A container multimodal transport scheduling method and system based on time effectiveness

Abstract

The application discloses a container multimodal transport scheduling method and system based on timeliness, and through optimization of a whale algorithm, scheduling timeliness and scheduling cost are taken into account, and scheduling efficiency is improved.The technical scheme is as follows: starting point and target point coordinates generate a global optimal path through an improved whale algorithm capable of avoiding local optimization, and the global optimal path is used as a container multimodal transport scheduling scheme to increase diversity and improve optimization capability.An optimal whale position updating parameter is introduced to expand a local optimization range, a regulation threshold parameter is classified and adjusted according to whale fitness conditions to adjust a decay degree, and a whale position selection strategy is additionally added to replace random selection, and a whale with high fitness is selected to update a population.In the application, an adaptive multimodal transport scheduling model is established, the model considers timeliness and cost, and weights of two models are determined first.In the application, customer special requirements are considered preferentially, if there is no special requirement of the customer, an experience switching mode is established, weights of a timeliness model and a cost model are set according to historical transportation goods experience, and the target function can be determined according to specific requirements when used.

Description

TECHNICAL FIELD

[0001] The application relates to a container multimodal transport technology, in particular to a container multimodal transport scheduling method and system based on timeliness. BACKGROUND

[0002] With the development of economic globalization, the original transportation mode cannot meet the current social demand, and the flow of goods in the region is further strengthened, which directly requires higher efficiency, lower transportation cost and stronger timeliness of container transportation, so various transportation modes need to be coordinated to complete the transportation task. The container multimodal transport system can improve the integrity and efficiency of transportation, fully exert the advantages of various transportation modes, and reduce the transportation cost.

[0003] In the scheduling process of the container multimodal transport, the whale optimization algorithm is usually used, and the whale optimization algorithm is an intelligent algorithm based on meta-heuristic biological groups. The algorithm has the problem of premature convergence, that is, the population converges to a local optimum point in advance. In general, early convergence is due to the lack of global search. The premature convergence problem of the whale optimization algorithm will affect the scheduling effect of the container multimodal transport system. SUMMARY

[0004] The following gives a brief summary of one or more aspects to provide a basic understanding of these aspects. This summary is not an exhaustive overview of all contemplated aspects, and neither is it intended to identify key or critical elements of all aspects nor to delineate the scope of any or all aspects. Its only purpose is to give some concepts of one or more aspects in a simplified form as a prelude to the more detailed description given later.

[0005] The application aims to solve the above problems, and provides a container multimodal transport scheduling method and system based on timeliness. The whale optimization algorithm is optimized, the timeliness and scheduling cost are considered, and the scheduling efficiency is improved.

[0006] The technical scheme of the application is as follows: the application discloses a container multimodal transport scheduling method based on timeliness, and the method comprises the following steps:

[0007] Step 1: define the type according to the characteristics of the goods, set the detention cost and delay cost of different types of goods, define different transportation modes, set the corresponding transportation schedule and transfer cost, establish the transportation network under each transportation mode according to the transportation mode and road data owned by the city, connect each city according to the connection information between nodes to form a multimodal transport network, and update the established multimodal transport network according to the weather factors in the time demand range which cannot be transported by the transportation mode according to the influence of the weather factors on the transportation mode.

[0008] Step two: establishing an adaptive multimodal transport scheduling model, wherein the adaptive on-time target function is determined by determining the weights of the time efficiency model and the cost model;

[0009] Step three: generating a global optimal path by an improved whale algorithm capable of avoiding local optimum according to the starting point and target point coordinates, using the generated global optimal path as the multimodal transport scheduling scheme of the goods, wherein an optimal whale position updating parameter is introduced into the improved whale algorithm to expand the local optimization range, the adjustment threshold parameter is classified and adjusted according to the whale fitness condition to adjust the attenuation degree, a whale judgment operator is added to select the whale with high fitness to update the population, and then the improved whale algorithm is used to solve the adaptive multimodal transport scheduling model established in step two.

[0010] According to an embodiment of the container multimodal transport scheduling method based on time efficiency, the pre-estimation according to the weather factors within the time requirement range in step one further comprises:

[0011] Setting the decision variable of the automobile transportation mode between the nodes ij:

[0012]

[0013] Setting the decision variable of the railway transportation mode between the nodes ij:

[0014]

[0015] Setting the decision variable of the air transportation mode between the nodes ij:

[0016]

[0017] Setting the decision variable of the waterway transportation mode between the nodes ij:

[0018]

[0019] Setting the decision variable of each transportation mode between the nodes ij under the condition of extreme weather natural disasters:

[0020]

[0021] According to an embodiment of the container multimodal transport scheduling method based on time efficiency, in step two, the customer's special requirements are given priority when determining the weights, and if the customer has no special requirements, an experience switching mode is established, that is, the weights of the time efficiency model and the cost model are set according to the historical transport goods experience.

[0022] According to the container multimodal transport scheduling method based on time effectiveness provided by the application, step two further comprises: in the process of setting weights for the time effectiveness model and the cost model, an adaptive cargo punctuality objective function is determined according to specific requirements, the time effectiveness model considers cargo retention and cargo delay, and the cost model considers transportation cost, transfer cost and external cost.

