Dual-objective Medical Waste Transportation Method Based on Improved RSA-VNS

Abstract

The present invention provides a dual-objective medical waste transportation method based on improved RSA-VNS, which relates to the technical field of path optimization. In the present invention, first, the medical waste transportation tasks and vehicle scheduling resources are obtained; secondly, based on the medical waste transportation tasks and vehicle scheduling resources, with the minimization of cost and storage risk as the optimization objectives, a dual-objective medical waste transportation model is constructed; finally, the improved RSA-VNS algorithm is used to solve the dual-objective medical waste transportation model, and the Pareto external archive is output to decode and obtain the optimal medical waste transportation plan. By comprehensively considering the start service time of medical institutions, the type, intensity and the possibility of spreading diseases of medical waste, the storage risk of medical waste is effectively quantified, and the model is more in line with the characteristics of the recycling and transportation problem of medical waste. In addition, the proposed hybrid algorithm effectively improves the algorithm convergence speed and solution diversity, and also strengthens the local convergence ability in the late stage of the algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of path optimization, and particularly relates to a dual-objective medical waste transportation method based on improved RSA-VNS. Background Art

[0002] The dual-objective path optimization problem is a specific application of multi-objective problems and has received extensive attention and research in recent years. It widely exists in various industries of modern transportation, such as the network communication industry, the medical industry, the logistics industry, etc. Different from the traditional path optimization problem, in the dual-objective path optimization problem, the two optimization objectives often conflict with each other. In the transportation operations of the medical industry, due to the special nature of the transported items, not only the cost issue is often considered, but also the risk factors brought by medical waste should be taken into account. Therefore, it has strong practical significance to study the dual-objective path optimization problem of recycling waste in the medical industry.

[0003] In related technologies, intelligent algorithms have been widely used to solve various dual-objective optimization problems. For example, the paper (Menares, F., Montero, E., Paredes-Belmar, G.,&Bronfman, A. (2023). A bi-objective time-dependent vehicle routing problem with delivery failure probabilities. Computers&Industrial Engineering , 185 , 109601.) implements a multi-objective genetic algorithm NSGA-II to solve a bi-objective time-dependent vehicle routing problem with delivery failure probabilities (TDVRPDFP).

[0004] Again, the paper (Ren, X., Huang, H., Feng, S.,&Liang, G. (2020). An improved variable neighborhood search for bi-objective mixed-energy fleet vehicle routing problem. Journal of Cleaner Production, 275, 124155.) proposed an improved variable neighborhood search (VNS) with a selection mechanism on the basis of the former's research to find the Pareto boundary of the model. The calculation results based on the Solomon benchmark show that the VNS with the selection mechanism has better performance and efficiency than the original VNS.

[0005] However, there is little research considering both cost and storage risk at the same time. Solving this complex problem is the key to the recycling and transportation of medical institution waste, while the traditional path optimization problem cannot solve this problem. Summary of the Invention

[0006] (I) Technical Problem to be Solved

[0007] In view of the deficiencies of the prior art, the present invention provides a dual-objective medical waste transportation method based on improved RSA-VNS, which solves the technical problem of ignoring the risk factors brought by medical waste in the process of optimizing the waste transportation path.

[0008] (II) Technical Solution

[0009] To achieve the above object, the present invention is realized through the following technical solutions:

[0010] A dual-objective medical waste transportation method based on improved RSA-VNS includes:

[0011] Obtaining medical waste transportation tasks and vehicle scheduling resources;

[0012] Based on the medical waste transportation tasks and vehicle scheduling resources, taking minimizing cost and storage risk as the optimization objectives, constructing a dual-objective medical waste transportation model; wherein, based on the start service time of the medical institution and combining the type, intensity and the possibility of spreading diseases of the medical waste, the storage risk is quantified;

[0013] Using the improved RSA-VNS algorithm to solve the dual-objective medical waste transportation model, outputting the Pareto external archive, and decoding to obtain the optimal medical waste transportation plan.

