A method for allocating empty containers in multiple locations based on multiple uncertain factors

By building an empty box allocation model containing multiple uncertain factors, the cost efficiency and low problem caused by uncertain factors in container allocation is solved, the cost-minimized scheduling scheme and more realistic transportation simulation are realized, and the transportation path is optimized and container reserve.

CN118378862BActive Publication Date: 2024-11-12ZHEJIANG SIGANG LINKAGE DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410814164.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-24
Publication Date
2024-11-12
Estimated Expiration
2044-06-24

AI Technical Summary

Technical Problem

When faced with a variety of uncertain factors, existing container allocation methods lead to problems of high allocation costs and low efficiency, such as uncertain transportation mode and uncertain port loading and unloading containers, which affect the transit timeliness.

Method used

A method of empty boxes for multiple places based on various uncertain factors is proposed. By defining decision variables and constructing constraints and models for empty boxes for allocation costs, a scheduling scheme with minimal cost is formed, considering railways, ports, and sea-rail transport centers The storage capacity, transportation path, transportation mode, and washing box demand are many uncertain factors.

Benefits of technology

It realizes that while meeting customer needs, the operating costs of empty box allocation are reduced, and more realistic transportation and storage scenarios are simulated, the transportation path and box storage strategy are optimized, and the transportation time and cost are reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118378862B_ABST
    Figure CN118378862B_ABST
Patent Text Reader

Abstract

The present invention relates to a multi-location empty container allocation method based on multiple uncertain factors, belonging to the technical field of container scheduling, comprising the following steps: S1. defining decision variables, the decision variables include t The number of empty containers transported from the origin to the destination during the time period, t The number of empty containers at the origin at the beginning of the time period and t The number of empty containers called from outside during the time period; S2. Based on the decision variables, the constraints of the empty container allocation cost and the constraints based on various uncertain factors are constructed; S3. Based on the constraints of the empty container allocation cost, an empty container allocation model is constructed; S4. A scheduling plan with minimum cost is formed through the empty container allocation model. The present invention can provide an optimal solution that meets customer needs and minimizes costs, reduces the operating cost of empty container allocation, takes various uncertain factors into consideration, and can simulate a more realistic situation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of container dispatching, and in particular relates to a multi-location empty container dispatching method based on multiple uncertain factors. Background Art

[0002] Intermodal transport refers to the transportation completed by two or more means of transportation connecting and transferring with each other. In order to ensure that customers' container needs can be met in a timely manner and the transportation process is not interrupted due to insufficient containers, empty containers in various places need to be allocated from time to time.

[0003] At present, the container allocation methods are all focused on path planning. For example, the Chinese patent application with application publication number CN117193211A discloses a path planning method for a composite navigation IGV of a sea-rail intermodal U-shaped automated container terminal, which includes the following steps: S1. Analyze the key production organization and distribution layout of the organizational facilities of the sea-rail intermodal U-shaped automated container terminal; S2. Sort out and analyze the sea-rail intermodal operation process of the composite navigation IGV; S3. Construct a global path planning model for the operation of the composite navigation IGV of the sea-rail intermodal U-shaped automated container terminal; S4. Based on the global path planning algorithm of differential evolution, obtain the global operation path of multiple IGVs; S5. Predict conflicts and congestion for the composite navigation IGV, and perform local path optimization and adjustment.

[0004] However, in actual operation, many uncertain factors will affect the decision-making of empty container allocation, such as maintenance, economic benefit considerations, technical difficulties, etc. Specifically, the maximum number of empty containers that can be loaded in a certain transportation mode is uncertain, because container transportation generally gives priority to the transportation of heavy containers; the number of containers that can be loaded and unloaded at a port at the same time is also uncertain, so there will be many uncertainties in the transit time. These uncertain factors will lead to increased costs and low efficiency in the container scheduling process. Summary of the invention

[0005] The present invention provides a multi-location empty container allocation method based on multiple uncertain factors to solve the problem of high allocation cost and low efficiency caused by many uncertain factors in the existing container allocation planning.