[0023] According to the container multimodal transport scheduling method based on time effectiveness provided by the application, the processing steps of the improved whale optimization algorithm in step three comprise:

[0024] (1) Algorithm parameter initialization, the independent variable of the objective function is taken as the position information X of the whale individual, the population position is randomly initialized in the solution space, and parameters including the population number N, the logarithmic spiral shape constant b, the random number l, the iteration number t and the maximum iteration number T are initialized max ;

[0025] (2) Calculate the fitness of the population, denoted as g(t), find and record the optimal individual position in the population

[0026] (3) Enter the iteration stage, sort the fitness of the current population, divide the sorted population into n groups, extract a member from each group, calculate the position of the member, calculate the average fitness g(t) a , compare the difference between the fitness of the whale individual and the average fitness, select different convergence factor functions, if t max , update the average value a, the randomly changed system vector A, the system vector C, the random number l in [-1, 1] and the random number p between 0 and 1;

[0027] (4) According to the iteration number, calculate the optimal whale position update parameter a, when p r < 1, the whale position is re-determined; if A≥1, the whale individual position X is determined within the current population range, and the current whale position is updated;

[0028] (5) When p≥0.5, the whale individual position is re-determined;

[0029] (6) Record the position of the best whale individual at this time and its fitness, if t max , go to step (7); otherwise, t=t+1, repeat steps (3) to (6) until the condition is met;

[0030] (7) Output the optimal individual position and its fitness.

[0031] The application also discloses a container multimodal transport scheduling system based on timeliness.

[0032] The multimodal transport network establishing module is configured to define types according to cargo characteristic classification, set the detention cost and delay cost of different types of cargos, define different transport modes, set corresponding transport schedules and transfer cost, establish transport networks under different transport modes according to the transport modes and road data of cities, connect the cities according to the connection information between nodes, and form a multimodal transport network, and update the established multimodal transport network by pre-estimating weather factors within a time requirement range, screening transport modes that cannot be performed within the time requirement range according to the influence of the weather factors on the transport modes, and updating the established multimodal transport network.

[0033] The multimodal transport scheduling model establishing module is configured to establish an adaptive multimodal transport scheduling model, wherein the adaptive cargo punctuality objective function is determined by determining the weights of timeliness models and cost models.

[0034] The multimodal transport network scheduling model solving module is configured to generate a global optimal path by using an improved whale algorithm capable of avoiding local optimization according to the coordinates of the starting point and the target point, and use the generated global optimal path as the multimodal transport scheduling scheme of the cargo, wherein the optimal whale position updating parameter is introduced into the improved whale algorithm to expand the local optimization range, the adjustment threshold parameter is classified and adjusted according to the whale fitness to adjust the attenuation degree, the whale judgment operator is added to select the whale with high fitness to update the population, and then the improved whale algorithm is used to solve the adaptive multimodal transport scheduling model established in step two.

[0035] According to an embodiment of the container multimodal transport scheduling system based on timeliness, the pre-estimation of the weather factors within the time requirement range in the multimodal transport network establishing module further includes:

[0036] The decision variable between the nodes ij under the automobile transport mode is set as follows:

[0037]

[0038] The decision variable between the nodes ij under the railway transport mode is set as follows:

[0039]

[0040] The decision variable between the nodes ij under the air transport mode is set as follows:

[0041]

[0042] The decision variable between the nodes ij under the waterway transport mode is set as follows:

[0043]

[0044] In the case of extreme weather natural disasters, the decision variables between the nodes ij under each mode of transport are set:

[0045]

[0046] According to an embodiment of the container multimodal transport scheduling system based on time effectiveness provided by the application, in the multimodal transport scheduling model establishment module, the weight is determined by giving priority to the special requirements of the customer, and if the customer has no special requirements, an experience switching mode is established, that is, the weight of the time effectiveness model and the cost model is set according to historical transport experience.

[0047] According to an embodiment of the container multimodal transport scheduling system based on time effectiveness provided by the application, in the process of setting the weight of the time effectiveness model and the cost model in the multimodal transport scheduling model establishment module, an adaptive goods punctuality objective function is determined according to specific requirements, the time effectiveness model considers goods retention and delay, and the cost model considers transport cost, transfer cost and external cost.

[0048] According to an embodiment of the container multimodal transport scheduling system based on time effectiveness provided by the application, in the multimodal transport network scheduling model solution module, the processing steps of the improved whale algorithm include:

[0049] (1) Algorithm parameter initialization, the independent variable of the objective function is taken as the position information X of the whale individual, the population position is randomly initialized in the solution space, and parameters including the population number N, the logarithmic spiral shape constant b, the random number l, the iteration number t and the maximum iteration number T are initialized max ;

[0050] (2) Calculate the fitness of the population, denoted as g(t), find and record the optimal individual position in the population

[0051] (3) Enter the iteration stage, sort the fitness of the current population, divide the sorted population into n groups, extract a member in each group, calculate the position, calculate the average fitness g(t) a , compare the difference between the fitness of the whale individual and the average fitness, select different convergence factor functions, if t max , update the average value a, the randomly changed system vector A, the system vector C, the random number l in [-1, 1] and the random number p between 0 and 1;

[0052] (4) According to the number of iterations, the optimal whale position update parameter a is calculated, when p < 0.5, if A < 1, the whale position is re-determined; if A >= 1, the whale individual position X is determined within the current group range r , the current whale position is updated at the same time;

[0053] (5) When p >= 0.5, the whale individual position is re-determined;

[0054] (6) The position of the best whale individual at this time is recorded , and the fitness thereof, if t > T max , step (7) is entered; otherwise, t = t + 1, steps (3) to (6) are repeated until the condition is met;

[0055] (7) The optimal individual position and the fitness thereof are output.