[0014] Preferably, the medical waste transportation tasks and vehicle scheduling resources include:

[0015] v represents the set of all nodes, including the recycling point v0 and the set of medical institutions v c ;

[0016] B represents the selected subset of medical institutions, B v c ;

[0017] r i represents the waste collection volume of medical institution i;

[0018] ts i represents the loading service time of the vehicle at medical institution i;

[0019] d ij represents the distance between medical institutions i and j;

[0020] ati Denote the start service time of medical institution i;

[0021] h i Denote the medical waste storage risk of medical institution i;

[0022] K represents the vehicle group;

[0023] p k Denote the path label set of the p-th transportation task of vehicle k;

[0024] v el Denote the average vehicle speed;

[0025] w represents the maximum vehicle load;

[0026] v z Denote the loading speed of the vehicle;

[0027] δ1, δ2, δ3 respectively denote the fixed cost, unit fuel consumption cost, and cooling cost per unit time of the vehicle;

[0028] FC ij Denote the fuel consumption of the vehicle when traveling between medical institutions i and j;

[0029] D represents the set of medical waste types;

[0030] β d 、ρ d respectively denote the intensity of medical waste d and the possibility of spreading diseases.

[0031] Preferably, the double-objective medical waste transportation model includes the objective function:

[0032]

[0033] Among them, Min represents the minimization function, and f1, f2 respectively represent the cost and storage risk; is the decision variable. If the p-th transportation task of vehicle k includes traveling from medical institution i to j, take 1, otherwise take 0.

[0034] Preferably, the double-objective medical waste transportation model includes the constraint conditions:

[0035]

[0036] Among them, Equation (3) restricts that only one vehicle is allowed to provide service to each medical institution once;

[0037] Equation (4) determines that the vehicle leaves the recycling point as an empty vehicle; is the vehicle load of vehicle k from medical institution i to j in the p-th transportation task;

[0038] Equation (5) restricts that vehicle k must return to the recycling point after completing the p-th transportation task;

[0039] Equation (6) ensures the continuity of the path;

[0040] Equation (7) eliminates the sub-route constraint;

[0041] Equations (8)-(9) restrict that the total amount of waste transported shall not exceed the maximum load capacity of the vehicle, and the waste generated by medical institutions shall not exceed the maximum load capacity of the vehicle;

[0042] Equations (10)-(11) represent the constraints on variables; is a decision variable. If vehicle k has completed transporting the waste generated by medical institution i, it takes 1, otherwise it takes 0;

[0043] Equations (12)-(13) ensure that each vehicle starts from the recycling point, and the first transportation of each vehicle also starts from the recycling point; is a decision variable. If vehicle k is used, it takes 1, otherwise it takes 0.

[0044] Preferably, the improved RSA-VNS algorithm is used to solve the bi-objective medical waste transportation model, and the Pareto external archive is output, including:

[0045] S31. Initialize the population, and let the current iteration number t = 0; where the coding rule is to randomly generate decimals between [0,1) and assign them to the medical institution order vector and the vehicle allocation vector respectively;

[0046] S32. Based on the bi-objective medical waste transportation model, calculate the objective function of the particle and store the non-dominated solutions in the Pareto external archive;

[0047] S33. Randomly select individuals from the Pareto external archive and perform the first cycle judgment to select the update method, including:

[0048] If t is less than 0.25T, perform high gait walking to update the individual; if t is less than or equal to 0.5T and greater than 0.25T, perform abdominal gait walking to update the individual; where T is the maximum number of iterations;

[0049] S34. Perform external archive update, let t = t + 1. If t is equal to 0.5T, complete the RSA part of the reptile search algorithm and transfer to S35, otherwise return to S33;

[0050] S35. Use the correction strategy to convert the current population and the Pareto external archive into discrete coding;

[0051] S36. Randomly select the discretized individuals from the Pareto external archive and perform the second cycle judgment to select the update method, including:

[0052] If \(t\leq0.75T\) and \(t > 0.5T\), then update the discretized individual using the VNS - 1 operator; if \(t\leq T\) and \(t > 0.75T\), then update the discretized individual using the VNS - 2 operator;

[0053] S37. Perform external archive update, let \(t=t + 1\). If \(t>T\), then complete the variable neighborhood search (VNS) part and transfer to S38; otherwise, return to S36;

[0054] S38. Output the final Pareto external archive.

[0055] Preferably, during the update process of the reptile search algorithm (RSA), for the case where the elements in the new solution exceed 1 or are less than 0, the correction strategy refers to repairing this part of the solution, and the repair method is:

[0056] For each element \(a\) in the solution, let \(a=(a - a min ) / (a max -a min ) (1 - 10e -10 ); where \(a min and \(a max represent the minimum and maximum values of the medical institution order vector or vehicle allocation vector respectively, \(e\) is a mathematical constant, and \(1 - 10e -10 is used to prevent the calculation error of the decoding intermediate vector caused by the numerical value equal to 1.