[0006] In order to solve the above technical problems, the technical solution provided by the present invention is:

[0007] The present invention relates to a method for allocating empty containers at multiple locations based on multiple uncertain factors, which comprises the following steps:

[0008] S1. Define decision variables, which include t The number of empty containers transported from the origin to the destination during the time period, t The number of empty containers at the origin at the beginning of the time period andt The number of empty containers called from outside during the time period;

[0009] S2. Construct the constraint condition of empty container allocation cost based on decision variables, the expression is:

[0010] (1),

[0011] in, μ is the confidence level, N is a collection of time periods, c For container boxes, C It is a collection of container boxes. u is the container condition, U It is a collection of container conditions. l As the place of origin, l’ For the destination, l,l’ is one of the paths from the origin to the destination. A is the path from the origin to the destination that can be used to transport empty containers, d For transport mode, D A collection of transport modes. From the origin to the destination via transportation mode d exist t The shipping cost per box during the time period, for t Transport mode within the time period d The types of containers transported from the origin to the destination are c Box condition is u The amount of empty containers, i For customers who need empty containers, I For customers who need empty boxes, S is the set of locations that can supply empty containers, for t The storage cost of a single container at the origin within the time period, for t The types of boxes available at the departure point at the initial time of the time period are: k And the box condition is u The amount of empty containers, To satisfy customers within the time period t i Abnormal transportation costs incurred during transportation due to the demand for boxes. for t Due to customer i The box class that needs to be called from outside is c Box condition is u The number of empty boxes, W is the budget, α A confidence level threshold for the probability that the empty container allocation cost exceeds the budget;

[0012] S3. Construct an empty container allocation model based on the constraint condition of empty container allocation cost;

[0013] S4. Form a cost-minimizing scheduling plan through the empty container allocation model.

[0014] Preferably, S2 also constructs the restriction conditions of the total amount of containers that can be stored in railways, ports and sea-rail transport centers and the restriction conditions of the total amount of empty containers that can be transported under each path based on the decision variables.

[0015] The expression for the restriction condition of the total container quantity that can be stored in the railway, port and sea-rail transport center is:

[0016] (2),

[0017] in, for t The amount of box storage at the start of the time period, R A collection of railway stations. P It is a collection of ports and terminals. V It is a collection of sea-rail transport centers. β is the confidence level threshold of the box storage volume at the origin;

[0018] The expression for the restriction condition of the total amount of empty containers that can be transported under each path is:

[0019] (3),

[0020] in, for t By mode of transport during the time period d The maximum number of containers that can be loaded from the origin to the destination. γ The confidence level threshold for the maximum number of boxes that can be transported.

[0021] Preferably, S2 further constructs a limiting condition for satisfying the good container demand within the scope of the transportation plan, and the limiting condition for satisfying the good container demand within the scope of the transportation plan is expressed as:

[0022] (4),

[0023] in, A is the path from the origin to the destination that can be used to transport empty containers, for t The types of boxes that can be called between the starting point and the destination within the time period are c And the box condition is the empty box quantity of good boxes, for t Customers within time period i The box type is cThe demand for empty boxes in good condition, for t Due to customer i The box class that needs to be called from outside is c The number of empty boxes in good condition. o The box condition is good.

[0024] Preferably, the S2 further constructs a restriction condition for transporting dirty tanks out of the destination, and the expression of the restriction condition for transporting dirty tanks out of the destination is:

[0025] (5),

[0026] in, A is the path from the origin to the destination that can be used to transport empty containers, τ The time required for the box washing station. τ sld For pass mode d Empty boxes from the supply point s To destination l The time required, For Representative t + τ sld Time period through mode d From the supply point s To destination l The box type shipped is c The amount of dirt tank, For time period t Internal box supply point s Available boxes are: c And the box condition is the empty box quantity of the dirty box, m Indicates a dirt tank.

[0027] Preferably, the S2 further constructs a dirty tank inventory restriction condition of the tank washing station and a good tank inventory restriction condition of the tank washing station, and the expression of the dirty tank inventory restriction condition of the tank washing station is:

[0028] (6),

[0029] in, for t Container washing station within the time period k The storage box types are c And the box condition is a dirty box m The amount of empty containers, A is the path from the origin to the destination that can be used to transport empty containers, K For the collection of box washing stations, t’ for t Another time period before the time period, τThe time required for the box washing station. for t’ Transport mode used during time period d From the place of origin l Arrival at the container washing station k The box type is c The box condition is a dirty box m The number of empty containers;

[0030] The expression of the good box inventory restriction condition of the box washing station is:

[0031] (7),

[0032] in, for t Container washing station within the time period k The storage box types are c And the box condition is good o The amount of empty containers, for t’ Transport mode used during time period d From the place of origin l Arrival at the container washing station k The box type is c Good box condition o The amount of empty containers, for t’ Transport mode used during time period d From the place of origin l Arrival at the container washing station k The box type is c The box condition is a dirty box m The amount of empty containers, τ kld For pass mode d Empty containers from the container washing point k To destination l The time required, for t’ + τ kld Time period through mode d From the box wash point k The box types shipped are c A good amount of boxes.

