A medical waste collection and transportation method based on interval multi-objective hybrid frog leap algorithm

Abstract

The application discloses a medical waste collection and transportation method based on an interval multi-objective hybrid frog leap algorithm, and comprises the following steps: obtaining an initial collection and transportation request and road condition information; in an initial scheduling period, based on the initial collection and transportation request and the road condition information, an interval multi-objective hybrid frog leap algorithm is used to solve a pre-constructed medical waste collection and transportation path planning model to obtain an initial vehicle collection and transportation path; the vehicle collects garbage according to the initial vehicle collection and transportation path until the garbage collection and transportation task is completed; if the garbage collection and transportation task is not completed and a next periodic scheduling time point is reached, according to a dynamic request set obtained in a previous scheduling period and the medical waste collection and transportation path planning model, a structure recombination method is used to adjust the initial vehicle collection and transportation path to obtain a vehicle collection and transportation sub-path; a new vehicle is started from a garage to execute the vehicle collection and transportation sub-path until the garbage collection and transportation task is completed. The application reduces the total scheduling cost and waste collection and transportation risk and improves the collection and transportation efficiency of the vehicle.

Description

Technical Field

[0001] This invention relates to the fields of computational intelligence and vehicle routing technology, and more specifically to a medical waste collection method based on a multi-objective hybrid frog-jumping algorithm. Background Technology

[0002] The background of medical waste collection and transportation can be traced back to the 1980s, when people realized that the generation of medical waste was a serious ecological problem. With the continuous advancement of medical technology, the amount of medical waste has also been increasing. Medical waste contains various hazardous substances, such as chemicals, radioactive materials, and infectious viruses. If not properly handled, it can pose a significant threat to the environment and human health. To collect and transport this medical waste efficiently, at low cost, and with low risk, a comprehensive medical waste collection and treatment system needs to be established. Unlike general garbage collection and dispatching, considering the influence of factors such as traffic congestion and weather during actual dispatching, vehicle speeds are often uncertain, easily affecting the actual arrival time of vehicles and the completion time of tasks. Therefore, research on medical waste collection and transportation with uncertain vehicle speeds is essential.

[0003] Numerous studies have emerged in vehicle routing planning. The main methods employed include genetic algorithms and ant colony algorithms. Genetic algorithms are widely applicable and can handle large-scale problems well, but their drawback is that they are prone to getting trapped in local optima. Ant colony algorithms excel in solution efficiency and performance, but they require sophisticated initialization and parameter tuning of the population. Summary of the Invention

[0004] The hybrid frog leapfrog algorithm (SFLA) is a metaheuristic algorithm. Compared with traditional optimization algorithms for vehicle routing problems, SFLA's unique search mechanism allows it to better maintain the balance between local and global searches. The interval multi-objective hybrid frog leapfrog algorithm is an improved version of the hybrid frog leapfrog algorithm. It integrates the problem information of medical waste transportation with the structural characteristics of the hybrid frog leapfrog algorithm, generating a scheduling scheme with strong robustness to uncertain environments at the initial scheduling time, reducing the negative impact of uncertainty on scheduling performance. Furthermore, the scheduling scheme needs to be adjusted promptly for new customer requests or changes in demand.

[0005] Therefore, this invention provides a medical waste collection method based on a multi-objective hybrid frog-jumping algorithm, which can minimize the total scheduling cost and waste collection risk by optimizing the vehicle collection route under constraints such as vehicle capacity and working time.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0007] Firstly, a method for medical waste disposal based on a multi-objective hybrid frog-leaping algorithm is provided, including:

[0008] Obtain the initial waste removal request and road condition information;

[0009] In the initial scheduling cycle, based on the initial collection request and road condition information, the pre-constructed medical waste collection route planning model is solved using the interval multi-objective hybrid frog leap algorithm to obtain the initial vehicle collection route;

[0010] The vehicles collect garbage according to the initial vehicle collection route until the garbage collection task is completed;

[0011] If the waste collection task is not completed and the next periodic scheduling time point is entered, the initial vehicle collection path is adjusted using the structural reorganization method based on the dynamic request set obtained in the previous scheduling cycle and the medical waste collection path planning model to obtain the vehicle collection sub-path; a new vehicle departs from the depot to execute the vehicle collection sub-path until the waste collection task is completed.

[0012] In some embodiments, the medical waste collection method based on the interval multi-objective hybrid frog-leaping algorithm further includes: when a dynamic event request occurs during periodic scheduling, updating the vehicle collection route and the dynamic request set using a dynamic response mechanism; wherein the dynamic response mechanism includes:

[0013] Dynamic event requests refer to requests generated by customers during the scheduling process, including: new medical waste collection requests from medical institutions and adjustments to customers' expected processing times; among them, new medical waste collection requests from medical institutions refer to new customer requests that occur during the scheduling process, including the time, location, and demand of the new customer; adjustments to customers' expected processing times refer to customers' proactive requests to advance or delay the expected processing time for medical waste.

