A method for optimizing the distribution path of urban emergency materials

Abstract

The present invention provides a method for optimizing the distribution path of urban emergency materials, including: establishing an evaluation index system for the urgency of demand points of emergency materials according to the newly inflow and outflow population in the jurisdiction, newly confirmed cases in the jurisdiction, hospital beds, etc.; constructing an optimization model for the distribution path of emergency materials with the goal of maximizing vehicle utilization and minimizing vehicle transportation costs; using the AHP hierarchical analysis method to calculate the weight of each indicator in the demand urgency evaluation index system regarding the demand urgency; introducing a universe difference operator and a random search enhancement mechanism based on Tent chaotic mapping to improve the multiverse algorithm, using a hybrid multiverse algorithm based on differential evolution to solve the path optimization model, and generating an optimized urban emergency material distribution path. The present invention improves the existing evaluation index system, studies the distribution of heterogeneous emergency materials, and the established optimization method for the distribution path of urban emergency materials can better cope with cities with epidemic outbreaks.

Description

Technical Field

[0001] The present invention relates to the field of emergency material distribution, and in particular to a method for optimizing the path of urban emergency material distribution. Background Art

[0002] With the rapid development of the economy, the size of cities in my country is increasing. When dealing with the problem of urban emergency material distribution, the general VRP problem solution is no longer applicable. Therefore, how to efficiently distribute urban emergency materials during an epidemic outbreak, while ensuring that emergency materials can be delivered in time and fairly distributed to various demand points, is the primary problem to be solved in urban emergency material distribution in response to epidemic outbreaks.

[0003] The current urban emergency material distribution has the following characteristics: 1) frequent logistics activities and large amounts of information; 2) short transportation distances; 3) high timeliness requirements; and 4) a wide variety of transportation items and small batches.

[0004] At present, there are few studies specifically on the distribution of emergency supplies during epidemic outbreaks. Due to the limited emergency supplies and limited transportation capacity of distribution companies, the urgency of the needs of various hospitals for various emergency supplies is not taken into consideration when distributing emergency supplies during epidemic outbreaks. The existing evaluation system is not perfect and reasonable. Summary of the invention

[0005] Due to the lack of research on the optimization of emergency material distribution in the event of an epidemic outbreak, emergency materials cannot be fairly distributed and delivered in a timely manner. For this reason, the present invention provides an optimization method for the urban emergency material distribution path. By quantifying the urgency of each hospital's demand for various materials and considering the utilization rate of vehicles and their transportation costs, an urban emergency material distribution path optimization model for responding to epidemic outbreaks is constructed to achieve efficient distribution of emergency materials.

[0006] The present invention provides a method for optimizing the urban emergency material distribution path, comprising the following steps:

[0007] Step 1: Obtain the infection status of patients and medical supplies reserves at each demand point, where each demand point can be a hospital, assess the specific needs of each hospital for various emergency supplies, and incorporate specific information such as new floating population, newly confirmed cases, and hospital beds in the jurisdiction into the evaluation system, so as to quantify the urgency of each hospital's demand for various supplies, establish a hospital demand urgency evaluation index system, and then construct the hospital's demand urgency model.

[0008] Step 2: With the goal of maximizing vehicle utilization and minimizing vehicle transportation costs, an emergency material distribution route optimization model is constructed;

[0009] Step 3: Calculate the weight of each indicator in the demand urgency evaluation index system established in step 1 using the AHP hierarchical analysis method.

[0010] Step 4: Apply intelligent algorithms to solve the model in step 2 to generate an optimized urban emergency material distribution route.

[0011] Furthermore, constructing the demand urgency model of the hospital in step 1 includes:

[0012] The infection situation, medical infrastructure situation and emergency material reserve situation of the demand points are obtained to construct an emergency material demand urgency evaluation index system table. Based on the indicator data in the table, the urgency of each hospital's demand for various materials is obtained, and a hospital's demand urgency model for materials is constructed.

