CN107220737B - An optimization method for branch line network of port container liner in hub-spoke mode - Google Patents

Abstract

The invention discloses a method for optimizing a branch line network of a port container liner in a Hub-spoke mode, comprising the following steps: step 1, establishing a model description; step 2, establishing model assumptions; step 3, establishing model parameters and symbols; step 4 , establish a mathematical model. Considering the constraints of multiple ship types, ship capacity, liner time, two-way flow and port water depth, and guided by the lowest transportation cost, build a container hub-spoke branch line transportation model of the hub-feeding port to achieve the best ship capacity. configuration. Compared with similar models, the model proposed by the invention is closer to the actual container transportation, and can realize the optimal transportation configuration of the container transportation network.

Description

An optimization method for branch line network of port container liner in hub-spoke mode

technical field

The invention relates to the technical field of waterway transportation, in particular to a method for optimizing a branch line network of a port container liner in a Hub-spoke mode.

Background technique

As a necessary means of international trade and global economic integration, container ocean transportation has the advantages of large capacity and low cost, and plays a key role in international economic development.

Container ocean shipping has high requirements on ports and berths, both sufficient supply of goods and sufficient water depth conditions. Small ports (including small coastal ports, inland ports and dry land ports) often do not have freight organizations that open ocean routes. and berthing capacity, which, in the form of feeding large ports, collect containers to large seaports and then carry out ocean transportation. At present, container ocean shipping routes mainly include pendulum type, multi-port docking type and hub-spoke type, among which the hub-spoke mode has more cost and time advantages.

At present, most of the research on the optimization model of the two-way transportation scheduling of port container branch lines in the hub-spoke mode is based on considering factors such as empty container allocation, ship route speed, ship configuration, liner time, ship capacity, inventory capacity, etc. An optimization model of port container feeder transportation scheduling under the hub-spoke mode is constructed. However, there are few research results on how to carry out transportation optimization and ship scheduling when considering container loading and launching two-way flow and port water depth constraints, especially models that consider both two-way flow and port water depth constraints.

Therefore, considering the constraints of multiple ship types, ship capacity, liner time, two-way flow and port water depth, it has become an urgent problem to build a container hub-spoke feeder transport model from a hub port to a feeding port.

SUMMARY OF THE INVENTION

In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is to provide a method for optimizing a branch line network of a port container liner in a Hub-spoke mode, so as to solve the deficiencies of the prior art.

In order to achieve the above purpose,

A method for optimizing a branch line network of a port container liner in a hub-spoke mode, comprising the following steps:

Step 1. Create a model description;

Step 2: Establish model assumptions;

Step 3: Establish model parameters and symbols;

The fourth step is to establish a mathematical model.

In step 1, the model description is established as follows:

In setting a container hub-spoke mode ocean route, construct a hub port v ₀ on the hub-spoke network side of an ocean route, and there are n feed ports in the hub port v ₀ as {vi | _i =1,2 ,...,n}, the period of the route is T, and the shipping network is defined as a directed and authorized hub-spoke network composed of hub ports and feed ports G=(V,E,W), V={ v _i |i=0,1,...,n} represents the port in the network, E=(e _ij )=(i,j) represents the route between v _i and v _j , i,j∈(1,2 ,...,n), i≠j, W=d _ij represents the weight of the route e _ij .

The second step is to establish the model assumptions as follows:

(2-1) Assuming that the amount of containers fed into and launched from the feeding port is known, the hub port transports the containers from the feeding port to the hub port through differentiated shipping, and transports containers through large ships at the hub port. ocean shipping;

(2-2) Only one ship is accepted for the loading and unloading service of each feeding port, and the loading capacity of the ship cannot exceed the rated loading capacity;

(2-3) From the feeding port to the hub port, the loading capacity, capacity, transportation speed and cost of different transportation methods are different, and the hub port needs to complete the transportation task within the time window;

(2-4) The ship departs from the hub port, and all container logistics services that feed the port must be completed, and finally return to the hub port.

