Railway station container scheduling optimization method, system, equipment and medium

Abstract

The invention discloses a railway station container scheduling optimization method, system, equipment and medium, and belongs to the technical field of railway station container scheduling, and the method comprises the steps: building a railway container yard container space assignment optimization model based on a container space freedom degree yard information representation method and a container stacking optimization process, stacking the to-be-stacked boxes by using a box stacking algorithm from a box yard to a train to obtain the optimal stacking positions of the to-be-stacked boxes; extracting the to-be-fetched box by using a box lifting algorithm from a train to a box yard to obtain an optimal target box position of the to-be-fetched box; using a heuristic algorithm to solve the railway container yard container quantity balanced distribution model to obtain the optimal division of the container area container quantity; and solving the railway container yard container space assignment optimization model by using a heuristic algorithm in combination with optimal division of container area container quantity to obtain an optimal container space assignment mode.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of railway station container scheduling, and particularly relates to a railway station container scheduling optimization method, system, device and medium. BACKGROUND

[0002] In order to facilitate operation, in the case of relatively abundant space resources of a railway container yard, the yard generally adopts the mode of respectively stacking arrival containers and sending containers. The arrival time of sending containers at the yard has great uncertainty. According to this feature, the yard stacking and picking strategy generally adopts the following two modes: one is to pre-arrange the containers before they are loaded onto the departure train; the other is to sort and store the containers after they arrive at the yard. The pre-arrangement strategy has the advantages of simplifying the stacking work of the sending container yard and saving the yard space. However, the large amount of pre-arrangement work increases the burden of the railway container center station operation. The sorting and storage strategy has the advantage of avoiding a large amount of pre-arrangement work before the departure of the train, but a relatively complex management system and a wide storage space are required during the sorting and storage process.

[0003] The arrival containers are generally arrived at the railway container center station according to the operation plan deadline, which is contrary to the situation of sending containers. According to this feature, a series of work such as arrival container loading, unloading, stacking and picking can be prepared in advance, so that the arrival containers can be picked up by the consignee as soon as possible, avoiding the occupation of the container yard stacking resources.

[0004] In recent years, with the rapid growth of railway container transportation demand, some major international container yards have experienced explosive growth in operation volume. Many railway container centers are limited by the area of the yard, and the traditional mode of respectively stacking arrival containers and sending containers cannot meet the actual operation demand. Therefore, for some railway container centers with limited resources, the mixed stacking mode of arrival containers and sending containers is adopted to improve the utilization rate of the yard space. However, in this stacking mode, there is no buffer zone, and the arrival containers are directly placed in the container area of the yard, and this stacking mode generally adopts a larger stacking height, which leads to an increase in the container turnover rate and a decrease in the work efficiency of the yard equipment. If the mixed stacking mode is adopted, the container yard information collection equipment and intelligent technology equipment need to be fully configured to minimize the problems of container turning and misloading, and fully utilize the advantages of mixed stacking.

[0005] Arrival and sending containers are respectively stacked to effectively reduce the complexity of operation, reduce the rehandling rate and improve the utilization efficiency of equipment, but the utilization rate of the yard is relatively low, and this stacking mode is not suitable for the yard with tight space resources. On the contrary, the arrival and sending containers are mixedly stacked, which is suitable for the container yard with tight space resources, but puts forward higher requirements for the related supporting facilities and equipment of the container yard and the operation management technology.

[0006] Railway container yard intelligent management is a system engineering, which not only includes the stacking strategy of the yard, but also needs the cooperation of other facilities and equipment of the container yard, such as railway gantry, highway gantry, loading and unloading equipment, etc. The intelligent management of the railway container yard is also a closed-loop management system integrating collection, decision-making, execution and feedback, which is jointly built by the above hardware and intelligent management system. Combined with intelligent technology equipment, the vehicle / train feeds back the information such as car number, container number and transportation demand to the management platform after passing through the gantry, and the intelligent loading and unloading equipment executes the operation scheme calculated and decided by the platform, and feeds back the execution result to the management platform.

[0007] The intelligent management of the railway container yard should realize the transformation from manual input to automatic identification by machines and from experience judgment to intelligent decision-making by systems combined with intelligent technology equipment, but there are problems of low efficiency, misloading, missing loading and pressing and overturning caused by manual operation in the intelligent management of the railway container yard. SUMMARY

[0008] The purpose of the present application is to overcome the problems of low efficiency, misloading, missing loading and pressing and overturning caused by manual operation in the intelligent management of the railway container yard, and a railway yard container scheduling optimization method, system, equipment and medium are proposed.

[0009] To achieve the above purpose, the technical scheme is as follows: In a first aspect, the present application provides a railway yard container scheduling optimization method, comprising the following steps: A yard information representation method based on the degree of freedom of the container location and a container stacking optimization process are used to construct a railway container yard location assignment optimization model, the constraint conditions of the railway container yard location assignment optimization model are configured according to the container yard stacking rules and the container stacking form, and the container yard stacking rules and the container stacking form are preset according to the container attributes; The stacking algorithm from the container yard to the train and the unloading algorithm from the train to the container yard are obtained in combination with the railway container yard location assignment optimization model; the stacking algorithm from the container yard to the train is used to stack the to-be-stacked container to obtain the optimal stacking position of the to-be-stacked container, and the unloading algorithm from the train to the container yard is used to unload the to-be-unloaded container to obtain the optimal target location of the to-be-unloaded container; According to the classification of the yard and the optimization of the storage space resources, a container type construction railway container yard container quantity balanced distribution model is built, a heuristic algorithm is used to solve the railway container yard container quantity balanced distribution model to obtain the optimal division of the container area container quantity, and a heuristic algorithm is used to solve the railway container yard container quantity balanced distribution model to obtain the optimal container quantity balanced distribution model.

[0010] Further, the container yard storage rules include: Before the yard floor is full, all containers are free containers, and the stacking operation is performed according to the priority top principle; When the bottom layer of the yard is full, the remaining containers generate a container pressing operation, and the stacking operation is performed according to the order of the increasing number of containers pressed in the yard; The target position of the container and the movement path of the equipment are in the equipment operation area; 20-foot containers and 40-foot containers are stacked in different multiples, and containers cannot be stacked above the upper part of the non-pressing container type; The target position of the container to be stacked or extracted is obtained within the space range of the yard; The container stacking form includes vertical, flat and stepped, and the single container is operated one by one during stacking and extraction.

[0011] Further, the container stacking optimization process includes: In the first stage, when the bottom layer of the yard is not full, there is remaining space on the yard, and containers with the same attribute are stacked according to the priority top principle; In the second stage, when the attribute of the newly entered container is different from that of the containers on the yard, the newly entered container is placed in the space position where the container with the same attribute can be placed; When there is no position of the container with the same attribute for the newly entered container on the yard or the attribute of the newly entered container is different from all the attributes on the yard, a container pressing operation is generated, and the position or position set of the newly entered container is obtained by maximizing the sum of the total degrees of freedom of the containers on the yard; In the third stage, when the yard cannot satisfy one-time pressing of containers, two-time pressing of containers is generated, and the position of the container that can be stacked or extracted is obtained by maximizing the degree of freedom of the container on the yard; If the stacking state of the yard returns to the state of the previous stage during the container extraction operation, the operation of the next stacking or extraction operation is performed according to the previous stage; The degree of freedom of the container position is:

[0012] wherein, x ∈X, y ∈Y,x , y )∈S, S∈X × Y, X represents the set of the positive integer horizontal coordinates of the bottom layer projection coordinates of the container location, Y represents the set of the positive integer vertical coordinates of the bottom layer projection coordinates of the container location, x , y ) is the bottom layer projection coordinates of the container location, z is the stacking height of the container; S1 represents the freedom of the container location when the container location has a container placed and is not pressed by other containers, but the stack is not placed to the highest layer; f = 1; S2 represents the freedom of the container location when the container location has a container placed, but is pressed by other containers, and is not in the stack of the same attribute container, and the stack is not placed to the highest layer, which is k times the freedom of the container location pressed by the container; S3 represents the freedom of the container location when the container location has no container placed, and has a container placed below or has a temporary mark on the location, which is f = 0; S4 represents the freedom of the container location when the container location has a container placed, and is in the stack of the same attribute container, and the stack is placed to the highest layer, which is f = 2; S5 represents the freedom of the container location when the container location and its upper and lower positions have no container placed, which is f = 3; When the adjacent container placed on the container is the same attribute container, k = α , α ∈[0, 1] or k = β , β ∈[0, 1], β < α ; The objective function of the railway container yard location assignment optimization model is:

