Shared automobile website multi-objective optimization site selection method and system considering environmental protection factors
Through the supply and demand balance method and NSGA-II algorithm combined with green sensitivity factors, the location selection of shared car outlets is solved, and the problem of unreasonable layout of shared car outlets is achieved, efficient and environmentally friendly shared car outlet planning is achieved, user experience and operational efficiency is improved, and environmental pollution is reduced.
Patent Information
- Application Number
- CN202510327655.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-19
- Publication Date
- 2025-07-29
AI Technical Summary
The unreasonable layout of shared car outlets, inaccurate user demand forecasts, and neglect of green environmental protection factors, resulting in poor user experience, low operational efficiency and serious environmental pollution.
The supply and demand balance method is used to predict the demand scale of shared car outlets, and combined with green sensitivity factors such as pollution index and congestion index, a multi-objective optimization module is built, and the NSGA-II algorithm is used to optimize the location selection of shared car outlets to generate the final shared car outlet site selection scheme.
Accurately plan shared car outlets, improve user experience and operational efficiency, reduce environmental pollution and energy consumption, alleviate urban traffic pressure, and improve transportation infrastructure and resource allocation.
Smart Images
Figure CN120387848A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of shared travel, and particularly relates to a multi-objective optimization site selection method and system for shared cars considering environmental protection factors. Background Art
[0002] With the annual increase in the number of automobiles in China, the problem of urban traffic congestion has become increasingly prominent, and the problems of air pollution and energy consumption caused by automobile exhaust emissions have also become more serious. According to the latest statistical data of the Traffic Management Bureau of the Ministry of Public Security, as of the end of 2024, the number of motor vehicles in China has climbed to 453 million, of which cars account for 353 million, further increasing the environmental pressure on cities. Against this background, shared cars, as a new travel mode, have emerged. By improving the vehicle utilization rate and reducing the private car ownership rate, they effectively alleviate the above problems. In recent years, the "Four Modernizations" trend of electrification, networking, intelligence, and sharing is profoundly changing the pattern of the automotive industry. Especially against the background of the rapid development of the sharing economy, shared cars have gradually become one of the important choices for urban residents' short-distance travel. However, although shared cars show great development potential in the Chinese market, their actual application still faces many challenges, including unreasonable network layouts, inaccurate prediction of user needs, and other problems. These problems not only affect the user experience but also limit the healthy development of the shared car industry to a certain extent.
[0003] In the development of the shared car industry, site selection is a crucial link. A reasonable network layout can not only ensure that users can conveniently pick up and return vehicles, but also optimize resource allocation and reduce operating costs. At the same time, it has a positive impact on alleviating traffic congestion, reducing environmental pollution, and saving energy. Currently, most shared car operators mainly rely on experience and intuition for site setting. The research on shared car site selection mainly focuses on objectives such as capital constraints and traffic factors and the optimization of site selection schemes, but these studies often ignore the important perspective of green, low-carbon environmental protection. For example, an unreasonable network layout may lead to too long a time for users to find vehicles and difficult vehicle allocation, thus affecting the user experience and operating efficiency. In addition, the existing site selection methods fail to fully consider the impact of green sensitivity factors such as traffic congestion degree and network coverage rate on the environment and cannot meet the urgent needs of the city for green travel.
[0004] Therefore, the present invention proposes a multi-objective optimization site selection method and system for shared car networks considering environmental protection factors, combining the green concept with the site selection of shared car networks. From the perspective of environmental protection, a set of scientific and reasonable site selection modules are constructed, and the site selection of shared car networks is restricted by green sensitivity factors to obtain a more low-carbon, environmentally friendly and energy-saving site selection plan. This site selection method can more accurately plan the shared car networks, promote the popularization of shared cars, improve the utilization rate of shared cars, and at the same time help reduce environmental pollution. By alleviating the urban traffic road pressure and improving the service level of the shared industry, it effectively improves the urban traffic infrastructure and resource allocation. Summary of the Invention
[0005] In view of the technical problems such as unreasonable layout of shared car networks, inaccurate prediction of user demand, and neglect of green environmental protection factors proposed above, a multi-objective optimization site selection method and system for shared car networks based on green sensitivity analysis are provided. The present invention mainly uses the supply-demand balance method to predict the demand scale of shared cars, constructs a multi-objective optimization module in combination with green sensitivity factors (such as pollution index, congestion index), and uses the NSGA-II algorithm to solve, so as to optimize the layout of shared car networks, improve the coverage rate of user demand, and reduce environmental pollution and energy consumption.
