Dangerous goods transportation path optimization method
By calculating the differentiation and risk of dangerous goods transportation paths, establishing a dual-objective optimization model, and optimizing vehicle scheduling using non-dominant sorting genetic algorithms, solving the problem of path risks superposition in the existing technology, and achieving safer and more effective transportation of dangerous goods.
Patent Information
- Application Number
- CN202510219903.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-26
- Publication Date
- 2025-07-11
AI Technical Summary
The prior art often chooses the shortest path in the transportation of dangerous goods, resulting in the superposition of path risks, lack of optimization of transportation paths, and it is impossible to minimize risks and maximize path differences at the same time, resulting in unsafe and inefficient transportation.
By calculating the differentiation and risk between dangerous goods transport paths, a two-objective optimization model is established, and the model is solved using the non-dominant sorting genetic algorithm to obtain the optimal path, and vehicle scheduling and path selection are optimized.
Significantly reduce the risk of transporting dangerous goods, increase path differentiation, and achieve safer and more effective transportation of dangerous goods.
Smart Images
Figure CN120297842A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of path planning, and more particularly, to a method for optimizing the transportation path of dangerous goods. Background Art
[0002] With the rapid growth of the economy and the acceleration of the industrialization process, the demand and production volume of hazardous chemicals have been increasing year by year. In recent years, the hazardous goods industry in China has developed vigorously, which has promoted the rapid growth of the demand for hazardous goods logistics, and the market scale of the industry has continued to expand. According to the data, between 2015 and 2022, the transportation volume of hazardous chemicals in China increased from 1.355 billion tons to 2.03 billion tons, and this growth trend highlights the vitality and potential of the hazardous goods logistics industry. In the transportation network system of dangerous goods, transportation enterprises are faced with a key problem: how to select a suitable path to safely transport dangerous goods from the warehouse to the demand point. Selecting different transportation paths will not only affect the risk level of goods transportation, but also change the dissimilarity between transportation paths. When designing the transportation path plan of dangerous goods, transportation enterprises need to consider two goals simultaneously: one is to minimize the risk of goods transportation, and the other is to maximize the dissimilarity of paths.
[0003] However, traditional transportation enterprises often choose the shortest transportation path to reduce costs. Although this approach is simple, it will lead to the superposition of path risks. With the increase in the number of transportation vehicles, the mutual influence between vehicles and the possibility of potential harm will also increase. Only by optimizing the driving route and scientifically dispatching vehicles can the transportation cost and total risk of completing all transportation tasks be minimized, while the number of allocated vehicles is the least and the fairness of vehicle transportation tasks is ensured. Therefore, how to optimize the path and how to optimize vehicle dispatching to achieve safer and more efficient transportation of dangerous goods are the most urgent technical problems to be solved currently. Summary of the Invention
[0004] The purpose of the present invention is to provide a method for optimizing the transportation path of dangerous goods, which can significantly reduce the transportation risk of dangerous goods, improve the transportation dissimilarity, and achieve safer and more efficient transportation of dangerous goods.
[0005] The technical solution of the present invention is as follows:
[0006] The present application provides a method for optimizing the transportation path of dangerous goods, which includes the following steps:
[0007] S1. Obtain the node data of the transportation network and calculate the dissimilarity between different paths of dangerous goods transportation;
[0008] S2. Calculate the risk degree of simultaneously transporting dangerous goods on different paths based on the dissimilarity between different paths;
[0009] S3. Establish a bi-objective optimization model with the maximum dissimilarity between different paths of dangerous goods transportation and the minimum risk degree of simultaneous transportation on different paths as the objectives;
[0010] S4. Solve the bi-objective optimization model based on the non-dominated sorting genetic algorithm to obtain the optimal solution, and output the optimal solution to get the best path for dangerous goods transportation.
