Product oil distribution path planning method and system

Abstract

The invention relates to the technical field of dangerous goods transportation, and discloses a refined oil product distribution path planning method and system, and the method comprises the following steps: calculating a comprehensive safety risk value of each road section in a road network according to a calculation formula: the comprehensive safety risk value = risk concentration * road section length * passing time; performing weighted sum on the comprehensive safety risk values of all passing road sections of one candidate path to obtain a total safety risk of the candidate path; constructing a vehicle path planning function of multi-objective planning; setting constraint conditions for the vehicle path planning function; and solving the vehicle path planning function, and taking the candidate path corresponding to the solution of the vehicle path planning function when the target function value is minimum as the optimal path. According to the method, the problem that path optimization is difficult to realize on the basis of considering the social public security risk in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of dangerous goods transportation, and particularly relates to a finished oil distribution path planning method and system. BACKGROUND

[0002] Finished oil transportation is a key link connecting a refinery and an end user, and the core thereof lies in balancing efficiency, cost and safety. Since a transportation mode directly affects logistics cost and market response speed, the transportation mode needs to be flexibly selected according to oil product characteristics, distance and demand. A finished oil transportation scheme combines multiple transportation modes, highlights a path scheme strategy of optimizing economic cost, time cost and safety risk, and ensures efficient distribution.

[0003] Finished oil distribution optimization targets mainly focus on economic cost (such as shortest total mileage, least number of vehicles and lowest fuel consumption) and time efficiency (such as punctuality), and basically do not consider social public safety risk of a distribution path, which leads to an accident. SUMMARY

[0004] In order to overcome the defects of the prior art, the present application provides a finished oil distribution path planning method and system, which solves the problem that the prior art cannot realize path optimization on the basis of considering social public safety risk.

[0005] The technical solution of the present application for solving the above technical problem is as follows: A finished oil distribution path planning method comprises the following steps: calculating a comprehensive safety risk value of each road section in a road network, and the calculation formula is: comprehensive safety risk value = risk concentration x road section length x passing time; weighting and summing the comprehensive safety risk values of all road sections passed by a candidate path to obtain a total safety risk of the candidate path; constructing a vehicle path planning function of multi-objective programming, and the formula of the vehicle path planning function is: Minimize Z = [α*Total_Cost, β*Total_Time, γ*Total_Risk]; wherein Minimize represents a minimum operation, Z represents a target function value, α represents a first weight, Total_Cost represents a total economic cost, β represents a second weight, Total_Time represents a total time cost, γ represents a third weight, and Total_Risk represents a total safety risk; setting a constraint condition for the vehicle path planning function; wherein the constraint condition comprises one or more of the following: a station time window, a vehicle loading rate, a vehicle fixed cost, a number of vehicles, a customer point, a road and bridge fee and a fuel cost; Solve the vehicle path planning function, and take the candidate path corresponding to the solution of the vehicle path planning function that minimizes the objective function value as the optimal path.

[0006] This invention calculates the comprehensive safety risk value of each road segment in the road network based on risk concentration, road segment length, and travel time. Then, it further weights and sums the comprehensive safety risk values ​​of all road segments along a candidate route to obtain the total safety risk of the candidate route. Next, it constructs a vehicle route planning function by comprehensively considering total economic cost, total time cost, and total safety risk, taking station time windows, vehicle loading rate, vehicle fixed costs, number of vehicles, customer points, road and bridge tolls, and fuel costs as constraints to optimize the route. Thus, considering the social public safety risks of the delivery route, it obtains the optimal route scheme in terms of total economic cost, total time cost, and total safety risk, making route planning more scientific and more suitable for the characteristics of the refined oil delivery industry.

[0007] As a preferred technical solution, γ=0 is set, and safety risk is used as a constraint condition for the vehicle path planning function.

[0008] In addition to using safety risks as risk factors, safety risks can also be used as constraints on vehicle path planning functions, thus realizing a second method of using safety risks for path planning.