[0006] The technical means adopted by the present invention are as follows:
[0007] The multi-objective optimization site selection method for shared car networks considering environmental protection factors includes the following steps:
[0008] Obtain residents' travel information and shared car travel information;
[0009] Based on residents' travel information and shared car travel information, generate alternative site selections for shared car networks;
[0010] Construct a site demand prediction module, which predicts the demand scale of shared car networks through the supply-demand balance method based on residents' travel information and shared car travel information;
[0011] Construct a multi-objective site selection module, which optimizes the site selection layout of shared car networks based on the alternative site selections of shared car networks and the demand scale of shared car networks. The multi-objective site selection module includes a first objective function with the shortest total distance as the goal, a second objective function with the largest demand satisfaction as the goal, and a third objective function with the smallest pollution congestion coefficient as the goal;
[0012] Construct a solution module, which uses the second-generation non-dominated sorting genetic algorithm to solve the site selection module and generate the final site selection of shared car networks.
[0013] Further, the network demand prediction module predicts the demand scale of shared car networks based on residents' travel information and shared car travel information, including:
[0014] Based on residents' travel information and shared car travel information, the total number of shared cars is predicted using the supply-demand balance method. The calculation formula for predicting the total number of shared cars is:
[0015]
[0016] Among them, N is the predicted total number of shared cars, R is the total resident population, A is the average number of daily trips per resident, P is the travel ratio occupied by shared cars, D is the average driving distance per shared car trip, S is the average passenger capacity per shared car trip, t is the average driving time per shared car trip, v is the average driving speed per shared car trip, and δ is the normal operation percentage of shared cars.
[0017] Further, the first objective function with the total distance and shortest distance as the goal is specifically:
[0018]
[0019] Among them, min f1 is the first objective function, y j is the first decision variable indicating whether the j-th alternative network is selected as a shared car network, d ij is the distance between the shared car demand point i and the alternative network j, j is a single alternative shared car network, i is a single shared car demand point, I is the set of shared car demand points, and J is the set of alternative shared car networks.
[0020] The second objective function with the goal of meeting the maximum demand is specifically:
[0021]
[0022] Among them, minf2 is the second objective function, y j is the first decision variable indicating whether the j-th alternative network is selected as a shared car network, x ij is the second decision variable indicating whether the shared car demand point i can be covered by the alternative network j, ρ i is the population density of the shared car demand point i, j is a single alternative shared car network, i is a single shared car demand point, I is the set of shared car demand points, and J is the set of alternative shared car networks.
[0023] The third objective function with the goal of minimizing the pollution congestion coefficient is specifically:
[0024]
[0025] Among them, min f3 is the third objective function, y j is the first decision variable indicating whether the j-th alternative network point is selected as a shared car network point. θ1 is the weight coefficient of the pollution index of the shared car alternative network point i, θ2 is the weight coefficient of the congestion index of the shared car alternative network point i, c j is the pollution index of the shared car alternative network point i, t j is the congestion index of the shared car alternative network point i, j is a single shared car alternative network point, i is a single shared car demand point, I is the set of shared car demand points, and J is the set of shared car alternative network points.
[0026] Furthermore, the constraint conditions of the first objective function, the second objective function, and the third objective function include:
[0027]
[0028] Among them, N min is the lower limit of the number of shared car network points, y j is the first decision variable indicating whether the j-th alternative network point is selected as a shared car network point, N max is the upper limit of the number of shared car network points, y m is the third decision variable indicating whether the m-th alternative network point is selected as a shared car network point, y n is the fourth decision variable indicating whether the n-th alternative network point is selected as a shared car network point, d min is the minimum allowable distance between any two selected shared car network points, d mn is the actual distance between alternative network point m and alternative network point n, x ij is the second decision variable indicating whether the shared car demand point i can be covered by the alternative network point j, d ij is the distance between the shared car demand point i and the alternative network point j, d0 is the service radius of the shared car network point, j is a single shared car alternative network point, i is a single shared car demand point, I is the set of shared car demand points, and J is the set of shared car alternative network points.