[0056] The present application has the following beneficial effects compared with the prior art: in the scheme of the present application, the starting point and target point coordinates generate a globally optimal path through the improved whale algorithm for avoiding local optimum, which is used as a container multimodal transport scheduling scheme, so as to increase diversity and improve optimization ability. An optimal whale position update parameter is introduced to expand the local optimization range, and the adjustment threshold parameter is classified and adjusted according to the whale fitness to adjust the decay degree, and the whale position selection strategy is also added to replace random selection, and the population is updated by selecting whales with high fitness. In the present application, an adaptive multimodal transport scheduling model is established, which considers time efficiency and cost, determines the target function, and first determines the weights of the two models. The present application also gives priority to customer special needs, and if the customer has no special needs, the present application establishes an experience switching mode. According to historical transportation experience, the weights of the time efficiency model and the cost model are set. The target function can be determined according to specific requirements. The time efficiency model considers goods retention and delay, and the cost model considers transportation cost, transfer cost and external cost. This adaptive multimodal transport scheduling model not only considers customer special needs, but also considers time efficiency and cost, and has practical significance. BRIEF DESCRIPTION OF DRAWINGS

[0057] The above features and advantages of the present application can be better understood after reading the detailed description of embodiments of the present application in conjunction with the following drawings. In the drawings, components are not necessarily drawn to scale, and components having similar related properties or features can have the same or similar reference numerals.

[0058] Figure 1 A flowchart of an embodiment of a container multimodal transport scheduling method based on time efficiency of the present application is shown.

[0059] Figure 2 A flowchart of the improved whale optimization algorithm in the illustrated embodiment is shown. Figure 1 A flowchart of the improved whale optimization algorithm in the illustrated embodiment is shown.

[0060] Figure 3 A schematic diagram of an embodiment of the container multimodal transport scheduling system based on time effectiveness of the present application is shown. DETAILED DESCRIPTION

[0061] The present application will be described in detail below with reference to the accompanying drawings and specific embodiments. Note that the aspects described below in conjunction with the accompanying drawings and specific embodiments are only exemplary and should not be understood as limiting the scope of protection of the present application in any way.

[0062] Before describing the solutions of the present application, the symbols involved in the text and their meanings are explained as follows.

[0063] a represents the type of goods;

[0064] b represents the mode of transport;

[0065] represents the occurrence of a mode of transport conversion at node i, converted to mode of transport b;

[0066] represents the unit detention time penalty cost of a type of goods a;

[0067] represents the unit delay time penalty cost of a type of goods a;

[0068] NT b represents the next departure time of mode of transport b;

[0069] T represents the specified arrival time of the goods.

[0070] is the actual traffic accident risk probability (times / vehicle) of a specific c type of traffic mode within a unit link;

[0071] C a is the corresponding loss cost of c specific type of accident within a unit link;

[0072] c=1,2,3 respectively represent minor traffic accidents, ordinary traffic accidents, fatal traffic accidents.

[0073] direct economic loss cost in traffic accidents.

[0074] C n : management or rescue cost of traffic law enforcement departments, fire departments, public security departments, insurance companies and rescue centers in traffic accidents;

[0075] Ct : medical cost of injured person in traffic accident;

[0076] : delay cost of goods in traffic accident;

[0077] : decision variable indicating whether there is a b transportation mode from node i to node j;

[0078] : unit transportation cost of using b transportation mode from node i to node j;

[0079] : distance of using b transportation mode from node i to node j;

[0080] n: number of containers;

[0081] : decision variable indicating changing transportation mode from b to c at node i;

[0082] : unit transshipment cost of changing transportation mode from b to c at node i;

[0083] c ld : unit container loading and unloading cost.

[0084] Figure 1 The flow of an embodiment of the time-sensitive based container multimodal transport scheduling method of the present application is shown. Please refer to Figure 1 The implementation steps of the method of the present embodiment are described as follows.

[0085] Step 1: define types according to cargo characteristic classification, set the delay cost and delay cost of different types of goods, define different transportation modes, set corresponding transportation schedule and transshipment cost, etc.; according to the known transportation modes and road data (these transportation modes and road data include transportation path and transshipment point set) of the city, establish the transportation network under each transportation mode, connect each city according to the connection information between nodes to form a multimodal transport network; according to the weather factors within the time demand range, make prior estimation, according to the influence of weather factors on transportation mode, screen the transportation modes that cannot be carried out within the time demand range, and update the established multimodal transport network.

[0086] In the traditional intermodal network, the consideration of weather factors is applied to calculate the carbon emission cost to measure the economic loss. In this embodiment, the consideration of weather factors is to measure the delay cost of weather on the transportation mode. Weather factors are another important external factor that affects the intermodal process. First, the selection of transportation mode will be affected by weather factors, such as bad weather leading to the termination of air transportation. Therefore, in this embodiment, a prior estimation scheme is introduced to screen the transportation modes that cannot be performed within the time demand range according to the influence of weather factors on the transportation mode, so as to update the intermodal network.

[0087] At the same time, because of the prior estimation screening, it can be ensured that there is no situation that the next link transportation cannot be obtained within the demand range when the subsequent intermodal scheduling scheme selection is performed.

[0088] The decision variables according to the prior estimation of weather factors are as follows.

[0089] Rain (snow) weather, wet road surface, reduced friction coefficient of car, easy to cause brake out of control, cause vehicle skidding collision accident, even cause big traffic accident.

[0090]

[0091] Rail transportation is less affected by climate. As long as the climate is not too bad, rail transportation can be regular, accurate and on time. As long as the railway can be built, transportation can be carried out.

[0092]

[0093] Fog weather, visibility decreases, driver's judgment of vehicle spacing is difficult, potential danger is difficult to discover, and traffic signs and road facilities are difficult to identify effectively, resulting in traffic accidents.