[0057] Preferably, the VNS - 1 operator includes swapping hospitals served by different vehicles, inserting hospitals served by different vehicles, and global - scale reversal;

[0058] The VNS - 2 operator includes swapping hospitals served by the same vehicle, inserting hospitals served by the same vehicle, and global - scale reversal.

[0059] A two - objective medical waste transportation system based on improved RSA - VNS, comprising:

[0060] An acquisition module, used to acquire medical waste transportation tasks and vehicle scheduling resources;

[0061] A construction module, used to construct a two - objective medical waste transportation model based on the medical waste transportation tasks and vehicle scheduling resources, with the optimization objectives of minimizing cost and storage risk; wherein, based on the start service time of medical institutions, and combined with the type, intensity, and disease - spreading possibility of medical waste, the storage risk is quantified;

[0062] A solution module, which is used to solve the dual-objective medical waste transportation model by using the improved RSA-VNS algorithm, and output a Pareto external archive to decode and obtain the optimal medical waste transportation plan.

[0063] A storage medium stores a computer program for dual-objective medical waste transportation based on the improved RSA-VNS. Among them, the computer program enables a computer to execute the dual-objective medical waste transportation method as described above.

[0064] An electronic device includes:

[0065] One or more processors; a memory; and one or more programs, where the one or more programs are stored in the memory and are configured to be executed by the one or more processors. The programs include those for executing the dual-objective medical waste transportation method as described above.

[0066] (3) Beneficial effects

[0067] The present invention provides a dual-objective medical waste transportation method based on the improved RSA-VNS. Compared with the prior art, it has the following beneficial effects:

[0068] In the present invention, first, the medical waste transportation tasks and vehicle scheduling resources are obtained; secondly, based on the medical waste transportation tasks and vehicle scheduling resources, with the minimization of cost and storage risk as the optimization objectives, a dual-objective medical waste transportation model is constructed; finally, the improved RSA-VNS algorithm is used to solve the dual-objective medical waste transportation model, and a Pareto external archive is output to decode and obtain the optimal medical waste transportation plan. By comprehensively considering the start service time of medical institutions, the types, intensities of medical waste, and the possibility of spreading diseases, the storage risk of medical waste is effectively quantified, and the modeling is more in line with the characteristics of the recycling and transportation problems of medical waste. In addition, the proposed hybrid algorithm effectively improves the algorithm convergence speed and solution diversity, and also enhances the local convergence ability in the late stage of the algorithm. Description of the drawings

[0069] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0070] Figure 1 It is a block diagram of a dual-objective medical waste transportation method based on the improved RSA-VNS provided by an embodiment of the present invention;

[0071] Figure 2A flowchart of an improved RSA-VNS algorithm provided by an embodiment of the present invention;

[0072] Figure 3 A coding schematic diagram provided by an embodiment of the present invention;

[0073] Figures 4 to 6 Schematic diagrams of the swap operator swapop1, the insertion operator insertop1, and the global range reverse inverseop1 provided by embodiments of the present invention respectively;

[0074] Figures 7 to 9 Schematic diagrams of the swap operator swapop2, the insertion operator insertop2, and the global range reverse inverseop2 provided by embodiments of the present invention respectively. Detailed implementation manners

[0075] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0076] The embodiments of the present application provide a dual-objective medical waste transportation method based on an improved RSA-VNS, which solves the technical problem of ignoring the risk factors brought by medical waste in the process of optimizing the waste transportation path.

[0077] The technical solutions in the embodiments of the present application to solve the above technical problems are generally as follows:

[0078] The present invention also aims to solve the path optimization problem considering storage risks in the dual-objective scenario. The purpose is to determine which specific transportation vehicle each medical institution will be assigned to, the order in which the transportation vehicle visits medical institutions, and the arrival time points, so as to minimize the transportation cost and storage risks. Based on the characteristics of the problem, an effective hybrid algorithm is proposed through theoretical analysis and mathematical derivation to solve this combinatorial optimization problem, providing a new method for the medical waste transportation problem in a complex environment.

[0079] Supplementary explanations on terms:

[0080] (1) Reptile Search Algorithm, abbreviated as RSA in English.

[0081] (2) Variable Neighborhood Search, abbreviated as VNS in English.