[0029] Furthermore, the working process of the solving module is as follows:
[0030] S1. Population initialization: According to the number of objective functions M and the number of decision variables V, using the real number coding method, initialize an N×(M + V) matrix by setting the population size N to generate the initial population;
[0031] S2. Non-dominated sorting: Use the non-dominated sorting algorithm to stratify the population of size N. By continuously comparing and marking non-dominated individuals, remove them layer by layer and execute in a loop until the population is fully stratified;
[0032] S3. Crowdedness calculation: By calculating the average distance between a certain solution in the population and its adjacent left and right points, its crowdedness value is determined to ensure population diversity, and the sorting and weighting methods are used to determine the weight of the objective function in the total weighted average crowdedness to select the final solution set;
[0033] S4. Selection, mutation and crossover: Individuals are selected through the roulette wheel selection method, and the individual solutions are randomly changed through the mutation operation and the better solutions are retained. A new generation of population is generated through the crossover operation, and finally a new population containing 2N individuals is formed;
[0034] S5. Elite retention: The new population in the t-th generation and the parent generation are merged into a population R with a size of 2N t , after non-dominated sorting and crowdedness calculation, the individuals in the non-dominated set are first put into the new parent population. If the number is less than N, the next-level non-dominated set is continuously supplemented until it exceeds N, and then it is adjusted to N by the crowdedness operator. Finally, a new offspring population is generated through the genetic operator;
[0035] S6. Terminate iteration: When the total number of iterations reaches the set value, the iteration is terminated to generate the final location selection of the shared car network points.
[0036] The multi-objective optimization location selection system for shared car network points considering environmental protection factors is used to implement the above-mentioned multi-objective optimization location selection method for shared car network points considering environmental protection factors, including:
[0037] The data acquisition module is used to acquire residents' travel information and shared car travel information;
[0038] The network point demand prediction module is used to predict the demand scale of shared car network points based on residents' travel information and shared car travel information;
[0039] The multi-objective network point location selection module is used to construct the first objective function with the shortest total distance as the goal, the second objective function with the largest demand satisfaction as the goal, and the third objective function with the smallest pollution congestion coefficient as the goal based on the alternative shared car network point location selection and the shared car network point demand scale, and optimize the layout of the shared car network points;
[0040] The solution module is used to solve the multi-objective network point location selection module based on the second-generation non-dominated sorting genetic algorithm to generate the final location selection of the shared car network points.
[0041] Compared with the existing technology, the present invention has the following advantages:
[0042] 1. The present invention incorporates green sensitivity factors (such as pollution index, congestion index) into the multi-objective network site selection module. By quantifying the environmental impact, it avoids setting up network sites in areas with severe pollution or traffic congestion. This method can effectively reduce the negative impact of shared car operation on the environment, reduce exhaust emissions and energy consumption, and contribute to the realization of the urban green and low-carbon travel goal.
[0043] 2. The present invention predicts the demand scale of shared cars through the supply-demand balance method and combines a multi-objective optimization module (shortest total distance, largest demand coverage rate, smallest pollution and congestion coefficient) to accurately plan the layout of shared car network sites. This method avoids the unreasonable network site distribution problem caused by the traditional site selection method relying on experience, ensures that users can conveniently pick up and return vehicles, and at the same time optimizes the vehicle allocation efficiency, significantly improving the user experience and operation efficiency.
[0044] 3. The present invention solves the multi-objective optimization module through the NSGA-II algorithm to generate the final shared car network site selection. This method can not only reduce the operation cost of shared cars, but also improve the vehicle utilization rate, relieve urban traffic pressure, and improve urban traffic infrastructure and resource allocation.
[0045] Based on the above reasons, the present invention can be widely promoted in the fields of shared travel and the like. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.
[0047] Figure 1 It is a flowchart of the multi-objective optimization site selection method for shared car network sites considering environmental protection factors of the present invention.
[0048] Figure 2 It is a schematic diagram of the change in air quality after the launch of shared cars of the present invention.
[0049] Figure 3 It is a basic flowchart of the NSGA-II algorithm of the present invention.
[0050] Figure 4 It is a schematic diagram of the result of the network site location in the embodiment of the present invention.
[0051] Figure 5 It is a schematic diagram of the result of the service range of the network site in the embodiment of the present invention.
[0052] Figure 6 It is a schematic diagram of the result of the fitness evolution curve in the embodiment of the present invention. Detailed implementation mode
[0053] To enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.
[0054] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.
[0055] As Figure 1 shown, the present invention provides a multi-objective optimization site selection method for shared car outlets considering environmental protection factors, and the steps include:
[0056] Obtain residents' travel information and shared car travel information.
[0057] Generate alternative shared car outlet site selections based on residents' travel information and shared car travel information.