Step 3: Establish model parameters and symbols as follows:

(3-1) Establish ship parameters and symbols

为船舶集合；

共有K种船舶，m_k为第k种船舶的数量，

为船舶的总数量；(3-1a) In this hub-spoke network, the means of container transportation are ships, which are divided into inland ships and near-ocean ships. Define

Assemble for ships;

There are K kinds of ships, m _k is the number of k-th ships,

is the total number of ships;

(3-1b) Define the ship attribute set B _k , B _k =(s _k ,c _k ,d _k ,u _k ,p _k ,f _k ,e _k ,λ _k ),where k=1,2,…, K, B _k represent the attribute set of the k-th type of ship, s _k is the depreciation cost of the ship, _{ck represents the maximum capacity of the ship, d k represents} the required water depth of the ship, _uk represents the cost of the ship sailing unit distance, p _k Represents the unit time cost of berthing and departure of the ship, f _k represents the starting cost of the ship, e _k represents the unit time cost when the ship is docked; λ _k is a variable of 0, 1, when the kth ship is a coastal shipping ship _k = 1, otherwise λ _k = 0;

(3-2) Establish port parameters and symbols

和

为k类型船舶中的第l艘船舶上行离开喂给港i驶向喂给港j时的船舶箱量；

为k类型船舶中的第l艘船舶下行离开喂给港i驶向喂给港j时的船舶箱量；(3-2a) It is known that the quantity of containers fed to port i for loading and launching are respectively:

and

It is the container volume of the lth ship in the k type of ships when it leaves the feeding port i and goes to the feeding port j;

is the container volume of the lth ship among the k-type ships when it leaves the feeding port i and heads to the feeding port j;

为k类型船舶中的第l艘船舶上行过程中服务喂给港i的服务时间；

为k类型船舶中的第l艘船舶下行过程中服务喂给港i的服务时间；

为k类型船舶的第l艘船舶由喂给港i驶向喂给港j的航行时间；

为k类型船舶中的第l艘船舶服务喂给港i的靠泊和离港时间；(3-2b) Define hi as the docking depth of the ship feeding port _i ; define d _ij as the sailing distance from feeding port i to feeding port j; define

The service time of feeding port i for the l-th ship in the k-type ship during its ascent;

The service time of feeding port i for the l-th ship in the k-type ship during the descending process;

The sailing time of the lth ship of type k from feeding port i to feeding port j;

Feed the berthing and departure times of port i for the lth ship of type k;

喂给港i由k类船舶中的第l艘完成上行配送，则

否则

喂给港i由k类船舶中的第l艘完成下行配送，则

否则

k类型船舶中的第l艘由喂给港i驶向喂给港j，则x_ijkl＝1，否则x_ijkl＝0。(3-2c) Define control variables:

Feeding port i is completed by the lth ship in the k category of ships, then

otherwise

Feeding port i is completed by the lth ship in the k-category ship, then

otherwise

The lth ship among k-type ships sails from feeding port i to feeding port j, then x _ijkl =1, otherwise x _ijkl =0.

In step 4, the mathematical model is established as follows:

Determine the ship's path set to transport n loading and launching containers to the port, the objective function of the container port Hub-spoke branch transport model is:

Constraints St.

Equation (1) is the objective function, which requires the minimum delivery cost of all ships, μP is the penalty term, μ is the stage function, when the total time spent by each ship is less than the route period, μ=0, otherwise μ=1; P is a positive number (the value range of P is 1000000-10000000);

Equation (2) is a constraint condition, which means that all the upstream containers fed to the port have been served by ships, and only by one ship;

Equation (3) is a constraint condition, which means that all the downstream containers fed to the port have been served by ships, and only by one ship;

Equation (4) is a constraint condition, indicating that the flow of the ship is closed-loop;

Equation (5) is a constraint condition, which means that the loading box of each ship does not exceed its maximum capacity;

Equation (6) is a constraint condition, which means that all ships should depart from the hub port and return to the hub port;

Equation (7) is a constraint condition, indicating that the water depth of the feeding port meets the requirements of docked ships;

Equation (8) is a constraint condition, the service completion time of all ships is less than the period of the route, and the loading and unloading of containers cannot be completed at the same time.

The beneficial effects of the present invention are:

The invention solves the model with an algorithm through the constructed hub port-feeding port container Hub-spoke branch line transport model, and can obtain the optimal transport configuration of the container Hub-spoke transport network.

The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings, so as to fully understand the purpose, characteristics and effects of the present invention.

Description of drawings

FIG. 1 is an iterative diagram of branch network optimization of the present invention.

Detailed ways

As shown in Figure 1, a method for optimizing a branch line network of a port container liner in a hub-spoke mode includes the following steps:

Step 1. Create a model description;

Step 2: Establish model assumptions;

Step 3: Establish model parameters and symbols;

The fourth step is to establish a mathematical model.