[0011] Further, in step S1, the expression for calculating the dissimilarity between different paths of dangerous goods transportation includes:
[0012]
[0013] In the formula, a represents the a-th path, b represents another path different from a, w is the buffer radius of the vehicle's impact on the surrounding environment, dis(a, b, w) is the dissimilarity between path a and path b, is the length of the road section (i, j) in the a-th path, is the length of the road section (i, j) in the b-th path, N is the total length of the road sections, K is the set of paths, is the road section (i, j) in the a-th path, is the road section (i, j) in the b-th path, i is the starting point of the road section, and j is the ending point of the road section.
[0014] Further, in step S2, the process of calculating the risk degree of simultaneous transportation of dangerous goods on different paths includes:
[0015]
[0016] In the formula, r(a, b) represents the risk degree that path a poses to path b, is the probability of no accident occurring on path a, pop b is the population exposure of path b, e -ε[d(a,b)]2 is the distance-sensitive damage function, where ε is the influence coefficient of the dangerous goods, d(a, b) is the Euclidean distance between paths a and b, represents a certain road section in path b of the risk degree, r b represents the total risk degree of path b, r a,b represents the risk degree when transporting simultaneously on paths a and b, a represents the a-th path, b represents another path different from a, is the length of the road section (i, j) in the a-th path, is the length of the road section (i, j) in the b-th path, N is the total length of the road sections, K is the set of paths, is the road section in the a-th path, is the road section in the b-th path, i is the starting point of the road section, and j is the ending point of the road section.
[0017] Furthermore, in step S3, the calculation formula of the above-mentioned bi-objective optimization model includes:
[0018]
[0019] In the formula, maxf1 is the maximum dissimilarity, a represents the a-th path, b represents another path different from a, is the length of the road section (i, j) in the a-th path, is the length of the road section (i, j) in the b-th path, N is the total length of the road sections, K is the set of paths, is the road section in the a-th path, is the road section in the b-th path, w is the buffer radius of the vehicle's impact on the surrounding environment, is the dissimilarity of a certain road section in paths a and b, minf2 is the minimum risk, Q is the vehicle capacity limit, L(a) is the total length of the a-th path, i is the starting point of the road section, j is the ending point of the road section, s and t respectively represent the starting point and the ending point of the transportation; is the decision variable; q ij is the transportation volume from node i to j.
[0020] Compared with the prior art, the present invention has at least the following advantages or beneficial effects:
[0021] A method for optimizing the transportation path of dangerous goods according to the present invention calculates the dissimilarity between different paths of transporting dangerous goods and the risk degree of transporting dangerous goods simultaneously on different paths, and constructs a bi-objective optimization model with the maximum dissimilarity between different paths of transporting dangerous goods and the minimum risk degree of simultaneous transportation on different paths as the objectives, realizing the minimization of the risk of transporting dangerous goods and the maximization of the dissimilarity at the same time, significantly reducing the risk of transporting dangerous goods, and achieving safer and more effective transportation of dangerous goods. Description of the Drawings
[0022] Figure 1 is the step diagram of a method for optimizing the transportation path of dangerous goods according to the present invention;
[0023] Figure 2 is the schematic diagram of the buffer zone;
[0024] Figure 3 is the schematic diagram of chromosome coding, crossover strategy and gene segment mutation;
[0025] Figure 4 is the schematic diagram of the dangerous goods transportation network;
[0026] Figure 5 is the risk and dissimilarity trend diagram. Detailed Embodiments
[0027] Example 1
[0028] Please refer to Figure 1 , Figure 1 which shows the step diagram of a method for optimizing the transportation route of dangerous goods provided by an embodiment of the present application.
