CN117010778A - Data management method based on multi-mode intermodal - Google Patents

Abstract

The invention provides a data management method based on multi-mode intermodal, which comprises the following steps: acquiring a first historical data set in the multi-type intermodal transportation process; performing cluster analysis on the distance between the departure place and the destination of the historical transportation under the set time length in the first historical data set to generate a first cluster set based on the departure place and a second cluster set based on the destination; determining a normal historical data set, an abnormal historical data set, a minimized cost objective function and a minimized transportation time objective function based on the first clustering set and the second clustering set; determining an optimal path according to the normal historical data set, the abnormal historical data set, the minimized cost objective function and the minimized transportation time objective function; and establishing a diversified supply chain network and a data platform based on the optimal path to realize data sharing. The system can improve the transportation efficiency of the supply chain, reduce the transportation cost of goods, enhance the flexibility and toughness of the supply chain and improve the data management and analysis capability.

Description

Data management method based on multi-mode intermodal

Technical Field

The invention relates to the technical field of data management, in particular to a data management method based on multi-mode intermodal.

Background

At present, along with the continuous expansion of the total economic quantity and the importance of people on living development environment, energy and efficiency, higher requirements are also put forward on logistics transportation, and multi-mode intermodal transportation is used as a transportation mode for jointly completing the transportation process by mutually connecting and transporting two or more transportation means, so that the advantages of different transportation modes can be fully exerted, the defects of unreachable, uneconomical and poor reliability of the traditional transportation modes are avoided, and the path planning and data management of the multi-mode intermodal transportation are more complex.

Disclosure of Invention

The invention provides a data management method based on multi-mode intermodal, which is characterized in that a normal historical data set, an abnormal historical data set, a minimized cost objective function and a minimized transportation time objective function are determined based on a first clustering set of a departure place and a second clustering set of a destination, and an optimal path is determined to establish a diversified supply chain network and a data platform, so that data sharing is realized, the transportation efficiency of the supply chain is improved, the transportation cost of goods is reduced, the flexibility and toughness of the supply chain are enhanced, and the data management and analysis capability is improved.

The invention provides a data management method based on multi-mode intermodal, which comprises the following steps:

Step 1: acquiring a first historical data set in the multi-joint transportation process, wherein the first historical data set comprises a transportation route, transportation time, transportation mode and transportation cost of each time of historical transportation under a set time length, departure places under the transportation route, corresponding cargo transportation amount, a plurality of destinations under the transportation route and corresponding cargo demand respectively;

step 2: performing cluster analysis on longitude and latitude coordinates of a departure place and a destination of historical transportation under a set time length in a first historical data set to generate a first cluster set based on the departure place and a second cluster set based on the destination;

step 3: determining a normal historical data set, an abnormal historical data set, a minimized cost objective function and a minimized transportation time objective function based on the first clustering set and the second clustering set;

step 4: determining an optimal path according to the normal historical data set, the abnormal historical data set, the minimized cost objective function and the minimized transportation time objective function;

step 5: and establishing a diversified supply chain network and a data platform based on the optimal path to realize data sharing.

Preferably, generating the first set of clusters based on the departure place and the second set of clusters based on the destination includes:

Randomly selecting N1 objects from a plurality of departure places under the set time length in a first historical data set as a first clustering center, and randomly selecting N2 objects from a plurality of destinations as a second clustering center;

calculating the distance between each first clustering center and the rest departure place in the first historical data set respectively, and determining a first subset based on each first clustering center;

judging whether a first cluster center based on the first subset needs to be replaced or not;

if not, regarding the corresponding first cluster center as a third cluster center;

otherwise, taking the point of the first subset meeting the nearest principle as a third class center;

calculating the distance between each second aggregation center and the rest destination in the first historical data set, and determining a second subset based on each second aggregation center;

determining whether a second hub based on the second subset needs replacement;

if not, regarding the corresponding second clustering center as a fourth clustering center;

otherwise, taking the points of the second subset meeting the principle of closest distance as a fourth clustering center.

Preferably, determining the normal historical data set and the abnormal historical data set based on the first clustering set and the second clustering set comprises:

Counting a second historical data set based on N1 third cluster centers in the first cluster set and N2 fourth cluster centers in the second cluster set;

analyzing the second historical data set, and determining a transportation capacity constraint condition of each third cluster center, and a time constraint condition, a distance constraint condition and a transportation mode constraint condition corresponding to each third cluster center to each fourth cluster center;

determining the first decision variable of each constraint condition asWherein, the method comprises the steps of, wherein,the method comprises the steps of carrying out a first treatment on the surface of the Determining the second decision variable of each constraint condition asWherein->Wherein N1>1 and N2>1；

Wherein, the transportation capability constraint condition of each third class center is:

wherein k represents the transport mode selection of each third cluster center to the fourth cluster center j,/I>Representing the number of modes of transportation, +.>Representing the cargo demand of the fourth cluster center j, < +.>Representing the maximum transport capacity of each third cluster center under all transport modes from reaching all fourth cluster centers; />Representing the maximum transport capacity of each third cluster center to reach the fourth cluster center j using the transport mode k;

the time constraint conditions and the distance constraint conditions corresponding to the third cluster center to all the fourth cluster centers are respectively as follows:

Wherein, the->Representing the transport time using transport means k from the third cluster center i to the fourth cluster center j, +.>Weights representing path optimization from the third cluster center i to the fourth cluster center j, +.>Limited transportation time, + representing the use of transportation means k from third cluster center i to fourth cluster center j>Representing the transportation distance from the third cluster center i to the fourth cluster center j using transportation means k, #>Representing a limited transportation distance from the third cluster center i to the fourth cluster center j using the transportation means k; />Representing the optimization time for path optimization from the third cluster center i to the fourth cluster center j; />Representing an optimized distance for path optimization from the third cluster center i to the fourth cluster center j, wherein +.>The method comprises the steps of carrying out a first treatment on the surface of the Grouping the historical data related to the corresponding third class center in the second historical data set, wherein the historical data simultaneously meets the transportation capacity constraint condition, the time constraint condition, the distance constraint condition and the transportation mode constraint conditionAnd (3) classifying the rest data which do not meet the constraint condition at the same time into the normal data set and classifying the rest data into the abnormal data set.