[0082] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0083] Example 1:

[0084] As Figure 1 shown, an embodiment of the present invention provides a dual-objective medical waste transportation method based on improved RSA-VNS, including:

[0085] S1. Obtain medical waste transportation tasks and vehicle scheduling resources;

[0086] S2. Based on the medical waste transportation tasks and vehicle scheduling resources, with minimizing cost and storage risk as the optimization objectives, construct a dual-objective medical waste transportation model; wherein, based on the start service time of medical institutions, and in combination with the type, intensity, and disease transmission possibility of medical waste, quantify the storage risk;

[0087] S3. Use the improved RSA-VNS algorithm to solve the dual-objective medical waste transportation model, output the Pareto external archive, and decode to obtain the optimal medical waste transportation plan.

[0088] The embodiment of the present invention effectively quantifies the storage risk of medical waste, and the modeling is more in line with the characteristics of the recycling and transportation problems of medical waste. In addition, the proposed RSA-VNS algorithm effectively improves the algorithm convergence speed and solution diversity, and also strengthens the local convergence ability at the end of the algorithm.

[0089] Next, each step of the above solution will be introduced in detail:

[0090] In step S1, obtain medical waste transportation tasks and vehicle scheduling resources.

[0091] In this step, the obtained medical waste transportation tasks and vehicle scheduling resources include:

[0092] v represents the set of all nodes, including the recycling point v0 and the medical institution set v c ;

[0093] B represents the selected subset of medical institutions, B v c ;

[0094] r i represents the waste collection volume of medical institution i;

[0095] ts i represents the loading service time of the vehicle at medical institution i;

[0096] d ijDenote the distance between medical institutions \(i\) and \(j\);

[0097] at i Denote the start service time of medical institution \(i\);

[0098] h i Denote the medical waste storage risk of medical institution \(i\);

[0099] Let \(K\) denote the vehicle group;

[0100] p k Denote the path label set of the \(p\)-th transportation task of vehicle \(k\);

[0101] v el Denote the average vehicle speed;

[0102] Let \(w\) denote the maximum vehicle load;

[0103] v z Denote the loading speed of the vehicle;

[0104] Let \(\delta_1\), \(\delta_2\), \(\delta_3\) denote the fixed cost, unit fuel consumption cost, and cooling cost per unit time of the vehicle respectively;

[0105] FC ij Denote the fuel consumption of the vehicle when traveling between medical institutions \(i\) and \(j\);

[0106] Let \(D\) denote the set of medical waste types;

[0107] β d 、ρ d Denote the intensity and the possibility of spreading diseases of medical waste \(d\) respectively.

[0108] In step S2, based on the medical waste transportation task and vehicle scheduling resources, with the goal of minimizing cost and storage risk, a bi-objective medical waste transportation model is constructed; among them, based on the start service time of medical institutions, and combined with the type, intensity, and the possibility of spreading diseases of medical waste, the storage risk is quantified.

[0109] It should be added that the embodiment of the present invention considers the bi-objective path optimization problem of medical waste transportation, and the goal is to minimize the transportation cost and the medical waste storage risk of the hospital. The problem description is as follows:

[0110] Given a set \(v\) of medical institutions containing known amounts of medical waste c which need to be recycled and processed by multiple transport vehicles. Different medical wastes have different storage risks in different medical institutions, but the recycling time is the same.

[0111] The problem assumptions are as follows:

[0112] (1) The recycling center has a sufficient number of medical waste recycling vehicles of the same type, with known vehicle capacities and a constant speed during transportation.

[0113] (2) The locations of each medical institution are known, the amount of medical waste generated by each medical institution can be scientifically estimated, and the amount recycled does not exceed the rated load of a single vehicle.

[0114] (3) All drivers of the medical waste recycling vehicles have received unified and strict training, and the fuel consumption does not change due to subjective factors. The energy required for refrigeration is provided by the engine.

[0115] (4) The amount of medical waste generated by each small and medium-sized medical institution is small, and the recycling time is the same.

[0116] (5) Each vehicle can visit multiple medical institutions, and each medical institution can only be visited by one vehicle once.

[0117] (6) During the path optimization process, each vehicle can perform multiple transportation tasks.

[0118] (7) Force majeure factors such as road congestion and weather impacts are not considered.

[0119] Correspondingly, the two-objective medical waste transportation model in this step includes

[0120] Objective function:

[0121]

[0122] Among them, Min represents the minimization function, and f1 and f2 represent cost and storage risk respectively; is a decision variable. If the p-th transportation task of vehicle k includes going from medical institution i to j, it takes 1, otherwise it takes 0.