[0009] As a preferred technical solution, solving the vehicle path planning function and selecting the candidate path corresponding to the solution of the vehicle path planning function that minimizes the objective function value as the optimal path includes the following steps: Set initialization parameters; The initialization parameters include one or more of the following: Constraints, site information, vehicle information, function execution information; Generate an initial population, each initial population consisting of one or more individuals, each individual corresponding to a candidate path; Calculate the fitness of each individual; A roulette wheel selection is performed based on fitness to select individuals to enter the offspring; The selected individuals are cross-linked pairwise to generate new individuals; Add new individuals to the population and calculate their fitness. Individuals with high fitness are selected from both the parent and offspring generations to form the next generation of the population. Repeat the roulette wheel selection based on fitness to select individuals to enter the offspring until the maximum number of iterations is reached or the optimal solution for consecutive generations remains unchanged, then proceed to the next step. The individual with the highest fitness in the population is output as the optimal path.

[0010] The periodic vehicle routing problem is solved by using the PVRP solution algorithm.

[0011] As a preferred technical solution, the site information includes one or more of the following: Site address, site time window, site fuel quantity, site fuel volume, site fuel type.

[0012] The content included in the site information is further clarified, making the path planning effect better.

[0013] As a preferred technical solution, the vehicle information includes one or more of the following: number of vehicles, fixed cost, loading capacity, loading volume, loading energy type.

[0014] The content included in the vehicle information is further clarified, making the path planning effect better.

[0015] As a preferred technical solution, the function execution information includes one or more of the following: crossover probability, mutation probability, maximum number of iterations.

[0016] The content included in the function execution information is further clarified, making the path planning effect better.

[0017] As a preferred technical solution, roulette wheel selection is performed according to the fitness to select individuals entering the offspring, including the following steps: Based on the fitness of individuals, the total fitness is calculated. The calculation formula is: F = Σf_i; where, F represents the total fitness, i represents the individual number, i is an integer and 1 ≤ i ≤ N, N represents the total number of individuals in the population, and f_i represents the fitness of the i-th individual; Calculate the selection probability of each individual. The calculation formula is: P_i = f_i / F; where, P_i represents the selection probability of the i-th individual; Calculate the cumulative probability of each individual. The calculation formula is: Q_i = ΣP_i; where, Q_i represents the selection probability of the i-th individual; Generate a random number r uniformly distributed in the interval [0, 1); Select the individuals entering the offspring according to the random number r. The selection method is: If r ≤ Q_1, then select the first individual; If Q_{i - 1} < r ≤ Q_i, then select the i-th individual.

[0018] The content included in selecting the individuals entering the offspring by roulette wheel selection according to the fitness is further clarified, making the path planning effect better.

[0019] As a preferred technical solution, the risk concentration is the weighted sum of risk factors; where the risk factors include static risk factors and dynamic risk factors.

[0020] The content of risk concentration (especially risk factors) has been further clarified, which makes the path planning more effective.

[0021] As a preferred technical solution, static risk factors include one or more of the following: sensitive areas, infrastructure, and historical accident data; dynamic risk factors include one or more of the following: real-time population density, real-time traffic flow, weather conditions, and manual dynamic adjustments.

[0022] The content of static and dynamic risk factors has been further clarified, which makes the path planning more effective.

[0023] A refined oil product distribution route planning system includes a processor, a memory, and a computer program stored in the memory. When the computer program is executed by the processor, it implements the steps of the refined oil product distribution route planning method.

[0024] Compared with the prior art, the present invention has the following advantages: This invention calculates the comprehensive safety risk value of each road segment in the road network based on risk concentration, road segment length, and travel time. Then, it further weights and sums the comprehensive safety risk values ​​of all road segments along a candidate route to obtain the total safety risk of the candidate route. Finally, it constructs a vehicle route planning function by comprehensively considering the total economic cost, total time cost, and total safety risk. Thus, considering the social public safety risks of the delivery route, it obtains the optimal route scheme in terms of total economic cost, total time cost, and total safety risk, making route planning more scientific and more suitable for the characteristics of the refined oil delivery industry. Attached Figure Description