[0014] Determine whether the dynamic event request can be processed. If it can, update the dynamic event request to the existing path using the dynamic patching method, and the vehicle continues to travel along the updated path; otherwise, the vehicle continues to travel along the original path, and the request is placed in the dynamic request set ΔH, waiting for unified processing at the next periodic scheduling time point.

[0015] The dynamic repair method includes: at time t (t∈[0,T]), there are newly added medical institutions h. u To meet the waste collection and transportation needs, the medical waste collection and transportation route planning model will be used to insert new medical institutions into the current collection and transportation routes, so as to minimize route changes and adapt to the dynamic needs of medical institutions.

[0016] In some embodiments, the initial collection request and road condition information include the number of medical institutions that the medical waste transport vehicle needs to visit, the coordinates of the medical institutions, the coordinates of the medical waste treatment station, the coordinates of the vehicle yard, and the expected processing time.

[0017] The dynamic request set stores dynamic requests generated during the previous scheduling cycle. These requests cannot be processed immediately but are processed uniformly at the start of the next scheduling cycle, and completed cleanup requests are removed from the dynamic request set.

[0018] In some embodiments, the medical waste disposal route planning model comprises three strongly coupled sub-problems:

[0019] (1) Multiple trip determination: Within a scheduling cycle, the number of trips for each vehicle and the medical institutions that each trip passes through are determined;

[0020] (2) Assignment of service vehicles to medical institutions;

[0021] (3) Vehicle selection: Select vehicles according to the waste transportation route;

[0022] The scheduling of subproblems is represented by three decision variables:

[0023] Multiple itineraries confirmed

[0024] Medical institution allocation

[0025] Vehicle Selection

[0026] In some embodiments, in the medical waste disposal route planning problem, there are a total of n medical institutions h u m waste treatment stations q There is one vehicle depot d, where u = 1, 2, ..., n, q = 1, 2, ..., m; vehicles depart from the depot at the initial time, sequentially collect waste from various medical institutions, then transport it to any appropriate waste treatment station for processing, and finally return to the depot to end the task; during the collection process, vehicles are allowed to have multiple trips, Trips k Let k be the set of trips. For the w-th journey of vehicle k, L k It is the last journey of vehicle k; the set of medical facilities, waste disposal stations, and parking lots is represented by V = {h1, h2, ..., h}. n ,r1,...,r m The path lengths between these locations are represented by the edge set E = {(p, d}}. i ,p j )|p i ,pj ∈V,i≠j} represents, where p i p j p is the node accessed by the vehicle. i p j ∈H∪R∪{d}, where H is the medical institution h u The set of R, where R is the waste treatment station r q The set; considering the uncertainty of road conditions, for each edge e ij Each has a corresponding range of driving speeds. in v ij and These represent the vehicle on edge e respectively. ij The minimum and maximum driving speeds; the working time period is [0, TM], where TM is the maximum working time; t l (l=1,2,...,|T s |) represents the scheduling time point for periodic scheduling, T s Let the scheduling time t be l A set;

[0027] The medical waste transportation route planning model includes an objective function and constraints.

[0028] The objective function, which aims to minimize scheduling cost and infection risk, is expressed as:

[0029] min f1=Cost1+Cost2+Cost3+Cost4

[0030] min f2=Risk1+Risk2

[0031] The scheduling cost f1 includes fixed cost Cost1, fuel consumption cost Cost2, overtime cost Cost3, and pollution cost Cost4; the infection risk f2 includes transportation risk Risk1 and waste disposal risk Risk2.

[0032]

[0033] Among them, z k This indicates whether vehicle k is selected for use, where K is the set of all vehicles, and CV represents the driver compensation and startup cost required to call a vehicle once.

[0034]

[0035] in, FC ij Indicates that the vehicle passed through road segment e ij fuel consumption cost, α k and β kCharacteristic parameters of road conditions and vehicles; Q0, v ij and d ij These are the vehicle's net weight, vehicle load, vehicle speed, and distance traveled, respectively. ij For point p i Point P j The European distance, CF is the price of fuel; x ijkw For vehicle k, the w-th journey starts from point p. i Point P j ;

[0036]

[0037] Where CO represents overtime pay, tr k The time when vehicle k returns to the depot during its last trip;

[0038]

[0039] Where CP stands for pollution cost. Let vehicle k reach point p. j At that moment, ET j The expected processing time for the waste treatment plant;

[0040]

[0041] in, For vehicles from p i Click to p j Time spent (in hours) Let vehicle k leave point p i Load capacity (tons) at that time, P ij For path e ij Population size in the vicinity;

[0042]

[0043] Where R is the waste treatment station r j The set of M j Waste treatment station j Waste treatment volume, VD j Waste treatment station j The speed of waste processing Waste treatment station j Population in the vicinity.

[0044] In some embodiments, the constraints of the medical waste disposal route planning model include:

[0045] (1) All vehicles will depart from the depot for the first trip and arrive at any waste disposal station to unload after serving the last medical facility on the trip.