[0123] And the constraint conditions:

[0124]

[0125] Among them, equation (3) restricts that each medical institution only allows one vehicle to provide service once;

[0126] Equation (4) determines that the vehicle leaves the recycling point as an empty vehicle; is the vehicle load of vehicle k from medical institution i to j in the p-th transportation task;

[0127] Equation (5) restricts that vehicle k must return to the recycling point after completing the p-th transportation task;

[0128] Equation (6) ensures the continuity of the path;

[0129] Equation (7) eliminates the sub-route constraints;

[0130] Equations (8)-(9) limit the total amount of waste transported not to exceed the maximum load capacity of the vehicle, and the waste generated by medical institutions shall not exceed the maximum load capacity of the vehicle;

[0131] Equations (10)-(11) represent the constraints on variables; is a decision variable. If vehicle k has completed transporting the waste generated by medical institution i, take 1, otherwise take 0;

[0132] Equations (12)-(13) ensure that each vehicle starts from the recycling point, and the first transportation of each vehicle also starts from the recycling point; is a decision variable. If vehicle k is used, take 1, otherwise take 0.

[0133] In step S3, the improved RSA-VNS algorithm is used to solve the dual-objective medical waste transportation model, and the Pareto external archive is output to decode and obtain the optimal medical waste transportation plan.

[0134] It should be noted that the embodiments of the present invention have at least the following improvements on the existing RSA algorithm:

[0135] First, introduce the Pareto external archive

[0136] In the existing RSA, the optimal solution is needed to guide the evolution. However, due to the multi-objective problem dealt with in the embodiments of the present invention, the optimal solution cannot be selected. Therefore, a Pareto external archive is maintained in the program. The archive stores the non-dominated solutions obtained currently. In each iteration, non-dominated sorting is performed on the current population. For the obtained non-dominated solutions, first judge whether they dominate some solutions in the external archive. If they do, delete these solutions in the external archive. Secondly, judge whether they are dominated by some solutions in the external archive. If they are dominated, they cannot be added to the external archive. If they are not dominated by all solutions, add the obtained non-dominated solutions to the external archive. If the archive is full, randomly replace. When optimizing, a solution is randomly selected from the external archive as the optimal solution to guide the evolution.

[0137] Second, repair illegal solutions

[0138] In the update process of the RSA algorithm, the situation may occur that the elements in the solution exceed 1 or are less than 0. At this time, the solution needs to be repaired. Specifically, re-normalization operations are adopted in the code.

[0139] Third, design of neighborhood operators

[0140] After converting the continuous encoding to discrete encoding in the hybrid algorithm, neighborhood operations are then performed. When t is less than or equal to 0.75T and greater than 0.5T, a neighborhood with a large difference from the original solution is considered for design to facilitate the search in a larger solution space. Different from this, when t is less than T and greater than 0.75T, a neighborhood with a small difference from the original solution is considered for design to facilitate local exploration.

[0141] As Figure 2 shown, Figure 2 a flowchart of an improved RSA-VNS algorithm is disclosed.

[0142] Referring to Figure 2 , the solution process of this step specifically includes:

[0143] S31. Initialize the population and set the current iteration number t = 0; where:

[0144] The encoding rule is as follows:

[0145] As Figure 3 shown, random decimals are generated between [0, 1), and are respectively assigned to the medical institution order vector and the vehicle allocation vector, Figure 3 where NH in

[0146] is the number of medical institutions.

[0147] Correspondingly, the decoding rule is supplemented as follows:

[0148] First, based on the vehicle allocation vector, confirm the vehicles of each hospital. If the decimal belongs to [0, 1 / NK), it represents selecting the first vehicle. If the decimal belongs to [1 / NK, 2 / NK), it represents selecting the second vehicle, and so on. Here, NK is the number of vehicles.

[0149] Secondly, add the obtained vehicle numbers and the order vector, then sort them in ascending order, record the obtained hospital sequence, and add 0 as a marker between different vehicles.

[0150] S32. Based on the double-objective medical waste transportation model, calculate the objective function of the particle and store the non-dominated solutions in the Pareto external archive.

[0151] S33. Randomly select an individual from the Pareto external archive and perform the first cycle judgment to select the update method, including:

[0152] If t is less than 0.25T, then perform high gait walking to update the individual;

[0153]

[0154] If t is less than or equal to 0.5T and greater than 0.25T, then execute the abdominal gait walking to update the individual;

[0155]

[0156] where T is the maximum number of iterations; x (i,j) represents the j-th position of the i-th solution (corresponding to the j-th column of the encoded individual); Best j (t) is the j-th position in the optimal solution obtained so far; η (i,j) represents the hunting operator for the j-th position of the i-th solution; β is a sensitive parameter used to control the exploration accuracy (i.e., high walking) in the surrounding stage during the iteration process and can be fixed at 0.1; the reduction function R (i,j) is the value used to reduce the search area; r1 represents a random position; ES represents the evolution factor, which is a probability ratio and takes a randomly decreasing value between 2 and -2 throughout the number of iterations; rand represents a random number between 0 and 1.