[0025] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0026] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0027] The principles and features of the present invention are described below. The embodiments given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0028] Example 1 like Figure 1 As shown, a method for planning the delivery route of refined oil products includes the following steps: The comprehensive safety risk value of each road segment in the road network is calculated using the following formula: Comprehensive safety risk value = Risk concentration × Road segment length × Travel time; The total safety risk of a candidate path is obtained by weighting and summing the comprehensive safety risk values ​​of all road segments it passes through. Construct a multi-objective programming vehicle routing function. The formula for the vehicle routing function is: Minimize Z=[α*Total_Cost,β*Total_Time,γ*Total_Risk]; Where Minimize represents the minimization operation, Z represents the objective function value, α represents the first weight, Total_Cost represents the total economic cost, β represents the second weight, Total_Time represents the total time cost, γ represents the third weight, and Total_Risk represents the total security risk. Set constraints for the vehicle routing function; The constraints include one or more of the following: site time window, vehicle loading rate, vehicle fixed cost, number of vehicles, customer points, road and bridge tolls, and fuel costs. Solve the vehicle path planning function, and take the candidate path corresponding to the solution of the vehicle path planning function that minimizes the objective function value as the optimal path.

[0029] This invention calculates the comprehensive safety risk value of each road segment in the road network based on risk concentration, road segment length, and travel time. Then, it further weights and sums the comprehensive safety risk values ​​of all road segments along a candidate route to obtain the total safety risk of the candidate route. Next, it constructs a vehicle route planning function by comprehensively considering total economic cost, total time cost, and total safety risk, taking station time windows, vehicle loading rate, vehicle fixed costs, number of vehicles, customer points, road and bridge tolls, and fuel costs as constraints to optimize the route. Thus, considering the social public safety risks of the delivery route, it obtains the optimal route scheme in terms of total economic cost, total time cost, and total safety risk, making route planning more scientific and more suitable for the characteristics of the refined oil delivery industry.

[0030] As a preferred technical solution, γ=0 is set, and safety risk is used as a constraint condition for the vehicle path planning function.

[0031] In addition to using safety risks as risk factors, safety risks can also be used as constraints on vehicle path planning functions, thus realizing a second method of using safety risks for path planning.

[0032] As a preferred technical solution, solving the vehicle path planning function and selecting the candidate path corresponding to the solution of the vehicle path planning function that minimizes the objective function value as the optimal path includes the following steps: Set initialization parameters; The initialization parameters include one or more of the following: Constraints, site information, vehicle information, function execution information; Generate an initial population, each initial population consisting of one or more individuals, each individual corresponding to a candidate path; Calculate the fitness of each individual; A roulette wheel selection is performed based on fitness to select individuals to enter the offspring; The selected individuals are cross-linked pairwise to generate new individuals; Add new individuals to the population and calculate their fitness. Individuals with high fitness are selected from both the parent and offspring generations to form the next generation of the population. Repeat the roulette wheel selection based on fitness to select individuals to enter the offspring until the maximum number of iterations is reached or the optimal solution for consecutive generations remains unchanged, then proceed to the next step. The individual with the highest fitness in the population is output as the optimal path.

[0033] A solution to the periodic vehicle routing problem using the PVRP algorithm was implemented.

[0034] As a preferred technical solution, the site information includes one or more of the following: Site address, site time window, site fuel quantity, site fuel volume, site fuel type.

[0035] The content of the site information has been further clarified, resulting in better route planning.

[0036] As a preferred technical solution, vehicle information includes one or more of the following: number of vehicles, fixed cost, load capacity, load volume, and type of energy source.

[0037] The content of vehicle information has been further clarified, resulting in better route planning.

[0038] As a preferred technical solution, the function execution information includes one or more of the following: crossover probability, mutation probability, and maximum number of iterations.

[0039] The content of function execution information has been further clarified, resulting in better path planning.

[0040] As a preferred technical solution, roulette wheel selection is performed based on fitness to select individuals for offspring, including the following steps: The total fitness is calculated based on the fitness of an individual, using the formula: F = Σf_i; where F represents the total fitness, i represents the individual number (i is an integer and 1 ≤ i ≤ N), N represents the total number of individuals in the population, and f_i represents the fitness of the i-th individual. The selection probability of each individual is calculated using the formula: P_i = f_i / F; where P_i represents the selection probability of the i-th individual. Calculate the cumulative probability of each individual, and the calculation formula is: Q_i = ΣP_i; where, Q_i represents the selection probability of the i-th individual; Generate a random number r uniformly distributed in the interval [0, 1); Select the individuals entering the offspring according to the random number r, and the selection method is: If r ≤ Q_1, then select the first individual; If Q_{i - 1} < r ≤ Q_i, then select the i-th individual.

[0041] Further clarify the content of the roulette selection according to the fitness, and the individuals selected to enter the offspring, so as to make the path planning effect better.