[0046]

[0047] Where, x uqkw For vehicle k's wth journey from medical institution h u To the waste treatment station q ;

[0048] (2) The vehicle’s next journey will depart from the waste treatment station to which it is currently en route;

[0049]

[0050] (3) After the vehicle unloads at the waste treatment station on its last trip, it must return to the depot;

[0051]

[0052] in, For the last journey L of vehicle k k From waste treatment plant q Return to parking lot d;

[0053] (4) The vehicle arrives at the medical institution. u Then must be from h u Leave, and the vehicle can only arrive at and leave the medical facility at most. u once;

[0054]

[0055]

[0056] Where, x jukw For vehicle k, the w-th journey starts from point p. j to medical institutions u ;x ujkw For vehicle k's wth journey from medical institution h u Point P j ;y uk For medical institutions h u Vehicle K is responsible for it;

[0057] (5) There is one and only one vehicle serving a particular medical institution;

[0058]

[0059] (6) The amount of waste transported by the vehicle on each trip shall not exceed the vehicle capacity Q;

[0060]

[0061] Among them, g u This refers to the amount of waste collected by medical institutions.

[0062] (7) The number of vehicles departing from the parking lot shall not exceed the maximum number of available vehicles;

[0063]

[0064] Where, x dukw For vehicle k's w-th journey from parking lot d to medical facility h u .

[0065] In some embodiments, an interval multi-objective hybrid frog-leap algorithm is used to solve the pre-constructed medical waste collection route planning model, including:

[0066] S301: Initialize the population;

[0067] S302: Perform interval non-dominated sorting on the population and divide it into subgroups based on the ordinal value;

[0068] S303: Generate new individuals for each subgroup using a gene recombination and greedy crossover strategy;

[0069] S304: Use goal-enhancing mechanisms to search for non-dominant individuals;

[0070] S305: Screening for the next generation of the population;

[0071] S306: Determine whether the maximum number of target evaluations has been reached. If it has, terminate, output the optimal individual, decode the individual, and the decoded path is the medical waste collection path; otherwise, go to step S302.

[0072] In some embodiments, generating new individuals for each subgroup using a gene recombination greedy and crossover strategy specifically includes:

[0073] (1) Greedy cross strategy

[0074] The greedy crossover strategy is an individual update strategy that constructs a path step by step from the starting point. Based on the greedy idea, among the individuals to be crossed, the nearest point from the medical institutions adjacent to the starting point of the new individual is selected as the next medical institution for the new individual, so as to retain the effective information in the crossed individuals. The complete transfer path is generated by repeatedly executing the greedy selection step.

[0075] (2) Gene recombination strategy

[0076] The optimal solution obtained by the greedy crossover strategy described above is used to replace the suboptimal solution, and the encoding of the suboptimal solution is adjusted so that the starting point and waste treatment station in each column of medical institutions are consistent with the original encoding, thus obtaining an individual; then, the starting point is updated; the distance from the starting point to the medical institution in each column of the new individual is calculated, and the first rand1 starting points and medical institutions with the farthest distance are selected, where rand1 is a random number between 1 and the number of medical institutions n; the original starting point is replaced with the starting point corresponding to the medical institution in the optimal solution; finally, the waste treatment station is updated in the same way.

[0077] In some embodiments, based on the dynamic request set obtained in the previous scheduling cycle and the medical waste collection route planning model, the initial vehicle collection route is adjusted using a structural reorganization method to obtain a vehicle collection sub-route, including:

[0078] The medical institutions in the dynamic request set are shuffled to generate an encoded string;

[0079] One of the two-element optimization operators, embedding optimization operators, and exchange optimization operators is randomly selected as the strategy. The encoded string is then optimized and solved according to the medical waste transportation route planning model to obtain the adjusted optimal vehicle transportation sub-route.

[0080] In a second aspect, the present invention provides a medical waste collection device based on an interval multi-objective hybrid frog-jumping algorithm, including a processor and a storage medium;

[0081] The storage medium is used to store instructions;

[0082] The processor is configured to operate according to the instructions to execute the method according to the first aspect.

[0083] Thirdly, the present invention provides an apparatus comprising,

[0084] Memory;

[0085] processor;

[0086] as well as

[0087] Computer programs;

[0088] The computer program is stored in the memory and configured to be executed by the processor to implement the method described in the first aspect above.

[0089] Fourthly, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0090] Beneficial effects:

[0091] This invention provides a medical waste removal method based on a multi-objective hybrid frog-jumping algorithm, which has the following advantages compared with the prior art:

[0092] (1) Considering the uncertainty of vehicle speed, a model for the medical waste collection vehicle routing problem with dynamic uncertainty characteristics was established.

[0093] (2) A dynamic scheduling mechanism was designed, which can dynamically adjust the vehicle's transportation route according to real-time transportation requests or needs. Attached Figure Description

[0094] Figure 1 This is a schematic diagram of a medical waste collection method based on a multi-objective hybrid frog-jumping algorithm according to an embodiment of the present invention;

[0095] Figure 2 This is a schematic diagram comparing the overall mean and the best value on the hypervolume ratio HVR according to an embodiment of the present invention.