[0157] S34. Perform external archive update, let t = t + 1. If t is equal to 0.5T, then complete the reptile search algorithm RSA part and transfer to S35; otherwise, return to S33.

[0158] S35. Adopt a correction strategy to convert the current population and the Pareto external archive into discrete coding.

[0159] During the update process of the reptile search algorithm RSA, for the case where the elements in the new solution exceed 1 or are less than 0, the correction strategy refers to repairing this part of the solution, and the repair method is:

[0160] For each element a in the solution, let a = (a - a min ) / (a max - a min ) (1 - 10e -10 ); where a min and a max represent the minimum and maximum values of the medical institution order vector or the vehicle allocation vector respectively, e is the mathematical constant, and 1 - 10e -10 is used to prevent the numerical value from being equal to 1 and causing errors in calculating the decoded intermediate vector.

[0161] S36. Randomly select a discretized individual from the Pareto external archive and perform the second cycle judgment to select the update method, including:

[0162] If t is less than or equal to 0.75T and greater than 0.5T, then the discretized individual is updated using the VNS-1 operator; exemplarily, the VNS-1 operator includes swapping hospitals served by different vehicles, inserting hospitals served by different vehicles, and global reverse; specifically including:

[0163] As Figure 4 shown, the swap operator swapop1: Given a solution S; obtain the hospitals that each vehicle needs to visit according to the solution S; randomly select two vehicles, and randomly select a hospital of the two vehicles; swap the hospital positions to obtain a new solution.

[0164] As Figure 5 shown, the insertion operator insertop1: Given a solution S; obtain the hospitals that each vehicle needs to visit according to the solution S; randomly select two vehicles, and randomly select a hospital i and j of the two vehicles; insert hospital i after hospital j to obtain a new solution.

[0165] As Figure 6 shown, the global reverse inverseop1: Given a solution S; randomly select two non-repeating positions l1, l2; reverse the elements between [l1, l2] (including l1, l2).

[0166] If t is less than or equal to T and greater than 0.75T, then the discretized individual is updated using the VNS-2 operator; exemplarily, the VNS-2 operator includes swapping hospitals served by the same vehicle, inserting hospitals served by the same vehicle, and global reverse; specifically including:

[0167] As Figure 7 shown, the swap operator swapop2: Given a solution S; obtain the hospitals that each vehicle needs to visit according to the solution S; randomly select a vehicle, and randomly select two hospitals of the vehicle; swap the hospital positions to obtain a new solution.

[0168] As Figure 8 shown, the insertion operator insertop2: Given a solution S; obtain the hospitals that each vehicle needs to visit according to the solution S; randomly select a vehicle, and randomly select two hospitals i and j of the vehicle; insert hospital i after hospital j to obtain a new solution.

[0169] As Figure 9 shown, the global reverse inverseop2: Given a solution S; obtain the hospitals that each vehicle needs to visit according to the solution S; randomly select a vehicle, and randomly select two non-repeating positions l1, l2 of the vehicle; reverse the elements between [l1, l2] (including l1, l2).

[0170] S37. Perform external file update, set t = t + 1. If t is greater than T, complete the variable neighborhood search (VNS) part and proceed to S38; otherwise, return to S36.

[0171] S38. Output the final Pareto external file.

[0172] S39. Based on the Pareto external file, those skilled in the art can select the optimal solution according to actual needs and preferences, and perform decoding operations to obtain the optimal medical waste transportation plan.

[0173] It can be understood that in multi-objective optimization problems, there is often no single optimal solution, but multiple conflicting optimal solutions. Each point on the Pareto front represents a possible solution, and these solutions achieve the best trade-off among multiple objectives. It can provide comprehensive information and multiple choices for decision-makers, enabling them to select appropriate solutions according to actual needs and preferences on the basis of fully understanding various possibilities.

[0174] So far, the embodiment of the present invention has completed all the processes of the dual-objective medical waste transportation method based on the improved RSA-VNS.