[0013] Wherein, maxZ is the objective function of the railway container yard location assignment optimization model, representing the sum of the freedoms of all locations in the yard is the highest; e is a 0-1 variable, 1 represents the container is placed from the train to the yard operation, 0 represents the container is picked up from the yard to the train operation, f represents the freedom of the container location, p (y) is a decision variable, representing the output of the container placement or the container pickup location; q ( d ) is the container location set of the container block d ; D is the set of all container blocks,d ∈D; h ( y ) is a 0-1 variable, indicating whether the container is a container of a certain attribute, taking the value 1 if yes, and 0 if no; g ( R ( d ), y ) is the box cell d in the box cell y the serial number of R ( d ), R ( d ) is the stacking order of the box cell d , the stacking order of the box cell y , according to the vertical stacking rule, starting from the smallest element, taking the value g ( R ( d ), y )=1,2,3,…, n ; L (y) is a temporary withholding flag constraint; The constraint conditions of the railway container yard box assignment optimization model are as follows:

[0014] m 2> m 1>0

[0015] Among them, L (y) indicates that the empty box is preferentially coded without a temporary withholding mark, m 1 indicates that the container is coded with a temporary withholding mark, m 2 indicates that the container is coded without a temporary withholding mark; h ( b ) indicates the generalized attribute of the box b , h (p( y) ) indicates the generalized classification attribute of the box; The generalized classification attribute is a set of classification attributes of the box area and the box cell; the box area includes the sending box area, the arrival box area, the international box area, the refrigerated box area, the special box area, the empty box area, the standby box area, the maintenance box area, the cleaning and disinfection box area; each box area is divided into several box cells according to the classification attribute combined with the past cargo flow; the vertical stacking form is adopted when stacking in the box cell.

[0016] Further, the container coding algorithm from the yard to the train includes the following steps: S11, input the container attribute to be coded, and determine the box cell to which the container to be coded belongs; S12, in the found box cell, according to the code box sequence table of the box cell, searching from the minimum element to the maximum element, for the empty box position with the temporary mark encountered in the search, removing the temporary mark until the first real box position is found, marking the real box minimum element on the real box position; S13, according to the code box sequence table of the box cell, searching from the minimum element to the maximum element, finding the first box position available for the code box in the yard, and the box position has no temporary mark, and recording the sum of the total degrees of freedom of each box position on the yard in the new code position state after the to-be-coded box is placed in the box position as the first sum of degrees of freedom, if all the box positions available for the code box have temporary marks, go to S15; S14, according to the code box sequence table of the box cell, searching for the empty box position available for the code box without the temporary mark one by one, and calculating the sum of the degrees of freedom of each box position on the yard after the to-be-coded box is placed in the box position as the second sum of degrees of freedom, if the second sum of degrees of freedom is greater than the first sum of degrees of freedom, the priority position of the code box is replaced by the current box position, otherwise, the original position is still kept as the priority code position, go to S16; S15, if no box position available for the code box without the temporary mark is found in S12, then according to the box position sequence table, all the box positions with the temporary mark are found and the temporary mark on the box position is removed, go to S13; S16, updating the degrees of freedom of the box position of the box cell being operated in the yard; S17, outputting the coordinates of the found box position available for the code, if not found, outputting no information; The box picking algorithm from the train to the container yard includes the following steps: S21, inputting the attributes of the to-be-picked box, finding the box cell to which the to-be-picked box belongs, otherwise, outputting no solution information; S22, in the found box cell, according to the box sequence table of the box cell, searching from the minimum element to the maximum element, finding the first container with the same attributes as the to-be-picked box and completely free without the temporary mark, and setting the temporary mark on the box position for picking, if no container meeting the conditions is found, go to S24; S23, updating the degrees of freedom of the box position of the box cell being operated in the yard; S24, outputting the coordinates of the found box position, if not found, outputting no information.

[0017] Further, the yard is classified into box areas, and the box areas include a sending box area, an arrival box area, an international box area, a customs supervision area, a refrigerated box area, a special box area, an empty box area, a standby box area, a repair box area, a cleaning and disinfection box area; The container type of the storage space resource optimization includes a DX type to-be-unloaded box, a DT type to-be-picked box, a DF type to-be-sent box, and a DY type to-be-transported box; The DX type to be unloaded indicates that the container is still on the train, waiting to be unloaded and stacked in the yard; The DT type to be picked up indicates that the container has been stacked in the yard, waiting for the truck to pick up the container; The DF type to be sent indicates that the to-be-sent container has not been allocated to the yard and is waiting to be transported to the yard for stacking; The DY type to be shipped indicates that the to-be-shipped container has been waiting for loading on the train in the yard.

[0018] Further, the objective function of the railway container yard box quantity balancing allocation model is as follows:

[0019] wherein w1 represents the weight coefficient of the container yard area truck operation box quantity balancing, w2 represents the weight coefficient of the container yard area train operation box quantity balancing, and w3 represents the weight coefficient of the container yard area overall box quantity balancing; represents t the period of time when the container is transported to the container yard and allocated to the box area i the box quantity of the DF type container loaded on the train, represents t the period of time when the container is unloaded from the train to the box area i the box quantity of the DX type container picked up, represents t the box quantity of the DT type container picked up in the period of time, represents t the box quantity of the DY type container loaded on the train in the period of time, represents the box quantity of the DY type container stacked in the box area i loaded on the train in the period of time, t represents the box quantity of the DT type container picked up, represents the box quantity of the DT type container picked up in the box area in the period of time, i represents the box quantity of the DT type container picked up in the box area t in the period of time; The constraint conditions of the railway container yard box quantity balancing allocation model are as follows: t =1,2… T ; K =0,1… T - t t =1,2… T ; i =0,1… B , t =1,2… T ;K =0,1… T - t , t =1,2… T ; i =0,1… B , t =1,2… T ; i =0,1… B , t =1,2… T ; i =0,1… B

[0020] in, express t The containers are transported to the container yard and allocated to the container area during the designated time period. i The capacity of the DF type box express t The containers are transported to the container yard and allocated to the container area during the designated time period. i In t + k The quantity of DF-type containers loaded onto the train during a given time period. B Indicates the number of container areas within the storage yard. T This indicates the time periods contained within a planning period. express t The time period is from unloading from the train to the container area. i The capacity of the DX type box. express t The time period is from unloading from the train to the container area. i In t + k The number of DX-type containers picked up during the specified time period. Indicates that it is stored in the container area. i In China t The number of DT-type containers picked up during the specified time period. This indicates that the container area has been allocated at the start of the planning cycle. i In, and in t The number of DT-type containers picked up during the specified time period. Indicate t The time period is from unloading from the train to the container area. i In t - k The number of DX-type containers picked up during the specified time period. Indicates that it is stored in the container area. iThe box amount of DY type box loaded on train in t The box amount of DY type box loaded on train in Indicates the box amount allocated in the box area i in the planning period, t The box amount of DF type box loaded on train in Indicates the box amount allocated in the box area t in the planning period, i Indicates the total box amount in the box area t in the planning period, k The box amount of DF type box loaded on train in Indicates the total box amount in the box area t in the planning period, i Indicates the total box amount in the box area Indicates the stacking capacity of the box area. i

[0021] Further, the heuristic algorithm comprises the following steps: S31, initializing the planning period parameters, setting the time variable t=1, and inputting the container information; S32, calculating the operation amount of DF type box in each box area in the yard in the t period, distributing it to each box area according to the constraint condition, and calculating the total operation amount of the four types of containers in the yard, to obtain the operation amount sorting of each box area; S33, limiting the operation amount of each box area from low to high to be distributed to each box area in turn, and judging whether to end; S34, calculating the total amount of the four types of containers in each period, calculating and storing the objective function value; S35, judging whether the box area with the largest operation amount has DX type container, if yes, entering S36, otherwise entering S37; S36, judging whether the DX type box amount of the yard in the period is allocated in the box area because the operation amount of the box area is the smallest in the allocation period, if yes, entering S38, otherwise returning to S35; S37, judging whether the box area with the largest operation amount has DX type container in the period, if no, entering S39, otherwise returning to S35; S8, setting t=t-1, judging whether the current period is the starting period in the planning period, calculating the objective function value, and entering S39; S39, comparing the objective function values calculated twice, and taking the minimum value as the objective function of the model.