[0096] Figure 3 This is a schematic diagram comparing the overall mean and the best value on the reverse generation distance IGD according to an embodiment of the present invention. Detailed Implementation

[0097] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0098] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0099] Example 1

[0100] Firstly, this embodiment provides a medical waste removal method based on an interval multi-objective hybrid frog-jumping algorithm, including:

[0101] Obtain the initial waste removal request and road condition information;

[0102] In the initial scheduling cycle, based on the initial collection request and road condition information, the pre-constructed medical waste collection route planning model is solved using the interval multi-objective hybrid frog leap algorithm to obtain the initial vehicle collection route;

[0103] The vehicles collect garbage according to the initial vehicle collection route until the garbage collection task is completed;

[0104] If the waste collection task is not completed and the next periodic scheduling time point is entered, the initial vehicle collection path is adjusted using the structural reorganization method based on the dynamic request set obtained in the previous scheduling cycle and the medical waste collection path planning model to obtain the vehicle collection sub-path; a new vehicle departs from the depot to execute the vehicle collection sub-path until the waste collection task is completed.

[0105] In some embodiments, the medical waste collection method based on the interval multi-objective hybrid frog-leaping algorithm further includes: when a dynamic event request occurs during periodic scheduling, updating the vehicle collection route and the dynamic request set using a dynamic response mechanism; wherein the dynamic response mechanism includes:

[0106] Dynamic event requests refer to requests generated by customers during the scheduling process, including: new medical waste collection requests from medical institutions and adjustments to customers' expected processing times; among them, new medical waste collection requests from medical institutions refer to new customer requests that occur during the scheduling process, including the time, location, and demand of the new customer; adjustments to customers' expected processing times refer to customers' proactive requests to advance or delay the expected processing time for medical waste.

[0107] Determine whether the dynamic event request can be processed. If it can, update the dynamic event request to the existing path using the dynamic patching method, and the vehicle continues to travel along the updated path; otherwise, the vehicle continues to travel along the original path, and the request is placed in the dynamic request set ΔH, waiting for unified processing at the next periodic scheduling time point.

[0108] The dynamic repair method includes: at time t (t∈[0,T]), there are newly added medical institutions h. u To meet the waste collection and transportation needs, the medical waste collection and transportation route planning model will be used to insert new medical institutions into the current collection and transportation routes, so as to minimize route changes and adapt to the dynamic needs of medical institutions.

[0109] In some specific embodiments, such as Figure 1 As shown, a medical waste removal method based on an interval multi-objective hybrid frog-leaping algorithm includes the following steps:

[0110] S1, initialize the waste collection request and road condition information, and clear the dynamic request set;

[0111] S2. Considering the uncertainty of driving speed, establish a medical waste transportation route planning model;

[0112] S3, the interval multi-objective hybrid frog jumping algorithm is used to determine the initial vehicle removal route that meets the constraints;

[0113] S4. If the garbage collection task is not completed, the vehicle will collect garbage according to the vehicle collection route; otherwise, the vehicle collection route will be output.

[0114] S5. When a garbage collection vehicle arrives at a periodic scheduling time point, a structural reorganization method is used to obtain the vehicle collection sub-path based on the dynamic request set generated at the current time. A new vehicle departs from the depot to execute the vehicle collection sub-path, update the vehicle collection path, and clear the dynamic request set until the garbage collection task is completed. When a dynamic event occurs during periodic scheduling, a dynamic response mechanism is used to update the vehicle collection path and the dynamic request set.

[0115] In the medical waste collection vehicle route planning problem, there are n medical institutions h u (u = 1, 2, ..., n), m waste treatment plants r q (q = 1, 2, ..., m) and one vehicle depot d; vehicles depart from the depot at the initial moment, collect waste from various medical institutions in sequence, then transport it to any appropriate waste treatment station for processing, and finally return to the depot to end the mission; during the collection process, vehicles are allowed to have multiple trips, Trips k Let k be the set of trips. For the w-th journey of vehicle k, L k It is the last journey of vehicle k; the set of medical facilities, waste disposal stations, and parking lots is represented by V = {h1, h2, ..., h}. n ,r1,...,r m The path lengths between these locations are represented by the edge set E = {(p, d}}. i ,p j )|p i ,p j ∈V,i≠j} represents, where p i p j p is the node accessed by the vehicle. i p j ∈H∪R∪{d}, where H is the medical institution h u The set of R, where R is the waste treatment station r q The set; considering the uncertainty of road conditions, for each edge e ij Each has a corresponding range of driving speeds. in v ij and These represent the vehicle on edge e respectively. ij The minimum and maximum driving speeds; the working time period is [0, TM], where TM is the maximum working duration; when a dynamic event occurs during periodic scheduling, t l (l=1,2,...,|T s |) represents the scheduling time point for periodic scheduling, Ts Let the scheduling time t be l The set (e.g., a period of 2 hours, medical institution working hours from 8:00 to 16:00, medical waste collection from 10:00 to 16:00, then T) s ={10,12,14,16}); while dynamic events refer to requests generated by customers during the scheduling process, including: new requests for medical waste collection from medical institutions and adjustments to customers' expected processing times; among them, new dynamic requests refer to new customer requests that occur during the scheduling process, including the time, location and demand of new customers; adjustments to customers' expected processing times refer to customers' proactive requests to advance or delay the expected processing time for medical waste.