[0013] Furthermore, the data in the demand urgency evaluation index system table is normalized:

[0014]

[0015] Among them, θ max is the maximum value of each indicator in the table, θ min Indicates the minimum value of each indicator in the table, C * is the standardized value of each indicator after processing;

[0016] Through the expert scoring method, we collected the opinions of several experts in this field and asked them to score the weights of each indicator that needed to be judged. The obtained weights were weighted and calculated to further obtain the urgency values ​​of each hospital's demand for various materials:

[0017]

[0018] Among them, ω r is the weight of each indicator r,(r=1,2,···,k) in the indicator layer to the urgency of demand, k is the total number of indicators; C jir is the standardized value of evaluation index r of hospital j regarding material i;

[0019] Since the scope of the demand urgency ratio target problem is unknown, the value of the relative demand urgency of hospital j for material i is calculated as ji , as the condition for judging the range; taking the minimum demand urgency value as the benchmark, we have:

[0020]

[0021] Furthermore, the vehicle utilization rate in step 2 mainly considers the load utilization rate and volume utilization rate; the vehicle load is generally optimized based on the experience of the warehouse manager; the evaluation standard of vehicle utilization rate is to select the larger value of the two when loading the same batch of goods:

[0022]

[0023] τ=∑η An (5)

[0024] Among them, W maxn is the maximum load of vehicle type n, V maxn represents the maximum loading volume of vehicle model n, m is the total number of hospitals delivered by vehicle A, and τ is the total utilization rate of the vehicle;

[0025] Transportation costs for vehicle A:

[0026]

[0027] In formula (6),

[0028]

[0029]

[0030] Furthermore, an optimization model for urban emergency material distribution paths is established:

[0031]

[0032]

[0033]

[0034]

[0035]

[0036]

[0037]

[0038]

[0039]

[0040]

[0041] Among them, N nmax is the maximum number of vehicles of model n, H SUM is the total number of hospitals, is the time it takes for the vehicle to actually deliver the emergency supplies to hospital j, T j The latest time for a vehicle to deliver emergency supplies to the hospital.

[0042] Formulas (11) and (12) are the maximum weight limit and maximum volume limit of the vehicle, respectively. Formula (13) indicates that a standard van can deliver emergency supplies to a maximum of five hospitals at a time. Formula (14) indicates that the number of vehicles of each type cannot exceed its maximum number. Formula (15) indicates that all supplies must be delivered to each hospital without omission. Formulas (10), (16), (17), and (18) are 0-1 variables.

[0043] Furthermore, the weighted data of the expert scores in step 1 are averaged by removing the maximum and minimum scores, and the judgment matrix is ​​listed. The weights are calculated using the sum-product method, and a consistency test is performed to identify whether the judgment matrix is ​​acceptable.

[0044] Furthermore, by drawing on the mechanism of differential evolution algorithm, the universe difference operator and the random search enhancement mechanism based on Tent chaotic mapping are introduced to improve the multiverse algorithm (MVO).

[0045] Furthermore, the model in step 2 is solved by using a hybrid multiverse algorithm based on differential evolution (DE-IMVO), and the specific steps include:

[0046] Step 4.1: Initialize algorithm parameters: including universe dimension D; crossover probability CR; difference coefficient F; number of universes NP; number of iterations G; search space upper limit x max ; Search space lower limit x min ; Vehicle speed v; Freight per unit time M;

[0047] Step 4.2: Multiverse initialization;

[0048] According to the total number of orders N o Loop and generate a value between 1 and N for each order. v After the loop is completed, calculate the number of orders for each vehicle and all the routes for delivering the orders, calculate the one with the shortest time, and determine whether it meets the constraints. If it meets the constraints, it will be added to the multiverse. If it does not meet the constraints, repeat this step until the multiverse has NP universes that meet the constraints.

[0049] Step 4.3: Calculate the expansion of the universe and normalize the expansion;

[0050] The expansion degree of each universe is calculated according to the value of the objective function. In this model, the universe with the largest expansion degree represents the optimal solution, so the universe that does not meet the constraints is assigned a negative infinity value; the expansion degree of the universe is standardized using the standardized formula:

[0051]

[0052] Step 4.4: Update the global extremum and save the universe with the largest expansion as

[0053] Step 4.5: The universe exchanges matter through black holes and white holes;

[0054] For each universe Generate a random number r between 0 and 1 i 1 ,like The universe that produces the black hole is selected by roulette A universe with a black hole sends matter into this universe:

[0055]

[0056] Step 4.6: The universe receives matter sent from the optimal universe;

[0057] like Then a random number r between 0 and 1 is generated for the universe i 2 , if r i 2 <WEP, then generate a random number r i 3 , the optimal universe sends matter to this universe through the wormhole:

[0058]

[0059] in, is the standard expansion rate of the universe, WEP is the probability of wormhole existence;

[0060] Save the universe following the optimal universe search;

[0061] Step 4.7: Random search based on Tent chaotic map;

[0062] If r i 2 >WEP, then perform a chaotic search on the universe:

[0063] ①Generate interval in [(x max -x min ),2(x max -x min )] i 4 , let the minimum search step length be:

[0064]

[0065] ②Generate a random number r between -0.5 and 0.5 i 5 , the cosmic chaos search strategy is:

[0066]

[0067] ③Update search step

[0068]

[0069] ④ Calculate the expansion of the universe and perform individual updates

[0070]

[0071] Step 4.8: Determine whether the maximum number of iterations has been reached. If so, exit the optimization. If not, return to step 4.3.

[0072] Step 4.9: Output the result that satisfies the constraints and objective function;

[0073] The above differential evolution-based hybrid multiverse algorithm (DE-IMVO) is used to generate the optimized urban emergency material distribution path.

[0074] Beneficial effects of the present invention:

[0075] (1) The present invention provides a method for optimizing the distribution path of urban emergency supplies. When considering the problem of distributing emergency supplies in a city, the urgency of the demand at each demand point is different, and each demand point can be a hospital. Furthermore, the information of new floating population, newly confirmed cases, hospital beds, etc. in the jurisdiction is included in the evaluation system, and the evaluation system is improved to make it more scientific and reasonable.

[0076] (2) The present invention takes into account the transportation capacity constraints of emergency supplies and distribution companies, effectively quantifies the urgency of each hospital's demand for various supplies, considers the utilization rate of vehicles and their transportation costs, and designs an emergency supply distribution path optimization model, thereby effectively coping with the effective distribution of urban emergency supplies in the event of an epidemic outbreak. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 This is a flow chart of the optimization method provided by the present invention.

[0078] Figure 2 This is the consistency weight result test chart.

[0079] Figure 3 Visualize the vehicle transport-order diagram.

[0080] Figure 4 This is a flow chart of the improved universe algorithm of the present invention. DETAILED DESCRIPTION

[0081] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0082] The present invention provides a method for optimizing the distribution path of urban emergency materials, which mainly includes the following steps:

[0083] Step 1: Obtain the patient infection situation and medical supplies reserves of each demand point, where each demand point can be a hospital, evaluate the specific needs of each hospital for various emergency supplies, and incorporate specific information such as new floating population, newly confirmed cases, and hospital beds in the jurisdiction into the evaluation system, establish an evaluation index system for the urgency of demand for each hospital, as shown in Table 1, and then construct a demand urgency model for each hospital.

[0084]

[0085] Table 1: Evaluation index system for the urgency of demand for emergency supplies

[0086] Since the data units of beds, personnel, etc. in the above table are not unified, the indicators in Table 1 are first put into SPSS software for normalization:

[0087]

[0088] Among them, θ max is the maximum value of each indicator in the above table, θ min Indicates the minimum value of each indicator in the above table, C * are the standardized values ​​of each indicator after processing.

[0089] Through the expert scoring method, we collected the opinions of several experts in this field and asked them to score the weights of each indicator that needed to be judged. The obtained weights were weighted and calculated to further obtain the urgency values ​​of each hospital's demand for various materials:

[0090]

[0091] Among them, ω r is the weight of each indicator r,(r=1,2,···,k) in the indicator layer to the urgency of demand, k is the total number of indicators; C jir is the standardized value of evaluation index r of hospital j regarding material i.