[0033] Preferably, S2 further constructs limiting conditions for the quantity of various types of empty containers that can be provided by each land transport station, railway station, port, and sea-rail transport center in each time period. The limiting conditions for the quantity of various types of empty containers that can be provided by each land transport station, railway station, port, and sea-rail transport center in each time period are expressed as follows:

[0034] (8),

[0035] in, t’ for t Another time period before the time period, for t’ Transport mode used during time period d From the starting point l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, τ ll’d For pass mode d Empty containers from the origin l To destination l’ The time required, for t’ + τ ll’d Transport mode used during time period d From the place of origin l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, J is the set of sites in the transportation network, P It is a collection of ports and terminals. R A collection of railway stations. V It is a collection of sea-rail transport centers. u Indicates the box condition. U A collection of empty bin conditions, including dirty bins m Reconciliation Box o .

[0036] Preferably, the S2 also constructs t The number of containers that can be shipped out of each type and condition within a time period is limited. t The expression for the limit condition of the number of outbound containers of each type and condition within a time period is:

[0037] (9),

[0038] in, τ ll’d For pass mode m Empty containers from the starting point l To destination l’ The time required, for t’ + τ ll’d Transport mode used during time period m From the place of origin l Arrival at destination l’ The box type is c And the box condition is uThe amount of empty containers, J is the set of sites in the transportation network, P It is a collection of ports and terminals. R A collection of railway stations. C For the collection of box washing stations, V It is a collection of sea-rail transport centers.

[0039] Preferably, S2 further constructs a restriction condition that the empty container is only transported in a given logistics network, and the restriction condition that the empty container is only transported in a given logistics network is expressed as:

[0040] (10),

[0041] in, for t Transport mode within the time period d The types of containers transported from the origin to the destination are c Box condition is u The number of empty containers at a time that does not belong to the time period set t The empty box quantity needs to be 0.

[0042] Preferably, S2 also constructs the limiting conditions for the number of boxes shipped out of each location, and the expression for the limiting conditions for the number of boxes shipped out of each location is:

[0043] (11),

[0044] in, For pass mode d Empty containers from the starting point l To destination l’ The time required, δ Any time period, △ The storage time allowed at each location for multimodal transport, for t + δ + τ ll’d Transport mode used during time period d From the place of origin l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, P It is a collection of ports and terminals. R A collection of railway stations. V It is a collection of sea-rail transport centers. for t The types of boxes transported from the starting point l during the time period are c And the box condition is u The box volume.

[0045] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0046] 1. The multi-location empty container allocation method based on multiple uncertain factors involved in the present invention constructs an empty container allocation model with constraints including the empty container allocation cost. Through this model, the optimal solution that meets customer needs and has the lowest cost can be given, thereby reducing the operating cost of empty container allocation.

[0047] 2. The multi-location empty container allocation method based on multiple uncertain factors involved in the present invention also constructs restrictions on multiple uncertain factors, taking into account the demand, transportation time between points in each mode + storage time requirements at various transit points + time spent on washing boxes, etc., the number of empty containers of various types and conditions in various places, the maximum number of containers that can be loaded in a single trip in each mode, dynamic comparison of import and export container volumes at each point in the same time period, and multiple uncertain factors involved in transportation, storage, and washing boxes in various empty container allocation schemes, so as to simulate a more realistic situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of the multi-location empty container allocation method based on multiple uncertain factors. DETAILED DESCRIPTION

[0049] In order to further understand the content of the present invention, the present invention is described in detail in conjunction with examples. The following examples are used to illustrate the present invention but are not used to limit the scope of the present invention.