In this embodiment, in step 1, the description of the established model is specifically:

In a container hub-spoke mode ocean route, construct a hub port v ₀ on the hub-spoke network side of an ocean route, and there are n feed ports in the hub port v ₀ as {vi | _i =1,2, ...,n}, the period of the route is T, and the shipping network is defined as a directed and authorized hub-spoke network composed of hub ports and feed ports G=(V,E,W), V={v _i |i=0,1,...,n} represents the port in the network, E=(e _ij )=(i,j) represents the route between v _i and v _j , i,j∈(1,2, ..., n), i≠j, W=d _ij represents the weight of the route e _ij , which in the present invention refers to the distance between the routes between ports.

In this embodiment, in step 2, the assumption of establishing the model is specifically:

(2-1) Assuming that the amount of containers fed into and launched from the feeding port is known, the hub port transports the containers from the feeding port to the hub port through differentiated shipping, and transports containers through large ships at the hub port. ocean shipping;

(2-2) Only one ship is accepted for the loading and unloading service of each feeding port, and the loading capacity of the ship cannot exceed the rated loading capacity;

(2-3) From the feeding port to the hub port, the loading capacity, capacity, transportation speed and cost of different transportation methods are different, and the hub port needs to complete the transportation task within the time window;

(2-4) The ship departs from the hub port, and all container logistics services that feed the port must be completed, and finally return to the hub port.

In this embodiment, in step 3, the parameters and symbols of the established model are specifically:

(3-1) Establish ship parameters and symbols

为船舶集合；

共有K种船舶，m_k为第k种船舶的数量，

为船舶的总数量。(3-1a) In this hub-spoke network, the main vehicle for container transportation is ships, which are divided into inland ships and near-ocean ships. Definition

Assemble for ships;

There are K kinds of ships, m _k is the number of k-th ships,

is the total number of ships.

(3-1b) Define the ship attribute set B _k , B _k =(s _k ,c _k ,d _k ,u _k ,p _k ,f _k ,e _k ,λ _k ),where k=1,2,…, K, B _k represent the attribute set of the k-th type of ship, s _k is the depreciation cost of the ship, _{ck represents the maximum capacity of the ship, d k represents} the required water depth of the ship, _uk represents the cost of the ship sailing unit distance, p _k Represents the cost per unit time of the ship's berthing and departure, f _k represents the starting cost of the ship, and e _k represents the cost per unit time when the ship is docked. λ _k is a variable of 0 and 1, when the k-th ship belongs to the coastal shipping ship, λ _k =1, otherwise λ _k =0.

(3-2) Establish port parameters and symbols

和

为k类型船舶中的第l艘船舶上行离开喂给港i驶向喂给港j时的船舶箱量；

为k类型船舶中的第l艘船舶下行离开喂给港i驶向喂给港j时的船舶箱量。(3-2a) It is known that the quantity of containers fed to port i for loading and launching are respectively:

and

It is the container volume of the lth ship in the k type of ships when it leaves the feeding port i and goes to the feeding port j;

It is the container volume of the lth ship in the k type of ships when it leaves the feeding port i and goes to the feeding port j.

为k类型船舶中的第l艘船舶上行过程中服务喂给港i的服务时间；

为k类型船舶中的第l艘船舶下行过程中服务喂给港i的服务时间；

为k类型船舶的第l艘船舶由喂给港i驶向喂给港j的航行时间；

为k类型船舶中的第l艘船舶服务喂给港i的靠泊和离港时间；(3-2b) Define hi as the docking depth of the ship feeding port _i ; define d _ij as the sailing distance from feeding port i to feeding port j; define

The service time of feeding port i for the l-th ship in the k-type ship during its ascent;

The service time of feeding port i for the l-th ship in the k-type ship during the descending process;

The sailing time of the lth ship of type k from feeding port i to feeding port j;

Feed the berthing and departure times of port i for the lth ship of type k;

喂给港i由k类船舶中的第l艘完成上行配送，则

否则

喂给港i由k类船舶中的第l艘完成下行配送，则

否则

k类型船舶中的第l艘由喂给港i驶向喂给港j，则x_ijkl＝1，否则x_ijkl＝0。(3-2c) Define control variables:

Feeding port i is completed by the lth ship in the k category of ships, then

otherwise

Feeding port i is completed by the lth ship in the k-category ship, then

otherwise

The lth ship among k-type ships sails from feeding port i to feeding port j, then x _ijkl =1, otherwise x _ijkl =0.

In the present embodiment, step 4, establishing a mathematical model is specifically:

Determine the ship's path set to transport n loading and launching containers to the port, the objective function of the container port Hub-spoke branch transport model is:

Constraints St.