[0094] Air transportation is greatly affected by climate. Aircraft generally take off and land against the wind, so that the aircraft has great lift and the shortest sliding distance, which is the safest. If the wind speed is too large, it is not safe to take off and land against the wind. When the snowfall reaches above moderate snow, the pilot cannot guarantee the safety of the flight due to too low visibility. Heavy snow will cause snow accumulation on the aircraft body, and once the snow accumulates and freezes, it will cause the internal parts such as the engine inlet duct to freeze, affecting flight safety.

[0095]

[0096] Water transportation is greatly affected by natural conditions. Inland waterways and some ports are greatly affected by seasons. In winter, the water level is low in dry season, and it is difficult to ensure all-year navigation. Moreover, transportation is difficult in heavy storm and rain weather.

[0097]

[0098] At the same time, extreme weather natural disasters will cut off the transport network and signal, affect the mode of transport.

[0099]

[0100] Step two: establish an adaptive intermodal transport scheduling model, the adaptive intermodal transport scheduling model considers the transport timeliness and transport cost, the adaptive goods punctuality objective function first to determine the timeliness model and cost model weight.

[0101] In determining the weight, priority is given to customer special needs, if the customer has no special requirements, the embodiment establishes an experience switching mode, that is, according to the historical transport goods experience (historical transport goods experience including goods type and customer demand time level), the weight of the timeliness model and the cost model is set, and the adaptive goods punctuality objective function is determined according to the specific requirements. The timeliness model considers the goods retention and delay, and the cost model considers the transportation cost, transfer cost and external cost.

[0102] The specific processing process of step two is as follows.

[0103] First, for the intermodal transport of goods, the consignor mainly considers the minimum value of the transportation cost; the consignee pays more attention to the timeliness of the goods, because saving time will produce higher time value, and the goods may have time limit, and the consignee benefit can be guaranteed. However, the transportation cost and the transportation timeliness are in a certain extent contradictory relationship, therefore, the scheme comprehensively considers the goods type and the time demand of the consignor, in the actual intermodal transport, the adaptive strategy is introduced, the weight parameter is set according to the actual demand level, so as to select the timeliness model or the cost minimization model, the expression function is:

[0104] W = ω1f (t) + ω2C (1)

[0105] Where ω1 and ω2 are weight parameters, ω1 + ω2 = 1, ω1, ω2 ∈ [0, 1]

[0106] In the above formula (1), W represents the adaptive intermodal transport scheduling objective function, f (t) represents the goods transport punctuality objective function, and C represents the total cost.

[0107] In the scheme, for the specific setting of the weight parameters ω1 and ω2 in the total model (i.e. the adaptive intermodal transport scheduling model), first, whether the customer has special requirements is considered, if the customer has requirements, the customer's setting is adopted, if not, the scheme establishes an experience switching mode function. According to the historical transport goods experience, the weight of the timeliness model and the cost model is set. The specific operation is as follows:

[0108] (1) Index evaluation on goods type and customer demand time.

[0109] (2) Normalization of data to obtain normalized matrix [U ij ] n×2 , n represents the number of historical transportation goods records.

[0110] (3) Calculation of attribute weight, the weight of the jth item is

[0111]

[0112] In the above formula (2), u ij represents the value of the ith row and jth column in the normalized matrix, that is, the attribute weight information corresponding to a historical transportation goods record.

[0113] Therefore, the objective function of the adaptive multimodal transport switching mode of the present scheme can be defined as:

[0114]

[0115] In the above formula (3), f(t) represents the goods transportation punctuality objective function, C represents the total cost, and ω0 represents the customer requirement.

[0116] Then a new transportation punctuality model is established, and the risk cost of waiting for the arrival of the class and the delay penalty cost of different goods transportation modes are introduced.

[0117] Due to the special value nature and time value of different goods, especially for some high-value time-sensitive goods such as vaccines and medicines, transportation needs to consider punctuality first, so a transportation punctuality model needs to be established. In order to be more in line with the actual situation, first, different goods need to be classified and the corresponding detention time penalty cost and delay time penalty cost need to be established on the basis of different goods types.

[0118] In the multimodal transport scheduling problem, for transportation punctuality, two aspects can be considered: first, if transportation mode conversion occurs at the node, the class needs to be considered, such as conversion to waterway, aviation, and railway class time fixed transportation mode, then the risk cost of waiting for the arrival of the class is introduced; second, the goods itself has a specified arrival time, and the delay penalty cost needs to be introduced. Therefore, the goods transportation punctuality objective function f(t) can be defined as:

[0119]

[0120] Among them, the total cost needs to consider the transportation cost, the transfer cost and the external cost. Its expression is:

[0121] Total cost C = C1 + C2 + C3 (5)

[0122]

[0123]

[0124] In the above formula, M and B in c∈B represent the set of nodes in the transportation network and the set of transportation modes, respectively.

[0125] In actual transportation, there are a large number of external costs that will affect the multimodal transport process. One of the important factors is the cost of traffic accidents, which not only causes direct material property losses such as transport tools and goods in the multimodal transport process, but also affects transportation activities due to traffic delays.

[0126]

[0127] In the above formula, d ij represents the distance of node ij.

[0128] For the selection of transportation modes, the application introduces a selection method considering operation frequency and waiting time, which is described in detail as follows.

[0129] Different transportation modes have different transportation frequencies, i.e., there are shifts, especially for water transport, railway, and aviation, which have long waiting time. Highway transportation has high frequency and strong flexibility. Therefore, when encountering a waiting situation, the time required for switching to highway transportation can be measured to improve timeliness. The expression is as follows:

[0130]

[0131] In the above formula, t1 and v1 represent the time required for highway transportation and the speed of highway transportation, respectively.