[0116] Specifically, the collection requests and road condition information mentioned in step S1 include the number of medical institutions that the medical waste transport vehicle needs to visit, the coordinates of the medical institutions, the coordinates of the medical waste treatment station, the coordinates of the parking lot, and the expected processing time; the dynamic request set stores the dynamic requests generated during the scheduling process. These requests cannot be processed immediately, but are processed uniformly at the start of the next scheduling cycle.

[0117] Specifically, the medical waste disposal route planning model described in step S2 includes three strongly coupled sub-problems:

[0118] (1) Multiple trip determination: Within a scheduling cycle, the number of trips for each vehicle and the medical institutions that each trip passes through are determined;

[0119] (2) Assignment of service vehicles to medical institutions;

[0120] (3) Vehicle selection: Select vehicles according to the waste transportation route.

[0121] Specifically, the scheduling of subproblems is represented by three decision variables:

[0122] Multiple itineraries confirmed

[0123] Medical institution allocation

[0124] Vehicle Selection

[0125] Specifically, the objective function of the medical waste collection route planning model described in step S2 is to minimize scheduling costs and infection risks.

[0126] (1) Scheduling cost

[0127] The scheduling cost f1 includes fixed cost Cost1, fuel consumption cost Cost2, overtime cost Cost3, and pollution cost Cost4, which are defined as follows:

[0128]

[0129] Where k and K are the vehicle index and the set of all vehicles, respectively, and CV represents the driver fee and startup cost required to call a vehicle once;

[0130]

[0131] in, FC ij Indicates that the vehicle passed through road segment e ij fuel consumption cost, α k and β k Characteristic parameters of road conditions and vehicles; Q0, v ij and d ij These are the vehicle's net weight (kg), vehicle load (kg), vehicle speed (m / s), and distance traveled (m). ij For point p i Point P j The distance in European style (km), CF is the price of fuel (yuan / liter);

[0132]

[0133] Where CO represents overtime pay (yuan / hour), tr k The time when vehicle k returns to the depot during its last trip;

[0134]

[0135] Where CP represents pollution cost (yuan / hour), Let vehicle k reach point p. j At that moment, ET j The expected processing time for the waste treatment plant.

[0136] min f1=Cost1+Cost2+Cost3+Cost4

[0137] (2) Risk of infection

[0138] Infection risk f2 includes transportation risk Risk1 and waste disposal risk Risk2, which are defined as follows:

[0139]

[0140] in, For vehicles from p i Click to p j Time spent (in hours) Let vehicle k leave point p iLoad capacity (tons) at that time, P ij For path e ij Population size in the vicinity;

[0141]

[0142] Where R is the waste treatment station r j The set of M j Waste treatment station j Waste treatment capacity (tons), VD j Waste treatment station j Waste processing speed (tons / hour), Waste treatment station j Population in the vicinity.

[0143] min f2=Risk1+Risk2

[0144] Specifically, the initial vehicle removal route determined in step S3 needs to meet the following constraints:

[0145] (1) All vehicles will depart from the depot for the first trip and arrive at any waste disposal station to unload after serving the last medical facility on the trip.

[0146]

[0147] (2) The vehicle’s next journey will depart from the waste treatment station to which it is currently en route;

[0148]

[0149] (3) After the vehicle unloads at the waste treatment station on its last trip, it must return to the depot;

[0150]

[0151] (4) The vehicle arrives at the medical institution. u Then must be from h u The vehicle can only arrive at and leave the medical facility once at most;

[0152]

[0153]

[0154] (5) There is one and only one vehicle serving a particular medical institution;

[0155]

[0156] (6) The amount of waste transported by the vehicle on each trip shall not exceed the vehicle's capacity;

[0157]

[0158] (7) The number of vehicles departing from the parking lot shall not exceed the maximum number of available vehicles;

[0159]

[0160] Specifically, the steps in step S3 for solving the vehicle waste removal route based on the interval multi-objective hybrid frog-jumping algorithm are as follows:

[0161] S301: Initialize the population;

[0162] S302: Perform interval non-dominated sorting on the population and divide it into subgroups based on the ordinal value;

[0163] S303: Generate new individuals for each subgroup using a gene recombination and greedy crossover strategy;

[0164] S304: Use goal-enhancing mechanisms to search for non-dominant individuals;

[0165] S305: Screening for the next generation of the population;

[0166] S306: Determine if the maximum target number of evaluations has been reached. If so, terminate, output the optimal individual, decode the individual, and the decoded path is the medical waste disposal path. Otherwise, proceed to step S302.

[0167] Specifically, step S303, which describes generating new individuals for each subgroup using a gene recombination greedy and crossover strategy, includes:

[0168] (1) Greedy cross strategy

[0169] The greedy crossover strategy is an individual update strategy that constructs a path step by step from the starting point. Based on the greedy idea, among the individuals to be crossed, the nearest point from the medical institutions adjacent to the starting point of the new individual is selected as the next medical institution for the new individual, so as to retain the effective information in the crossed individuals. The complete transfer path is generated by repeatedly executing the greedy selection step.

