based on demands, the prediction in a particular time period can be performed, such as the predicted workload in the next one month.
Determining models, wherein predicted values from multiple selected models are combined. Preferably, models can be selected such as xgboost, random forest, lightgbm, LR, etc. Due to low performance of regression by LR wherein relatively similar performances of regression by the other three models are obtained, xgboost, random forest, lightgbm are selected and the final exporting value is obtained by combining the predicted values of the three models.
S200, adjusting warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes.
In an embodiment of the present invention, according to the prediction exported from S100, data dimensions such as throughput capacity, correlation with other warehouses and geographical locations are considered to dynamically plan the warehouse levels and numbers of warehouses for each level.
S300, building a basic network, wherein the basic network comprises a portion of warehouses.
In an embodiment of the present invention, after adjusting warehouse levels and numbers of warehouses for each level by S200, certain proportions of higher-level warehouses are selected to build a basic network.
Preferably, building the basic network aims at a maximum coverage range by the network.
In particular, the basic network is built according to the following equation, wherein the target function implies the freight volume of the maximized basic network:
MAX (1 Distii * (wgtij + wgtji) *x11) wherein, xi] is a variable of 0 or 1, representing that if a path is added into the basic network; wgtij is the OD demand freight volume from warehouse i to warehouse j; wgt ji is the OD
demand freight volume from warehouse j to warehouse i; and Distii is the distance between warehouse i and warehouse j.
Date Recue/Date Received 2021-09-30 ratio * linenum 0 xij = 0 wgtij < loadration * minioad 0 The two equations above are two constraint conditions, wherein the constraint condition 0 implies the basic network satisfying the network proportion constraint, and the constraint condition 0 implies the limited freight volume for adding a path y into the basic network.
S400, allocating collecting and distributing centers for the rest of warehouses.
In an embodiment of the present invention, besides the warehouses selected to build the basic network in S300, the rest of warehouses are allocated with collecting and distributing centers. Preferably, collecting and distributing centers are allocated strategically to minimize collecting and distributing costs.
In particular, collecting and distributing centers are allocated according to the following equation, wherein the target function is the minimized costs for collecting and distributing centers.
Xin_ab XOUt ab DiStOut b * ) minload - minload a b wherein a implies the at OD demand pair, and b implies the selected collecting or distributing warehouse b.
= out Xi nab ab 0 Xi nab leftb 0 Xou tab lefty 0 a E
b Xinab = transW gta 0 ELI Xmitab = transW gta 0 The five equations above are five constraint conditions, wherein the constraint condition 0 implies the main freight volume of the OD demand pair a to transfer in and leave from the warehouse b; the constraint condition 0 and the constraint condition 0 imply the freight volumes passing the warehouse b being equal or less than the remaining freight capacity in the warehouse b;
and the constraint condition 0 and the constraint condition 0 indicate that the overall freight volume for the OD demand a to pass each warehouses is equal to the overall freight volume required by the OD
demand a.
S500, exporting a logistic transportation network.
In an embodiment of the present invention, after the forementioned procedures, an optimized logistic transportation network is exported.
Date Recue/Date Received 2021-09-30 The forementioned logistic transportation network optimization method predicts the transportation network conditions in the next period of time based on history order routing data, history transportation data, routing plan data and so on. According to the predicted data, the warehouse levels are adjusted and a basic network is built based on an optimized model, wherein the network modifications are controlled in line with constraint conditions, the final optimized network is computed and exported.
In an embodiment of the present invention, as shown in Fig. 2, a logistic transportation network optimization device adopting the forementioned method is provided, comprising:
a prediction module 100, an adjustment module 200, a building module 300, an allocation module 400, and an export module 500.
In particular, the prediction module 100, configured to create a prediction model for exporting predicted freight transportation volumes.
In an embodiment of the present invention, the described prediction model 100 also includes:
a feature collecting unit, configured to collect target features, wherein the described target features comprises at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure;
a volume selection unit, configured to select a data volume, wherein the data volume is the overall number of the described target features;
a target setting unit, configured to set a predicted target, wherein the described predicted target is workload; and a combination unit, configured to determine models, wherein predicted values from multiple selected models are combined.
In particular, three models including xgboost, random forest, and lightgbm are selected and the final exporting value is obtained by combining the predicted values of the three models.