[0022] In a second aspect, the present application provides a railway station container scheduling optimization system, comprising: ​The railway container yard container assignment optimization model module is used for constructing a railway container yard container assignment optimization model based on a yard information representation method of container freedom and a container stacking optimization process, constraint conditions of the railway container yard container assignment optimization model are configured according to container yard stacking rules and container stacking forms, and the container yard stacking rules and the container stacking forms are preset according to container attributes; The optimal stacking position of the container to be stacked and the optimal target container position of the container to be taken are obtained, a container stacking algorithm from the yard to the train and a container taking algorithm from the train to the yard are obtained by combining the railway container yard container assignment optimization model, the container stacking algorithm from the yard to the train is used to stack the container to be stacked to obtain the optimal stacking position of the container to be stacked, and the container taking algorithm from the train to the yard is used to take the container to be taken to obtain the optimal target container position of the container to be taken. The optimal container assignment mode is obtained, a railway container yard container quantity balanced distribution model is constructed according to yard classification and stacking space resource optimization of container types, a heuristic algorithm is used to solve the railway container yard container quantity balanced distribution model to obtain optimal division of container area container quantity, and the heuristic algorithm is used to solve the railway container yard container assignment optimization model to obtain the optimal container assignment mode in combination with the optimal division of container area container quantity.

[0023] In a third aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the railway station container scheduling optimization method.

[0024] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, wherein the computer program is executed by a processor to implement the railway station container scheduling optimization method.

[0025] Compared with the prior art, the present application has the following beneficial technical effects: The railway station container scheduling optimization method provided by the application can more reasonably utilize the yard space, avoid the confusion and interference between containers with different attributes, ensure the uniform distribution of the container quantity of each yard, avoid the situation that some yards are overcrowded while some yards are idle, help maximize the overall storage capacity of the yard, improve the space utilization, reduce the idle container positions, improve the storage density of the yard, help reduce the number of container rehandling in the yard, reduce the operation cost and improve the operation efficiency, reduce the equipment wear and tear and energy consumption, form a more scientific and reasonable operation process through model optimization, help reduce the waiting time and idle time in the operation process, improve the overall operation efficiency, use the heuristic algorithm suitable for large-scale complex scenarios to solve the model for the mixed stacking mode, reduce the waiting time of the equipment, make the container stacking and extraction operation smooth by avoiding rehandling or reducing the number of rehandling, improve the utilization rate of the equipment, thereby shorten the container dispatching time, avoid the conflict and interference between different operations by optimizing the constraint conditions in the model, ensure the safety and reliability of the operation process, improve the yard utilization, improve the operation efficiency, reduce the operation cost, enhance the safety and reliability, improve the service capacity of the railway container yard and the quality of the container position assignment decision, optimize the operation process, enhance the management decision-making ability, and solve the problems of low efficiency, misloading, missing loading and container rehandling caused by manual operation in the intelligent management of the railway container yard. BRIEF DESCRIPTION OF DRAWINGS

[0026] The drawings described herein are for illustrative purposes only and are not intended to limit the scope of the present disclosure in any way. In addition, the shapes and proportions of the components in the drawings are only illustrative and are used to help understand the present application, and are not specific limitations on the shapes and proportions of the components. In the drawings: Figure 1 The flowchart of the railway station container scheduling optimization method of the present application.

[0027] Figure 2 The structure diagram of the railway station container scheduling optimization system of the present application.

[0028] Figure 3 The electronic device diagram of the railway station container scheduling optimization method of the present application.

[0029] Figure 4The optimization idea diagram for the railway container yard stacking scheme.

[0030] Figure 5 The vertical container stacking sequence diagram.

[0031] Figure 6 The flat container stacking sequence diagram.

[0032] Figure 7 The stepped container stacking sequence diagram.

[0033] Figure 8 The container yard bottom layer planning diagram.

[0034] Figure 9 The container yard container location degree of freedom diagram.

[0035] Figure 10 The railway container yard stacking resource optimization problem idea diagram.

[0036] Figure 11 The railway container yard container location intelligent assignment solution idea. DETAILED DESCRIPTION

[0037] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0038] Embodiment one Reference Figure 1 A railway station container dispatching optimization method, comprising the following steps: A container yard location assignment optimization model is constructed based on a yard information representation method and a container stacking optimization process based on the container yard location degree of freedom, the constraint conditions of the container yard location assignment optimization model are configured according to the container yard stacking rules and the container stacking form, and the container yard stacking rules and the container stacking form are preset according to the container attributes; The container stacking algorithm from the container yard to the train and the container picking algorithm from the train to the container yard are obtained in combination with the container yard location assignment optimization model, the optimal stacking position of the container to be stacked is obtained by using the container stacking algorithm from the container yard to the train for stacking the container to be stacked, and the optimal target container location of the container to be picked is obtained by using the container picking algorithm from the train to the container yard for picking the container to be picked; According to the classification of the yard and the optimization of the storage space resources, a container type construction railway container yard container quantity balanced allocation model is constructed, and a heuristic algorithm is used to solve the railway container yard container quantity balanced allocation model to obtain the optimal division of the container area container quantity. Combined with the optimal division of the container area container quantity, a heuristic algorithm is used to solve the railway container yard container location assignment optimization model to obtain the optimal container location assignment mode.

[0039] The embodiment improves the utilization rate of the yard by optimizing space allocation and reducing the number of container turning, and can more reasonably utilize the yard space by pre-setting the container yard rules and the container stacking form, avoid confusion and interference between containers with different properties, and minimize the number of container turning and container turning by the container location assignment optimization model, thereby reducing the additional time and cost in the operation process. The operation efficiency is improved by shortening the transportation time and optimizing the operation process, the scheduling optimization method can reduce the waiting time of the equipment and improve the utilization rate of the equipment, thereby shortening the container transportation time; by constructing the railway container yard container location assignment optimization model, a more efficient container stacking and container picking algorithm can be obtained, and the operation process is more smooth and efficient. By avoiding operation conflicts and improving emergency response capability, the safety and reliability are enhanced, and by optimizing the constraint conditions in the optimization model, the conflicts and interference between different operations can be avoided, and the safety and reliability of the operation process are ensured; in an emergency, the position of the required container can be quickly and accurately located, and the emergency response speed and efficiency are improved. By accurately balancing the container quantity, the efficient utilization of the yard space is ensured, the idle area is reduced, and the storage capacity is maximized; the container location assignment model further optimizes the specific storage position of the container, reduces the space waste, and improves the overall operation efficiency of the yard; balanced container quantity allocation helps to simplify loading and unloading operations, reduce the frequent movement of containers between different yards, and thereby reduce operation cost and time; reasonable container location assignment reduces the moving distance and waiting time of cranes or forklifts and other equipment, and improves the loading and unloading efficiency; the scheduling optimization method provides scientific data support to help management personnel make more accurate and efficient decisions and optimize resource allocation.

[0040] The intelligent technology equipment is combined to realize the change from manual input to automatic recognition by machines, from manual experience judgment to intelligent decision-making by systems in the yard, and the technical equipment mainly includes electronic tags, readers and other devices related to data information collection and identification; RFID devices related to container positioning in the yard; gate cameras, sensors and other devices related to vehicle / train access information collection, and wireless transmission devices for data interaction in the yard. Through the above hardware equipment, combined with the container yard container location intelligent optimization management method, the intelligent management of the railway container yard is realized, and the problems of low efficiency, misloading, container pressing and container turning caused by manual operation are reduced.

[0041] By analyzing the business characteristics of the container yard, its business can be summarized as five business scenarios: loading trains, unloading trains, loading cars, unloading cars, and intra-yard moving. Further, in combination with the facilities of the yard, tracks, roads, etc., the business scenarios can be summarized into three categories: from truck / train to container yard, container yard to truck / train, and container yard to container yard.

[0042] For truck / train to container yard, when a heavy truck or heavy train enters the container yard through the gate, i.e., when containers are stored in the yard from outside, the gate transmits the information such as license plate number, car number, order number, and container number to the intelligent management platform. For yards with gantry cranes as the main loading and unloading equipment, attention should be paid to balancing the workloads of each gantry crane to fully utilize the storage space and improve the loading and unloading efficiency. At the same time, attention should be paid to the relationship between container properties to optimize the degree of freedom of the container location and reduce the rehandling operation caused by stacking.