[0170] (2) Gene recombination strategy

[0171] The optimal solution obtained using the greedy crossover strategy described above replaces the suboptimal solution, and the encoding of the suboptimal solution is adjusted so that the starting point and waste disposal station in each column of medical institutions are consistent with the original encoding, thus obtaining the individual. Then, the starting points are updated. The distance from the starting point to the medical institution in each column of the new individual is calculated, and the first rand1 starting points and medical institutions with the farthest distances are selected, where rand1 is a random number between 1 and the number of medical institutions n. The original starting points are replaced with the starting points corresponding to the medical institutions in the optimal solution. Finally, the waste disposal stations are updated in the same way.

[0172] Specifically, the dynamic response mechanism described in step S5 consists of the following steps:

[0173] Determine whether the request can be processed. If it can, update the existing path with the new request using dynamic patching, and the vehicle continues to travel along the updated path. Otherwise, the vehicle continues to travel along the original path, and the request is placed in the dynamic request set ΔH, waiting for the next scheduling point to process it.

[0174] Among them, the dynamic repair method refers to the method where, at time t (t∈[0,T]), there is a newly added medical institution h. u To meet the waste disposal needs, new medical institutions are integrated into the existing disposal routes to minimize route changes and thus adapt to the dynamic needs of these institutions.

[0175] Specifically, the structural recombination method described in step S5 includes the following steps:

[0176] First, the medical institutions in the dynamic request set are shuffled to generate an encoded string;

[0177] Set the current optimal solution to the encoded string, and randomly select a strategy from the optimization operators O = {2-opt, Insert, Exchange} to optimize the encoded string and generate a solution S;

[0178] If the solution S dominates the encoded string, then the encoded string is equal to the solution S;

[0179] If solution S dominates the optimal solution, then the optimal solution is equal to S;

[0180] When the new solution S cannot dominate the encoded string after three consecutive optimizations, output the current optimal solution.

[0181] Verification Example

[0182] Taking hospitals at or above level II in five districts of Nanjing (Gulou, Jianye, Qinhuai, Qixia, and Xuanwu) as examples, this paper constructs a case study of the dynamic vehicle routing problem for medical waste collection. There are 59 hospitals in the five districts, with three waste treatment stations. Initially, 40 hospitals have already submitted collection requests, and the remaining 19 requests will appear during the scheduling process. Furthermore, based on the dynamic characteristics of the actual medical waste collection process, six simulation examples of progressively increasing size were selected from the commonly used Cordeau-MDVRP test set and modified. The number of stations, station coordinates, and other information remain consistent with the static examples in the original test set. It is assumed that work begins at 7:00 AM, with a maximum continuous working time of 4 hours. The locations of newly added medical institutions are randomly generated, the amount of waste (tons) of each medical institution is randomly generated in the range [0.5, 2], and its expected processing time is randomly generated during the working hours. The overtime pay is set at 20 yuan / hour, the pollution cost at 50 yuan / hour, the maximum load capacity of the collection vehicle is 2 tons, the fixed cost is 100 yuan / vehicle, and the unloading time of the vehicle at the garbage treatment station is 0.1 hours.

[0183] Figure 2 This is a schematic diagram comparing the overall average value (mean) and the best value (Best) on the hypervolume ratio (HVR) according to this embodiment.

[0184] Figure 3 This is a schematic diagram comparing the overall mean and the best value on the reverse generation distance IGD according to this embodiment.

[0185] This embodiment uses the hypervolume ratio (HVR) and inverse generation distance (IGD), commonly used in multi-objective optimization, as evaluation metrics for the algorithm. A larger HVR indicates higher convergence accuracy and a wider distribution range of the Pareto front obtained by the algorithm. A smaller IGD indicates better convergence of the Pareto front and better diversity of solution distribution.

[0186] Depend on Figures 2-3 It can be seen that, compared with the six existing algorithms, the interval multi-objective hybrid frog-leaping algorithm achieves the best overall value (HVR) and the best overall mean (IGD) across all examples. Statistical test results also show that the interval multi-objective algorithm significantly outperforms the comparison algorithms. The superior performance of the proposed algorithm is attributed to the comprehensive application of multiple strategies. In the individual update stage, the algorithm's exploration ability is enhanced by the idea of ​​greedy selection and heuristic replacement of variable scale. In the objective enhancement stage, an optimized objective-driven enhanced search mechanism is adopted to fully utilize the heuristic information of the problem and strengthen the algorithm's utilization ability, thereby achieving a balance between exploration and utilization.

[0187] This invention establishes a medical waste transportation route planning model. This model considers the dynamic demands generated during transportation and the uncertainty of vehicle speed. Under constraints such as vehicle capacity and working time, it optimizes vehicle transportation routes to minimize total scheduling costs and waste transportation risks. To solve this model, a medical waste transportation method based on a multi-objective hybrid frog-leaping algorithm is proposed.

[0188] Example 2

[0189] Secondly, based on Embodiment 1, this embodiment provides a medical waste collection device based on a multi-objective hybrid frog-jumping algorithm, including a processor and a storage medium;

[0190] The storage medium is used to store instructions;

[0191] The processor is configured to operate according to the instructions to execute the method according to Embodiment 1.