Preferably, it comprises:

the minimized transportation cost objective function Y1 and the minimized transportation time objective function Y2 are determined as follows:

The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing arrival at the ij0 th destination within the cluster range belonging to the fourth cluster center j from the third cluster center i and using transport means +.>Is>Represents the ij 0-th destination within the cluster range from the third cluster center i to the fourth cluster center j and uses transportation means +.>Min represents the minimum value symbol.

Preferably, determining the optimal path from the normal historical data set, the abnormal historical data set, and the minimized cost objective function, the minimized transit time objective function, comprises:

establishing a transportation model for a plurality of departure places, a plurality of destinations, a third cluster center and a fourth cluster center under a historical transportation line in a normal data set through a GIS (geographic information system), and simulating a multi-type intermodal transportation process through an ant colony;

determining a plurality of middle points from the third cluster center to the fourth cluster center as middle nodes;

determining a two-dimensional array according to the third cluster center and the fourth cluster centerRepresenting the number of path selections from each third cluster center to each fourth cluster center;

determining a two-dimensional array according to a plurality of middle points corresponding to the third cluster center and the fourth cluster center Wherein->Representing the number of intermediate nodes corresponding to the arrival of the third cluster center i at the fourth cluster center j, na being the maximum number of intermediate nodes corresponding to the arrival of the third cluster center i at the fourth cluster center j,/being the maximum number of intermediate nodes corresponding to the arrival of the third cluster center i at the third cluster center j,/being the maximum number of intermediate nodes corresponding to the arrival of the>Representing the number of the intermediate node where the current node is located;

determining an arrayWherein->Indicating that the node has been accessed and,representing nodes that have not been accessed;

initializing commandThe number of paths from the third cluster center i to the fourth cluster center j is 1;

path number calculation using depth-first search, starting from third class center, satisfiesUnder the condition of accessing each node in turn, and marking the accessed intermediate nodes asUpdating the array after reaching the fourth cluster center>The method comprises the steps of carrying out a first treatment on the surface of the When (when)When (I)>Representing the number of path selections from the third cluster center i to the fourth cluster center j and assigning a corresponding +.>The method comprises the steps of carrying out a first treatment on the surface of the The number of ants placed in each third class center isWherein m is an integer value determined from the total number of paths from all third cluster centers to all fourth cluster centers, and the number of ants in the path from each third cluster center to each fourth cluster center is limited to +.>The method comprises the steps of carrying out a first treatment on the surface of the Initializing the concentration of pheromones among the third cluster center, the intermediate node and the fourth cluster center to be 1;

Constructing a mixed integer linear model according to the first decision variable, the second decision variable, the transportation capacity constraint condition of each third cluster center, the time constraint condition, the distance constraint condition and the minimum transportation cost objective function corresponding to each third cluster center to all fourth cluster centers;

according to the mixed integer linear model, determining the violation degree of the transportation capability of each third class center, the violation degree of the time from each third class center to each fourth cluster center, the violation degree of the distance, the first contribution degree of the minimized transportation cost objective function corresponding to the first decision variable and the second contribution degree of the minimized transportation cost objective function corresponding to the second decision variable;

taking the transport capacity violation degree, the time violation degree, the distance violation degree, the first contribution degree and the second contribution degree as heuristic information and carrying out normalization processing;

determining a next node according to the current node and the pheromone concentration among the nodes and heuristic information after normalization processing;

updating the pheromone concentration between the corresponding current node and the determined next node according to the crawling time after the next node is selected;

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing the current node when local pheromone is updated +. >And the determined next node->Pheromone concentration between->Indicating the pheromone exertion rate->Representing the constant pheromone increment,>representing the ant at the current node->And the determined next node->The corresponding crawling time; />Representing the current node +.>Pheromone concentration of (2);

setting the incremental range of pheromone between the current node and the determined next node asIf (if)Then determine +.>Then determine +.>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing a minimum value of the corresponding pheromone increment range; />Representing the maximum value of the corresponding pheromone increment range;

after each iteration is finished, the global pheromone concentration is updated according to the volatilization rate of the pheromone and the pheromone left by ants;

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing the concentration of pheromone between the third cluster center i and the fourth cluster center j,/->Representing the shortest crawling time corresponding from the third cluster center i to the fourth cluster center j, +.>Representation->Crawling time used by ants from third cluster center i to fourth cluster center j only on average, +.>Representing a balance factor; />Representing the updated pheromone concentration;

wherein, all ants reach the fourth clustering center and are regarded as the end of one iteration;

when the maximum iteration times are reached, determining a path with highest pheromone concentration between each third cluster center i and each fourth cluster center j as a corresponding first transportation path according to the pheromone concentration distribution of the last iteration;

Determining a transportation abnormality reason according to the transportation route, transportation time, transportation mode and transportation cost from each third cluster center to each fourth cluster center in the abnormal data set;

and according to the reasons of the abnormal transportation and the actual transportation demands of the corresponding transportation routes in the abnormal data set, adjusting the corresponding first transportation paths to obtain the optimal paths.

Preferably, establishing a diverse supply chain network and a data platform based on the optimal path includes:

setting a network node based on the departure place, the destination and the neutral point in all the optimal paths;

and formulating unified standards and interface specifications for each network node, and constructing and obtaining a diversified supply chain network and a data platform.

Preferably, it comprises:

real-time tracking and monitoring all transportation processes between the first cluster set and the second cluster set based on the diversified supply chain network and the data platform;

and establishing a corresponding emergency mechanism according to the reasons of the transportation abnormality, and timely responding and processing when the transportation process tracked and monitored in real time is abnormal.

Preferably, it comprises:

when (when)When the value is 0, the third cluster center i to the fourth cluster center j do not use the path of the transportation mode k, and the third cluster center i is +. >When the value is 1, the paths from the third cluster center i to the fourth cluster center j by using the transportation mode k are shown;

when (when)When the value is 0, the third cluster center i to the fourth cluster center j do not perform path optimization according to the transportation cost, and are in the presence of->And when the value is 1, the third cluster center i to the fourth cluster center j are used for carrying out path optimization according to the transportation cost.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.