[0050] See attached Figure 1 As shown, the multi-location empty container allocation method based on multiple uncertain factors of the present invention comprises the following steps:

[0051] S1. Define decision variables, which include t The number of empty containers transported from the origin to the destination during the time period, t The number of empty containers at the origin at the beginning of the time period and t The number of empty containers called from outside during the time period;

[0052] S2. Construct constraints for various uncertain factors, including:

[0053] Constraints 1. Based on the decision variables, the constraints of the empty container allocation cost are constructed, and the expression is:

[0054] (1),

[0055] in, μ is the confidence level, N is a collection of time periods, c For container boxes, CIt is a collection of container boxes. u is the container condition, U It is a collection of container conditions. l As the place of origin, l’ For the destination, l,l’ is one of the paths from the origin to the destination. A is the path from the origin to the destination that can be used to transport empty containers, d For transport mode, D A collection of transport modes. From the origin to the destination via transportation mode d exist t The shipping cost per box during the time period, for t Transport mode within the time period d The types of containers transported from the origin to the destination are c Box condition is u The amount of empty containers, i For customers who need empty containers, I For customers who need empty boxes, S is the set of locations that can supply empty containers, for t The storage cost of a single container at the origin within the time period, for t The types of boxes available at the departure point at the initial time of the time period are: k And the box condition is u The amount of empty containers, To satisfy customers within the time period t i Abnormal transportation costs incurred during transportation due to the demand for boxes. for t Due to customer i The box class that needs to be called from outside is c Box condition is u The number of empty boxes, W is the budget, α A confidence level threshold for the probability that the empty container allocation cost exceeds the budget;

[0056] Constraint 1 ensures that the decision maker wants to reduce the budget for empty container transportation. W And hope that the total cost exceeds the budget W The probability is lower than the confidence level α This demand.

[0057] Constraint 2: The total container volume that can be stored in railways, ports, and sea-rail transport centers is limited by the following expression:

[0058] (2),

[0059] in, for t The amount of box storage at the start of the time period, R A collection of railway stations. P It is a collection of ports and terminals. V It is a collection of sea-rail transport centers. β is the confidence level threshold of the box storage volume at the origin;

[0060] This constraint ensures that the total volume of containers that can be stored at railways, ports, and sea-rail transport centers is not less than the confidence level. β The storage limit will not be exceeded.

[0061] Constraint 3. The total amount of empty containers that can be transported under each path is limited by the following expression:

[0062] (3),

[0063] in, for t By mode of transport during the time period d The maximum number of containers that can be loaded from the origin to the destination. γ is the confidence level threshold for the maximum number of containers that can be transported,

[0064] This constraint ensures that the total amount of empty containers that can be transported under each path is not less than the confidence level. γ In this case, the maximum loading box quantity limit must not be exceeded.

[0065] Constraint 4. The limiting condition for satisfying the demand for good containers within the scope of the transportation plan is expressed as:

[0066] (4),

[0067] in, A is the path from the origin to the destination that can be used to transport empty containers, for t The types of boxes that can be called between the starting point and the destination within the time period are c And the box condition is the empty box quantity of good boxes, for t Customers within time period i The box type is c The demand for empty boxes in good condition, for t Due to customer i The box class that needs to be called from outside is c The number of empty boxes in good condition. o The box condition is good;

[0068] This constraint indicates that the demand for containers in good condition can be met within the scope of the transportation plan. If the demand for containers in good condition cannot be met, it is necessary to borrow containers in good condition from sources outside the system.

[0069] Constraint 5. The destination dirty tank transport restriction condition is expressed as:

[0070] (5),

[0071] in, A is the path from the origin to the destination that can be used to transport empty containers, τ The time required for the box washing station. τ sld For pass mode d Empty boxes from the supply point s To destination l The time required, For Representative t + τ sld Time period through mode d From the supply point s To destination l The box type shipped is c The amount of dirt tank, For time period t Internal box supply point s Available boxes are: c And the box condition is the empty box quantity of the dirty box, m Indicates a dirt tank;

[0072] Due to the uncertainty of intermediate transportation, there is a probability that the container will be contaminated. The dirty container will affect the transportation safety and the quality of the delivered products. Once contaminated, it needs to be immediately transported to the container washing station for cleaning. This constraint indicates that once a dirty container is unloaded at the customer's container supply location, it must be immediately transported out of the location.