[0042] As a preferred technical solution, the risk concentration is the weighted sum of risk factors; where, the risk factors include static risk factors and dynamic risk factors.

[0043] Further clarify the content of the risk concentration (especially the risk factors), so as to make the path planning effect better.

[0044] As a preferred technical solution, the static risk factors include one or more of the following: sensitive areas, infrastructure, historical accident data, and the dynamic risk factors include one or more of the following: real-time population density, real-time traffic flow, weather conditions, manual dynamic adjustment.

[0045] Further clarify the content of the static risk factors and dynamic risk factors, so as to make the path planning effect better.

[0046] A refined oil distribution path planning system includes a processor, a memory, and a computer program stored in the memory. When the computer program is executed by the processor, the steps of the refined oil distribution path planning method described above are implemented.

[0047] Embodiment 2 As Figure 1 shown, on the basis of Embodiment 1, this embodiment provides a more refined implementation manner.

[0048] The present invention includes the following contents: 1. Quantification and modeling of safety risk indicators (1) Risk factor identification 1) Static risk factors: Sensitive areas: schools, hospitals, nursing homes, large transportation hubs, core business districts, etc. Bind the POI data with the attached road network, and assign different sensitive risk coefficients (R1) to the bound roads.

[0049] Infrastructure: tunnels, bridges (with weight and age limits), steep slopes, and sharp bends. An inherent risk factor (R2) is assigned to this section of the road.

[0050] Historical accident data: Based on the acquired historical real-time traffic data, accident-prone road sections are identified and defined as accident risk coefficients (R3).

[0051] 2) Dynamic risk factors: Real-time population density: The risk coefficient of sensitive areas is dynamically adjusted during specific time periods (such as peak school hours / after school / weekends and holidays).

[0052] Real-time traffic flow: Congested road sections are identified through real-time traffic data. Congestion increases the exposure time of vehicles in risk areas, thereby dynamically increasing the risk value.

[0053] Weather conditions: Severe weather such as rain, snow, and fog can affect road safety and increase the weather risk factor (R4) for the entire route or a specific section of the road.

[0054] Manual dynamic adjustment: The risk coefficient of a road segment is manually and dynamically adjusted (e.g., forced to pass through or avoid a certain point).

[0055] (2) Safety risk calculation model: Calculate a comprehensive safety risk value Risk_i for each road segment i in the road network.

[0056] An example calculation formula: Risk_i = (Σ(Sensitive Area R1 * Impact Weight) + R2 + R3 * Historical Accident Weight + R4) * Road Segment Length * Travel Time R2, R3, R4: Fixed values ​​are used in this embodiment.

[0057] Part 1: Core Risk Factors (Risk "Concentration") This part is the core of the formula, calculating the "basic risk density" per unit road segment and unit time. It typically includes: R1 (Sensitive Area Risk): Refers to sensitive points along the road section, such as schools, hospitals, nursing homes, commercial areas, and train stations. These areas have high pedestrian traffic and complex traffic behavior (such as frequent street crossings), resulting in higher risks.

[0058] Σ(Sensitive Area R1 * Influence Weight): For each sensitive area traversed by the road segment, its inherent risk value (R1) is multiplied by an influence weight (e.g., the weight for schools might be 1.5, while for ordinary residential areas it might be 1.0), and then summed. This reflects the differentiated contribution of different sensitive points to the risk.

[0059] R2 (Inherent Static Risk): This refers to the inherent risk attributes of the road itself, such as: few lanes, poor road surface conditions, sharp curves, steep slopes, insufficient nighttime lighting, lack of guardrails, etc. This part is usually a fixed value.

[0060] R3 (Historical Accident Risk): This refers to the frequency and severity of traffic accidents that have occurred on this road section in the past. Historical data is one of the most powerful indicators for predicting future risks.

[0061] Historical accident weighting: Used to amplify or reduce the impact of historical accident data. For example, a recent serious accident (fatal accident) might be weighted at 2.0, while a minor scrape accident might be weighted at 0.5.

[0062] R4 (Dynamic or Environmental Risk): refers to variable risk factors, such as average vehicle speed (the higher the speed, the greater the risk), weather conditions (higher risk in rain, snow, and fog), and time of day (risk may be higher at night than during the day).