The technical scheme of the invention is further described in detail through the drawings and the embodiments.

Drawings

The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:

fig. 1 is a flowchart of a method for managing data based on multi-modal intermodal in an embodiment of the invention.

Detailed Description

The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.

Example 1:

the embodiment of the invention provides a flow chart of a data management method based on multi-mode intermodal, which is shown in fig. 1 and comprises the following steps:

step 1: acquiring a first historical data set in the multi-joint transportation process, wherein the first historical data set comprises a transportation route, transportation time, transportation mode and transportation cost of each time of historical transportation under a set time length, departure places under the transportation route, corresponding cargo transportation amount, a plurality of destinations under the transportation route and corresponding cargo demand respectively;

step 2: performing cluster analysis on longitude and latitude coordinates of a departure place and a destination of historical transportation under a set time length in a first historical data set to generate a first cluster set based on the departure place and a second cluster set based on the destination;

step 3: determining a normal historical data set, an abnormal historical data set, a minimized cost objective function and a minimized transportation time objective function based on the first clustering set and the second clustering set;

step 4: determining an optimal path according to the normal historical data set, the abnormal historical data set, the minimized cost objective function and the minimized transportation time objective function;

Step 5: and establishing a diversified supply chain network and a data platform based on the optimal path to realize data sharing.

In this embodiment, the multi-mode intermodal means that different transport sub-sections under the transport route in the cargo transportation process are joined to each other by different transport modes, forming an overall transport mode.

In this embodiment, the set length of time in the first set of historical data may be one quarter, one year, two years, etc.

In this embodiment, the transportation mode refers to a transportation mode adopted by different transportation sub-sections under a transportation route corresponding to the goods from the departure place to the destination, and includes land transportation, sea transportation, air transportation, railway transportation, and the like.

In this embodiment, a transportation route refers to a collection of transportation sub-segments corresponding to a cargo from a departure place to a destination.

In this embodiment, the first cluster set refers to a set of the historic transportation departure points for a set time period, and the second cluster set refers to a set of the historic transportation destinations for a set time period.

In this embodiment, the normal history data set refers to a history transportation process that satisfies the constraint condition for a set time period, and the abnormal history data set refers to a history transportation process that does not satisfy the constraint condition for a set time period.

In this embodiment, the optimal path refers to an optimal transport path from the third cluster center to the fourth cluster center, which is obtained according to the minimized cost objective function, the minimized transport time objective function, the reason for transport anomalies, and the actual transport requirements of the corresponding transport route in the anomaly data set.

In this embodiment, the diversified supply chain network refers to a supply chain network which is built by various transportation modes and nodes and includes elements such as transportation modes and transfer schemes of different transportation routes and different transportation sub-sections under the transportation routes in the cargo transportation process, so as to select an optimal transportation mode and an optimal path according to characteristics, transportation requirements, geographical conditions and the like of the cargo.

In this embodiment, data sharing refers to sharing transportation data and information among related parties such as transportation service providers, owners, forwarders, customs clearance lines, etc. in various supply chain networks and data platforms.

The beneficial effects of the technical scheme are as follows: by determining a normal historical data set, an abnormal historical data set, a minimized cost objective function, a minimized transportation time objective function and an optimal path based on a first clustering set of departure places and a second clustering set of destinations to establish a diversified supply chain network and a data platform, data sharing is realized, so that the transportation efficiency of the supply chain can be improved, the transportation cost of goods can be reduced, the flexibility and toughness of the supply chain can be enhanced, and the data management and analysis capability can be improved.

Example 2:

the embodiment of the invention provides a data management method based on multi-mode intermodal, which generates a first cluster set based on a departure place and a second cluster set based on a destination, and comprises the following steps:

randomly selecting N1 objects from a plurality of departure places under the set time length in a first historical data set as a first clustering center, and randomly selecting N2 objects from a plurality of destinations as a second clustering center;

calculating the distance between each first clustering center and the rest departure place in the first historical data set respectively, and determining a first subset based on each first clustering center;

judging whether a first cluster center based on the first subset needs to be replaced or not;

if not, regarding the corresponding first cluster center as a third cluster center;

otherwise, taking the point of the first subset meeting the nearest principle as a third class center;

calculating the distance between each second aggregation center and the rest destination in the first historical data set, and determining a second subset based on each second aggregation center;

determining whether a second hub based on the second subset needs replacement;

if not, regarding the corresponding second clustering center as a fourth clustering center;

Otherwise, taking the points of the second subset meeting the principle of closest distance as a fourth clustering center.

In this embodiment, the first cluster center refers to N1 departure place objects randomly selected according to the number of a plurality of departure places under a set time length, and the second cluster center refers to N2 destination objects randomly selected according to the number of a plurality of destinations under the set time length.

In this embodiment, each departure point in the first subset based on the first cluster centers is at a smaller distance from the corresponding first cluster center than from the other first cluster centers.

In this embodiment, if the first cluster center is a point satisfying the closest principle, that is, the sum of the distances between each departure point in the first subset of the first cluster centers and the first cluster center is smaller than the sum of the distances between any departure point except the first cluster center and all other departure points and the first cluster center, the first cluster center is regarded as the third cluster center.

In this embodiment, if the first cluster center is not a point satisfying the closest principle, that is, the sum of distances between each departure point in the first subset of the first cluster centers and the first cluster center is not smaller than the sum of distances between any departure point except the first cluster center and all other departure points and the first cluster center, the departure point with the smallest sum of distances between all other departure points and the first cluster center is taken as the third cluster center.

In this embodiment, each destination in the second subset based on each second hub is at a smaller distance from the corresponding second hub than from the other second hubs.

In this embodiment, the second cluster center is considered a fourth cluster center if the second cluster center is the point satisfying the closest principle of distance from the second cluster center, i.e. the sum of the distances of each destination in the second subset of the second cluster center from the second cluster center is smaller than the sum of the distances of any destination other than the second cluster center from all other destinations and the second cluster center.

In this embodiment, if the second cluster center is not a point satisfying the closest principle, that is, the sum of the distances from the second cluster center to each destination in the second subset of the second cluster centers is not smaller than the sum of the distances from any destination other than the second cluster center to all other destinations and the second cluster center, the destination where the sum of the distances from all other destinations and the first cluster center is smallest is taken as the fourth cluster center.