[0175] Embodiment 2:

[0176] The embodiment of the present invention provides a dual-objective medical waste transportation system based on the improved RSA-VNS, including:

[0177] An acquisition module, configured to acquire medical waste transportation tasks and vehicle scheduling resources;

[0178] A construction module, configured to construct a dual-objective medical waste transportation model with minimizing cost and storage risk as optimization objectives based on the medical waste transportation tasks and vehicle scheduling resources; wherein, based on the start service time of medical institutions, and in combination with the type, intensity of medical waste and the possibility of spreading diseases, quantify the storage risk;

[0179] A solution module, configured to solve the dual-objective medical waste transportation model by using the improved RSA-VNS algorithm, output the Pareto external file, and decode to obtain the optimal medical waste transportation plan.

[0180] Embodiment 3:

[0181] The embodiment of the present invention provides a storage medium, which stores a computer program for dual-objective medical waste transportation based on the improved RSA-VNS, wherein the computer program enables a computer to execute the dual-objective medical waste transportation method as described in Embodiment 1.

[0182] Embodiment 4:

[0183] An embodiment of the present invention provides an electronic device, including:

[0184] One or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the programs include those for executing the dual-objective medical waste transportation method as described in Embodiment 1

[0185] It can be understood that the dual-objective medical waste transportation system, storage medium, and electronic device provided by the embodiments of the present invention correspond to the dual-objective medical waste transportation method provided by the embodiments of the present invention. For the explanations, examples, beneficial effects, etc. of the relevant content, reference can be made to the corresponding parts in the dual-objective medical waste transportation method, which will not be elaborated here.

[0186] In summary, compared with the prior art, the following beneficial effects are achieved:

[0187] 1. The embodiment of the present invention proposes a hybrid algorithm for the cost and storage risk considered in the medical waste transportation process, ensuring that after the process of assigning medical institutions to transportation vehicles is completed, the transportation capacity of each vehicle can be fully utilized, and at the same time, ensuring that the risk brought by the storage of medical waste in medical institutions is greatly reduced.

[0188] 2. General intelligent algorithms are prone to falling into local optima. However, it has been found through research that the reptile search algorithm exhibits good performance when combined with multiple algorithms in applications. The embodiment of the present invention designs an adaptive neighborhood search mechanism for the disadvantages and advantages of the reptile search algorithm and the specific application in this problem, and ensures the update of the solution set by performing different search rates at different times. This not only effectively improves the algorithm convergence speed and solution diversity, but also strengthens the local convergence ability of the algorithm in the late stage to a certain extent.

[0189] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or device including the element.

[0190] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dual-objective medical waste transportation method based on improved RSA-VNS, characterized in that, Including: Obtaining medical waste transportation tasks and vehicle scheduling resources; Based on the medical waste transportation tasks and vehicle scheduling resources, with the goal of minimizing costs and storage risks, constructing a bi-objective medical waste transportation model; among them, based on the start service time of medical institutions, and combining the type, intensity of medical waste and the possibility of spreading diseases, quantifying the storage risk; Using an improved RSA-VNS algorithm to solve the bi-objective medical waste transportation model, outputting a Pareto external archive, and decoding to obtain the optimal medical waste transportation plan; The step of using an improved RSA-VNS algorithm to solve the bi-objective medical waste transportation model and outputting a Pareto external archive includes: S31. Initialize the population, and let the current iteration number t = 0; where the encoding rule is to randomly generate decimals between [0,1), and assign them to the medical institution order vector and vehicle allocation vector respectively; S32. Based on the bi-objective medical waste transportation model, calculate the objective function of the particle, and store the non-dominated solutions in the Pareto external archive; S33. Randomly select an individual from the Pareto external archive, and perform the first cycle judgment to select the update method, including: If t is less than 0.25T, then perform high gait walking to update the individual; if t is less than or equal to 0.5T and greater than 0.25T, then perform abdominal gait walking to update the individual; where T is the maximum number of iterations; S34. Perform external archive update, let t = t + 1, if t is equal to 0.5T, then complete the reptile search algorithm RSA part, transfer to S35, otherwise return to S33; S35. Use a correction strategy to convert the current population and the Pareto external archive into discrete coding; S36. Randomly select a discretized individual from the Pareto external archive, and perform the second cycle judgment to select the update method, including: If t is less than or equal to 0.75T and greater than 0.5T, then use the VNS-1 operator to update the discretized individual; if t is less than or equal to T and greater than 0.75T, then use the VNS-2 operator to update the discretized individual; S37. Perform external archive update, let t = t + 1, if t is greater than T, then complete the variable neighborhood search VNS part, transfer to S38, otherwise return to S36; S38. Output the final Pareto external archive; The VNS-1 operator includes swapping hospitals served by different vehicles, inserting hospitals served by different vehicles, and reversing in the global range; The VNS-2 operator includes swapping hospitals served by the same vehicle, inserting hospitals served by the same vehicle, and reversing in the global range.