[0063] Parts Two and Three: Exposure (Total Risk).

[0064] The core risk factor calculates the “concentration” of risk, but the total “dose” of risk also depends on the scale and duration of your exposure.

[0065] Road length: The longer the road segment, the larger the area of ​​risk, and the more potential danger points or accident locations there are.

[0066] Travel time: The longer it takes for vehicles or pedestrians to pass through a section of road, the longer they are exposed to the risky environment, and the higher the probability of an incident occurring.

[0067] The logic for multiplying these three parts is: Overall risk = risk concentration × exposure scale × exposure duration.

[0068] Total safety risk of a path: For a candidate path P, its total safety risk Total_Risk_P is the weighted sum of the risk values ​​of all the road segments it passes through.

[0069] 2. Path optimization model incorporating security risks 1) Construction of multi-objective optimization function: Define the objective function, for example: Minimize Z=[α*Total_Cost,β*Total_Time,γ*Total_Risk] Where Total_Cost is the total economic cost (mileage, fuel consumption, etc.), Total_Time is the total time cost, and Total_Risk is the total safety risk. α, β, and γ are weighting coefficients that can be manually adjusted.

[0070] Merge multiple objectives into one scalar: Z = α * Total_Cost + β * Total_Time + γ * Total_Risk. When optimizing the algorithm, that is, stop the calculation when the value of Z is the smallest, and α, β, and γ are manually adjusted according to business needs.

[0071] Alternative solution: Use safety risk as a constraint condition, such as Total_Risk_P < Risk_Threshold (risk threshold). In the case of the smallest Z, if the risk control coefficient is controlled, the purpose of this constraint is to ensure that the overall risk of the selected path does not exceed an acceptable upper limit.

[0072] 2) Solving algorithm: Use the PVRP solving algorithm to solve the periodic vehicle routing problem, considering factors (i.e., constraint conditions): time window, risk coefficient, vehicle loading rate, vehicle fixed cost, number of vehicles, customer points, road and bridge tolls, fuel cost, etc. The specific process steps of the algorithm are as follows: Step 1: Initialize parameters Set the site address, site time window, site fuel quantity, site fuel volume, site fuel type, vehicle basic information (including number of vehicles, fixed cost, loading capacity, loading volume, loading energy type, etc.), crossover probability, mutation probability, maximum number of iterations, etc.

[0073] Step 2: Generate the initial population Generate n individuals, and each individual represents a path plan. Generation method: Randomly arrange the customer points, and then insert 0 (warehouse) according to the capacity constraint and time window constraint to form multiple paths. If the generated individuals do not meet the constraints, repair them (such as adjusting the customer point order or adding vehicles).

[0074] Customer point: Refers to the coordinates of each distribution point.

[0075] Capacity constraint: For example, a vehicle can only load 50T, and the loading rate is 90%, but the demand of 10 distribution points is 55T, then one vehicle cannot load it, and two vehicles are needed for distribution.

[0076] Time window constraint: For example, the distribution station can only distribute between 8:00 and 9:00, so the vehicle needs to prioritize this distribution point during distribution. The first step is just to simply construct an initial solution, and then optimize it according to the objective function.

[0077] Step 3: Calculate the fitness For each individual, decode to obtain the path of each vehicle, calculate the total cost (vehicle fixed cost + fuel cost + road and bridge tolls) and penalty cost (time window violation, overloading, safety risk coefficient, etc.), and the fitness takes the reciprocal of the total cost.

[0078] Decoding: Obtain the path information of the vehicle, which is to call the path planning service based on the point information, obtain the mileage, time, road and bridge toll information, risk coefficient, and calculate the cost.

[0079] Step 4: Selection Perform roulette wheel selection according to the fitness, and select n individuals to enter the next generation.

[0080] Roulette wheel selection is a commonly used selection mechanism in genetic algorithms. It simulates the game of roulette and determines the probability of an individual being selected according to the proportion of its fitness value to the total fitness value. The specific details are as follows: Basic principle: The probability of each individual being selected is proportional to its fitness value. The higher the fitness value, the greater the probability of being selected.

[0081] Suppose there are N individuals in the population, and the fitness value of each individual is f_i (i = 1, 2,..., N), then the probability P_i of individual i being selected is: P_i = f_i / Σ_{j = 1}^{N} f_j.