The adjustment module 200, configured to adjust warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes.
The building module 300, configured to build a basic network, wherein the basic network comprises a portion of warehouses.
The allocation module 400, configured to allocate collecting and distributing centers for the rest of warehouses.
The export module 500, configured to export a logistic transportation network.
The detailed specifications of the logistic transportation network optimization device can refer to the forementioned specifications for the logistic transportation network optimization method, which is not explained in detail hereby. Each the described module in the logistic transportation network optimization device, in whole or in part, may be accomplished by software, hardware, and their combinations. Each the
7 Date Recue/Date Received 2021-09-30 described module may be independent hardware or hardware embedded in the processors of the computer apparatus. In the meanwhile, each the described module may be software that is stored in the memory unit of the computer apparatus, to invoke and execute the corresponding performances of the described modules.
In an embodiment of the present invention, a computer apparatus is provided, wherein the described computer apparatus can be a server with the internal structure diagram shown in Fig. 4. The computer apparatus comprises a processor, a memory unit, a network connection port, and a database connected by system bus control. The memory unit of the computer apparatus includes a nonvolatile storage medium and an internal memory. The operating system, computer programs, and databases are stored in the nonvolatile storage medium. The internal memory provides the operation environment for the execution of the operating system and the computer programs stored in the nonvolatile storage medium. The database of the computer apparatus is configured to store the message execution results. The network connection port of the computer apparatus is configured for communication with the external terminals via network connection.
The execution of the computer apparatus by the processor permits the method of logistic transportation network optimization.
It is comprehensible for those skilled in the art that the structure shown in Fig. 4 represents only a portion of structure associated with the applications of the present invention. The computer apparatus associated with the applications of the present invention are not restricted or limited by the structure. An exact computer apparatus may include more components or less components than that is shown in the drawings, possibly with combinations of some components or different component layouts.
In an embodiment of the present invention, a computer apparatus is provided, comprising a memory unit, a processor, and computer programs stored in the memory unit and executable on the processor. When the computer programs are executed on the processor, the forementioned procedures of the logistic transportation optimization method are performed.
All or portions of the forementioned procedures are comprehensible for those skilled in the art, and may be achieved by the computer program configured for sending command to the related hardware. The computer programs are stored in the computer readable nonvolatile memory unit.
When the computer programs are executed on the processor, the forementioned procedure of the embodiments may be included.
In particular, all the referred memory units, storages, databases, as well as any other media in the embodiments provided in the present invention, may include nonvolatile and/or volatile memory units. The nonvolatile memory units may include read-only memory (ROM), programmable ROM
(PROM), electrical programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory.
The volatile memory units may include random access memory (RAM) or external cache memory. To describe RAM without limiting, RAM may be different formats such as static RAM
(SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced
In an embodiment of the present invention, a computer apparatus is provided, wherein the described computer apparatus can be a server with the internal structure diagram shown in Fig. 4. The computer apparatus comprises a processor, a memory unit, a network connection port, and a database connected by system bus control. The memory unit of the computer apparatus includes a nonvolatile storage medium and an internal memory. The operating system, computer programs, and databases are stored in the nonvolatile storage medium. The internal memory provides the operation environment for the execution of the operating system and the computer programs stored in the nonvolatile storage medium. The database of the computer apparatus is configured to store the message execution results. The network connection port of the computer apparatus is configured for communication with the external terminals via network connection.
The execution of the computer apparatus by the processor permits the method of logistic transportation network optimization.
It is comprehensible for those skilled in the art that the structure shown in Fig. 4 represents only a portion of structure associated with the applications of the present invention. The computer apparatus associated with the applications of the present invention are not restricted or limited by the structure. An exact computer apparatus may include more components or less components than that is shown in the drawings, possibly with combinations of some components or different component layouts.
In an embodiment of the present invention, a computer apparatus is provided, comprising a memory unit, a processor, and computer programs stored in the memory unit and executable on the processor. When the computer programs are executed on the processor, the forementioned procedures of the logistic transportation optimization method are performed.
All or portions of the forementioned procedures are comprehensible for those skilled in the art, and may be achieved by the computer program configured for sending command to the related hardware. The computer programs are stored in the computer readable nonvolatile memory unit.
When the computer programs are executed on the processor, the forementioned procedure of the embodiments may be included.