Equation (1) is the objective function, which requires the minimum delivery cost of all ships, μP is the penalty term, μ is the stage function, when the total time spent by each ship is less than the route period, μ=0, otherwise μ=1; P is a positive number (the value range of P is 1000000-10000000);

Equation (2) is a constraint condition, which means that all the upstream containers fed to the port have been served by ships, and only by one ship;

Equation (3) is a constraint condition, which means that all the downstream containers fed to the port have been served by ships, and only by one ship;

Equation (4) is a constraint condition, indicating that the flow of the ship is closed-loop;

Equation (5) is a constraint condition, which means that the loading box of each ship does not exceed its maximum capacity;

Equation (6) is a constraint condition, which means that all ships should depart from the hub port and return to the hub port;

Equation (7) is a constraint condition, indicating that the water depth of the feeding port meets the requirements of docked ships;

Equation (8) is the constraint condition, T is the route period, which means that the service completion time of all ships is less than the period of the route, and container loading and unloading and launching can not be completed at the same time.

Example 1

The method for optimizing the branch line network of port container liner under the Hub-spoke mode proposed by the present invention is applied to the Hub-spoke transport network of Shanghai Port to the United States and West containers as an example, and the implementation is as follows:

Shanghai Port is selected as the hub port, Nantong, Zhenjiang, Yangzhou, Nanjing and Changzhou are the ports along the Yangtze River. The time window requires serving the feeding port, unloading the import container to the corresponding port, and loading the export container, and all ships return to Shanghai Port within the specified time.

The transportation distance between the hub port Shanghai port and the feeder port is shown in Table 1.

Table 1 Shipping distance between Shanghai port and feeder port (unit: nautical miles)

The required water depth of each feeding port channel is shown in Table 2.

Table 2 Water depth of each feeding port channel (unit: meter)

The volume of import and export containers fed to each port within a week is shown in Table 3.

Table 3 The volume of import and export containers fed to each port (unit: TEU)

The loading and unloading costs of container ships at each feeding port can be obtained from the volume of import and export containers, as shown in Table 4. The berthing and unloading charges include container port charges and container handling charges.

Table 4 Loading and unloading charges for each feeding port (unit: yuan)

According to the volume of import and export containers in the feeder port, 350TEU, 500TEU, 800TEU and 1000TEU are selected as representative ship types. The basic parameters of the four types of ships are shown in Table 5.

Table 5 represents the basic parameters of the ship type

The voyage costs of the four types of ships are shown in Table 6. Among them, the initial cost is calculated based on the depreciation cost of the container ship, calculated at 5% of the value of the ship, and the depreciation period is 15 years. Navigation costs are based on the price of heavy and light fuel oil consumed by the ship and the wages of the crew. The berthing and departure fees are based on the net tonnage of the ship to obtain port charges, ship tonnage tax, pilotage fees, berthing fees, mooring and un-mooring fees and tugboat fees.

Table 6 represents the voyage cost of the ship type

(Data source: Liu Jianqiu, "Research on Network Optimization of Feeder Container Liners in Uncertain Environment")

According to the above data, the multi-agent genetic algorithm is used to solve the model. The model requires container ships to call at the feeder port to meet the draft requirements, and the overall shipping network to meet the time window requirements (the time window is limited to 1 week). The correlation coefficients are shown in Table 7.

Table 7 Multi-agent genetic algorithm correlation coefficient

Under the win8 operating system, the memory is 4G, the CPU is Intel(R) Core(TM) i5-4210U hardware platform, the matlab2017a platform is used for programming, and the multi-agent genetic algorithm and the genetic algorithm are used to optimize the iteration, the process is shown in the figure 1 shown.

The running time of the multi-agent genetic algorithm is 12.38 seconds, the algorithm converges to the extreme point of 2 336 787 in about 25 times, and the algorithm is relatively stable. The running time of the genetic algorithm is 26.848 seconds, and the algorithm converges to the extreme point of 2 336 787 in about 75 times. It can be seen from the optimization process that the multi-agent genetic algorithm has obvious advantages compared with the genetic algorithm, and the convergence speed and convergence time are significantly improved.

Statistical optimization results are obtained several times, and the optimal branch line transportation network is obtained as shown in the table.

Table 8 Optimal feeder transportation network

The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and changes according to the concept of the present invention without creative efforts. Therefore, all technical solutions that can be obtained by those skilled in the art through logical analysis, reasoning or limited experiments on the basis of the prior art according to the concept of the present invention shall fall within the protection scope determined by the claims.