[0058] Construct a site demand prediction module. The site demand prediction module predicts the demand scale of shared car outlets through the supply-demand balance method based on residents' travel information and shared car travel information.
[0059] Specifically, based on residents' travel information and shared car travel information, and based on the supply-demand balance method, predict the total number of shared cars. The calculation formula for predicting the total number of shared cars is:
[0060]
[0061] Among them, N is the total number of predicted shared cars, R is the total resident population, A is the average number of daily trips per resident, P is the travel ratio occupied by shared cars, D is the average driving distance per shared car trip, S is the average passenger capacity per shared car trip, t is the average driving time per shared car trip, v is the average driving speed per shared car trip, and δ is the normal operation percentage of shared cars.
[0062] The total number of predicted shared cars N is the basic input parameter of the multi-objective network location selection module, which affects the optimization process in the following ways:
[0063] 1. Constraining the number of network points: Ensure that the total number of vehicles matches the network capacity.
[0064] 2. Driving the objective function: Maximizing the coverage rate depends on sufficient vehicles and a reasonable network layout.
[0065] 3. Balancing environmental protection and efficiency: By controlling the total number of vehicles and the network location, reduce environmental pollution and congestion.
[0066] In the following embodiments, the number of optimized network points remains 11, but by adjusting the location, the pollution index is reduced by 14.3% and the coverage rate is increased by 4.5%, verifying the key supporting role of N in the optimization results.
[0067] According to economic principles, shared cars can also be regarded as a kind of commodity, which conforms to the supply and demand relationship: oversupply will cause idle waste of resources, occupation of public resources, environmental pollution and unfair competition; undersupply will reduce the service level of shared cars and cannot meet the travel needs of travelers. Therefore, the supply-demand balance method comprehensively predicts the demand scale of shared car network points based on travel characteristics such as population, travel mode ratio and the number of trips of travelers.
[0068] Construct a multi-objective network location selection module. The multi-objective network location selection module optimizes the location layout of shared car network points based on the alternative shared car network point locations and the demand scale of shared car network points. The multi-objective network location selection module includes a first objective function with the total distance and shortest as the goal, a second objective function with the maximum demand satisfaction as the goal, and a third objective function with the minimum pollution and congestion coefficient as the goal.
[0069] First, based on the analysis of the influencing factors of shared car location selection in the above text, in order to simplify the module and obtain better calculation results, the following assumptions are made for the module:
[0070] Assumption 1: By default, the service range of each network point is the same, all set to 2 km.
[0071] Hypothesis 2: The alternative sites include the areas near eight types of land use, namely shopping malls, residential areas, subway stations, bus stops, hospitals, schools, government agencies, and scenic spots, and include four attributes: geographical coordinates, population density, pollution index, and congestion index.
[0072] Hypothesis 3: Each demand point is also an alternative site, and its location is known.
[0073] The present invention analyzes the site selection factors of shared cars on the basis of considering the green concept. First, it analyzes from four aspects: travel demand, traffic conditions, geographical conditions, and environmental protection impact.
[0074] Analyzing the travel demand factor, for the convenience of travelers, the distance between the shared car site and the demand point should be as small as possible.
[0075] Analyzing the traffic condition factor, the alternative site should be selected in an area with convenient access, comply with the urban public facilities planning specifications, and coordinate the connection with transportation hubs.
[0076] Analyzing the geographical conditions, the alternative site should preferably be selected in an area with flat and vast terrain and relatively high population density.
[0077] Analyzing the environmental protection impact, pay attention to selecting alternative sites far away from areas such as parks and forests and areas with serious pollution.
[0078] In order to quantify the green environmental protection factors for constraints in modeling, the environmental protection factors are emphasized by using green sensitivity analysis, which includes two aspects: one is the congestion degree of the road section where the site is set up, and the other is the site coverage rate.
[0079] Congestion degree of the road section where the site is set up: By analyzing the impact of the congestion degree on the comprehensive air index and pollutant emissions, the green sensitivity factor of the congestion degree of the road section where the site is set up is obtained. The specific data is shown in Table 1.
[0080] Table 1 Relationship between traffic operation status and pollutant emissions
[0081]
[0082] According to Table 1, in the case of smooth traffic, the vehicle flow reaches the maximum value, and at the same time, the accumulation amount of tail gas per unit time is relatively high, but the total emission amount remains the smallest. When the vehicle density reaches a certain level, the traffic flow will increase sharply. As the traffic congestion level increases, the generated pollutant emissions also increase, resulting in an increase in the total pollutant amount. In the case of serious traffic congestion, the total pollutant emissions of motor vehicles on the same road section show a growth trend of about 5.7 times compared with the smooth traffic situation. It can be known that the road congestion degree is a green sensitivity factor for air pollution.