The beneficial effects of the technical scheme are as follows: the first clustering set based on the departure place and the second clustering set based on the destination are generated by calculating the distances between the departure place and the destination under a plurality of transportation routes, so that basis can be provided for building a diversified supply chain network and a data platform, and information sharing and communication are promoted.

Example 3:

the embodiment of the invention provides a data management method based on multi-mode intermodal, which determines a normal historical data set and an abnormal historical data set based on a first clustering set and a second clustering set, and comprises the following steps:

counting a second historical data set based on N1 third cluster centers in the first cluster set and N2 fourth cluster centers in the second cluster set;

analyzing the second historical data set, and determining a transportation capacity constraint condition of each third cluster center, and a time constraint condition, a distance constraint condition and a transportation mode constraint condition corresponding to each third cluster center to each fourth cluster center;

determining the first decision variable of each constraint condition asWherein, the method comprises the steps of, wherein,the method comprises the steps of carrying out a first treatment on the surface of the Determining the second decision variable of each constraint condition asWherein->Wherein N1>1 and N2>1, a step of; wherein, the transportation capability constraint condition of each third class center is:

wherein k represents the transport mode selection of each third cluster center to the fourth cluster center j,/I>Representing the number of modes of transportation, +.>Representing the cargo demand of the fourth cluster center j, < +.>Representing the maximum transport capacity of each third cluster center under all transport modes from reaching all fourth cluster centers; / >Representing the maximum transportation of each third cluster center to the fourth cluster center j using transportation means kCapacity of transportation;

the time constraint conditions and the distance constraint conditions corresponding to the third cluster center to all the fourth cluster centers are respectively as follows:

wherein, the->Representing the transport time using transport means k from the third cluster center i to the fourth cluster center j, +.>Weights representing path optimization from the third cluster center i to the fourth cluster center j, +.>Limited transportation time, + representing the use of transportation means k from third cluster center i to fourth cluster center j>Representing the transportation distance from the third cluster center i to the fourth cluster center j using transportation means k, #>Representing a limited transportation distance from the third cluster center i to the fourth cluster center j using the transportation means k; />Representing the optimization time for path optimization from the third cluster center i to the fourth cluster center j; />Representing an optimized distance for path optimization from the third cluster center i to the fourth cluster center j, wherein +.>The method comprises the steps of carrying out a first treatment on the surface of the Simultaneously meeting transportation capacity constraint conditions, time constraint conditions and distance constraint strips in the second historical data setHistorical data related to the corresponding third class center of the piece and the transportation mode constraint condition are classified into a normal data set, and the rest data which do not meet the constraint condition at the same time are classified into an abnormal data set.

In this embodiment, the second historical data set includes a transportation route, a transportation time, a transportation mode, a transportation cost, and a cargo transportation amount corresponding to the third cluster center under the transportation route from each third cluster center to each fourth cluster center, and a cargo demand corresponding to each fourth cluster center under the transportation route.

In this embodiment, the cargo traffic corresponding to the third class center under the transportation route includes the sum of the cargo traffic corresponding to the third class center and the cargo traffic of all departure places except the third class center in the first subset.

In this embodiment, the respective corresponding cargo demands of each fourth cluster center under the transportation route include a sum of the cargo demands of the fourth cluster center and the cargo demands of all destinations in the second subset except the fourth cluster center.

In this embodiment, the transportation capability constraint of each third cluster center means that the cargo transportation amount of all transportation modes of each third cluster center does not exceed the sum of all cargo demands corresponding to all fourth cluster centers.

In this embodiment, the time constraint means that the transportation time of all transportation modes from each third cluster center to all fourth cluster centers does not exceed the limited transportation time of all transportation modes from each third cluster center to all fourth cluster centers.

In this embodiment, the distance constraint refers to that the transportation distance of all transportation manners from each third cluster center to all fourth cluster centers does not exceed the limited transportation distance of all transportation manners from each third cluster center to all fourth cluster centers.

In this embodiment of the present invention, the process is performed,for each third cluster center to each fourth cluster centerTransportation means constraints, wherein->Indicating that only one transportation mode can be selected from each third cluster center i to the fourth cluster center j.

In this embodiment, if any one or more of the transportation capability constraint, the time constraint, the distance constraint, and the transportation mode constraint is not satisfied, the abnormal data is classified as an abnormal data set.

The beneficial effects of the technical scheme are as follows: and determining a normal data set and an abnormal data set through the transportation capacity constraint condition, the time constraint condition, the distance constraint condition and the transportation mode constraint condition, and providing a data basis for determining an optimal path, reducing transportation distance and time and reducing transportation cost.

Example 4:

the embodiment of the invention provides a data management method based on multi-mode intermodal, which comprises the following steps:

The minimized transportation cost objective function Y1 and the minimized transportation time objective function Y2 are determined as follows:

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing arrival at the ij0 th destination within the cluster range belonging to the fourth cluster center j from the third cluster center i and using transport means +.>Is>Represents the ij 0-th destination within the cluster range from the third cluster center i to the fourth cluster center j and uses transportation means +.>Min represents the minimum value symbolNumber (x).

In this embodiment, the minimized cost objective function refers to the minimized sum of transportation costs, transit costs, insurance costs, and additional costs used by different transportation sub-segments under a transportation route within a cluster range from the third cluster center to the fourth cluster center.

In this embodiment, the minimized transit time objective function refers to the shortest sum of loading, transporting, unloading, and transit times used by different transportation sub-segments under a transportation route within a cluster range from the third cluster center to the fourth cluster center.

The beneficial effects of the technical scheme are as follows: the transportation cost can be effectively controlled and reduced by determining the minimum transportation cost objective function and the minimum transportation time objective function, the utilization rate of transportation capacity resources is improved, the transportation time is shortened, and the transportation efficiency is improved.