[0043] For container yard to truck / train, when an empty truck or empty train enters the container yard to pick up containers, i.e., when containers are loaded and unloaded from the yard to the outside, as the information such as gantry crane operation balance and container property consistency has been considered when storing containers, the main consideration when developing the loading plan should be the maximization of loading and unloading efficiency. Therefore, when containers are loaded and unloaded from the yard to the truck / train, the essential problem becomes the generalized shortest path problem of containers or loading equipment to reduce operation cost and improve efficiency.

[0044] For container yard to container yard, when the container yard needs to move containers within the yard due to inventory or stacking, the problem of balancing the operation of each loading and unloading equipment within the yard also needs to be considered. In addition, when developing the moving plan, the impact of container properties on container locations needs to be considered.

[0045] In summary, the optimization idea for the storage plan of the railway container yard is as shown in Figure 4 .

[0046] The basic rule of the intelligent scheduling optimization method is to clearly define the information of each container and container location. First, define the relevant operation scenarios and location characteristics according to the properties of the containers, and then develop the container yard storage rules to further optimize the container assignment plan. The basic definitions are as follows: Property container, according to the property characteristics of the containers in the container yard, such as the consignee unit, arrival station name, type of goods loaded, size, etc., the internal freight clerk determines that a certain property A is the main property characteristic of the container. Such containers are called property container A.

[0047] Stacking, refers to a three-dimensional location in the container yard where the bottom layer has only one container, and other containers are vertically stacked on top to a certain height.

[0048] Stacking operation, refers to the operation of containers belonging to different stacks being stacked together in the model.

[0049] Free box, refers to the box which is not pressed by other classification attribute box.

[0050] Semi-free box, refers to the box which is pressed by other classification attribute box, but there is the same classification attribute.

[0051] Non-free box, refers to the box which is pressed by other classification attribute box, and there is no same classification attribute.

[0052] Degree of freedom, refers to the degree of freedom of a box in the yard, which will be different due to the characteristics of the box.

[0053] Priority principle, refers to the principle of stacking containers in the container stacking operation process, which is to stack containers belonging to the same attribute on the same stack, and as close as possible to the maximum stacking layer allowed by the yard. When the maximum stacking layer is reached, if there are still containers of the same attribute left, they will be stacked on another stack according to the same principle until all containers of the same attribute are stacked. The final result is that containers of the same attribute are stacked together and occupy the least area on the bottom layer of the yard.

[0054] Single compression box, refers to the stacking of two different attribute boxes in the same column of the stack, double compression box refers to the stacking of three different attribute containers in the same column, and so on.

[0055] Further, the container yard stacking rules include: Before the yard layer is full, all containers are free boxes, and the priority principle is used for stacking operation; When the bottom layer of the yard is full, the remaining containers will be compressed, and the stacking operation will be carried out according to the order of increasing the number of compressed boxes in each stack in the yard; The target position of the box and the moving path of the equipment are in the equipment operation area; 20-foot containers and 40-foot containers are stacked on different levels, and containers cannot be stacked above the non-compression type; The target position of the container to be stacked or extracted is obtained within the space range of the yard; Referring to Figure 5 , Figure 6 , Figure 7 , the container stacking form includes vertical, flat and stepped, and the single container is operated one by one when stacking and extracting. In actual container picking operation, whether the customer needs one or more containers, the demand can be converted into multiple one-time container picking demand, that is, the stacking and extraction are operated one by one, which is regarded as multiple operations of single container.

[0056] When the operation of stacking containers is inevitable, the containers are stacked in the order of the increasing number of stacks in the yard during the operation. That is, when the yard has already generated a stack of one container, it cannot have a free container position to satisfy the stack of one container, and a new stack of containers is generated to increase the number of stacks. The space of the yard must be able to meet the space requirement of the containers to be stacked or extracted in the yard. Therefore, the target position of the containers to be stacked or extracted cannot exceed the space range of the yard.

[0057] Further, the container stacking optimization process includes: In the first stage, when the bottom layer of the yard is not full, there is a remaining position space on the yard, and containers with the same attribute are stacked according to the priority principle; as many containers as possible are stacked on the yard to improve the utilization rate of the yard; In the second stage, when the attribute of the newly entered container is different from that of the container in the yard, the newly entered container is placed in the space position where the container with the same attribute can be placed; When there is no position of the container with the same attribute of the newly entered container in the yard or the attribute of the newly entered container is different from all the attributes of the containers in the yard, the operation of stacking containers is generated; the position or position set where the newly entered container can be placed is obtained by maximizing the sum of the total degrees of freedom of the positions of the containers in the yard; Since the containers with different attributes are stacked in the yard, they occupy the space of the bottom layer of the yard. When the attribute of the newly entered container is different from that of the container in the yard, the newly entered container is placed in the space position where the container with the same attribute can be placed; when there is no position of the container with the same attribute or the attribute of the newly entered container is different from all the attributes of the containers in the yard, the operation of stacking containers is generated. By defining the degrees of freedom of the positions of the containers in the yard and considering the influence of the stacking and extracting operations on the degrees of freedom, the position or position set where the newly entered container can be placed is obtained by maximizing the sum of the total degrees of freedom of the positions of the containers in the yard; In the third stage, when the yard cannot satisfy the stack of one container, the stack of two containers is generated, and the position where the container can be stacked or extracted is obtained by maximizing the degrees of freedom of the positions of the containers in the yard; similar to the second stage, the position where the container can be stacked or extracted is obtained by maximizing the degrees of freedom of the positions of the containers in the yard; If the stacking state of the yard in the extracting operation returns to the state of the previous stage, the operation of stacking or extracting is performed according to the situation of the previous stage; when the stacking state of the yard is returned to a certain stage before the current stage due to the influence of the extracting operation, such as a large number of containers extracted in a batch, the operation is performed according to the converted state stage. Due to the influence of the extracting operation, the stacking state of the containers in the yard in the second stage with the stack of one container is converted to the stacking state of the first stage without the stack of containers, and the operation of stacking or extracting is performed according to the situation of the first stage.

[0058] For a given container yard, the area of ​​the ground floor available for stacking containers is fixed, and the maximum stacking height C of the yard can be determined by factors such as the degree of ground hardening and the type and capacity of loading and unloading machinery. When a container is stacked to a certain height in the yard, the number of stacking layers is represented by a positive integer Z, where Z∈[1,C]. A schematic diagram of the ground floor container space planning in a container yard is shown below. Figure 8 As shown. It can be seen that in reality, storage yards can be regular or irregular. The numerical coordinates in each cell of the diagram represent the projection of that container location onto the bottom plane. Through the division of the bottom-level container locations and the representation of the plane coordinates of each location, for a given storage yard, let X represent the range of positive integer abscissas that the container location can take in the bottom-level projection coordinates, and let Y represent the range of positive integer ordinates that the container location can take in the bottom-level projection coordinates. X×Y represents the Cartesian product of the elements in the two sets X and Y. Then, the bottom-level projection coordinates of a certain container location ( x , y The value can be... x ∈X, y ∈Y, ( x , y )∈S, S∈X×Y, further determined by the container stacking height z This allows us to determine the specific location of containers within the yard and define... p(a) Indicates container a Position function, f Let the degrees of freedom represent the container's position. For any container position within the yard, the following relationship holds: The degrees of freedom for the container's location are:

[0059] in, x ∈X, y ∈Y, ( x , y )∈S, S∈X×Y, where X represents the range of positive integer x-coordinates that the box location can take in the bottom projection coordinates, and Y represents the range of positive integer y-coordinates that the box location can take in the bottom projection coordinates. x , y ) represents the bottom projection coordinates of the box location. z This refers to the stacking height of containers; S1 indicates that a container is placed in the storage location and is not blocked by other containers, but the stack is not at its highest level, giving the container location a degree of freedom. f =1; S2 indicates that a container is placed in a location, but it is pressed down by other containers and is not in a stack of containers with the same attribute, and the stack is not stacked to its highest point. In this case, the degree of freedom of the container location is k times the degree of freedom of the container location pressing down on it. S3 means that the location is empty and its lower part is stacked with containers or the location is marked with a temporary detention mark, and the degree of freedom of the location is f =0; S4 means that the location is stacked with containers and is in the stack of the same attribute, and the stack is stacked to the highest layer, and the degree of freedom of the location is f =2; S5 means that the location and its upper and lower parts are not stacked with containers, and the degree of freedom of the location is f =3; When the adjacent containers stacked on the container are containers of the same attribute, k = α , α ∈[0,1] or k = β , β ∈[0,1], β < α ; In the definition of the degree of freedom, it is defined for each location, and the space definition of the location x , y , z ) is the spatial position of a location in the box area, at this time the classification attribute set of the box area and the box area is called the generalized classification attribute, and h ( a ) represents the generalized attribute of the box a , h (p( a) ) represents the generalized classification attribute of the location. While considering the box partition, in order to avoid the simultaneous occurrence of multiple optimal solutions, the vertical stacking form is adopted when stacking the box area.