[0083] Therefore, it is necessary to avoid setting up pick-up and drop-off points on severely congested roads. On severely congested roads, the speed is slow, which will greatly increase the pollutants emitted by cars, causing unnecessary waste and pollution. Therefore, when selecting the location of shared car pick-up and drop-off points, the factor of vehicle speed is transformed into the constraint condition of "avoid setting up pick-up and drop-off points on severely congested roads" to avoid exacerbating road congestion.
[0084] Pick-up and drop-off point coverage rate: By comparing the energy utilization efficiency of electric vehicles and fuel vehicles and the breakpoint regression analysis of air quality before and after the launch of shared cars, as shown in Table 2, Figure 2 the pick-up and drop-off point coverage rate, a green sensitivity factor, is obtained.
[0085] Table 2 Energy consumption and emissions per 100 kilometers of electric vehicles and gasoline vehicles
[0086]
[0087] As can be seen from Table 2, the energy consumption per 100 kilometers of electric vehicles is only 41.5% of that of fuel vehicles, and the tailpipe emissions of electric vehicles are only 6.1% of those of fuel vehicles, and the carbon dioxide emissions are about 16 times that of electric vehicles. Compared with fuel vehicles, electric vehicles are of great significance in environmental protection and energy conservation.
[0088] Taking a certain city as an example, observe the change of the air quality index after the launch of shared cars, as Figure 2 shown. The horizontal axis is the execution variable, representing the number of days since the launch, with the launch day being 0, negative on the left and positive on the right; the vertical axis is the air quality index, represented by AQI. In this study, data for 60 days before and after the target date were selected, and greater weights were given to the samples near the breakpoint, and a quadratic polynomial model was used to more accurately reflect the changes at critical moments. To reflect the specific changes in air quality after the launch of shared cars, the launch date was set as the breakpoint, and the local weighted regression scatterplot smoothing method was used to make a trend chart of AQI changing with time. It can be clearly seen from Figure 2 that before the launch of shared cars, the AQI had a trend of first decreasing and then increasing, and after the launch, there was a significant downward trend, indicating that the air quality has been improved to a certain extent after the shared cars are put into use.
[0089] Based on the above conclusions, when selecting the location of shared car pick-up and drop-off points, this factor can be transformed into "maximizing the pick-up and drop-off point coverage rate of shared cars". The greater the coverage rate of shared cars, the more travel needs can be met, air pollution can be improved, and the utilization rate can be increased. At the same time, most shared cars are new energy vehicles, which can reduce the travel frequency of private cars and the driving distance of fuel vehicles, thereby reducing energy consumption and achieving the goal of low-carbon environmental protection.
[0090] Based on the analysis of the influencing factors for the location selection of shared cars in the above text, on the basis of considering the concept of environmental protection, it is also hoped to facilitate the car rental and return process for travelers, maximize the coverage rate of shared cars, and at the same time try to set up the service points on sections with relatively optimistic congestion and pollution conditions. Therefore, the objective function is defined from three aspects: the total distance is the shortest, the demand satisfaction is the largest (the coverage rate is the largest), and the pollution and congestion coefficient is the smallest.
[0091] The specific form of the first objective function with the shortest total distance as the goal is as follows:
[0092]
[0093] Among them, min f1 is the first objective function, and y j is the first decision variable indicating whether the j-th alternative service point is selected as a shared car service point, taking values of 0 or 1. 1 means the j-th alternative service point is selected as a shared car service point, and vice versa for 0. d ij is the distance between the shared car demand point i and the alternative service point j. j is a single alternative shared car service point, i is a single shared car demand point, I is the set of shared car demand points, and J is the set of alternative shared car service points.
[0094] The specific form of the second objective function with the largest demand satisfaction as the goal is as follows:
[0095]
[0096] Among them, min f2 is the second objective function, and y j is the first decision variable indicating whether the j-th alternative service point is selected as a shared car service point, taking values of 0 or 1. 1 means the j-th alternative service point is selected as a shared car service point, and vice versa for 0. x ij is the second decision variable indicating whether the shared car demand point i can be covered by the alternative service point j, taking values of 0 or 1. 1 means the shared car demand point i can be covered by the alternative service point j, and vice versa for 0. ρ i is the population density of the shared car demand point i. j is a single alternative shared car service point, i is a single shared car demand point, I is the set of shared car demand points, and J is the set of alternative shared car service points.