[0092] Since the scope of the demand urgency ratio target problem is unknown, the value of the relative demand urgency of hospital j for material i is calculated as ji , as the condition for judging the range; taking the minimum demand urgency value as the benchmark, we have:

[0093]

[0094] Step 2: With the goal of maximizing vehicle utilization and minimizing vehicle transportation costs, an emergency material distribution route optimization model is constructed;

[0095] Vehicle utilization mainly considers load utilization and volume utilization. The load of the vehicle is generally optimized based on the experience of the warehouse manager; the evaluation standard of vehicle utilization is to select the larger value of the two when loading the same batch of goods:

[0096]

[0097] τ=∑η An (5)

[0098] Among them, W maxn is the maximum load of vehicle type n, V maxn represents the maximum loading volume of vehicle model n, m is the total number of hospitals delivered by vehicle A, and τ is the total utilization rate of the vehicle.

[0099] Transportation costs for vehicle A:

[0100]

[0101] Among them, L oj represents the distance from distribution center O to hospital j; L jj′ represents the distance from hospital j to hospital j′.

[0102]

[0103]

[0104] Establish an optimization model for urban emergency material distribution routes:

[0105]

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112]

[0113]

[0114]

[0115] Among them, N nmax is the maximum number of vehicles of model n, H SUM is the total number of hospitals, is the time it takes for the vehicle to actually deliver the emergency supplies to hospital j, T j The latest time for a vehicle to deliver emergency supplies to the hospital.

[0116] Formulas (11) and (12) are the maximum weight limit and maximum volume limit of the vehicle respectively; Formula (13) indicates that a standard van can deliver emergency supplies to a maximum of five hospitals at a time; Formula (14) indicates that the number of vehicles of each type cannot exceed its maximum number; Formula (15) indicates that all kinds of supplies must be delivered to each hospital without omission; Formulas (10), (16), (17), and (18) are 0-1 variables.

[0117] Step 3: Calculate the weight of each indicator in the "evaluation index system" established in step 1 regarding the urgency of demand using the AHP hierarchical analysis method.

[0118] Remove the maximum and minimum scores from the weighted data of the experts' scores in step one and calculate the average value. List the judgment matrix, put the data into SPSS software, and use the sum-product method to calculate the weight.

[0119] The calculation results are shown in Table 2:

[0120]

[0121] Table 2: AHP hierarchical analysis results

[0122] In order to identify whether the judgment matrix is ​​acceptable, a consistency check is required.

[0123] Depend on Figure 2 It can be seen that the CR value is 0.030<0.1, which meets the conditions, indicating that the established judgment matrix is ​​acceptable and the obtained weights are consistent.

[0124] The solution is as follows:

[0125] The model is discretized to reduce the complexity of the problem and enhance the optimization effect of the algorithm. That is, the demand for a certain material in different hospitals can only be transported by one vehicle, and the vehicle is used as a variable for optimization. The number of hospitals multiplied by the type of material is the number of orders. For a batch of orders, a random integer between 1 and the total number of vehicles is generated for each order, representing the order transported by the vehicle with the number, as shown in the visualization. Figure 3 .

[0126] Step 4: Apply intelligent algorithms to solve the model.

[0127] Since the model problem in this paper is relatively complex, there are many constraints, the objective function is sensitive to individual changes, and the solution results using the global update strategy are unsatisfactory. Therefore, we refer to the mechanism of differential evolution algorithm and introduce Figure 4 The cosmic difference operator and the random search enhancement mechanism based on Tent chaotic mapping are used to improve the multiverse algorithm (MVO).

[0128] This paper uses the improved universe algorithm, namely the differential evolution based hybrid multiverse algorithm (DE-IMVO) to solve the model in step 2. The specific steps are as follows:

[0129] Step 4.1: Initialize algorithm parameters: including universe dimension D; crossover probability CR; difference coefficient F; number of universes NP; number of iterations G; search space upper limit x max ; Search space lower limit x min ; Vehicle speed v; Freight per unit time M.

[0130] Step 4.2: Multiverse initialization.

[0131] According to the total number of orders N o Loop and generate a value between 1 and N for each order. v After the loop is completed, calculate the orders that each vehicle has, and calculate all the paths for delivering the orders, calculate the one that takes the least time, and determine whether it meets the constraints. If it does, it will be added to the multiverse. If it does not, repeat this step until the multiverse has NP universes that meet the constraints.

[0132] Step 4.3: Calculate the expansion of the universe and normalize the expansion.