[0073] Constraint 6. The inventory limit of dirty boxes at the box washing station is expressed as:

[0074] (6),

[0075] in, for t Container washing station within the time period k The storage box types are c And the box condition is a dirty box m The amount of empty containers, A is the path from the origin to the destination that can be used to transport empty containers, K For the collection of box washing stations, t’ for tAnother time period before the time period, τ The time required for the box washing station. for t’ Transport mode used during time period d From the place of origin l Arrival at the container washing station k The box type is c The box condition is a dirty box m The number of empty containers;

[0076] This constraint states that the dirty bin inventory limit can be inferred from the number of dirty bins entering the station and the number of good bins leaving the station.

[0077] Constraint 7. The inventory limit of good boxes at the box washing station is expressed as:

[0078] (7),

[0079] in, for t Container washing station within the time period k The storage box types are c And the box condition is good o The amount of empty containers, for t’ Transport mode used during time period d From the place of origin l Arrival at the container washing station k The box type is c Good box condition o The amount of empty containers, for t’ Transport mode used during time period d From the place of origin l Arrival at the container washing station k The box type is c The box condition is a dirty box m The amount of empty containers, τ kld For pass mode d Empty containers from the container washing point k To destination l The time required, for t’ + τ kld Time period through mode d From the box wash point k The box types shipped are c The quantity of good boxes;

[0080] This constraint indicates that the good bin inventory can be determined by the number of good bins entering and leaving the station and the number of dirty bins entering the station.

[0081] Constraint 8. The limiting conditions for the quantity of various empty containers that can be provided by each land transport station, railway station, port, and sea-rail transport center in each time period are expressed as follows:

[0082] (8),

[0083] in, t’ for t Another time period before the time period, for t’ Transport mode used during time period d From the starting point l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, τ ll’d For pass mode d Empty containers from the starting point l To destination l’ The time required, for t’ + τ ll’d Transport mode used during time period d From the place of origin l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, J is the set of sites in the transportation network, P It is a collection of ports and terminals. R A collection of railway stations. V It is a collection of sea-rail transport centers. u Indicates the box condition. U A collection of empty bin conditions, including dirty bins m Reconciliation Box o ;

[0084] When initializing variables, it should be assumed that the initial container inventory at each location is 0. This constraint mainly reflects the quantity of various empty containers that can be provided by various locations in different time periods, such as land transportation stations, railway stations, ports, and sea-rail transport centers.

[0085] Constraint 9. t The number of containers that can be shipped out of each type and condition within a time period is limited. it The expression is:

[0086] (9),

[0087] in, τ ll’d For pass modem Empty containers from the starting point l To destination l’ The time required, for t’ + τ ll’d Transport mode used during time period m From the place of origin l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, J is the set of sites in the transportation network, P It is a collection of ports and terminals. R A collection of railway stations. C For the collection of box washing stations, V It is a collection of sea-rail transport centers;

[0088] This constraint states that the number of shipped containers of each type and condition during time period t cannot be greater than the number of empty containers available at each location during that time period.

[0089] Constraint 10. The constraint that empty containers are transported only in a given logistics network is expressed as:

[0090] (10),

[0091] in, for t Transport mode within the time period d The types of containers transported from the origin to the destination are c Box condition is u The number of empty containers at a time that does not belong to the time period set t The empty box quantity needs to be 0;

[0092] The main purpose of this constraint is to ensure that containers can only be transported within a given logistics network.

[0093] Constraint 11. The constraint condition of the number of boxes shipped out of each location is expressed as:

[0094] (11),

[0095] in, For pass mode d Empty containers from the starting point l To destination l’ The time required, δ Any time period, △ The storage time allowed for each multimodal transport location (port, railway, sea-rail station), fort + δ + τ ll’d Transport mode used during time period d From the place of origin l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, P It is a collection of ports and terminals. R A collection of railway stations. V It is a collection of sea-rail transport centers. for t The types of boxes transported from the starting point l during the time period are c And the box condition is u The box volume.

[0096] This constraint means that in any time period δ, the number of empty containers available in each location cannot be greater than the number of containers shipped out of each location, to avoid insufficient supply.

[0097] S3. Based on the constraints of various uncertain factors including the constraints of empty container allocation cost, an empty container allocation model is constructed. The empty container allocation model is:

[0098] .