[0060] In order to better illustrate the degree of freedom, the degree of freedom is defined based on each box area, that is, the x , y , z ) involved in the definition is a location in a box area. The following illustrates the degree of freedom of some locations. In the six subgraphs in the following figure, A and B represent two different attributes, which are marked on the upper part of the location, indicating that the container has been stacked there and has its own attribute category; The unmarked location means that it is an empty location. According to the above definition of the degree of freedom of the container location, the degree of freedom of each location can be obtained, and the degree of freedom of each location in the figure is marked on the location, and some corresponding scenes are listed, as shown in Figure 9 .

[0061] In the yard operation, in order to ensure the principle of "first in first out", the "temporary detention" mark is set for each box position, and the empty box position without "temporary detention" mark is preferred when loading the container, and if not found, the empty box position with "temporary detention" mark is considered. Under the principle of "first in first out", the degree of freedom of the entire yard is maximized.

[0062] The objective function of the railway container yard box assignment optimization model is:

[0063] Where maxZ is the objective function of the railway container yard box assignment optimization model, representing the sum of the degrees of freedom of all box positions in the container yard is the highest; e is a 0-1 variable, 1 represents the container loading from the train to the yard operation, and 0 represents the container unloading from the yard to the train operation, f is the degree of freedom of the container position, p (y) is a decision variable, representing the output loading or unloading position; q ( d ) is the set of box positions of the container cell d ; D is the set of all container cells, d ∈D; h ( y ) is a 0-1 variable, indicating whether the container is a container with a certain attribute, and if yes, the value is 1, and if not, the value is 0; g ( R ( d ), y ) is the box position of the container cell d About the serial number of y ( R d ), R ( d ) is the loading order of the container cell d , the loading order of the box position y , according to the vertical loading rule, starting from the minimum element, the value is g ( R ( d ), y )=1,2,3,…, n ; L (y) is the temporary detention flag constraint; The constraint conditions of the railway container yard box assignment optimization model are as follows:

[0064] m 2> m 1>0 ​

[0065] in, L (y) indicates that empty boxes without temporary detention markings are given priority when stacking boxes. m 1 indicates that the box is temporarily detained when it is being stored. m 2 indicates that there is no temporary detention mark when the box is stacked; h ( b ) indicates box b The general attributes of h (p( y) ) represents the generalized classification attribute of the container location; The broad classification attributes are a set of classification attributes from two aspects: container area and container sub-area. The container area includes the outbound container area, the inbound container area, the international container area, the refrigerated container area, the special container area, the empty container area, the spare container area, the maintenance container area, and the cleaning and disinfection container area. Each container area is divided into several container sub-areas based on the classification attributes and past cargo flow. Vertical stacking is used when stacking container sub-areas.

[0066] After defining the degrees of freedom for each container location, the stacking or retrieval operations of containers in the yard directly affect the degrees of freedom of the existing containers in the yard. Stacking operations reduce the degrees of freedom of the existing container locations in the yard, while adding degrees of freedom to newly added containers; retrieval operations increase the degrees of freedom of the existing container locations in the yard, and may lead to more instances of empty containers on stack rows. Therefore, the optimization objective of maximizing the overall degrees of freedom of the yard is achieved.

[0067] Based on the railway container space allocation optimization model, we designed container stacking and container retrieval algorithms for two scenarios: “container yard to (train)” and “(train) to container yard”.

[0068] Furthermore, the container stacking algorithm from the container yard to the train includes the following steps: S11. Input the box attributes of the box to be coded and determine the box community to which the box belongs; S12. In the found box cell, search from the smallest element to the largest element according to the box cell code order table. For empty box positions with temporary detention marks encountered in the search, remove the temporary detention marks until the first full box position is found. Mark the full box smallest element on the full box position. S13. According to the container stacking order table of the container area, search from the smallest element to the largest element step by step to find the first container position that can be used for stacking in the yard, and the container position has no temporary detention mark. Record the sum of the total degrees of freedom of each container position in the yard under the new stacking state after the container to be stacked is placed in the container position. Record it as the sum of the first degree of freedom. If all container positions that can be used for stacking have temporary detention marks, go to S15. S14, find the empty box position which can be used for coding box according to the box cell code box order table one by one, and calculate the sum of the freedom degrees of each box position on the yard after the to-be-coded box is placed in the box position, and mark it as the second sum of freedom degrees, if the second sum of freedom degrees is greater than the first sum of freedom degrees, the priority position of the code box is replaced by the current box position, otherwise the original position is still retained as the priority coding position, and turn to S16; S15, if no box position which can be used for coding box without the temporary detention mark is found in S12, find all the box positions with the temporary detention mark according to the box position order table, and remove the temporary detention mark on the box position, and turn to S13; S16, update the box position freedom degree of the box cell which is operated in the yard; S17, output the found box position coordinates, and if no box position is found, output no information; In the actual application of the box picking operation, if the boxes of a certain box cell cannot be completely picked up for a long time, it may cause the long-time stay of the boxes near the minimum element, thereby causing the "large point box" problem. Although the boxes with the same general classification attribute are equivalent, if the sum of the freedom degrees in the yard is to be maximized, the just-coded container will be picked up in the box picking operation, and the picking order is "last in, first out", which causes the previously coded containers to be long-stayed in the yard, thereby violating the "first in, first out" principle. In order to solve this problem, the 0-1 variable "x" is added to the objective function, which is 0 when the box picking operation is performed, that is, the container meeting the conditions can be directly picked up according to the box cell picking order table. e

[0069] The separated storage and box picking algorithm from the train to the yard includes the following steps: S21, input the attributes of the to-be-picked box, find the box cell to which the to-be-picked box belongs, otherwise output no solution information; S22, in the found box cell, search from the minimum element to the maximum element according to the box cell picking order table, find the first container with the same attribute as the to-be-picked box and completely free without the temporary detention mark, and set the temporary detention mark on the box position used for picking up, if no container meeting the conditions is found, turn to S24; S23, update the box position freedom degree of the box cell which is operated in the yard; S24, output the found box position coordinates, and if no box position is found, output no information.

[0070] ​If the location of the container to be picked up has been found in the yard, but the attributes of the container located at the location and the container to be picked up are inconsistent, it indicates that the containers with the same attribute as the container to be picked up are all pressed by containers with other attributes, or there is no container with the attribute in the yard. When the containers with the attribute are pressed by containers with other attributes, the number of containers with other attributes on the container is minimized in the yard space range. In this case, the container will also be picked up. When there is no container with the attribute in the yard, the operation fails directly.

[0071] Further, the yard is classified into container areas, including sending container area, arriving container area, international container area, customs supervision area, refrigerated container area, special container area, empty container area, standby container area, maintenance container area, cleaning and disinfection container area; The container types for optimizing storage space resources include DX type, DT type, DF type and DY type. The DX type indicates that the container is still on the train and is waiting to be unloaded and stored in the yard. The DT type indicates that the container has been stored in the yard and is waiting for a truck to pick it up. The DF type indicates that the container to be sent has not been allocated to the yard and is waiting to be transported to the yard for storage. The DY type indicates that the container to be sent has been waiting for loading on the train in the yard.

[0072] In order to fully utilize the overall efficiency of the yard logistics system, economic and practical loading and unloading machinery is selected. In the railway container yard, the yard can be divided into bulk cargo placement area and scattered cargo placement area, and then the container area, container location and yard road are reasonably divided according to the operation process to minimize the cross interference of operation.