[0132] The traffic accident cost is introduced in the scheme, mainly including the following aspects: direct economic loss cost, management cost, medical cost, delay cost, and the accident type is divided into levels. Considering that the severity of traffic accidents is different, the cost borne is greatly different, so the economic loss borne by different traffic accident types needs to be considered when modeling. In order to reduce calculation error, the traffic accident type is divided into three categories for calculation, minor accident (c=1): only the type of traffic accident with property loss; ordinary accident (c=2): traffic accident with personnel injury and property loss; fatal accident (c=3): traffic accident with death and serious property loss.

[0133]

[0134] Highway accidents account for the vast majority of traffic accidents, and railway and waterway transportation accidents are relatively much less, and aviation transportation accident rate is the lowest, so the accident probability of different transportation modes is different.

[0135] The total traffic accident cost C3 in multimodal transport can be expressed as:

[0136]

[0137]

[0138] Therefore, the expression of the minimum objective function is:

[0139]

[0140] The constraint conditions of the minimum objective function are:

[0141] (1) The starting point to the end point is a complete path

[0142]

[0143]

[0144]

[0145] (2) Only one mode of transport can be selected when transferring between two nodes

[0146]

[0147] (3) The number of containers does not exceed the maximum capacity of the transport mode

[0148]

[0149] (4) At most one transfer at the node

[0150]

[0151] In the above formula, m represents the transport node, r s represents the starting point, r e represents the end point, represents the maximum capacity when using mode b between nodes ij.

[0152] Step three: According to the starting point and target point coordinates, a global optimal path is generated by the improved whale optimization algorithm to avoid local optimization, and the global optimal path generated is used as the multimodal transport scheduling scheme of the goods. In the improved whale optimization algorithm, the optimal whale position update parameter is introduced to expand the local optimization range, the adjustment threshold parameter is classified and adjusted according to the whale fitness condition to adjust the decay degree, and the whale judgment operator is added to select the whale with high fitness to update the population, and then the improved whale optimization algorithm is used to solve the multimodal transport scheduling model

[0153] Whale optimization algorithm is a meta-heuristic intelligent algorithm based on biological population, which has the problem of premature convergence, that is, the population converges to a local optimal point in advance. Generally, early convergence is due to the lack of global search. The embodiment optimizes the traditional whale optimization algorithm to increase diversity and improve optimization ability.

[0154] First, in the whale optimization algorithm, the coefficient vector is used to represent the ability of local search and global search. With the linear reduction of the convergence factor, the later search is easy to fall into the local optimal situation. Whale optimization is updated based on the position of the optimal whale when updating, and the closer to the optimal position, the smaller the disturbance to the whale. In global optimization, in order to make the whale move greatly, the convergence factor should be set to low attenuation degree; in local optimization, in order to make the whale move slightly, the convergence factor should be set to high attenuation degree. The fitness of the whale individual and the average fitness of the whale are calculated, and the individual fitness is lower than the average fitness, which is the better whale. Therefore, according to the fitness change of the whale, the newly defined convergence factor can balance the diversity of the global search ability in the early stage and the convergence speed of the local search in the later stage.

[0155]

[0156] Therefore, the improved adjustment threshold of whale optimization algorithm is

[0157]

[0158]

[0159] In the above formula:

[0160] s t represents the current convergence factor,

[0161] s0 represents the initial convergence factor,

[0162] t represents the current iteration number,

[0163] T max represents the maximum iteration number,

[0164] g(t) represents the population fitness,

[0165] g(t) a represents the average population fitness,

[0166] represents a random vector between 0 and 1,

[0167] represents the coefficient vector.

[0168] Secondly, the whale algorithm realizes local optimization in the operation of surrounding the prey and bubble net attack. At this time, when the whale approaches the local optimal solution, it can only approach the local optimal solution and cannot change the range to better local optimization, which is easy to converge too fast. If the whale approaches the food, the position of the optimal whale is updated according to the iteration number of the whale itself, which can improve the local optimization ability of the whale. Therefore, a new optimal whale position update parameter η is introduced, so that the whale can move randomly in a small range to promote local optimization:

[0169] η∈[-α,α] (22)

[0170]

[0171] In the above formula, tanh represents the hyperbolic tangent function, and T is the maximum number of iterations.

[0172] Therefore, in the bubble net attack stage, the contraction mechanism and spiral position update expression are updated according to the iteration number of the whale itself:

[0173]

[0174] In the above formula:

[0175] represents the updated whale position,

[0176] η optimal whale position update parameter,

[0177] represents the distance from the i-th whale to the prey (the best solution obtained so far),

[0178] e bl b is a constant that defines the shape of the logarithmic spiral, and l is a random number in [-1, 1],

[0179] p is a random number between 0 and 1,

[0180] is the position vector of the best solution obtained so far,

[0181] l is a random number in [-1, 1].

[0182] Again, in the contraction mechanism, according to the fitness change of the whale, a grouping selection mechanism is introduced instead of random selection, and the grouping selection steps are as follows: first, sort the population according to the fitness value; then divide the sorted population into n groups. After that, a member is randomly selected from each group, and the following formula is used to calculate instead of random selection.

[0183]

[0184]

[0185] In the above equation:

[0186] represents a randomly selected position,

[0187] represents the current best position,

[0188] represents the current position.

[0189] The above equation is related to the surrounded prey, which is considered as the best target, while other whales try to update their positions relative to this better whale. Among them, is the position of the selected member in the group. t is the current iteration number, T max is the maximum iteration number, r is a dynamic random number between 0 and 1.