[0029] A method for optimizing the transportation route of dangerous goods provided by the present application includes the following steps:
[0030] S1. Obtain the node data of the transportation network and calculate the dissimilarity between different paths of dangerous goods transportation;
[0031] S2. Calculate the risk degree of transporting dangerous goods simultaneously on different paths based on the dissimilarity between different paths;
[0032] S3. Establish a bi-objective optimization model with the maximum dissimilarity between different paths of dangerous goods transportation and the minimum risk degree of simultaneous transportation on different paths as the objectives;
[0033] S4. Solve the bi-objective optimization model based on the non-dominated sorting genetic algorithm to obtain the optimal solution, and output the optimal solution to obtain the best path for dangerous goods transportation.
[0034] As a preferred implementation manner, in step S1, the expression for calculating the dissimilarity between different paths of dangerous goods transportation includes:
[0035]
[0036] In the formula, a represents the a-th path, b represents another path different from a, w is the buffer radius of the vehicle's impact on the surrounding environment, dis(a, b, w) is the dissimilarity between path a and path b, is the length of the road section (i, j) in the a-th path, is the length of the road section (i, j) in the b-th path, N is the total length of the road section, K is the set of paths, is the road section (i, j) in the a-th path, is the road section (i, j) in the b-th path, i is the starting point of the road section, and j is the ending point of the road section.
[0037] As Figure 2 shown is the buffer schematic diagram, which describes the buffer of the strip path, and the intersecting shaded part is the overlapping area of the buffers of the two paths. In Example 1 of the present application, the coordinates of the two paths are known as shown in Table 1:
[0038] Table 1 Node coordinate diagram
[0039]
[0040] The buffer radius is set to 100 m. The intersection area can be calculated to be 1560.7628 m2, the buffer area of path 1 is 3598.1993 m2, and the buffer area of path 2 is 3291.5361 m2. According to Equation (1), the dissimilarity between the two paths is calculated to be 0.55.
[0041] As a preferred implementation manner, in step S2, the process of calculating the risk degree of simultaneously transporting dangerous goods on different paths includes:
[0042]
[0043] In the formula, r(a, b) represents the risk degree generated by path a on path b. is the probability of no accident occurring on path a, pop b is the population exposure of path b, e -ε[d(a,b)]2 is the distance-sensitive damage function, where ε is the influence coefficient of the dangerous goods, and d(a, b) is the Euclidean distance between paths a and b. represents a certain section in path b of the risk degree, r b represents the total risk degree of path b, r a,b represents the risk degree when transporting simultaneously on paths a and b. a represents the a-th path, and b represents another path different from a. is the length of the section (i, j) in the a-th path. is the length of the section (i, j) in the b-th path. N is the total length of the sections, and K is the set of paths. is the section in the a-th path. is the section in the b-th path. i is the starting point of the section, and j is the ending point of the section.
[0044] As a preferred implementation manner, in step S3, the calculation formula of the bi-objective optimization model includes:
[0045]
[0046]
[0047] In the formula, maxf1 is the maximum dissimilarity. a represents the a-th path, and b represents another path different from a. is the length of the section (i, j) in the a-th path. is the length of the section (i, j) in the b-th path. N is the total length of the sections, and K is the set of paths. is the section in the a-th path. is the section in the b-th path. w is the buffer radius of the impact of the vehicle on the surrounding environment. Let \(d_{ab}\) be the dissimilarity of a certain section in paths \(a\) and \(b\), \(minf_2\) be the minimum risk, \(Q\) be the vehicle capacity limit, \(L(a)\) be the total length of the \(a\)-th path, \(i\) be the starting point of a section, \(j\) be the ending point of a section, and \(s\) and \(t\) represent the starting point and the ending point of transportation respectively; is a decision variable; \(q\) ij is the transportation volume from node \(i\) to node \(j\).
[0048] It should be noted that Equation (8) is the vehicle capacity constraint to ensure that the weight of each path does not exceed the maximum vehicle load; Equation (9) is the energy conservation constraint for modeling \(K\) paths from node \(s\) to node \(t\); Equation (10) is the section length constraint to ensure that the sum of the lengths of all sections on the path is less than the path; Equation (11) is a decision variable, which is 1 when the vehicle is on the \(k\)-th path from node \(i\) to node \(j\), and 0 otherwise.