2. The dual-target medical waste transportation method according to claim 1, characterized in that, The medical waste transportation tasks and vehicle scheduling resources include: Let \(v\) denote the set of all nodes, including the recycling point \(v_0\) and the set of medical institutions \(v\). c ; B represents a selected subset of medical institutions, B v c ; r i represents the waste collection volume of medical institution i; ts i Indicates the loading service time of the vehicle at medical institution i; d ij represents the distance between medical institutions i and j; at i Indicates the start service time of medical institution i; h i represents the medical waste storage risk of medical institution i; K represents the vehicle group; p k Denote the set of path labels for the p-th transportation task of vehicle k; v el represents the average vehicle speed; w represents the maximum vehicle load; v z represents the loading speed of the vehicle; δ1, δ2, δ3 respectively represent the fixed cost of the vehicle, the unit fuel consumption cost, and the cooling cost per unit time; FC ij Indicates the fuel consumption of the vehicle traveling between medical institutions i and j; D represents the set of medical waste types; β d and ρ d represent the intensity of medical waste d and the possibility of spreading diseases, respectively.

3. The dual-target medical waste transportation method according to claim 2, characterized in that, The bi-objective medical waste transportation model includes an objective function: Among them, Min represents the minimization function, and f1 and f2 represent cost and storage risk respectively; is a decision variable. If the p-th transportation task of vehicle k includes traveling from medical institution i to j, it takes the value of 1; otherwise, it takes the value of 0.

4. The dual-target medical waste transportation method according to claim 3, wherein The bi-objective medical waste transportation model includes constraint conditions: Among them, formula (3) restricts that each medical institution only allows one vehicle to provide service once; Equation (4) determines that the vehicle leaves the collection point as an empty vehicle; is the vehicle load of vehicle k from medical institution i to j in the p-th transportation task; Equation (5) restricts that vehicle k must return to the recycling point after completing the p-th transportation task; Equation (6) ensures the continuity of the path; Equation (7) eliminates the sub-route constraint; Equations (8)-(9) restrict that the total amount of waste transported shall not exceed the maximum load capacity of the vehicle, and the waste generated by medical institutions shall not exceed the maximum load capacity of the vehicle; Equations (10) to (11) represent the constraints on the variables; is a decision variable, which takes 1 if vehicle k completes the transportation of the waste generated by medical institution i, and 0 otherwise; Equations (12) to (13) ensure that each vehicle starts from the recycling point, and the first transportation of each vehicle also starts from the recycling point; is a decision variable, taking 1 if vehicle k is used and 0 otherwise.

5. The dual-target medical waste transportation method according to claim 1, wherein During the update process of the Reptile Search Algorithm (RSA), for the case where the elements in the new solution exceed 1 or are less than 0, the correction strategy refers to repairing this part of the solution, and the repair method is as follows: For each element a in the solution, let a = (a - a min ) / (a max - a min ) (1 - 10e -10 ) ; where a min and a max represent the minimum and maximum values of the medical institution order vector or the vehicle allocation vector respectively, e is a mathematical constant, and 1 - 10e -10 is used to prevent the numerical value from being equal to 1, which may cause errors in calculating the decoded intermediate vector.

6. A dual-objective medical waste transportation system based on improved RSA-VNS, characterized in that, For executing the dual-objective medical waste transportation method as described in Claim 1, including: An acquisition module, configured to acquire medical waste transportation tasks and vehicle scheduling resources; A construction module, configured to construct a dual-objective medical waste transportation model with minimizing cost and storage risk as the optimization objectives based on the medical waste transportation tasks and vehicle scheduling resources; wherein, based on the start service time of medical institutions, and in combination with the type, intensity, and the possibility of spreading diseases of medical waste, the storage risk is quantified; A solution module, configured to solve the dual-objective medical waste transportation model by using an improved RSA-VNS algorithm, and output a Pareto external archive to decode and obtain the optimal medical waste transportation plan.

7. A storage medium, characterized in that, It stores a computer program for dual-objective medical waste transportation based on the improved RSA-VNS, wherein the computer program enables the computer to execute the dual-objective medical waste transportation method as described in any one of Claims 1-5.

8. An electronic device, characterized in that, Including: One or more processors; A memory; And one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the programs include those for executing the dual-objective medical waste transportation method as described in any one of Claims 1-5.

Images