[0082] Specific steps: Calculate the fitness value: Calculate the fitness value of each individual in the population.

[0083] Calculate the total fitness value: Add up the fitness values of all individuals to obtain the total fitness value F = Σf_i.

[0084] Calculate the selection probability: Calculate the selection probability P_i = f_i / F of each individual.

[0085] Calculate the cumulative probability: Calculate the cumulative probability Q_i = ΣP_i. In this way, each individual occupies a sector area on the roulette, and the angle size is proportional to the fitness value.

[0086] Generate a random number: Generate a random number r uniformly distributed in the interval [0, 1).

[0087] Select an individual: If r ≤ Q_1, then select the first individual; Otherwise, if Q_{i - 1} < r ≤ Q_i, then select the i-th individual (i = 2, 3,..., N).

[0088] Repeat steps 5 and 6 until the required number of individuals is selected (usually the same as the population size, allowing repeated selection).

[0089] Step 5: Crossover The selected individuals are cross-crossed pairwise (with cross-cross probability pc) to generate new individuals. When crossing, care should be taken to avoid duplicate customer points and to maintain the feasibility of the path as much as possible.

[0090] Crossover probability is one of the most critical control parameters in genetic algorithms. It determines the probability that parent individuals will produce new individuals through crossover during the breeding process.

[0091] Exploration (Cross): High PC → More new combinations → Strong global search capability; Utilization (copying): Low Pc → Maintaining excellent individuals → Strong local search capability.

[0092] Step 6: Mutation For each individual, perform a mutation operation with a mutation probability pm. After mutation, check the feasibility (randomly select two customer points to swap positions; or randomly select a sub-path to reverse; or randomly select a customer point to insert into another random position).

[0093] Step 7: Calculate the fitness of the new individual Add new individuals to the population and calculate their fitness.

[0094] Step 8: Environment Selection Select n individuals with high fitness from the parent and offspring generations to form the next generation population.

[0095] High adaptability: equivalent to satisfying both hard and soft constraints. Soft constraints: Risk constraints may take precedence over cost constraints. Soft constraints mean that they can be disregarded. Some routes may not be able to be calculated under this constraint. Routes that meet the condition should be selected first.

[0096] Hard constraint 1: Time window constraint may be a hard requirement that must be met.

[0097] Hard constraint two: The vehicle capacity limit must be met; this is a constraint that must be satisfied.

[0098] Step 9: Consider constraint handling Time window: When calculating fitness, check the time window for each route, and add penalty costs if it is violated.

[0099] Safety risk coefficient: The risk coefficient can be converted into cost or used as a constraint (such as not exceeding a certain threshold). A constraint can be manually added to the objective function; if the risk exceeds this value, the route is considered high-risk and needs to be eliminated.

[0100] Vehicle loading rate: Calculate the loading capacity of each vehicle, and apply a penalty if the maximum loading capacity is exceeded.

[0101] Vehicle fixed costs: Fixed costs are added for each vehicle used.

[0102] Road and bridge tolls: calculated based on the specific road sections traversed by the route.

[0103] Fuel costs: calculated based on driving distance and vehicle fuel consumption.

[0104] Step 10: Check termination conditions If the maximum number of iterations is reached or the optimal solution has not improved for several consecutive generations, then terminate; otherwise, proceed to step 4.

[0105] The evaluation of the optimal solution is the objective function Z = α*Total_Cost + β*Total_Time + γ*Total_Risk.

[0106] The logic of the entire algorithm is to calculate the most reasonable solution that meets the constraints while minimizing the objective function value.

[0107] Step 11: Output the optimal solution Output the individual with the highest fitness, i.e., the path scheme with the optimal total economic cost, total time cost, and total safety risk.

[0108] As described above, the present invention can be implemented well.

[0109] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0110] In the description of this invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0111] In the description of this invention, although embodiments of the invention have been shown and described herein, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this invention.

[0112] In the description of this invention, all features disclosed in all embodiments of this specification, or steps in all methods or processes implied in the disclosure, may be combined and / or extended or replaced in any way, except for mutually exclusive features and / or steps.