[0192] Example 3

[0193] Thirdly, based on Embodiment 1, this embodiment provides a device, including,

[0194] Memory;

[0195] processor;

[0196] as well as

[0197] Computer programs;

[0198] The computer program is stored in the memory and configured to be executed by the processor to implement the method described in Embodiment 1.

[0199] Example 4

[0200] Fourthly, based on Embodiment 1, this embodiment provides a storage medium on which a computer program is stored, and when the computer program is executed by a processor, it implements the method described in Embodiment 1.

[0201] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0202] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0203] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0204] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0205] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A medical waste collection and transportation method based on interval multi-objective hybrid frog leaping algorithm, characterized in that, include: Obtain the initial waste removal request and road condition information; In the initial scheduling period, based on the initial collection request and road condition information, the pre-constructed medical waste collection route planning model is solved using an interval multi-objective hybrid frog-leaping algorithm to obtain the initial vehicle collection routes. The objective function of the medical waste collection route planning model, with the optimization objective of minimizing scheduling cost and infection risk, is expressed as: ； ； wherein the scheduling cost f1 comprises a fixed cost , a fuel cost , an overtime cost and a pollution cost ; the infection risk f2 comprises a transportation risk Risk1 and a waste disposal risk Risk2; A multi-objective hybrid frog-jumping algorithm is used to solve a pre-constructed medical waste collection route planning model, including: S301: initializing the population; S302: sorting the population into non-dominated groups based on the interval order value and dividing them into subgroups; S303: generating new individuals for each subgroup through gene recombination and greedy crossover strategies; S304: searching for non-dominated individuals using an objective enhancement mechanism; S305: selecting the next generation of the population; S306: determining whether the maximum number of objective evaluations has been reached. If it has, the process terminates, the optimal individual is output, and the decoded path is the medical waste collection route; otherwise, proceed to step S302. The vehicles collect garbage according to the initial vehicle collection route until the garbage collection task is completed; If the waste collection task is not completed and the next periodic scheduling time point is entered, the initial vehicle collection route is adjusted using the structural reorganization method based on the dynamic request set obtained in the previous scheduling cycle and the medical waste collection route planning model to obtain the vehicle collection sub-route; a new vehicle departs from the depot to execute the vehicle collection sub-route until the waste collection task is completed. When a dynamic event request occurs during periodic scheduling, a dynamic response mechanism is used to update the vehicle transport route and the dynamic request set. This dynamic response mechanism includes: a dynamic event request refers to a request generated by a customer during the scheduling process, including: newly added medical waste transport requests from medical institutions and adjustments to customers' expected processing times; where a newly added medical waste transport request refers to a new customer request that appears during the scheduling process, including the time, location, and demand of the new customer; and an adjustment to the customer's expected processing time refers to the customer's proactive request to advance or delay the expected processing time for medical waste. The system then determines whether the dynamic event request can be processed. If so, the dynamic event request is updated to the existing route using a dynamic patching method, and the vehicle continues along the updated route; otherwise, the vehicle continues along the original route, and the request is added to the dynamic request set, awaiting unified processing at the next periodic scheduling point.

2. The medical waste collection and removal method based on interval multi-objective mixed frog leap algorithm according to claim 1, characterized in that, The dynamic repairing method comprises the following steps of There are new medical institutions h u If there is a waste removal demand of the medical waste, the new medical institution is inserted into the current removal path based on the medical waste removal path planning model, so as to reduce the change of the path to the greatest extent, thereby adapting to the dynamic demand of the medical institution.

3. The medical waste collection and removal method based on interval multi-objective mixed frog leap algorithm according to claim 1, characterized in that, The initial collection request and road condition information include the number of medical institutions that the medical waste transport vehicle needs to visit, the coordinates of the medical institutions, the coordinates of the medical waste treatment station, the coordinates of the vehicle yard, and the expected processing time.

4. The medical waste collection and removal method based on interval multi-objective mixed frog leap algorithm according to claim 1, characterized in that, The dynamic request set stores dynamic requests generated during the previous scheduling cycle. These requests cannot be processed immediately but are processed uniformly at the start of the next scheduling cycle, and completed cleanup requests are removed from the dynamic request set.

5. The medical waste collection and removal method based on interval multi-objective mixed frog leap algorithm according to claim 1, characterized in that, The medical waste disposal route planning model comprises three strongly coupled sub-problems: (1) Multiple trip determination: Within a scheduling cycle, determine the number of trips for each vehicle and the medical institutions that each trip passes through; (2) Assignment of service vehicles to medical institutions; (3) Vehicle selection: Select vehicles according to the waste transportation route; The scheduling of subproblems is represented by three decision variables: Multi-pass determination ; Medical facility assignment ; Vehicle selection .