[0049] Embodiment 2
[0050] In Embodiment 2 of the present invention, when solving the bi-objective optimization model, a reasonable chromosome coding needs to be designed according to the characteristics of the model. For this bi-objective model, each path is denoted as \(i\), and the chromosome represents the departure order of the vehicles, with a length of the demand volume \(q\). Taking 9 vehicles on 9 paths as an example to illustrate the chromosome coding and genetic strategy, please refer to Figure 3 , Figure 3 which contains the departure order of 9 vehicles. The algorithm is globally and locally optimized through crossover and mutation. The roulette wheel method is used to select 2 chromosomes, and a crossover segment is randomly generated. Genes 4 - 6 in the figure are the crossover segments, and the crossover segments are swapped. After swapping, it may lead to a change in the number of vehicles resulting in an illegal chromosome, so a chromosome crossover repair operation is required. As Figure 3 shown, the first new element in the non-crossover segment of the first chromosome is 5, and element 5 also appears in the fourth position of the crossover segment, indicating that vehicle 5 appears twice, which is an illegal chromosome. The second element in the second chromosome is the same as the fourth element in the crossover segment, and the two elements are swapped. Similarly, the second element of the first chromosome is swapped with the eighth element of the second chromosome to generate a new chromosome. By selecting different mutation strategies according to the value of \(mutation\_type\), NSGA-II can more flexibly handle different optimization problems and evolution stages. This strategic mutation helps to maintain the population diversity while effectively searching the solution space. When \(mutation\_type < 0.2\), the exchange mutation of multiple genes is adopted. When \(0.2\leq mutation\_type\leq0.4\), the gene segment mutation is adopted. When \(mutation\_type > 0.4\), the exchange mutation of a single gene is adopted, which can conduct a more refined search of the local solution space while maintaining the population stability.
[0051] Embodiment 3
[0052] Please refer to Figure 4 , Figure 4 which shows a schematic diagram of a dangerous goods transportation network with 10 nodes. Node 1(0,0) is used as the warehouse, and node 10(80,0) is the demand point. The parameters of the NSGA-II algorithm are set as follows: population size NP = 200, maximum number of iterations NG = 200, crossover rate pc = 0.6, and mutation rate pm = 0.4. Since the accurate data of population exposure cannot be obtained, it is recorded as a constant 1. The transportation risk is calculated when the accident impact coefficient is ε = 0.003. The node coordinates in the transportation network are shown in Table 2:
[0053] Table 2 Node Coordinates
[0054]
[0055] There are multiple paths from node 1 to node 10 in this road network. For the conventional shortest path problem, only the path with the minimum cost or the shortest distance needs to be found, or when the demand does not exceed the load capacity of a single vehicle, the value of k is taken as 1.
[0056] In the dangerous goods transportation network, when the demand exceeds the load capacity of a single vehicle, an additional vehicle is required for every 40t exceeded. Since different vehicles may affect each other when transporting a batch of goods on the same path, different transportation paths need to be planned. Given the known demand, the k-shortest path algorithm is used to calculate the path with the minimum risk. In this embodiment, there are 15 such paths, as shown in Table 3:
[0057] Table 3 Paths with the Minimum Risk
[0058]
[0059] The dissimilarity and risk of these initial paths are evaluated. If the buffer range is set to 2km and the value of k ranges from 2 to 10, the calculation results of risk and dissimilarity are shown in Table 4:
[0060] Table 4 Calculation Results of Risk and Dissimilarity
[0061]
[0062]
[0063] Given that the calculated result is that the buffer radius is 2000m, the buffer radius is set to 1000m, 500m, and 250m respectively. Comparing with the above results, with k = 10, the optimal solutions of risk and dissimilarity are calculated, and the results are shown in Table 5:
[0064] Table 5 Calculation Results with k = 10
[0065]
[0066] Two line charts were drawn based on the data in Table 5, showing the trends of the two average values of the risk degree and the dissimilarity degree changing with the buffer radius, as Figure 5 shown; then the change rates between adjacent data points were calculated and trend analysis was carried out. The analysis results show that as the buffer radius decreases, the risk decreases, but the decreasing trend is not obvious, while the dissimilarity degree shows a significant decreasing trend. The specific change rates are as follows: the change rate of the risk is -2.21% between the radii of 2000m and 1000m, 0.74% between 1000m and 500m, and 2.32% between 500m and 250m. The change rate of the dissimilarity degree is 12.71% between the radii of 2000m and 1000m, 20.79% between 1000m and 500m, and 41.07% between 500m and 250m. According to different buffer radii, the path dissimilarity changes, resulting in different optimal path sets.