In particular, all the referred memory units, storages, databases, as well as any other media in the embodiments provided in the present invention, may include nonvolatile and/or volatile memory units. The nonvolatile memory units may include read-only memory (ROM), programmable ROM
(PROM), electrical programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory.
The volatile memory units may include random access memory (RAM) or external cache memory. To describe RAM without limiting, RAM may be different formats such as static RAM
(SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced
8 Date Recue/Date Received 2021-09-30 DDRSDRAM (EDDRSDRAM), Synchlink DRAM (SLDRAM), direct rambus RAM (RDRAM), direct rambus dynamic RAM (DRDRAM) and rambus dynamic RAM (RDRAM).
The method, the device, the computer apparatus, and the storage medium for logistic transportation optimization in the present invention are different form the currently available methods wherein the adjustments of dynamic routing plans are based on optimal solutions. In contrary, the present invention provides appropriate modifications of both static routing plans and dynamic routing plans based on the current network. The present invention uses prediction models to predict the freight volumes in the next period of time according to the history data, then properly adjust warehouse levels and relationships between warehouses based on features of each warehouse, as known as modifications on the static routing plans of the network. The dynamic routing plans of the network are modified and optimized based on the predicted freight volumes, wherein the overall network costs are reduced while OD demands at the current timeliness are ensured. However, due to constraints by hardware environment of the current network, the detailed transportation data for each sorting center is difficult to acquire.
As a result, the network is modeled for illustration. When the modeled network structures are close to the reality, extent of modifications, modification costs and feasibility of practical implementations of the optimizations are considered. In terms of the currently available methods, the optimal solutions are calculated, wherein the existing transportation capacity of the current network and the discrepancy of the optimal solutions from the current network are not considered, leading to difficulties of practical implementations. In the present invention, with comprehensive considerations of current network structures, based on center levels and OD demand information amongst centers, the present invention adopts a multi-network strategy to build the overall transportation network with sequential buildup and integral optimization process, to allow an optimization strategy for adjusting current networks. In the meanwhile, the present invention provides: appropriate modifications of static routing plans and dynamic routing plans based on the future freight volume, to lower the overall transportation costs; dynamic modifications on warehouse levels based on the future freight volume to improve feasibility of the functionalities of warehouses in the networks; construction of basic networks based on the future freight volume and optimizing models, to maximize basic network utility; and a transportation model based on the future freight volume, to yield more appropriate network transportation structures and cost structures and improve feasibility of the modified network.
The forementioned technical strategies can be combined as possible. To be concise, not all possible combinations of each technical strategy in the forementioned embodiments are explained. However, the combinations of these technical strategies without any conflict shall fall in the protection scope of the present invention.
The method, the device, the computer apparatus, and the storage medium for logistic transportation optimization in the present invention are different form the currently available methods wherein the adjustments of dynamic routing plans are based on optimal solutions. In contrary, the present invention provides appropriate modifications of both static routing plans and dynamic routing plans based on the current network. The present invention uses prediction models to predict the freight volumes in the next period of time according to the history data, then properly adjust warehouse levels and relationships between warehouses based on features of each warehouse, as known as modifications on the static routing plans of the network. The dynamic routing plans of the network are modified and optimized based on the predicted freight volumes, wherein the overall network costs are reduced while OD demands at the current timeliness are ensured. However, due to constraints by hardware environment of the current network, the detailed transportation data for each sorting center is difficult to acquire.
As a result, the network is modeled for illustration. When the modeled network structures are close to the reality, extent of modifications, modification costs and feasibility of practical implementations of the optimizations are considered. In terms of the currently available methods, the optimal solutions are calculated, wherein the existing transportation capacity of the current network and the discrepancy of the optimal solutions from the current network are not considered, leading to difficulties of practical implementations. In the present invention, with comprehensive considerations of current network structures, based on center levels and OD demand information amongst centers, the present invention adopts a multi-network strategy to build the overall transportation network with sequential buildup and integral optimization process, to allow an optimization strategy for adjusting current networks. In the meanwhile, the present invention provides: appropriate modifications of static routing plans and dynamic routing plans based on the future freight volume, to lower the overall transportation costs; dynamic modifications on warehouse levels based on the future freight volume to improve feasibility of the functionalities of warehouses in the networks; construction of basic networks based on the future freight volume and optimizing models, to maximize basic network utility; and a transportation model based on the future freight volume, to yield more appropriate network transportation structures and cost structures and improve feasibility of the modified network.