[0133] The expansion of each universe is calculated based on the value of the objective function. In this model, the universe with the largest expansion represents the optimal solution, so the universe that does not meet the constraints is assigned a negative infinity value. The expansion of the universe is standardized using the standardized formula:

[0134]

[0135] Step 4.4: Update the global extremum and save the universe with the largest expansion as

[0136] Step 4.5: The universe exchanges matter through black holes and white holes.

[0137] For each universe Generate a random number r between 0 and 1 i 1 ,like The universe that produces the black hole is selected by roulette A universe with a black hole sends matter into this universe:

[0138]

[0139] Step 4.6: The universe receives matter sent from the optimal universe.

[0140] like Then a random number r between 0 and 1 is generated for the universe i 2 , if r i 2 <WEP, then generate a random number r i 3 , the optimal universe sends matter to this universe through the wormhole:

[0141]

[0142] in: is the standard expansion rate of the universe, WEP is the probability of wormhole existence;

[0143] Save the universe following the optimal universe search;

[0144] Step 4.7: Random search based on Tent chaotic map.

[0145] If r i 2 >WEP, then perform a chaotic search on the universe:

[0146] ①Generate interval in [(x max -x min ),2(x max -x min )] i 4 , let the minimum search step length be:

[0147]

[0148] ②Generate a random number r between -0.5 and 0.5 i 5 , the cosmic chaos search strategy is:

[0149]

[0150] ③Update search step

[0151]

[0152] ④ Calculate the expansion of the universe and perform individual updates

[0153]

[0154] Step 4.8: Determine whether the maximum number of iterations has been reached. If so, exit the optimization. If not, return to step 4.3.

[0155] Step 4.9: Output the results that satisfy the constraints and objective function.

[0156] The model is solved by the above-mentioned differential evolution-based hybrid multiverse algorithm (DE-IMVO) to generate the optimized urban emergency material distribution path.

[0157] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for optimizing the distribution path of urban emergency materials, comprising the following steps: Step 1: Obtain the infection status of patients and the reserve status of medical supplies at each demand point, where each demand point is a hospital, evaluate the specific needs of each hospital for various emergency supplies, and incorporate the new population in the jurisdiction, new influx of population, confirmed cases, newly confirmed cases, hospital beds, number of medical staff, and the degree of shortage of a certain type of emergency supplies into the evaluation system, establish a hospital demand urgency evaluation index system, and then construct a hospital demand urgency model; Among them, the data in the demand urgency evaluation index system table is normalized: Among them, θ max is the maximum value of each indicator in the table, θ min Indicates the minimum value of each indicator in the table, C * is the standardized value of each indicator after processing; Through the expert scoring method, we collect the opinions of experts and score the weights of each indicator that needs to be judged. We perform weighted calculation on the obtained weights to obtain the urgency value of each hospital's demand for various materials: Among them, ω r is the weight of each indicator r,(r=1,2,···,k) in the indicator layer to the urgency of demand, k is the total number of indicators; C jir is the standardized value of evaluation index r of hospital j regarding material i; Calculate the value of the relative urgency of hospital j’s demand for material i, γ ji , as the condition for judging the range; taking the minimum demand urgency value as the benchmark, we have: Step 2: With the goal of maximizing vehicle utilization and minimizing vehicle transportation costs, an emergency material distribution route optimization model is constructed; The vehicle utilization rate in step 2 includes load utilization rate and volume utilization rate. The load of the vehicle is optimized based on the experience of the warehouse manager. The evaluation standard of vehicle utilization rate is to select the larger value of the two values ​​for loading the same batch of goods: τ=∑η An (5); Among them, W maxn is the maximum load of vehicle type n, V maxn represents the maximum loading volume of vehicle model n, m is the total number of hospitals delivered by vehicle A, and τ is the total utilization rate of the vehicle; Transportation costs for vehicle A: Among them, L oj represents the distance from distribution center O to hospital j; L jj' represents the distance from hospital j to hospital j′; In formula (6), Among them, an optimization model for urban emergency material distribution paths is established: Among them, N nmax is the maximum number of vehicles of model n, H SUM is the total number of hospitals, To deliver emergency supplies to hospitals by vehicle j Time, T j The latest time for vehicles to deliver emergency supplies to hospitals; Formulas (11) and (12) are the maximum weight limit and maximum volume limit of the vehicle, respectively; Formula (13) indicates that a standard van can deliver emergency supplies to a maximum of five hospitals at a time; Formula (14) indicates that the number of vehicles of each type cannot exceed its maximum number; Formula (15) indicates that various supplies must be delivered to each hospital without omission; Formulas (10), (16), (17), and (18) are 0-1 variables; Step 3: Calculate the weight of each indicator on demand urgency in the demand urgency evaluation index system established in step 1 using the AHP hierarchical analysis method; Step 4: Apply intelligent algorithms to solve the model in step 2 to generate an optimized urban emergency material distribution route.