Example 5:

the embodiment of the invention provides a data management method based on multi-mode intermodal, which determines an optimal path according to a normal historical data set, an abnormal historical data set, a minimized cost objective function and a minimized transportation time objective function, and comprises the following steps:

establishing a transportation model for a plurality of departure places, a plurality of destinations, a third cluster center and a fourth cluster center under a historical transportation line in a normal data set through a GIS (geographic information system), and simulating a multi-type intermodal transportation process through an ant colony;

determining a plurality of middle points from the third cluster center to the fourth cluster center as middle nodes;

determining a two-dimensional array according to the third cluster center and the fourth cluster centerRepresenting the number of path selections from each third cluster center to each fourth cluster center;

determining a two-dimensional array according to a plurality of middle points corresponding to the third cluster center and the fourth cluster centerWherein->Representing the number of intermediate nodes corresponding to the arrival of the third cluster center i at the fourth cluster center j, na being the maximum number of intermediate nodes corresponding to the arrival of the third cluster center i at the fourth cluster center j,/being the maximum number of intermediate nodes corresponding to the arrival of the third cluster center i at the third cluster center j,/being the maximum number of intermediate nodes corresponding to the arrival of the >Representing the number of the intermediate node where the current node is located;

determining an arrayWherein->Indicating that the node has been accessed and,representing nodes that have not been accessed;

initializing commandThe number of paths from the third cluster center i to the fourth cluster center j is 1;

path number calculation using depth-first search, starting from third class center, satisfiesUnder the condition of that each node is accessed in turn, and the accessed intermediate node is marked as +.>Updating the array after reaching the fourth cluster center>The method comprises the steps of carrying out a first treatment on the surface of the When (when)When (I)>Representing the number of path selections from the third cluster center i to the fourth cluster center j and assigning a corresponding +.>The method comprises the steps of carrying out a first treatment on the surface of the The number of ants placed in each third class center isWherein m is an integer value determined from the total number of paths from all third cluster centers to all fourth cluster centers, and the number of ants in the path from each third cluster center to each fourth cluster center is limited to +.>The method comprises the steps of carrying out a first treatment on the surface of the Initializing the concentration of pheromones among the third cluster center, the intermediate node and the fourth cluster center to be 1;

constructing a mixed integer linear model according to the first decision variable, the second decision variable, the transportation capacity constraint condition of each third cluster center, the time constraint condition, the distance constraint condition and the minimum transportation cost objective function corresponding to each third cluster center to all fourth cluster centers;

According to the mixed integer linear model, determining the violation degree of the transportation capability of each third class center, the violation degree of the time from each third class center to each fourth cluster center, the violation degree of the distance, the first contribution degree of the minimized transportation cost objective function corresponding to the first decision variable and the second contribution degree of the minimized transportation cost objective function corresponding to the second decision variable;

taking the transport capacity violation degree, the time violation degree, the distance violation degree, the first contribution degree and the second contribution degree as heuristic information and carrying out normalization processing;

determining a next node according to the current node and the pheromone concentration among the nodes and heuristic information after normalization processing;

updating the pheromone concentration between the corresponding current node and the determined next node according to the crawling time after the next node is selected;

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing the current node when local pheromone is updated +.>And the determined next node->Pheromone concentration between->Indicating the pheromone exertion rate->Representing the constant pheromone increment,>representing the ant at the current node->And the determined next node->The corresponding crawling time; />Representing the current node +.>Pheromone concentration of (2);

Setting the incremental range of pheromone between the current node and the determined next node asIf (if)Then determine +.>Then determine +.>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing a minimum value of the corresponding pheromone increment range; />Representing the maximum value of the corresponding pheromone increment range;

after each iteration is finished, the global pheromone concentration is updated according to the volatilization rate of the pheromone and the pheromone left by ants;

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing the concentration of pheromone between the third cluster center i and the fourth cluster center j,/->Representing the shortest crawling time corresponding from the third cluster center i to the fourth cluster center j, +.>Representation->Crawling time used by ants from third cluster center i to fourth cluster center j only on average, +.>Representing a balance factor; />Representing the updated pheromone concentration;

wherein, all ants reach the fourth clustering center and are regarded as the end of one iteration;

when the maximum iteration times are reached, determining a path with highest pheromone concentration between each third cluster center i and each fourth cluster center j as a corresponding first transportation path according to the pheromone concentration distribution of the last iteration;

determining a transportation abnormality reason according to the transportation route, transportation time, transportation mode and transportation cost from each third cluster center to each fourth cluster center in the abnormal data set;

And according to the reasons of the abnormal transportation and the actual transportation demands of the corresponding transportation routes in the abnormal data set, adjusting the corresponding first transportation paths to obtain the optimal paths.

In this embodiment, the reasons for abnormal transportation include traffic interruption caused by bad weather, cancellation or delay of flights, traffic accidents, road construction, traffic jam caused by insufficient road capacity, transportation equipment failure, supply chain link failure, emergency, etc.

In this embodiment, in determining the next node, the filter value is calculated based on the following formulaAnd locking the node corresponding to the maximum value from the screening value as the next node:

wherein,representing the current pheromone concentration corresponding to the third class center; />Representing the average pheromone concentration under the historical connecting line corresponding to the third class center; />Representing the maximum pheromone concentration under the historical connecting line corresponding to the third class center; />Representing the concentration of pheromones between nodes corresponding to the third class center; />Representing the violation degree of the transportation capacity after normalization treatment; />Representing the time violation degree after normalization processing; />Representing the distance violation degree after normalization processing; />Representing the first contribution degree after normalization processing; / >Representing the second contribution after normalization.

The beneficial effects of the technical scheme are as follows: and simulating paths from the third cluster center to the fourth cluster center through an ant colony algorithm, then determining the next node, analyzing the pheromone of the next node, finally screening to obtain a first transportation path, and further adjusting the first transportation path through determining the reason of abnormal transportation to obtain an optimal path.

Example 6:

the embodiment of the invention provides a data management method based on multi-mode intermodal, which establishes a diversified supply chain network and a data platform based on an optimal path and further comprises the following steps:

setting a network node based on the departure place, the destination and the neutral point in all the optimal paths;

and formulating unified standards and interface specifications for each network node, and constructing and obtaining a diversified supply chain network and a data platform.

In this embodiment, the neutral point is the point of transition in the transport mode during multi-modal transport at the optimal path.