Claims (1)

1. a method for optimizing a branch line network of a port container liner under a Hub-spoke mode, is characterized in that, comprises the following steps: Step 1. Create a model description; Step 2: Establish model assumptions; Step 3. Establish model parameters and symbols; Step 4. Establish a mathematical model; In step 1, the model description is established as follows: In setting a container hub-spoke mode ocean route, construct a hub port v ₀ on the hub-spoke network side of an ocean route, and there are n feed ports in the hub port v ₀ as {vi | _i =1,2 ,...,n}, the period of the route is T, and the shipping network is defined as a directed and authorized hub-spoke network composed of hub ports and feeding ports, G=(V,E,W), V={v _i |i=0,1,...,n} represents the port in the network, E=(e _ij )=(i,j) represents the route between v _i and v _j , i,j∈(1,2, ...,n), i≠j, W=d _ij represents the weight of the route e _ij , which refers to the distance between the ports; The second step is to establish the model assumptions as follows: (2-1) Assuming that the volume of containers fed into and launched from the feeder port is known, the hub port transports containers from the feeder port to the hub port through differentiated shipping, and transports containers from the feeder port to the hub port through large ships at the hub port. ocean shipping; (2-2) Only one ship is accepted for the loading and unloading service of each feeding port, and the loading capacity of the ship cannot exceed the rated loading capacity; (2-3) From the feeding port to the hub port, the loading capacity, capacity, transportation speed and cost of different transportation methods are different, and the hub port needs to complete the transportation task within the time window; (2-4) The ship departs from the hub port, and all container logistics services feeding the port must be completed, and finally return to the hub port; Step 3: Establish model parameters and symbols as follows: (3-1) Establish ship parameters and symbols (3-1a) In this hub-spoke network, the means of container transportation are ships, which are divided into inland ships and near-ocean ships. Define

Assemble for ships;

There are K kinds of ships, m _k is the number of k-th ships,

is the total number of ships; (3-1b) Define the ship attribute set B _k , B _k =(s _k ,c _k ,d _k ,u _k ,p _k ,f _k ,e _k ,λ _k ),where k=1,2,…, K, B _k represent the attribute set of the k-th type of ship, s _k is the depreciation cost of the ship, _{ck represents the maximum capacity of the ship, d k represents} the required water depth of the ship, _uk represents the cost of the ship sailing unit distance, p _k Represents the unit time cost of berthing and departure of the ship, f _k represents the starting cost of the ship, e _k represents the unit time cost when the ship is docked; λ _k is a variable of 0, 1, when the kth ship is a coastal shipping ship _k = 1, otherwise λ _k = 0; (3-2) Establish port parameters and symbols (3-2a) It is known that the quantity of containers fed to port i for loading and launching are respectively:

and

It is the container volume of the lth ship in the k type of ships when it leaves the feeding port i and goes to the feeding port j;

is the container volume of the lth ship among the k-type ships when it leaves the feeding port i and heads to the feeding port j; (3-2b) Define hi as the docking depth of the ship feeding port _i ; define d _ij as the sailing distance from feeding port i to feeding port j; define

The service time of feeding port i for the l-th ship in the k-type ship during its ascent;

The service time of feeding port i for the l-th ship in the k-type ship during the descending process;

The sailing time of the lth ship of type k from feeding port i to feeding port j;

Feed the berthing and departure times of port i for the lth ship of type k; (3-2c) Define control variables:

Feeding port i is completed by the lth ship in the k category of ships, then

otherwise

Feeding port i is completed by the lth ship in the k-category ship, then

otherwise

The lth ship among k-type ships sails from feeding port i to feeding port j, then x _ijkl = 1, otherwise x _ijkl = 0; In step 4, the mathematical model is established as follows: Determine the ship's path set to transport n loading and launching containers to the port, the objective function of the container port Hub-spoke branch transport model is:

Constraints St.

Equation (1) is the objective function, the purpose requires the minimum delivery cost of all ships, μP is the penalty term, μ is the stage function, when the total time spent by each ship is less than the route period, μ=0, otherwise μ=1; P is a positive number; Equation (2) is a constraint condition, which means that all the upstream containers fed to the port have been served by ships, and only by one ship; Equation (3) is a constraint condition, which means that all the downstream containers fed to the port have been served by ships, and only by one ship; Equation (4) is a constraint condition, indicating that the flow of the ship is closed-loop; Equation (5) is a constraint condition, which means that the loading box of each ship does not exceed its maximum capacity; Equation (6) is a constraint condition, which means that all ships should depart from the hub port and return to the hub port; Equation (7) is a constraint condition, indicating that the water depth of the feeding port meets the requirements of docked ships; Equation (8) is the constraint condition, T is the route period, indicating that the service completion time of all ships is less than the route period, and container loading and unloading and launching can not be completed at the same time; The value range of P is 1000000-10000000.

Images