6. The medical waste collection and removal method based on interval multi-objective mixed frog leap algorithm according to claim 5, characterized in that, In the problem of planning the transportation routes for medical waste, there are n medical institutions. m family waste treatment station and 1 parking lot ,in , The vehicle departs from the depot at the initial moment, collects waste from various medical institutions in sequence, then transports it to any appropriate waste treatment station for processing, and finally returns to the depot to complete the mission; during the collection process, the vehicle is allowed to make multiple trips. For vehicles The itinerary collection, , For vehicles The Next trip, It is a vehicle The last stop on the itinerary; a combination of medical facilities, waste disposal sites, and parking lots. The path length between these locations is represented by the edge set. It means that, among them, The node accessed by the vehicle. , For medical institutions The set, Waste treatment station The set; considering the uncertainty of road conditions, for each edge Each has a corresponding range of driving speeds. ,in and These represent the vehicles on the side. The minimum and maximum driving speeds; the working time period is [0, TM], where TM is the maximum working time; For periodic scheduling, the scheduling time points, For scheduling time A set; ； wherein, representing a vehicle selecting whether or not to use, is a set of all vehicles, CV represents the driver's remuneration and start-up costs that need to be paid out by the vehicle for one call; ； wherein, , represents the fuel consumption cost of vehicle k through road segment , and are characteristic parameters of road condition and vehicle; , , are the net weight of vehicle, the load of vehicle, the driving speed of vehicle, is the Euclidean distance from point to point , is the price of fuel; is the wth trip of vehicle k from point to point ; ； wherein, for overtime pay, for the vehicle the time of the last trip back to the depot; ； wherein CP is the pollution cost, for the vehicle of arrival point of the time, is the desired treatment time for the waste treatment station; ； wherein, for the vehicle from point to point, for the vehicle at the departure point at the departure point, for the route the number of people in the vicinity; ； wherein R is the waste treatment station , is the waste treatment station , is the waste treatment amount of the waste treatment station , is the speed of waste treatment of the waste treatment station , is the population number near the waste treatment station , is the waste treatment station .

7. The medical waste collection and removal method based on interval multi-objective mixed frog leap algorithm according to claim 6, characterized in that, The constraints of the medical waste collection route planning model include: (1) All vehicles will depart from the depot for the first trip and arrive at any waste disposal station to unload after serving the last medical facility on the trip. ； wherein, is the wth trip of vehicle k from a medical facility to a waste treatment station ; (2) The vehicle’s next journey will depart from the waste treatment station to which it is currently en route; ； (3) After unloading at the waste treatment station on its last trip, the vehicle must return to the depot; ； wherein, for the last trip of vehicle k from the waste treatment station back to the depot ; (4) the vehicle arrives at the medical facility must be from departure, and the vehicle can only arrive at and depart from the medical facility once; ； ； wherein, is the wth trip of vehicle k from point to a medical facility ; is the wth trip of vehicle k from the medical facility to point ; is the medical facility for which vehicle k is responsible; (5) There is one and only one vehicle serving a particular medical institution; ； (6) the amount of waste transported by the vehicle on each trip does not exceed the capacity of the vehicle ; ； wherein, is the amount of waste collected by the medical institution; (7) The number of vehicles departing from the parking lot shall not exceed the maximum number of available vehicles; ； wherein, is the wth trip of vehicle k from the depot to the medical facility .

8. The medical waste collection and removal method based on interval multi-objective mixed frog leap algorithm according to claim 1, characterized in that, In step S303, generating new individuals for each subgroup using a gene recombination greedy and crossover strategy specifically includes: (1) Greedy cross strategy The greedy crossover strategy is an individual update strategy that constructs a path step by step from the starting point. Based on the greedy idea, among the individuals to be crossed, the nearest point from the medical institutions adjacent to the starting point of the new individual is selected as the next medical institution for the new individual, so as to retain the effective information in the crossed individuals. The complete transfer path is generated by repeatedly executing the greedy selection step. (2) Gene recombination strategy The optimal solution obtained by the greedy crossover strategy described above is used to replace the suboptimal solution, and the encoding of the suboptimal solution is adjusted so that the starting point and waste treatment station in each column of medical institutions are consistent with the original encoding, thus obtaining an individual; then, the starting point is updated; the distance from the starting point to the medical institution in each column of the new individual is calculated, and the first rand1 starting points and medical institutions with the farthest distance are selected, where rand1 is a random number between 1 and the number of medical institutions n; the original starting point is replaced with the starting point corresponding to the medical institution in the optimal solution; finally, the waste treatment station is updated in the same way.

9. The medical waste collection and removal method based on interval multi-objective mixed frog leap algorithm according to claim 1, characterized in that, Based on the dynamic request set obtained in the previous scheduling cycle and the medical waste collection route planning model, the initial vehicle collection route is adjusted using the structural reorganization method to obtain vehicle collection sub-routes, including: The medical institutions in the dynamic request set are shuffled to generate an encoded string; One of the two-element optimization operators, embedding optimization operators, and exchange optimization operators is randomly selected as the strategy. The encoded string is then optimized and solved according to the medical waste transportation route planning model to obtain the adjusted optimal vehicle transportation sub-route.

10. A medical waste removal device based on interval multi-objective shuffled frog leaping algorithm, characterized in that, Including processor and storage media; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the method according to any one of claims 1 to 9.

Images