[0099] Taking a domestic container operating company as an example, the logistics supply chain data of the past year is used as the initial data, such as location information including all ports, railway stations, sea-rail transport centers, container washing stations, and the company's existing basic information on containers (including box type, size, quantity, etc.). The present invention mainly adopts linear, discrete, and empirical uncertainty distributions, and collects various information from domain experts through qualitative and quantitative analysis, interviews, and observations, including the transportation cost, cleaning time, storage time, storage cost, and storage capacity of multimodal terminals for each mode of transportation. By calculating the average price and transportation time between each pair of origin and destination, the transportation cost and transportation time of each mode of transportation can be obtained. For example, the distance between the origin and the destination is associated with the average speed of the truck (60 km / h), and the transportation time of road transportation can be obtained. Since domain experts are independent, the present invention sets confidence levels α, β, and γ.

[0100] Based on given historical data, and on the premise of meeting customers' timely container needs, the model can calculate the minimum total cost of empty container transportation required in any time period in the past. After comparing with the actual costs, it can be seen that the new transportation methods, route planning and container reserve strategies output by the model have produced certain effects on transportation timeliness and cost reduction.

[0101] S4. Input the parameters of the future time period and form a cost-minimizing scheduling plan through the empty container allocation model.

[0102] After verifying the capabilities of the empty container allocation model in S3, we input data such as the routes (origin and destination) of major shipping companies of a container operating company in the next three months, user demand for empty containers, and container conditions. Finally, we obtained the allocation path and transportation method of the empty containers with the lowest cost, and visualized the data in the map layer.

[0103] The present invention is described in detail above in conjunction with the embodiments, but the contents described are only preferred embodiments of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of application of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for allocating empty containers at multiple locations based on multiple uncertain factors, characterized by: It includes the following steps: S1. Define decision variables, which include t The number of empty containers transported from the origin to the destination during the time period, t The number of empty containers at the origin at the beginning of the time period and t The number of empty containers called from outside during the time period; S2. Construct the constraint condition of empty container allocation cost based on decision variables, the expression is: (1), in, μ is the confidence level, N is a collection of time periods, c For container boxes, C It is a collection of container boxes. u is the container condition, U It is a collection of container conditions. l As the place of origin, l’ For the destination, l,l’ is one of the paths from the origin to the destination. A is the path from the origin to the destination that can be used to transport empty containers, d For transport mode, D A collection of transport modes. From the origin to the destination by mode of transport d exist t The shipping cost per box during the time period, for t Transport mode within the time period d The types of containers transported from the origin to the destination are c Box condition is u The amount of empty containers, i For customers who need empty containers, I For customers who need empty boxes, S is the set of locations that can supply empty containers, for t The storage cost of a single container at the origin within the time period, for t The types of boxes available at the departure point at the initial time of the time period are: c And the box condition is u The amount of empty containers, To satisfy customers within the time period t i Abnormal transportation costs incurred during transportation due to the demand for boxes. for t During the period of time due to customers i The box class that needs to be called from outside is c Box condition is u The number of empty boxes, W is the budget, α A confidence level threshold for the probability that the empty container allocation cost exceeds the budget; S3. Construct an empty container allocation model based on the constraint condition of empty container allocation cost; S4. Form a cost-minimizing scheduling plan through an empty container allocation model; Based on the decision variables, S2 also constructs the constraints on the total amount of containers that can be stored in railways, ports, and sea-rail transport centers, as well as the constraints on the total amount of empty containers that can be transported under each path. The expression for the restriction condition of the total container quantity that can be stored in the railway, port and sea-rail transport center is: (2), in, for t The container storage capacity at the origin of the time period, R A collection of railway stations. P It is a collection of ports and terminals. V It is a collection of sea-rail transport centers. β is the confidence level threshold for the box storage quantity at the origin; The expression for the restriction condition of the total amount of empty containers that can be transported under each path is: (3), in, for t By mode of transport during the time period d The maximum number of containers that can be loaded from the origin to the destination. γ The confidence level threshold for the maximum number of containers that can be transported; S2 also constructs the limiting conditions for satisfying the good container demand within the transportation plan. The limiting conditions for satisfying the good container demand within the transportation plan are expressed as follows: (4), in, A is the path from the origin to the destination that can be used to transport empty containers, for t The available box types between the origin and the destination within the time period are: c And the box condition is the empty box quantity of good boxes, for t Customers within time period i The box type is c The demand for empty boxes in good condition, for t During the period of time due to customers i The box class that needs to be called from outside is c The number of empty boxes in good condition. o The box condition is good; The S2 also constructs the destination dirty tank transport out restriction condition, and the destination dirty tank transport out restriction condition is expressed as: (5), in, A is the path from the origin to the destination that can be used to transport empty containers, τ The time required for the box washing station. τ sl’d For pass mode d Empty boxes from the supply point s To destination l’ The time required, For Representative t + τ sl’d Time period through mode d From the supply point s To destination l’ The box type shipped is c The amount of dirt tank, For time period t Internal box supply point s Available boxes are: c And the box condition is the empty box quantity of the dirty box, m Indicates a dirt tank; The S2 also constructs the dirty tank inventory limit condition of the tank washing station and the good tank inventory limit condition of the tank washing station. The expression of the dirty tank inventory limit condition of the tank washing station is: (6), in, for t Container washing station within the time period k The storage box types are c And the box condition is a dirty box m The amount of empty containers, A is the path from the origin to the destination that can be used to transport empty containers, K For the collection of box washing stations, t’ for t Another time period before the time period, τ The time required for the box washing station. for t’ Transport mode used during time period d From the place of origin l Arrival at the container washing station k The box type is c The box condition is a dirty box m The number of empty containers; The expression of the good box inventory restriction condition of the box washing station is: (7), in, for t Container washing station within the time period k The storage box types are c And the box condition is good o The amount of empty containers, for t’ Transport mode used during time period d From the place of origin l Arrival at the container washing station k The box type is c Good box condition o The amount of empty containers, for t’ Transport mode used during time period d From the place of origin l Arrival at the container washing station k The box type is c The box condition is a dirty box m The amount of empty containers, τ kl’d For pass mode d Empty containers from the container washing point k To destination l’ The time required, for t’ + τ kl’d Time period through mode d From the box wash point k The box types shipped are c A good amount of boxes.