The forementioned technical strategies can be combined as possible. To be concise, not all possible combinations of each technical strategy in the forementioned embodiments are explained. However, the combinations of these technical strategies without any conflict shall fall in the protection scope of the present invention.
9 Date Recue/Date Received 2021-09-30 The forementioned contents of preferred embodiments of the present invention, and shall not limit the applications of the present invention. Therefore, all alternations, modifications, equivalence, improvements of the present invention fall within the scope of the present invention.
Date Recue/Date Received 2021-09-30
Date Recue/Date Received 2021-09-30
Claims (10)
1 . A logistic transportation network optimization method, comprises:
creating a prediction model to export predicted freight transportation volumes;
adjusting warehouse levels and numbers of warehouses or each level based on the predicted freight transportation volumes;
building a basic network, wherein the basic network comprises a portion of warehouses;
allocating collecting and distributing centers for the rest of warehouses; and exporting a logistic transportation network.
creating a prediction model to export predicted freight transportation volumes;
adjusting warehouse levels and numbers of warehouses or each level based on the predicted freight transportation volumes;
building a basic network, wherein the basic network comprises a portion of warehouses;
allocating collecting and distributing centers for the rest of warehouses; and exporting a logistic transportation network.
2. The method of claim 1, is characterized in that the predicting process of the described prediction model comprises:
collecting target features, wherein the described target features comprise at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure;
selecting data volume, wherein the data volume is the overall number of the described target features;
setting a predicted target, wherein the described predicted target is workload; and determining models, wherein predicted values from multiple selected models are combined.
collecting target features, wherein the described target features comprise at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure;
selecting data volume, wherein the data volume is the overall number of the described target features;
setting a predicted target, wherein the described predicted target is workload; and determining models, wherein predicted values from multiple selected models are combined.
3. The method of claim 2, is characterized in that three models including xgboost, random forest, and lightgbm are selected and the final exporting value is obtained by combining the predicted values of the three models.
4. The method of claim 3, is characterized in that the described basic network is constrained by network ratio limits and freight transportation volumes to yield the maximum coverage range of the described network.
5. The method of claim 3, is characterized in that the collecting and distributing centers with the lowest costs are allocated, by determining costs of collection and distributing from the rest of warehouses individually.
6. A logistic transportation network optimization device, comprises:
a prediction module, configured to create a prediction model for exporting predicted freight transportation volumes;
an adjustment module, configured to adjust warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes;
a building module, configured to build a basic network, wherein the basic network comprises a portion of warehouses;
an allocation module, configured to allocate collecting and distributing centers for the rest of warehouses; and an export module, configured to export a logistic transportation network.
a prediction module, configured to create a prediction model for exporting predicted freight transportation volumes;
an adjustment module, configured to adjust warehouse levels and numbers of warehouses for each level based on the predicted freight transportation volumes;
a building module, configured to build a basic network, wherein the basic network comprises a portion of warehouses;
an allocation module, configured to allocate collecting and distributing centers for the rest of warehouses; and an export module, configured to export a logistic transportation network.
7. The device of claim 6 is characterized in that the described prediction model also includes:
a feature collecting unit, configured to collect target features, wherein the described target features comprises at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure;
a volume selection unit, configured to select a data volume, wherein the data volume is the overall number of the described target features;
a target setting unit, configured to set a predicted target, wherein the described predicted target is workload; and a combination unit, configured to determine models, wherein predicted values from multiple selected models are combined.
a feature collecting unit, configured to collect target features, wherein the described target features comprises at least required freight transportation volumes, relationships between warehouses, functionality and level of each warehouse, and the current transportation network structure;
a volume selection unit, configured to select a data volume, wherein the data volume is the overall number of the described target features;
a target setting unit, configured to set a predicted target, wherein the described predicted target is workload; and a combination unit, configured to determine models, wherein predicted values from multiple selected models are combined.
8. The device of claim 7 and 8, is characterized in that three models including xgboost, random forest, and lightgbm are selected and the final exporting value is obtained by combining the predicted values of the three models.