Claims
1. A method for optimizing the transportation route of dangerous goods, characterized in that, It includes the following steps: S1. Obtain the node data of the transportation network and calculate the dissimilarity between different paths of dangerous goods transportation; S2. Calculate the risk degree of transporting dangerous goods simultaneously on different paths based on the dissimilarity between different paths; S3. Establish a bi-objective optimization model with the maximum dissimilarity between different paths of dangerous goods transportation and the minimum risk degree of simultaneous transportation on different paths as the objectives; S4. Solve the bi-objective optimization model based on the non-dominated sorting genetic algorithm to obtain the optimal solution, and output the optimal solution to obtain the best path of dangerous goods transportation.
2. The method for optimizing the transportation route of dangerous goods according to claim 1, wherein, In step S1, the expression for calculating the dissimilarity between different paths of dangerous goods transportation includes: In the formula, a represents the a-th path, b represents another path different from a, w is the buffer radius of the vehicle's impact on the surrounding environment, and dis(a, b, w) is the dissimilarity between path a and path b. is the length of the road segment (i, j) in the a-th path. is the length of the road segment (i, j) in the b-th path, N is the total length of the road segments, and K is the set of paths. is the road segment (i, j) in the a-th path. is the road segment (i, j) in the b-th path, where i is the starting point of the road segment and j is the ending point of the road segment.
3. The method for optimizing the transportation route of dangerous goods according to claim 2, wherein, In step S2, the process of calculating the risk degree of transporting dangerous goods simultaneously on different paths includes: Wherein, r(a, b) represents the risk degree generated by path a to path b, is the probability of no accident occurring on path a, pop b is the population exposure of path b, e -ε[d(a,b)]2 is the distance-sensitive damage function, where ε is the influence coefficient of dangerous goods, and d(a, b) is the Euclidean distance between paths a and b, represents a certain section in path b of the risk degree, r b represents the total risk degree of path b, r a,b represents the risk degree when transporting simultaneously on paths a and b, a represents the a-th path, and b represents another path different from a, is the length of section (i, j) in the a-th path, is the length of section (i, j) in the b-th path, N is the total length of the section, and K is the set of paths, is the section in the a-th path, is the section in the b-th path, i is the starting point of the section, and j is the ending point of the section.
4. The method for optimizing a dangerous goods transportation route according to claim 1, wherein In step S3, the calculation formula of the bi-objective optimization model includes: Where maxf1 is the maximum dissimilarity, a represents the ath path, and b represents another path different from a. is the length of the section (i, j) in the ath path. is the length of the section (i, j) in the bth path, N is the total length of the sections, and K is the set of paths. is the section in the ath path. is the section in the bth path, w is the buffer radius of the vehicle's impact on the surrounding environment. is the dissimilarity of a certain section in paths a and b, minf2 is the minimum risk, Q is the vehicle capacity limit, L(a) is the total length of the ath path, i is the starting point of the section, j is the ending point of the section, and s and t respectively represent the starting point and ending point of the transportation. is a decision variable; q ij is the transportation volume from node i to j.