[0073] Empty container storage area, mainly used for stacking containers without loading cargo, after the completion of the box, the empty container yard according to the railway container or the owner of the box is divided into code, before loading from the empty container yard to adjust the empty container according to the loading conditions. The result of the yard planning should be based on the selection of the commission, prediction and container box code, and the box area box management system should realize the automatic calculation of the incoming box which should be stacked into the box. If the calculation result does not meet the conditions of the box, it is prompted which several times need to be stacked. The condition of becoming an incoming box is to meet the box code and the box is not full. If there are multiple conditions, select the box with the least number of times; if there is no box, select other empty times, prefer to select the empty times that are consistent with the properties of the incoming box after the last empty time, and try to select the nearest times from the current incoming box. The principle of assigning the incoming box position mainly includes: 1) first, according to the customer's box position requirements, the planner can manually specify or the system can automatically generate the box position of the incoming box; 2) if not specified, the incoming box can only enter the incoming box times; 3) if the incoming box times are full, an alarm can be given and the condition can be reset to find the position.

[0074] Disassembling and assembling area, according to the different requirements of container transportation, the containers that need to be unpacked in the yard are transported to this area for disassembling operation. Then the goods are stacked in the specified position and waiting for the owner to pick up. Generally, the containers that need to be disassembled include two types: one is the goods transported by container, and the other is the whole container transported goods.

[0075] Residual box area, when the containers are inspected, the residual boxes are stored in the residual box area according to different sizes, and the mechanical operation is recorded after completion.

[0076] Repair area, when the incoming box is inspected and found to be damaged, the box management personnel and repair personnel arrange the incoming box to enter the repair area, and the dispatcher manually specifies the position. After the stacking operation is completed, record the box position, and repair the box in this area.

[0077] Heavy box storage area, when the heavy box enters, record the stacked box position after the operation is completed. Different sizes should be stored separately. When assigning the box position of the incoming heavy box, try to stack the containers according to the same container on the container lifting list. According to the business needs and the requirement of fast and efficient on-site operation, the Xi'an International Port container yard box management should design different properties of the container to be stacked separately.

[0078] Temporary storage area, the temporarily stored containers enter this area and do not perform unpacking operation. Record the position after the mechanical operation is completed, and store different sizes separately.

[0079] When the lower layer container of the same box position has a process time earlier than the departure time of the upper layer container, the situation of pressing the box occurs. For each layer container of the same box position, the total number of containers that constitute the upper layer container is called the number of pressing containers.

[0080] In the mixed mode, the scheduling process of the railway container yard storage space resources includes DX type and DF type, two steps of container allocation yard and assignment of specific container location. Mainly according to the yard classification standard and stacking rules to reasonably determine the number of containers in the yard and assign the container location, balance the work load of each container area, minimize the number of container inversion in a work task or a work cycle. Balance the work load between the container areas of the yard, balance the overall container quantity of the yard, improve the utilization efficiency of storage space resources and equipment resources. Minimize the number of containers in the yard, reduce the number of container inversion due to the pressure of the container, and shorten the waiting time of the container train and the truck.

[0081] In the container area container quantity optimization stage, the container quantity of each container area of the yard is balanced, and the DX type and DF type containers allocated to each container area in the container area container quantity optimization stage are allocated to the specific container location of each container area, so that the number of containers in the yard is minimized. DX type, DF type container in the planning period or in a certain statistical period, thereby reducing the number of container inversion during container lifting or loading operation, and improving the efficiency of yard operation. The optimization idea is as shown in Figure 10 .

[0082] Further, the objective function of the railway container yard container quantity balancing allocation model is as follows:

[0083] Wherein, w1 represents the weight coefficient of the container quantity balance of the container area truck operation of the yard, w2 represents the weight coefficient of the container quantity balance of the train operation of the container area of the yard, w3 represents the weight coefficient of the overall container quantity balance of the container area of the yard; represents t the container quantity of the DX type container allocated to the container area of the container yard in the time period, i the container quantity of the DF type container loaded on the train in the time period, represents t the container quantity of the DX type container unloaded from the train to the container area in the time period, i the container quantity of the DT type container lifted in the time period, represents t the container quantity of the DY type container loaded on the train in the time period, represents t the container quantity of the DY type container loaded on the train in the time period, represents the container quantity of the DY type container stacked in the container area i in the time period, t represents the container quantity of the DY type container loaded on the train in the time period, represents the container quantity of the DT type container lifted, represents the container quantity of the DT type container stacked in the container areai The box quantity of DT type box taken away in the time period t The box quantity of DT type box taken away in the time period The constraint conditions of the railway container yard box quantity balance distribution model are as follows: t =1,2… T ; K =0,1… T - t t =1,2… T ; i =0,1… B , t =1,2… T ; K =0,1… T - t , t =1,2… T ; i =0,1… B , t =1,2… T ; i =0,1… B , t =1,2… T ; i =0,1… B

[0084] wherein, represents t the box quantity of DF type box distributed to the box area in the container yard in the time period, i represents the box quantity of DF type box loaded onto the train in the time period, t represents i the box quantity of DF type box loaded onto the train in the time period, t + k the box quantity of DF type box loaded onto the train in the time period, B represents the number of box area in the yard, T represents the time period contained in a planning period, represents t the box quantity of DX type box unloaded from the train to the box area in the time period, i represents the box quantity of DX type box unloaded from the train to the box area in the time period, tThe time period is from unloading from the train to the container area. i In t + k The number of DX-type containers picked up during the specified time period. This indicates that the container is stored in the container area. i In China t The number of DT-type containers picked up during the specified time period. This indicates that the container area has been allocated at the start of the planning cycle. i In, and in t The number of DT-type containers picked up during the specified time period. Indicate t The time period is from unloading from the train to the container area. i In t - k The number of DX-type containers picked up during the specified time period. This indicates that the container is stored in the container area. i In China t The quantity of DY-type containers loaded onto the train during the specified time period. This indicates that the container area has been allocated at the start of the planning cycle. i In, and in t The quantity of DF-type containers loaded onto the train during a given time period. express t The containers are transported to the container yard and allocated to the container area during the designated time period. i In t - k The quantity of DF-type containers loaded onto the train during a given time period. express t container area at the start of the time period i Total number of boxes in the container. Indicates box area i Storage capacity.

[0085] The overall solution process of the heap space resource scheduling optimization model is as follows: Figure 11 As shown. Based on the objective and constraints of the intelligent allocation optimization model for container yard locations, a heuristic algorithm was designed to solve the model. The algorithm flow is as follows: S31. Initialize the planning cycle parameters, set the time variable t=1, and input the container information; S32. Calculate the workload of DF type containers in each container area of ​​the yard during time period t, allocate them to each container area according to the constraints, calculate the total workload of the four types of containers in the yard, and obtain the workload ranking of each container area. S33. Limit the workload of each container area to be allocated to each container area in ascending order, and determine whether to end the task. S34. Calculate the total quantity of the four types of containers in each time period, and calculate and store the objective function value; S35. Determine if the container area with the largest workload has DX type containers. If so, proceed to S36; otherwise, proceed to S37. S36, judging whether the DX type box volume appearing in the period is allocated in the box area with the minimum operation volume in the allocation period, if yes, turning to S38, otherwise returning to S35; S37, judging whether the box area with the maximum operation volume exists, if no, turning to S39, otherwise returning to S35; S8, setting t=t-1, judging whether the current period is the starting period in the planning period, calculating the objective function value, and turning to S39; S39, comparing the objective function values calculated twice, and taking the minimum value as the objective function of the model.

[0086] The embodiment designs a traversal search algorithm based on the container attributes for the separate stacking mode, improves the service capacity of the railway container yard and the decision quality of the box assignment, designs a heuristic algorithm for solving the model for the mixed stacking mode, and improves the service capacity of the railway container yard and the decision quality of the box assignment.