2. The method for optimizing the urban emergency material distribution path according to claim 1, characterized in that: Constructing the demand urgency model of the hospital in step 1, including: The infection situation, medical infrastructure situation and emergency material reserve situation of the demand points are obtained to construct an emergency material demand urgency evaluation index system table. Based on the indicator data in the table, the urgency of each hospital's demand for various materials is obtained, and a hospital's demand urgency model for materials is constructed.

3. The method for optimizing the urban emergency material distribution path according to claim 1, characterized in that: Remove the maximum and minimum scores from the weighted data of the expert scores in step 1 and calculate the average value. List the judgment matrix, use the sum-product method to calculate the weights, and perform a consistency test to determine whether the judgment matrix is ​​acceptable.

4. The method for optimizing the urban emergency material distribution path according to claim 3 is characterized in that: Drawing on the mechanism of differential evolution algorithm, the multiverse algorithm (MVO) was improved by introducing universe difference operator and random search enhancement mechanism based on Tent chaotic mapping.

5. The method for optimizing the urban emergency material distribution path according to claim 4, characterized in that: The model in step 2 is solved using the differential evolution based hybrid multiverse algorithm (DE-IMVO). The specific steps include: Step 4.1: Initialize algorithm parameters: including universe dimension D; crossover probability CR ; difference coefficient F; number of universes NP; number of iterations G; upper limit of search space x max ; Search space lower limit x min ; Vehicle speed v; Freight per unit time M; Step 4.2: Multiverse initialization; According to the total number of orders N o Loop and generate a value between 1 and N for each order. v After the loop is completed, calculate the number of orders for each vehicle and all the routes for delivering the orders, calculate the one with the shortest time, and determine whether it meets the constraints. If it meets the constraints, it will be added to the multiverse. If it does not meet the constraints, repeat this step until the multiverse has NP universes that meet the constraints. Step 4.3: Calculate the expansion of the universe and normalize the expansion; The expansion degree of each universe is calculated according to the value of the objective function. In this model, the universe with the largest expansion degree represents the optimal solution, so the universe that does not meet the constraints is assigned a negative infinity value; the expansion degree of the universe is standardized using the standardized formula: Step 4.4: Update the global extremum and save the universe with the largest expansion as Step 4.5: The universe exchanges matter through black holes and white holes; For each universe Generate a random number r between 0 and 1 i 1 ,like The universe that produces the black hole is selected by roulette A universe with a black hole sends matter into this universe: Step 4.6: The universe receives matter sent from the optimal universe; like Then a random number r between 0 and 1 is generated for the universe i 2 , if r i 2 <WEP, then generate a random number r i 3 , the optimal universe sends matter to this universe through the wormhole: in, is the standard expansion rate of the universe, WEP is the probability of wormhole existence; Save the universe following the optimal universe search; Step 4.7: Random search based on Tent chaotic map; If r i 2 >WEP, then perform a chaotic search on the universe: ①Generate interval in [(x max -x min ),2(x max -x min )] i 4 , let the minimum search step length be: ②Generate a random number r between -0.5 and 0.5 i 5 , the cosmic chaos search strategy is: ③Update search step ④ Calculate the expansion of the universe and perform individual updates Step 4.8: Determine whether the maximum number of iterations has been reached. If so, exit the optimization. If not, return to step 4.

3. Step 4.9: Output the results that satisfy the constraints and objective functions; use the above-mentioned differential evolution-based hybrid multiverse algorithm (DE-IMVO) to generate the optimized urban emergency material distribution path.

Images