[0190] Then, in the original whale algorithm, the selection of search predation method is based on probability random. For a period of time, some prey may escape from these two hunting mechanisms, so a new random position selection method is introduced to improve the population diversity and its search ability.

[0191]

[0192]

[0193]

[0194] where, and are the maximum and minimum values in the search domain of whale individuals, X r are uniformly distributed before and

[0195] In the above equation:

[0196] X i p , respectively represent the arithmetic mean position, the square mean position,

[0197] X r a randomly selected position,

[0198] X i the current position.

[0199] Therefore, the search predation expression is updated as:

[0200] ​

[0201] The intermodal transportation problem can be simply represented by a directed graph G = (V, E) with n points, where V = {1, 2, …, n} and E = {(i, j), i, j ∈ V}, and the objective is to solve the shortest path problem from the origin to the destination by traversing all points.

[0202] As shown in Figure 2 , the specific operation steps of the intermodal transportation problem, that is, the processing steps of the improved whale optimization algorithm, are as follows.

[0203] (1) Algorithm parameter initialization, the independent variable of the objective function is taken as the position information X of the whale individual, and the population position is randomly initialized in the solution space, and the parameters are initialized, including the population number N, the logarithmic spiral shape constant b, the random number l, the iteration number t, and the maximum iteration number T max .

[0204] (2) Calculate the fitness of the population, denoted as g(t), find and record the optimal individual position in the population

[0205] (3) Enter the iteration stage, sort the current population fitness, divide the sorted population into n groups, extract a member from each group, calculate its position, calculate the average fitness g(t) a , compare the difference between the fitness of the whale individual and the average fitness, and select different convergence factor functions, if t < T max , update the average value a, the randomly changing system vector A, the system vector C, the random number l in [-1, 1], and the random number p between 0 and 1.

[0206] (4) According to the iteration number, calculate the optimal whale position update parameter a, when p < 0.5, if A < 1, re-determine the whale position by formula (24); if A ≥ 1, the whale individual position X r needs to be determined within the current group range by (27), and the current whale position is updated by formula (30).

[0207] (5) When p ≥ 0.5, re-determine the whale individual position by formula (24).

[0208] (6) Record the position and fitness of the best whale individual at this time. If t > T max , go to step (7); otherwise, t = t + 1, repeat steps (3) to (6) until the condition is met.

[0209] (7) Output the optimal individual position and its fitness.

[0210] Figure 3 The principle of an embodiment of the container multimodal transport scheduling system based on time effectiveness of the present application is shown. Please refer to Figure 1 The system of the present embodiment includes: a multimodal transport network establishment module, a multimodal transport scheduling model establishment module, and a multimodal transport network scheduling model solving module.

[0211] The multimodal transport network establishment module is configured to: define types according to cargo characteristic classification, set the detention cost and delay cost of different types of cargo; define different transport modes, set corresponding transport schedules and transfer costs; establish a transport network under each transport mode according to the transport modes and road data possessed by the city, connect each city according to the connection information between nodes to form a multimodal transport network; make a prior estimate according to weather factors within the time demand range, filter the transport modes that cannot be carried out within the time demand range according to the influence of weather factors on transport modes, and update the established multimodal transport network accordingly.

[0212] The prior estimate according to weather factors within the time demand range in the multimodal transport network establishment module further includes:

[0213] Set the decision variable between nodes ij under automobile transport mode:

[0214]

[0215] Set the decision variable between nodes ij under railway transport mode:

[0216]

[0217] Set the decision variable between nodes ij under air transport mode:

[0218]

[0219] Set the decision variable between nodes ij under water transport mode:

[0220]

[0221] Under the condition of extreme weather natural disasters, set the decision variable between nodes ij under each transport mode:

[0222]

[0223] The remaining details of the multimodal transport network establishment module are similar to step one in the foregoing method embodiment, and will not be repeated here.

[0224] The multi-modal transport scheduling model establishing module is configured to establish an adaptive multi-modal transport scheduling model, wherein the adaptive freight punctuality objective function is determined by first determining the weights of the timeliness model and the cost model.

[0225] In the multi-modal transport scheduling model establishing module, the weights are determined by giving priority to the special requirements of the customer, and if the customer has no special requirements, an experience switching mode is established, i.e., the weights of the timeliness model and the cost model are set according to historical transport experience. In the process of setting the weights of the timeliness model and the cost model, the adaptive freight punctuality objective function is determined according to specific requirements, the timeliness model considers freight retention and delay, and the cost model considers transport cost, transfer cost and external cost.

[0226] The remaining details of the multi-modal transport scheduling model establishing module are similar to step two in the foregoing method embodiment, and will not be described here again.

[0227] The multi-modal transport network scheduling model solving module is configured to generate a globally optimal path according to the coordinates of the starting point and the target point by using an improved whale algorithm capable of avoiding local optimization, and use the generated globally optimal path as the multi-modal transport scheduling scheme of the freight, wherein an optimal whale position updating parameter is introduced into the improved whale algorithm to expand the local optimization range, the adjustment threshold parameter is classified and adjusted according to the fitness of the whale to adjust the decay degree, a whale judgment operator is added to select a whale with high fitness to update the population, and then the improved whale algorithm is used to solve the adaptive multi-modal transport scheduling model established in step two.