[0087] Embodiment two Referring to Figure 2 A railway station container scheduling optimization system, comprising: A railway container yard box assignment optimization model construction module is configured to construct a railway container yard box assignment optimization model based on a yard information representation method of box freedom degree and a container stacking optimization process, the constraint conditions of the railway container yard box assignment optimization model are configured according to a container yard stacking rule and a container stacking form, and the container yard stacking rule and the container stacking form are preset according to container attributes; A module for obtaining optimal stacking positions of containers to be stacked and optimal target box positions of containers to be picked up is configured to obtain a container stacking algorithm from a yard to a train and a container picking algorithm from the train to the yard in combination with the railway container yard box assignment optimization model, to obtain the optimal stacking positions of the containers to be stacked by using the container stacking algorithm from the yard to the train, and to obtain the optimal target box positions of the containers to be picked up by using the container picking algorithm from the train to the yard; A module for obtaining an optimal box assignment mode is configured to construct a railway container yard box volume balanced allocation model according to a container type constructed based on yard classification and stacking space resource optimization, to obtain optimal box volume division of a box area by using a heuristic algorithm to solve the railway container yard box volume balanced allocation model, and to obtain the optimal box assignment mode by using the heuristic algorithm to solve the railway container yard box assignment optimization model in combination with the optimal box volume division of the box area.

[0088] The embodiment can ensure that each yard is fully utilized and reduce the waste of idle space by accurately calculating the box volume allocation of each yard; the box location assignment model further optimizes the use of specific box locations, ensuring that the goods are stacked compactly and orderly, maximizing space utilization; balanced box volume allocation helps to reduce the transfer of goods between yards, thereby reducing handling costs and time; reasonable box location assignment can reduce the moving distance of cranes or other handling equipment, improving operational efficiency; automated and intelligent scheduling optimization models can help simplify management processes and reduce human errors; real-time monitoring and data analysis functions can help managers better understand the yard status and make more informed decisions; optimizing space utilization and operational efficiency can reduce energy consumption and labor costs; reducing the risk of damage and loss of goods further reduces operating costs.

[0089] Embodiment three Referring to Figure 3 An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the railway station container scheduling optimization method when executing the computer program.

[0090] Embodiment four A computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the railway station container scheduling optimization method.

[0091] Those skilled in the art will appreciate that embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0092] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The device that implements the functions specified in one or more flows and / or blocks. Figure 1 The device that implements the functions specified in one or more flows and / or blocks.

[0093] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flow Figure 1 one or more flow or block Figure 1 one or more blocks or blocks specified in the flow.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions executed on the computer or other programmable apparatus provide steps for implementing the flow Figure 1 one or more flow or block Figure 1 one or more blocks or blocks specified in the flow. Finally, it should be noted that the above-mentioned embodiments are merely intended to illustrate the technical solutions of the present application, but not to limit it. Although the present application has been described in detail with reference to the above-mentioned embodiments, it should be understood by those skilled in the art that the specific embodiments of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered within the protection scope of the claims of the present application.

Claims

1. A method for optimizing container scheduling at railway stations, characterized in that, Includes the following steps: Based on the yard information representation method of container position freedom and the container stacking optimization process, a railway container yard position allocation optimization model is constructed. The constraints of the railway container yard position allocation optimization model are configured according to the container yard storage rules and container stacking form. The container yard storage rules and container stacking form are preset according to the container attributes. By combining the railway container yard location allocation optimization model, we obtain the container stacking algorithm from the yard to the train and the container retrieval algorithm from the train to the yard. The container to be stacked is stacked using the container yard to the train stacking algorithm to obtain the optimal stacking position of the container to be stacked. The container to be retrieved is retrieved using the container retrieval algorithm from the train to the yard to obtain the optimal target location of the container to be retrieved. Based on the container type classification and storage space resource optimization of the container yard, a balanced allocation model for the container volume of the railway container yard is constructed. The optimal division of container area volume is obtained by solving the balanced allocation model of the railway container yard volume using a heuristic algorithm. Combined with the optimal division of container area volume, the optimal container location allocation method is obtained by solving the optimized model of container location allocation of the railway container yard using a heuristic algorithm.

2. The method for optimizing container scheduling at railway stations according to claim 1, characterized in that, The container yard storage rules include: Before the yard is fully occupied, all containers are free containers and are stacked according to the principle of priority placement on top. When the bottom layer of the yard is full, the remaining containers are stacked in order of increasing number of containers to be stacked in each stack in the yard. The target locations for stacking and picking up containers, as well as the equipment movement paths, are within the equipment's operating area. 20-foot and 40-foot containers should be stacked on different layers, and no containers should be stacked on top of non-compressible container types. Obtain the target location of containers to be stacked or picked up within the space of the yard; The container stacking methods include vertical, flat, and stepped stacking, with each container being handled individually and repeatedly during stacking and retrieval.

3. The method for optimizing container scheduling at railway stations according to claim 2, characterized in that, The container stacking optimization process includes: In the first stage, when the bottom layer of the yard is not full and there is remaining space in the yard, containers of the same type are stacked together according to the principle of prioritizing the top layer. In the second stage, when the attributes of newly arrived containers are different from those of containers in the yard, the newly arrived containers are placed in the space where containers of the same attribute can be placed. When there are no new containers of the same type in the yard, or when the attributes of the new container and all attributes in the yard do not simultaneously generate container pressing operations; by maximizing the sum of the total degrees of freedom of the container positions in the yard, we can obtain the possible positions or sets of positions for the new container. The third stage involves creating a second ballast when the yard can no longer meet the requirements of a single ballast. The maximum constraint on the degree of freedom of container position in the yard determines the position where the container can be stacked or retrieved. If the stacking status in the yard returns to the previous stage during the container retrieval operation, the subsequent stacking or retrieval operations will be carried out according to the previous stage. The degrees of freedom for the container's position are: in, x ∈X, y ∈Y, ( x , y )∈S, S∈X×Y, where X represents the range of positive integer x-coordinates that the box location can take in the bottom projection coordinates, and Y represents the range of positive integer y-coordinates that the box location can take in the bottom projection coordinates. x , y ) represents the bottom projection coordinates of the box location. z This refers to the stacking height of containers; S1 indicates that a container is placed in the storage location and is not blocked by other containers, but the stack is not at its highest level, giving the container location a degree of freedom. f =1; S2 indicates that a container is placed in a location, but it is pressed down by other containers and is not in a stack of containers with the same attribute, and the stack is not stacked to its highest point. In this case, the degree of freedom of the container location is k times the degree of freedom of the container location pressing down on it. S3 indicates that there are no containers placed in the container location, but there are containers stacked underneath it or there is a temporary detention mark at that location. The container location has 10 degrees of freedom. f =0; S4 indicates that a container is placed in a location within a stack consisting entirely of containers of the same attribute, and that the stack has been reached at its highest level. This represents the container's degree of freedom. f =2; S5 indicates the container position with no containers stacked above or below it, representing the container position's degree of freedom. f =3; When adjacent containers stacked on top of each other are containers of the same type k = α , α ∈[0,1] or k = β , β ∈[0,1], β < α ; The objective function of the railway container yard space allocation optimization model is: Where maxZ is the objective function of the railway container yard space allocation optimization model, which means that the sum of the degrees of freedom of all container spaces in the yard is the highest; e This represents a 0-1 variable, where 1 indicates the container is being moved from the train to the yard, and 0 indicates the container is being retrieved from the yard to the train. f Indicates the degrees of freedom in the container's position. p (y) is the decision variable, representing the output code pickup box location or pickup box position; q ( d ) for box community d D is the set of all container locations; D is the set of all container cells. d ∈D; h ( y The variable ) is a 0-1 variable, indicating whether the container is a container with a certain attribute. If it is, the value is 1, and if it is not, the value is 0. g ( R ( d ), y ) for box community d Middle box position y about R ( d The serial number of ) R ( d ) for box community d The stacking order, container positions y The stacking order follows the vertical stacking rule, starting from the smallest element and taking values ​​sequentially. g ( R ( d ), y )=1,2,3,…, n ; L (y) represents the temporary detention flag constraint; The constraints of the railway container yard space allocation optimization model are as follows: m 2＞ m 1＞0 in, L (y) indicates that empty boxes without temporary detention markings are given priority when stacking boxes. m 1 indicates that the box is temporarily detained when it is being stored. m 2 indicates that there is no temporary detention mark when the box is stacked; h ( b ) indicates box b The general attributes of h (p( y) This represents the generalized classification attribute of the container location; The generalized classification attribute is a set of classification attributes in two aspects: container area and container sub-area. The container area includes outgoing container area, arriving container area, international container area, refrigerated container area, special container area, empty container area, spare container area, maintenance container area, and cleaning and disinfection container area. Each container area is divided into several container sub-areas based on the classification attributes and past cargo flow. Vertical stacking is used when stacking container sub-areas.