In this embodiment, the diversified supply chain network is composed of the origin, destination and intermediate point corresponding network nodes of all the optimal paths from each third cluster center to each fourth cluster center.

The beneficial effects of the technical scheme are as follows: the visualization degree of the supply chain can be improved by constructing a diversified supply chain network and a data platform, the information asymmetry and risk are reduced, and the cargo safety and quality management level are improved.

Example 7:

the embodiment of the invention provides a data management method based on multi-mode intermodal, which comprises the following steps:

real-time tracking and monitoring all transportation processes between the first cluster set and the second cluster set based on the diversified supply chain network and the data platform;

and establishing a corresponding emergency mechanism according to the reasons of the transportation abnormality, and timely responding and processing when the transportation process tracked and monitored in real time is abnormal.

In this embodiment, all transportation processes between the first cluster set and the second cluster set are tracked and monitored in real time, and when the transportation is abnormal, abnormal information and corresponding reasons of the transportation abnormality are obtained at the first time, and the abnormal information and the corresponding reasons of the transportation abnormality are responded and processed in time.

In this embodiment, the emergency mechanism includes real-time monitoring of weather conditions and maintaining close contact with the weather department, real-time monitoring of optimal path conditions and planning of alternate routes, periodic inspection of maintenance and repair transportation equipment, preparation of alternate transportation equipment, reinforcement of cargo packaging and handling operations management, and the like.

In this embodiment, responding to and handling in time when an anomaly occurs includes switching alternate routes, adjusting flights, using alternate transport equipment, and the like.

The beneficial effects of the technical scheme are as follows: real-time cargo transportation information and track tracking are provided, and cargo whole-course traceability is realized, so that abnormal transportation conditions can be rapidly handled, and the risk of transportation interruption is effectively reduced.

Example 8:

the embodiment of the invention provides a data management method based on multi-mode intermodal, which comprises the following steps:

when (when)When the value is 0, the third cluster center i to the fourth cluster center j do not use the path of the transportation mode k, and the third cluster center i is +.>When the value is 1, the paths from the third cluster center i to the fourth cluster center j by using the transportation mode k are shown;

when (when)When the value is 0, the third cluster center i to the fourth cluster center j do not perform path optimization according to the transportation cost, and are in the presence of->And when the value is 1, the third cluster center i to the fourth cluster center j are used for carrying out path optimization according to the transportation cost.

In this embodiment of the present invention, the process is performed,as a first decision variable, for example: when i=i1, j=j1, k=k1, +.>When the value is 0, the third cluster center i1 to the fourth cluster center j1 do not use the path of the transportation mode k1,when the value is 1, the paths from the third cluster center i1 to the fourth cluster center j1 using the transportation method k1 are indicated.

In this embodiment of the present invention, the process is performed,as a second decision variable, for example: when i=i1, j=j1, < ->When the value is 0, the third cluster center i1 to the fourth cluster center are representedj1 does not carry out route optimization according to transportation cost, < > and so on>When the value is 1, the third cluster center i1 to the fourth cluster center j1 are used for carrying out path optimization according to the transportation cost.

The beneficial effects of the technical scheme are as follows: the scientificity and the accuracy of the optimal path can be improved through the first decision variable and the second decision variable, and the allocation and the utilization of transportation resources are optimized.

Example 9:

based on embodiment 6, a unified standard and interface specification is formulated for each network node, including:

acquiring a historical transportation mode associated with each network node and a transportation uploading end set based on the historical transportation mode;

obtaining port types of the transport uploading end set to obtain a port list of a corresponding network node, wherein the port list comprises uploading port types, uploading quantity based on different uploading port types and information criticality of each uploading information under different uploading port types;

carrying out port analysis on the port type, determining an active port and a passive port, and extracting a first communication standard of the active port and a corresponding network node and a second communication standard of the passive port and the corresponding network node;

Determining a key port in the port list, and extracting a first standard format which is most similar to the key port from a standardized database by combining the communication structure of the key port;

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing a key value corresponding to the port type;representing corresponding port types and network nodesIs used for the adaptive weight of the (a); />Representing the uploading number of the corresponding port type;representing a total number of uploads based on the port list; />Information criticality of the j01 th uploading information of the corresponding port type is represented; />Representing the total criticality of all uploaded information based on the port list;

taking the port corresponding to the port type with the criticality larger than the preset degree as a critical port;

respectively carrying out similar comparison on the remaining ports in the port list and a first standard format, and determining to obtain a change coefficient of each remaining port;

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing a transmission structure corresponding to the remaining ports; />A standard structure representing a first standardized format; />Representing similarity symbols; />Representing a modification coefficient;

if the change coefficient is smaller than the preset coefficient, adding a first format conversion window to the residual port, and carrying out format change on the port transmission structure of the corresponding residual port according to a first standard format;

Otherwise, extracting a second standard format with the highest similarity with the communication standard of the corresponding residual port from the standardized database, adding a second format conversion window to the residual port, and carrying out format change on the port transmission structure of the corresponding residual port according to the second standard format;

meanwhile, according to the node attribute of the corresponding network node, calling a communication conflict rule from an attribute-conflict database and randomly screening m01 conflict detection groups matched with the corresponding network node from a historical communication database;

performing conflict detection on each conflict detection group based on the communication conflict rule, and determining a conflict type and the number of conflicts in each conflict detection group;

and setting candidate standard formats to the corresponding network nodes according to the conflict types and the conflict quantity, and when a format changing conflict result exists at the same time, carrying out format changing on the conflict result of the communication interface to be converted according to the candidate standard formats based on a candidate window set in the network nodes.

In this embodiment, a conversion window is added to the port to ensure direct conversion from the port, avoid excessive workload of the node, and ensure standard standardization of the node.

In this embodiment, the node attribute refers to a supervision function performed by the corresponding node in the transportation process, and the smaller the supervision function is, the less severe the corresponding acquired communication conflict rule is, that is, the less constraint conditions of the conflict are.

In this embodiment, the history communication database includes a plurality of cases where data are transmitted from different uploading ends at the same time, and in this process, there may be too few standardized interfaces 1 deployed by the node to enable communication reception of a plurality of data that are uploaded at the same time and correspond to the same format, and at this time, candidate standard formats need to be set to alleviate the existing collision situation.