[0228] In the multi-modal transport network scheduling model solving module, the processing steps of the improved whale algorithm include:

[0229] (1) Algorithm parameter initialization, the independent variable of the objective function is taken as the position information X of the whale individual, the population position is randomly initialized in the solution space, and parameters including the population number N, the logarithmic spiral shape constant b, the random number l, the iteration number t and the maximum iteration number T are initialized max ;

[0230] (2) Calculate the fitness of the population, denoted as g(t), find and record the optimal individual position in the population

[0231] (3) Enter the iteration stage, sort the current population fitness, divide the sorted population into n groups, extract a member from each group, calculate the position of the member, calculate the average fitness g(t) a , compare the difference between the fitness of the whale individual and the average fitness, and select different convergence factor functions, if t max, update the average value a, the random change system vector A, the system vector C, the random number I in [-1, 1], the random number p between 0 and 1;

[0232] (4) According to the number of iterations, calculate the optimal whale position update parameter a, when p < 0.5, if A < 1, re-determine the whale position; if A ≥ 1, determine the whale individual position X within the current population range r , update the current whale position;

[0233] (5) When p ≥ 0.5, re-determine the whale individual position;

[0234] (6) Record the position of the best whale individual at this time and its fitness, if t > T max , go to step (7); otherwise, t = t + 1, repeat steps (3) to (6) until the condition is met;

[0235] (7) Output the optimal individual position and its fitness.

[0236] The remaining details of the multimodal transport network scheduling model are similar to step three in the foregoing method embodiment, and are not described here.

[0237] Although the above methods are illustrated and described as a series of acts, it will be understood and appreciated that the methods are not limited by the order of acts, as some acts may, in accordance with one or more embodiments, occur simultaneously or in different order than shown and described herein, or may be omitted entirely, depending on the circumstances.

[0238] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality, without limitation. Whether such functionality is implemented as hardware or software depends on the particular application and design constraints imposed on the overall system. Skilled artisans can implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

[0239] The various illustrative logical blocks, modules, and circuits described in connection with the embodiments disclosed herein can be implemented or performed with a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor can be a microprocessor, but in the alternative, the processor can be any conventional processor, controller, microcontroller, or state machine. A processor can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.

[0240] The steps of a method or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor. The processor and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor and the storage medium can reside as discrete components in a user terminal.

[0241] In one or more exemplary embodiments, the functions described can be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functions can be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media can be any available media that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. Disk and disc, as used herein, includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0242] The previous description of the disclosure is provided to enable any persons skilled in the art to make or use the disclosure. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other variations without departing from the spirit or scope of the disclosure. Thus, the disclosure is not intended to be limited to the example and designs described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A time-sensitive container multimodal transport scheduling method, characterized in that the method... include: Step 1: Define cargo types based on cargo characteristics, and set demurrage and delay costs for different cargo types; define different modes of transport, and set corresponding transport schedules and transshipment costs; establish a transport network for each mode of transport based on the transport modes and road data available in each city, and connect the cities based on the connectivity information between nodes to form a multimodal transport network; make advance predictions based on weather factors within the time demand range, and filter out transport modes that cannot be carried out within the time demand range based on the impact of weather factors on transport modes, thereby updating the established multimodal transport network; Step 2: Establish an adaptive multimodal transport scheduling model, in which determining the adaptive cargo timeliness objective function first requires determining the weights of the timeliness model and the cost model; Step 3: Based on the coordinates of the starting point and the target point, a globally optimal path is generated using an improved whale algorithm that avoids local optima. The generated globally optimal path is used as the multimodal transport scheduling scheme for the goods. In the improved whale algorithm, the optimal whale position update parameter is introduced to expand the local optimization range. At the same time, the decay degree of the adjustment threshold parameter is adjusted according to the whale fitness. A whale judgment operator is also added to select whales with high fitness to update the population. Then, the improved whale algorithm is used to solve the adaptive multimodal transport scheduling model established in Step 2. In step two, the weighting of the timeliness model and cost model based on historical cargo transportation experience is performed as follows: (1) Evaluate the indicators of goods type and customer demand time; (2) Normalize the data to obtain the normalized matrix. n represents the number of historical transport cargo records; (3) Calculate the attribute weights, and obtain the weight of the j-th item as follows: ， In the above formula, This represents the value in the i-th row and j-th column of the normalized matrix, which is the attribute weight information corresponding to a certain historical cargo transport record; The objective function for adaptive multimodal transport switching mode is defined as: ， In the above formula, The objective function representing the timeliness of goods transportation is... Represents the total cost. This indicates the customer's request.

2. The time-sensitive container multimodal transport scheduling method according to claim 1, characterized in that, Step one, which involves making advance forecasts based on weather factors within the time frame, further includes: Define the decision variables between nodes ij under the vehicle transportation method: ； Set the decision variables between nodes ij under the railway transportation mode: ； Set the decision variables between nodes ij under the air transport mode; ； Set the decision variables between nodes ij under the waterway transportation mode: ； In the event of extreme weather or natural disasters, set the decision variables between nodes ij under each mode of transportation: 。 3. The time-sensitive container multimodal transport scheduling method according to claim 1, characterized in that, In step two, when determining the weights, priority is given to the customer's special needs. If the customer has no special needs, an experience-based switching mode is established, that is, the weights of the timeliness model and the cost model are set based on historical experience in transporting goods.

4. The time-sensitive container multimodal transport scheduling method according to claim 1, characterized in that, Step two further includes: in the process of setting weights for the timeliness model and the cost model, an adaptive cargo timeliness objective function is determined according to specific requirements. The timeliness model considers cargo waiting and cargo delays, while the cost model considers transportation costs, transshipment costs, and external costs.