[0113] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Based on the technical essence of the present invention, any simple modifications, equivalent substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for planning delivery routes for refined oil products, characterized in that, Includes the following steps: The comprehensive safety risk value of each road segment in the road network is calculated using the following formula: Comprehensive safety risk value = Risk concentration × Road segment length × Travel time; The total safety risk of a candidate path is obtained by weighting and summing the comprehensive safety risk values ​​of all road segments it passes through. Construct a multi-objective programming vehicle routing function. The formula for the vehicle routing function is: Minimize Z=[α*Total_Cost,β*Total_Time,γ*Total_Risk]; Where Minimize represents the minimization operation, Z represents the objective function value, α represents the first weight, Total_Cost represents the total economic cost, β represents the second weight, Total_Time represents the total time cost, γ represents the third weight, and Total_Risk represents the total security risk. Set constraints for the vehicle routing function; The constraints include one or more of the following: site time window, vehicle loading rate, vehicle fixed cost, number of vehicles, customer points, road and bridge tolls, and fuel costs. Solve the vehicle path planning function, and take the candidate path corresponding to the solution of the vehicle path planning function that minimizes the objective function value as the optimal path.

2. The method for planning a refined oil product distribution route according to claim 1, characterized in that, Let γ=0, and use safety risk as a constraint condition for the vehicle path planning function.

3. The method for planning a refined oil product distribution route according to claim 1, characterized in that, Solving the vehicle routing function and selecting the candidate path corresponding to the solution that minimizes the objective function value as the optimal path includes the following steps: Set initialization parameters; The initialization parameters include one or more of the following: Constraints, site information, vehicle information, function execution information; Generate an initial population, each initial population consisting of one or more individuals, each individual corresponding to a candidate path; Calculate the fitness of each individual; A roulette wheel selection is performed based on fitness to select individuals to enter the offspring; The selected individuals are cross-linked pairwise to generate new individuals; Add new individuals to the population and calculate their fitness. Individuals with high fitness are selected from both the parent and offspring generations to form the next generation of the population. Repeat the roulette wheel selection based on fitness to select individuals to enter the offspring until the maximum number of iterations is reached or the optimal solution for consecutive generations remains unchanged, then proceed to the next step. The individual with the highest fitness in the population is output as the optimal path.

4. The method for planning a refined oil product distribution route according to claim 3, characterized in that, Site information includes one or more of the following: Site address, site time window, site fuel quantity, site fuel volume, site fuel type.

5. The method for planning a refined oil product distribution route according to claim 3, characterized in that, Vehicle information includes one or more of the following: number of vehicles, fixed costs, load capacity, load volume, and type of energy source.

6. The method for planning a refined oil product distribution route according to claim 3, characterized in that, Function execution information includes one or more of the following: crossover probability, mutation probability, and maximum number of iterations.

7. The method for planning a refined oil product distribution route according to claim 3, characterized in that, The selection process, which involves a roulette wheel selection based on fitness, involves the following steps: The total fitness is calculated based on the fitness of an individual, using the formula: F = Σf_i; where F represents the total fitness, i represents the individual number (i is an integer and 1 ≤ i ≤ N), N represents the total number of individuals in the population, and f_i represents the fitness of the i-th individual. Calculate the selection probability of each individual, and the calculation formula is: P_i = f_i / F; where, P_i represents the selection probability of the i-th individual; Calculate the cumulative probability of each individual, and the calculation formula is: Q_i = ΣP_i; where, Q_i represents the selection probability of the i-th individual; Generate a random number r uniformly distributed in the interval [0, 1); Select the individuals entering the offspring according to the random number r, and the selection method is: If r ≤ Q_1, then select the first individual; If Q_{i - 1} < r ≤ Q_i, then select the i-th individual.

8. A method for planning the distribution route of refined oil products according to any one of claims 1 to 7, characterized in that, The risk concentration is the weighted sum of risk factors; where, the risk factors include static risk factors and dynamic risk factors.

9. A method for planning the distribution route of refined oil products according to claim 8, characterized in that, The static risk factors include one or more of the following: sensitive areas, infrastructure, historical accident data, and the dynamic risk factors include one or more of the following: real-time population density, real-time traffic flow, weather conditions, and manual dynamic adjustment.

10. A refined oil product delivery route planning system, characterized in that, It includes a processor, a memory, and a computer program stored in the memory. When the computer program is executed by the processor, it implements the steps of a method for planning a refined oil distribution path described in any one of claims 1 to 9.

Images