In this embodiment, the conflict detection is to determine the conflict type and the conflict number, for example, under the condition of time 1, there are 4 types 1 of data to be uploaded, but at this time, the number of types 1 that can be received by the network node is 3, then there is an extra type 1 regarded as the conflict type, at this time, the format can be converted from the candidate window to implement uploading, and the candidate standard format is to construct a window that can meet the requirement as far as possible for the existing conflict type and the conflict number, so as to implement effective communication and effective standard standardization.

In this embodiment, the historical transportation means includes: shipping, air, land, etc., and the manner of communicating with the nodes may be different in different manners, and the communication terminals involved are also different, and further the port output communication formats corresponding to the different communication terminals are also different, that is, the uploading port types are different.

In this embodiment, the port resolution is to determine whether the port communicates with the network node actively, or passively, if it communicates actively, then it is an active port, or if it communicates passively, then it is a passive port.

In this embodiment, the first communication standard and the second communication standard refer to communication standards between the port and the node when in a connected state, and the communication standards are communication structures.

In this embodiment, the standard database includes several different communication standards, that is, various standard formats, and is mainly a standardized database set for the node.

In this embodiment, the preset degree is preset, and the value is 0.1.

In this embodiment, the preset coefficient is preset and takes a value of 0.7.

The beneficial effects of the technical scheme are as follows: starting from the port type, the port number and the information criticality of the uploading end, preliminarily determining the critical ports, further realizing conversion by extracting standard formats from a database, reducing the receiving complexity of the network node as much as possible by determining the change coefficient in order to ensure the interface standardization of the network node, and finally setting a candidate window by performing conflict detection, thereby effectively ensuring the network node to complete connection of different communication types and realizing effective supervision on the transportation process.

It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.

Claims (8)

1. A method of managing data based on multi-modal intermodal, comprising:

step 1: acquiring a first historical data set in the multi-joint transportation process, wherein the first historical data set comprises a transportation route, transportation time, transportation mode and transportation cost of each time of historical transportation under a set time length, departure places under the transportation route, corresponding cargo transportation amount, a plurality of destinations under the transportation route and corresponding cargo demand respectively;

step 2: performing cluster analysis on the distance between the departure place and the destination of the historical transportation under the set time length in the first historical data set to generate a first cluster set based on the departure place and a second cluster set based on the destination;

step 3: determining a normal historical data set, an abnormal historical data set, a minimized cost objective function and a minimized transportation time objective function based on the first clustering set and the second clustering set;

Step 4: determining an optimal path according to the normal historical data set, the abnormal historical data set, the minimized cost objective function and the minimized transportation time objective function;

step 5: and establishing a diversified supply chain network and a data platform based on the optimal path to realize data sharing.

2. The multi-modal based data management method of claim 1, wherein generating a first set of clusters based on departure locations and a second set of clusters based on destination locations includes:

randomly selecting N1 objects from a plurality of departure places under the set time length in a first historical data set as a first clustering center, and randomly selecting N2 objects from a plurality of destinations as a second clustering center;

calculating the distance between each first clustering center and the rest departure place in the first historical data set respectively, and determining a first subset based on each first clustering center;

judging whether a first cluster center based on the first subset needs to be replaced or not;

if not, regarding the corresponding first cluster center as a third cluster center;

otherwise, taking the point of the first subset meeting the nearest principle as a third class center;

Calculating the distance between each second aggregation center and the rest destination in the first historical data set, and determining a second subset based on each second aggregation center;

determining whether a second hub based on the second subset needs replacement;

if not, regarding the corresponding second clustering center as a fourth clustering center;

otherwise, taking the points of the second subset meeting the principle of closest distance as a fourth clustering center.

3. The multi-modal based data management method of claim 1, wherein determining the normal historical data set and the abnormal historical data set based on the first cluster set and the second cluster set includes:

counting a second historical data set based on N1 third cluster centers in the first cluster set and N2 fourth cluster centers in the second cluster set, wherein N1>1 and N2>1;

analyzing the second historical data set, and determining a transportation capacity constraint condition of each third cluster center, and a time constraint condition, a distance constraint condition and a transportation mode constraint condition corresponding to each third cluster center to each fourth cluster center;

determining the first decision variable of each constraint condition as Wherein, the method comprises the steps of, wherein,the method comprises the steps of carrying out a first treatment on the surface of the Determining the second decision variable of each constraint condition asWherein->The method comprises the steps of carrying out a first treatment on the surface of the Wherein, the transportation capability constraint condition of each third class center is:

wherein k represents the transport mode selection of each third cluster center to the fourth cluster center j,/I>Representing the number of modes of transportation, +.>Representing the cargo demand of the fourth cluster center j, < +.>Representing the maximum transport capacity of each third cluster center under all transport modes from reaching all fourth cluster centers; />Representing the maximum transport capacity of each third cluster center to reach the fourth cluster center j using the transport mode k;

the time constraint conditions and the distance constraint conditions corresponding to the third cluster center to all the fourth cluster centers are respectively as follows:

wherein, the->Representing the transport time using transport means k from the third cluster center i to the fourth cluster center j, +.>Weights representing path optimization from the third cluster center i to the fourth cluster center j, +.>Limited transportation time, + representing the use of transportation means k from third cluster center i to fourth cluster center j>Representing the transportation distance from the third cluster center i to the fourth cluster center j using transportation means k, #>Representing a limited transportation distance from the third cluster center i to the fourth cluster center j using the transportation means k; / >Representing the optimization time for path optimization from the third cluster center i to the fourth cluster center j; />Representing an optimized distance for path optimization from the third cluster center i to the fourth cluster center j, wherein +.>The method comprises the steps of carrying out a first treatment on the surface of the And classifying the historical data related to the corresponding third class center in the second historical data set, which simultaneously satisfies the transportation capacity constraint condition, the time constraint condition, the distance constraint condition and the transportation mode constraint condition, into a normal data set, and classifying the rest data which does not simultaneously satisfy the constraint condition into an abnormal data set.