5. The time-sensitive container multimodal transport scheduling method according to claim 1, characterized in that, The improved whale algorithm in step three includes the following processing steps: (1) Initialize the algorithm parameters by taking the independent variable of the objective function as the position information of the individual whale. Within the solution space, the population locations are randomly initialized, along with parameters including the population size. Logarithmic spiral shape constant Random numbers Number of iterations Maximum number of iterations ; (2) Calculate the fitness of the population, denoted as . Find and record the location of the optimal individual in the population. ; (3) Enter the iteration phase, sort the fitness of the current population, divide the sorted population into n groups, draw one member from each group, calculate its position, and calculate the average fitness. By comparing the differences between individual whale fitness and average fitness, different convergence factor functions can be selected. Update the average value Randomly varying system vectors System vector Random numbers in [-1,1] Random numbers between 0 and 1 ; (4) Calculate the optimal whale position update parameters based on the number of iterations. ,when At that time, if Then the whale's location will be re-determined; if Then the location of an individual whale can be determined within the current group. At the same time, update the current whale location; (5) When At that time, the location of the individual whale was re-determined; (6) Record the location of the best individual whale at this time. and its fitness, if If yes, proceed to step (7); otherwise, Repeat steps (3) to (6) until the condition is met; (7) Output the optimal individual position And its adaptability.

6. A time-sensitive container multimodal transport scheduling system, characterized in that the system... include: The multimodal transport network establishment module is configured as follows: It classifies and defines types of goods based on their characteristics, setting demurrage and delay costs for different types of goods; it defines different modes of transport, setting corresponding transport schedules and transshipment costs; it establishes transport networks for each mode of transport based on the transport modes and road data available in each city; it connects various cities based on the connectivity information between nodes to form a multimodal transport network; it performs advance forecasting based on weather factors within the time demand range, and filters out transport modes that cannot be carried out within the time demand range based on the impact of weather factors, thereby updating the established multimodal transport network. The multimodal transport scheduling model building module is configured to build an adaptive multimodal transport scheduling model. Determining the adaptive cargo timeliness objective function first requires determining the weights of the timeliness model and the cost model. The multimodal transport network scheduling model solution module is configured to generate a globally optimal path based on the coordinates of the starting point and the destination point using an improved whale algorithm that avoids local optima. The generated globally optimal path is used as the multimodal transport scheduling scheme for goods. The improved whale algorithm introduces an optimal whale position update parameter to expand the local optimization range. At the same time, the attenuation degree of the adjustment threshold parameter is adjusted according to the whale fitness. A whale judgment operator is also added to select whales with high fitness to update the population. Then, the improved whale algorithm is used to solve the adaptive multimodal transport scheduling model established in step two. In the multimodal transport scheduling model building module, the operation of setting weights for the timeliness model and cost model based on historical cargo transport experience is as follows: (1) Evaluate the indicators of goods type and customer demand time; (2) Normalize the data to obtain the normalized matrix. n represents the number of historical transport cargo records; (3) Calculate the attribute weights, and obtain the weight of the j-th item as follows: ， In the above formula, This represents the value in the i-th row and j-th column of the normalized matrix, which is the attribute weight information corresponding to a certain historical cargo transport record; The objective function for adaptive multimodal transport switching mode is defined as: ， In the above formula, The objective function representing the timeliness of goods transportation is... Represents the total cost. This indicates the customer's request.

7. The time-sensitive container multimodal transport scheduling system according to claim 6, characterized in that, The multimodal transport network establishment module further includes the pre-assessment of weather factors within the time demand range, which includes: Define the decision variables between nodes ij under the vehicle transportation method: ； Set the decision variables between nodes ij under the railway transportation mode: ； Set the decision variables between nodes ij under the air transport mode; ； Set the decision variables between nodes ij under the waterway transportation mode: ； In the event of extreme weather or natural disasters, set the decision variables between nodes ij under each mode of transportation: 。 8. The time-sensitive container multimodal transport scheduling system according to claim 6, characterized in that, In the multimodal transport scheduling model building module, customer special needs are given priority when determining weights. If the customer has no special needs, an experience switching mode is established, that is, weights are set for the timeliness model and the cost model based on historical cargo transportation experience.

9. The time-sensitive container multimodal transport scheduling system according to claim 6, characterized in that, In the multimodal transport scheduling model building module, when setting weights for the timeliness model and the cost model, an adaptive cargo timeliness objective function is determined according to specific requirements. The timeliness model considers cargo waiting and cargo delays, while the cost model considers transportation costs, transshipment costs, and external costs.

10. The time-sensitive container multimodal transport scheduling system according to claim 6, characterized in that, In the multimodal transport network scheduling model solution module, the improved whale algorithm's processing steps include: (1) Initialize the algorithm parameters by taking the independent variable of the objective function as the position information of the individual whale. Within the solution space, the population locations are randomly initialized, along with parameters including the population size. Logarithmic spiral shape constant Random numbers Number of iterations Maximum number of iterations ; (2) Calculate the fitness of the population, denoted as . Find and record the location of the optimal individual in the population. ; (3) Enter the iteration phase, sort the fitness of the current population, divide the sorted population into n groups, draw one member from each group, calculate its position, and calculate the average fitness. By comparing the differences between individual whale fitness and average fitness, different convergence factor functions can be selected. Update the average value Randomly varying system vectors System vector Random numbers in [-1,1] Random numbers between 0 and 1 ; (4) Calculate the optimal whale position update parameters based on the number of iterations. ,when At that time, if Then the whale's location will be re-determined; if Then the location of an individual whale can be determined within the current group. At the same time, update the current whale location; (5) When At that time, the location of the individual whale was re-determined; (6) Record the location of the best individual whale at this time. and its fitness, if If yes, proceed to step (7); otherwise, Repeat steps (3) to (6) until the condition is met; (7) Output the optimal individual position And its adaptability.

Images