9. A computer apparatus comprises a memory unit, a processor, and computer programs stored in the memory unit executable on the processor, wherein the methods of any one of claims 1 ¨ 5 are performed when the described processor executes the described computer programs.
10. A readable computer storage medium with computer programs stored, wherein the methods of any one of claims 1 ¨ 5 are performed when the described computer programs are executed on the described processor.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2020107590486 | 2020-07-31 | ||
| CN202010759048.6A CN114065988A (en) | 2020-07-31 | 2020-07-31 | Method and device for optimizing logistics transportation network, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA3126555A1 true CA3126555A1 (en) | 2022-01-31 |
Family
ID=80185553
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA3126555A Pending CA3126555A1 (en) | 2020-07-31 | 2021-07-30 | Optimization method, device, computer equipment and storage medium of logistics transportation network |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114065988A (en) |
| CA (1) | CA3126555A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116362646A (en) * | 2023-05-31 | 2023-06-30 | 北京京东乾石科技有限公司 | Logistics network upgrading method and device |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116452245B (en) * | 2023-06-15 | 2023-09-01 | 跨越速运集团有限公司 | Logistics station site selection method, device, equipment and storage medium |
-
2020
- 2020-07-31 CN CN202010759048.6A patent/CN114065988A/en active Pending
-
2021
- 2021-07-30 CA CA3126555A patent/CA3126555A1/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116362646A (en) * | 2023-05-31 | 2023-06-30 | 北京京东乾石科技有限公司 | Logistics network upgrading method and device |
| CN116362646B (en) * | 2023-05-31 | 2023-09-26 | 北京京东乾石科技有限公司 | Logistics network upgrading method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114065988A (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA3126555A1 (en) | Optimization method, device, computer equipment and storage medium of logistics transportation network | |
| CN108776885A (en) | A kind of logistics platform system based on Internet of Things safety | |
| CN101421953A (en) | Control service capacity | |
| CN107451363B (en) | Calculation method for multi-objective balanced network continuous optimization problem | |
| CN113615137B (en) | CDN optimization platform | |
| CN111539590A (en) | Emergency resource allocation optimization method based on regret theory | |
| US20230252490A1 (en) | Methods for determining smart gas inspection plans and internet of things systems thereof | |
| Candas et al. | Integrated location and inventory planning in service parts logistics with customer-based service levels | |
| Triki | Solution methods for the periodic petrol station replenishment problem | |
| CN102740341A (en) | Method and device for predicting network traffic | |
| CN116703651B (en) | Intelligent gas data center operation management method, internet of things system and medium | |
| CN111105050B (en) | Fan maintenance plan generation method, device, equipment and storage medium | |
| Sechi et al. | A cost-simulation approach to finding economic optimality in leakage reduction for complex supply systems | |
| CN111583046A (en) | Financial resource processing method, device, equipment and system | |
| Lasdon et al. | Primal-dual and primal interior point algorithms for general nonlinear programs | |
| Kumar et al. | Zone-wise optimal operation policy and evaluation of system performance measures | |
| CN115641166A (en) | Virtual resource allocation strategy creation method and system | |
| Iturriaga et al. | Evolutionary algorithms for optimizing cost and QoS on cloud-based content distribution networks | |
| Silva Filho et al. | A Distributed Virtual-Machine Placement and Migration Approach Based on Modern Portfolio Theory | |
| Bonvin et al. | Robust design of pumping stations in water distribution networks | |
| Kalimulina | Math modeling of the reliability control and monitoring system of complex network platforms | |
| US20240112157A1 (en) | Liquidation management operations for improved ledger performance | |
| Kalimulina | A new approach for dependability planning of network systems | |
| Fernandes et al. | Downstream petroleum supply chains’ design and planning-contributions and roadmap | |
| CN116402238B (en) | Production scheduling method, device and equipment for packaging order and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request |
Effective date: 20220916 |
|
| EEER | Examination request |
Effective date: 20220916 |
|
| EEER | Examination request |
Effective date: 20220916 |
|
| EEER | Examination request |
Effective date: 20220916 |
|
| EEER | Examination request |
Effective date: 20220916 |
|
| EEER | Examination request |
Effective date: 20220916 |
|
| EEER | Examination request |
Effective date: 20220916 |
|
| EEER | Examination request |
Effective date: 20220916 |