4. The multi-modal based data management method of claim 3, further comprising:

the minimized transportation cost objective function Y1 and the minimized transportation time objective function Y2 are determined as follows:

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing arrival at the ij0 th destination within the cluster range belonging to the fourth cluster center j from the third cluster center i and using transport means +.>Is>Represents the ij 0-th destination within the cluster range from the third cluster center i to the fourth cluster center j and uses transportation means +.>Min represents the minimum value symbol.

5. The multi-modal based data management method of claim 1, wherein determining the optimal path based on the normal historical data set, the abnormal historical data set, and the minimized cost objective function, the minimized transit time objective function includes:

Establishing a transportation model for a plurality of departure places, a plurality of destinations, a third cluster center and a fourth cluster center under a historical transportation line in a normal data set through a GIS (geographic information system), and simulating a multi-type intermodal transportation process through an ant colony;

determining a plurality of middle points from the third cluster center to the fourth cluster center as middle nodes;

determining a two-dimensional array according to the third cluster center and the fourth cluster centerRepresenting the number of path selections from each third cluster center to each fourth cluster center;

determining a two-dimensional array according to a plurality of middle points corresponding to the third cluster center and the fourth cluster centerWherein->Representing the number of intermediate nodes corresponding to the arrival of the third cluster center i at the fourth cluster center j, na being the maximum number of intermediate nodes corresponding to the arrival of the third cluster center i at the fourth cluster center j,/being the maximum number of intermediate nodes corresponding to the arrival of the third cluster center i at the third cluster center j,/being the maximum number of intermediate nodes corresponding to the arrival of the>Representing the number of the intermediate node where the current node is located;

determining an arrayWherein->Indicating that the node has been accessed and,representing nodes that have not been accessed;

initializing commandThe number of paths from the third cluster center i to the fourth cluster center j is 1;

Path number calculation using depth-first search, starting from third class center, satisfiesUnder the condition of accessing each node in turn, and marking the accessed intermediate nodes asUpdating the array after reaching the fourth cluster center>The method comprises the steps of carrying out a first treatment on the surface of the When (when)When (I)>Representing the number of path selections from the third cluster center i to the fourth cluster center j and assigning a corresponding +.>The method comprises the steps of carrying out a first treatment on the surface of the The number of ants placed in each third class center isWherein m is an integer value determined from the total number of paths from all third cluster centers to all fourth cluster centers, and the number of ants in the path from each third cluster center to each fourth cluster center is limited to +.>The method comprises the steps of carrying out a first treatment on the surface of the Initializing the concentration of pheromones among the third cluster center, the intermediate node and the fourth cluster center to be 1;

constructing a mixed integer linear model according to the first decision variable, the second decision variable, the transportation capacity constraint condition of each third cluster center, the time constraint condition, the distance constraint condition and the minimum transportation cost objective function corresponding to each third cluster center to all fourth cluster centers;

according to the mixed integer linear model, determining the violation degree of the transportation capability of each third class center, the violation degree of the time from each third class center to each fourth cluster center, the violation degree of the distance, the first contribution degree of the minimized transportation cost objective function corresponding to the first decision variable and the second contribution degree of the minimized transportation cost objective function corresponding to the second decision variable;

Taking the transport capacity violation degree, the time violation degree, the distance violation degree, the first contribution degree and the second contribution degree as heuristic information and carrying out normalization processing;

determining a next node according to the current node and the pheromone concentration among the nodes and heuristic information after normalization processing;

updating the pheromone concentration between the corresponding current node and the determined next node according to the crawling time after the next node is selected;

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing current node at local pheromone updateAnd the determined next node->Pheromone concentration between->Indicating the pheromone exertion rate->Representing the constant pheromone increment,>representing the ant at the current node->And the determined next node->The corresponding crawling time; />Representing the current node +.>Pheromone concentration of (2);

setting the incremental range of pheromone between the current node and the determined next node asIf (if)Then determine +.>Then determine +.>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing a minimum value of the corresponding pheromone increment range; />Representing the maximum value of the corresponding pheromone increment range;

after each iteration is finished, the global pheromone concentration is updated according to the volatilization rate of the pheromone and the pheromone left by ants;

The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing the concentration of pheromone between the third cluster center i and the fourth cluster center j,/->Representing the shortest crawling time corresponding from the third cluster center i to the fourth cluster center j, +.>Representation->Crawling time used by ants from third cluster center i to fourth cluster center j only on average, +.>Representing a balance factor; />Representing the updated pheromone concentration;

wherein, all ants reach the fourth clustering center and are regarded as the end of one iteration;

when the maximum iteration times are reached, determining a path with highest pheromone concentration between each third cluster center i and each fourth cluster center j as a corresponding first transportation path according to the pheromone concentration distribution of the last iteration;

determining a transportation abnormality reason according to the transportation route, transportation time, transportation mode and transportation cost from each third cluster center to each fourth cluster center in the abnormal data set;

and according to the reasons of the abnormal transportation and the actual transportation demands of the corresponding transportation routes in the abnormal data set, adjusting the corresponding first transportation paths to obtain the optimal paths.

6. The multi-modal based data management method of claim 1, wherein the creating of the diverse supply chain network and data platform based on the optimal path includes:

Setting a departure place, a destination and a neutral point based on all the optimal paths as network nodes;

and formulating unified standards and interface specifications for each network node, and constructing and obtaining a diversified supply chain network and a data platform.

7. The multi-modal based data management method of claim 1, further comprising:

real-time tracking and monitoring all transportation processes between the first cluster set and the second cluster set based on the diversified supply chain network and the data platform;

and establishing a corresponding emergency mechanism according to the reasons of the transportation abnormality, and timely responding and processing when the transportation process tracked and monitored in real time is abnormal.

8. The method for managing data based on multi-modal transportation as set forth in claim 3, wherein,

when (when)When the value is 0, the third cluster center i to the fourth cluster center j do not use the path of the transportation mode k,when the value is 1, the paths from the third cluster center i to the fourth cluster center j by using the transportation mode k are shown;

when (when)When the value is 0, the third cluster center i to the fourth cluster center j do not perform path optimization according to the transportation cost, and are in the presence of->And when the value is 1, the third cluster center i to the fourth cluster center j are used for carrying out path optimization according to the transportation cost.