4. The method for optimizing container scheduling at railway stations according to claim 3, characterized in that, The container stacking algorithm from the container yard to the train includes the following steps: S11. Input the box attributes of the box to be coded and determine the box community to which the box belongs; S12. In the found box cell, search from the smallest element to the largest element according to the box cell code order table. For empty box positions with temporary detention marks encountered in the search, remove the temporary detention marks until the first full box position is found. Mark the full box smallest element on the full box position. S13. According to the container stacking order table of the container area, search from the smallest element to the largest element step by step to find the first container position that can be used for stacking in the yard, and the container position has no temporary detention mark. Record the sum of the total degrees of freedom of each container position in the yard under the new stacking state after the container to be stacked is placed in the container position. Record it as the sum of the first degree of freedom. If all container positions that can be used for stacking have temporary detention marks, go to S15. S14. According to the container retrieval order table, find empty container positions without temporary detention marks that can be used for container stacking one by one, and calculate the sum of the degrees of freedom of each container position in the yard after the container to be stacked is placed in the container position. Record it as the sum of the second degrees of freedom. If the sum of the second degrees of freedom is greater than the sum of the first degrees of freedom, the priority position of the container to be stacked is replaced by the current container position. Otherwise, the original position is retained as the priority stacking position, and go to S16. S15. If no box location without a temporary detention mark is found in S12, then find all box locations with temporary detention marks according to the box location sequence table, remove the temporary detention marks on the box locations, and go to S13. S16. Update the positional degrees of freedom of the container units in the storage yard; S17. Output the coordinates of the box that can be stacked. If no box is found, output no information. The container retrieval algorithm from the train to the container yard includes the following steps: S21. Input the attributes of the box to be picked up, find the box community to which the box belongs, otherwise there is no solution information; S22. In the found container area, search from the smallest element to the largest element according to the container area retrieval order table, find the first container with the same attributes as the container to be retrieved and that is completely free and has no temporary detention mark, and set a temporary detention mark on the container position used for retrieval. If no container that meets the conditions is found, go to S24. S23. Update the positional degrees of freedom of the container units in the storage yard; S24. Output the coordinates of the found box location. If not found, output no information.

5. The method for optimizing container scheduling at railway stations according to claim 1, characterized in that, The yard is classified into container zones, which include a shipping container zone, an arriving container zone, an international container zone, a customs-supervised area, a refrigerated container zone, a dedicated container zone, an empty container zone, a spare container zone, a maintenance container zone, and a cleaning and disinfection container zone. The container types optimized for storage space resources include DX type for unloading, DT type for pickup, DF type for dispatch, and DY type for transport. The DX type of container to be unloaded indicates that the container is still on the train and is waiting to be unloaded and stored in the yard. The DT type of container to be picked up indicates that the container has been stacked in the yard and is waiting for the truck to enter the yard to pick up the container; The DF type of container to be dispatched indicates a container that has not yet been allocated to the yard and is waiting to be transported to the yard by truck for storage. The DY type of the waiting container indicates a container that is already in the yard waiting to be loaded onto a train.

6. The method for optimizing container scheduling at railway stations according to claim 5, characterized in that, The objective function of the railway container yard capacity equalization allocation model is shown below: Where w1 represents the weighting coefficient for the balance of container volume in the container yard truck operations, w2 represents the weighting coefficient for the balance of container volume in the container yard train operations, and w3 represents the weighting coefficient for the overall balance of container volume in the container yard. express t The containers are transported to the container yard and allocated to the container area during the designated time period. i The quantity of DF-type containers loaded onto the train. express t The time period is from unloading from the train to the container area. i The quantity of DX-type boxes that were taken away. express t The number of DT-type containers picked up during the specified time period. express t The quantity of DY-type containers loaded onto the train during a given time period. Indicates that it is stored in the container area. i In China t The quantity of DY-type containers loaded onto the train during the specified time period. This indicates the quantity of DT-type boxes taken away. Indicates that it is stored in the container area. i In China t The number of DT-type containers picked up during the specified time period; The constraints of the railway container yard capacity equalization allocation model are as follows: t =1,2… T ; K =0,1… T - t t =1,2… T ; i =0,1… B , t =1,2… T ; K =0,1… T - t , t =1,2… T ; i =0,1… B , t =1,2… T ; i =0,1… B , t =1,2… T ; i =0,1… B in, express t The containers are transported to the container yard and allocated to the container area during the designated time period. i The capacity of the DF type box express t The containers are transported to the container yard and allocated to the container area during the designated time period. i In t + k The quantity of DF-type containers loaded onto the train during a given time period. B Indicates the number of container areas within the storage yard. T This indicates the time periods contained within a planning period. express t The time period is from unloading from the train to the container area. i The capacity of the DX type box. express t The time period is from unloading from the train to the container area. i In t + k The number of DX-type containers picked up during the specified time period. Indicates that it is stored in the container area. i In China t The number of DT-type containers picked up during the specified time period. This indicates that the container area has been allocated at the start of the planning cycle. i In, and in t The number of DT-type containers picked up during the specified time period. Indicate t The time period is from unloading from the train to the container area. i In t - k The number of DX-type containers picked up during the specified time period. Indicates that it is stored in the container area. i In China t The quantity of DY-type containers loaded onto the train during the specified time period. This indicates that the container area has been allocated at the start of the planning cycle. i In, and in t The quantity of DF-type containers loaded onto the train during a given time period. express t The containers are transported to the container yard and allocated to the container area during the designated time period. i In t - k The quantity of DF-type containers loaded onto the train during a given time period. express t container area at the start of the time period i Total number of boxes in the container. Indicates box area i Storage capacity.

7. The method for optimizing container scheduling at railway stations according to claim 6, characterized in that, The heuristic algorithm includes the following steps: S31. Initialize the planning cycle parameters, set the time variable t=1, and input the container information; S32. Calculate the workload of DF type containers in each container area of ​​the yard during time period t, allocate them to each container area according to the constraints, calculate the total workload of the four types of containers in the yard, and obtain the workload ranking of each container area. S33. Limit the workload of each container area to be allocated to each container area in ascending order, and determine whether to end the task. S34. Calculate the total quantity of the four types of containers in each time period, and calculate and store the objective function value; S35. Determine if the container area with the largest workload has DX type containers. If so, proceed to S36; otherwise, proceed to S37. S36. Determine whether the quantity of DX-type containers leaving during this period was allocated to this container area because the workload of this container area was the smallest during the allocation period. If so, proceed to S38; otherwise, return to S35. S37. Determine if the container area with the largest workload has DX type containers in this time period. If not, proceed to S39; otherwise, return to S35. S8. Let t = t-1, determine whether the current time period is the start time period within the planning cycle, calculate the objective function value, and go to S39; S39. Compare the two calculated objective function values ​​and take the minimum value as the objective function of the model.

8. A railway station container scheduling optimization system, characterized in that, include: A module for constructing a railway container yard space allocation optimization model is provided. This model is used to construct a railway container yard space allocation optimization model based on a yard information representation method with container space degrees of freedom and a container stacking optimization process. The constraints of the railway container yard space allocation optimization model are configured according to the container yard storage rules and container stacking forms. The container yard storage rules and container stacking forms are preset according to the container attributes. The module obtains the optimal stacking position of the container to be stacked and the optimal target container position of the container to be retrieved. It is used to combine the railway container yard container position assignment optimization model to obtain the stacking algorithm from the container yard to the train and the retrieval algorithm from the train to the container yard. The stacking algorithm from the container yard to the train is used to stack the container to be stacked to obtain the optimal stacking position of the container to be stacked. The retrieval algorithm from the train to the container yard is used to retrieve the container to obtain the optimal target container position of the container to be retrieved. The module for obtaining the optimal container location allocation method is used to construct a balanced allocation model for the container volume of railway container yards based on the container type of the yard classification and storage space resource optimization. The module uses a heuristic algorithm to solve the balanced allocation model for the container volume of railway container yards to obtain the optimal division of container area volume. Combining the optimal division of container area volume with the heuristic algorithm, the module uses a heuristic algorithm to solve the optimized model for container location allocation of railway container yards to obtain the optimal container location allocation method.

9. An electronic device, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements a railway station container scheduling optimization method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements a railway station container scheduling optimization method as described in any one of claims 1-7.

Images