2. The method for allocating empty containers at multiple locations based on multiple uncertain factors according to claim 1 is characterized in that: S2 also constructs the limiting conditions for the quantity of various empty containers that can be provided by railway stations, ports, and sea-rail transport centers in each time period. The limiting conditions for the quantity of various empty containers that can be provided by railway stations, ports, and sea-rail transport centers in each time period are expressed as follows: (8), in, t’ for t Another time period before the time period, for t’ Transport mode used during time period d From destination l’ Arrival at the place of departure l The box type is c And the box condition is u The amount of empty containers, τ ll’d For pass mode d Empty containers from the place of origin l To destination l’ The time required, for t’ + τ ll’d Transport mode used during time period d From the place of origin l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, P It is a collection of ports and terminals. R A collection of railway stations. V It is a collection of sea-rail transport centers. u Indicates the box condition. U A collection of empty bin conditions, including dirty bins m Reconciliation Box o .

3. The method for allocating empty containers at multiple locations based on multiple uncertain factors according to claim 1 is characterized in that: The S2 also builds t The number of containers that can be shipped out of each type and condition within a time period is limited. t The expression for the limit condition of the number of outbound containers of each type and condition within a time period is: (9), in, τ ll’d For pass mode d Empty containers from the place of origin l To destination l’ The time required, for t + τ ll’d Transport mode used during time period d From the place of origin l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, P It is a collection of ports and terminals. R A collection of railway stations. V It is a collection of sea-rail transport centers.

4. The method for allocating empty containers at multiple locations based on multiple uncertain factors according to claim 1 is characterized in that: S2 also constructs the restriction condition that empty containers are only transported in a given logistics network. The expression of the restriction condition that empty containers are only transported in a given logistics network is: (10), for t Transport mode within the time period d The types of containers transported from the origin to the destination are c Box condition is u The number of empty containers at a time that does not belong to the time period set t The empty box quantity needs to be 0.

5. The method for allocating empty containers at multiple locations based on multiple uncertain factors according to claim 1 is characterized in that: The S2 also constructs a restriction condition that does not allow storage at the transfer station. The expression of the restriction condition that does not allow storage at the transfer station is: (11), in, For pass mode d Empty containers from the place of origin l To destination l’ The time required, δ Any time period, △ The storage time allowed at each location for multimodal transport, for t + δ + τ ll’d Transport mode used during time period d From the place of origin l Arrival at destination l’ The box type is c And the box condition is u The amount of empty containers, P It is a collection of ports and terminals. R A collection of railway stations. V It is a collection of sea-rail transport centers.