[0097] The specific form of the third objective function with the smallest pollution and congestion coefficient as the goal is as follows:
[0098]
[0099] Among them, min f3 is the third objective function, and y jLet \(y_j\) be the first decision variable indicating whether the \(j\)-th alternative site is selected as a shared car site, taking values of 0 or 1. A value of 1 means the \(j\)-th alternative site is selected as a shared car site, and 0 otherwise. Let \(\theta_1\) be the weight coefficient of the pollution index of the \(i\)-th alternative site for shared cars, and \(\theta_2\) be the weight coefficient of the congestion index of the \(i\)-th alternative site for shared cars, \(c\) j is the pollution index of the \(i\)-th alternative site for shared cars, \(t\) j is the congestion index of the \(i\)-th alternative site for shared cars, \(j\) is a single alternative site for shared cars, \(i\) is a single shared car demand point, \(I\) is the set of shared car demand points, and \(J\) is the set of alternative sites for shared cars.
[0100] The constraint conditions of the first objective function, the second objective function, and the third objective function include:
[0101]
[0102] Among them, \(N\) min is the lower limit of the number of shared car sites, \(y\) j is the first decision variable indicating whether the \(j\)-th alternative site is selected as a shared car site, taking values of 0 or 1. A value of 1 means the \(j\)-th alternative site is selected as a shared car site, and 0 otherwise, \(N\) max is the upper limit of the number of shared car sites, \(y\) m Let \(y_m\) be the third decision variable indicating whether the \(m\)-th alternative site is selected as a shared car site, \(y\) n Let \(y_n\) be the fourth decision variable indicating whether the \(n\)-th alternative site is selected as a shared car site, \(d\) min is the minimum allowable distance between any two selected shared car sites, \(d\) mn is the actual distance between alternative site \(m\) and alternative site \(n\), \(x\) ij Let \(x_{ij}\) be the second decision variable indicating whether the \(i\)-th shared car demand point can be covered by the \(j\)-th alternative site, taking values of 0 or 1. A value of 1 means the \(i\)-th shared car demand point can be covered by the \(j\)-th alternative site, and 0 otherwise, \(d\) ij is the distance between the \(i\)-th shared car demand point and the \(j\)-th alternative site, \(d_0\) is the service radius of the shared car site, that is, the maximum distance that the site can cover the demand point, \(j\) is a single alternative site for shared cars, \(i\) is a single shared car demand point, \(I\) is the set of shared car demand points, and \(J\) is the set of alternative sites for shared cars.
[0103] The first constraint condition in the above formula represents the upper and lower limits of the number of shared car sites, and the second constraint condition represents that the distance between any two shared car sites should not be less than a certain interval.
[0104] Construct a solution module. The solution module uses the second-generation non-dominated sorting genetic algorithm to solve the site selection module and generate the final site selection of shared car sites.
[0105] The NSGA-II algorithm (Nondominated Sorting Genetic Algorithm II, the second-generation non-dominated sorting genetic algorithm) is a fast non-dominated sorting method used to solve multi-objective optimization problems. Its advantages are that it simplifies the complexity of algorithm calculation; it can arrange the individuals within the Paeto domain to the entire Paeto domain, effectively ensuring the diversity of the population; at the same time, it can adopt the elitist strategy, combine the offspring population with the parent population, and jointly compete to generate the next-generation population, so as to keep the excellent individuals in the parent population enter the next generation, and arrange the individuals of the population in layers to improve the population level.
[0106] As Figure 3 shown, the working process of the solution module is as follows:
[0107] S1. Population initialization: According to the number of objective functions M and the number of decision variables V, using the real-number coding method, initialize an N×(M + V) matrix by setting the population size N to generate the initial population;
[0108] S2. Non-dominated sorting: Use the non-dominated sorting algorithm to stratify the population of size N. By continuously comparing and marking non-dominated individuals, remove them layer by layer and execute in a loop until the population is fully stratified;
[0109] S3. Crowding degree calculation: Determine its crowding degree value by calculating the average distance between a certain solution in the population and its adjacent left and right points, so as to ensure the diversity of the population, and use sorting and weighting methods to determine the weights of the objective functions in the total weighted average crowding degree to select the final solution set;
[0110] S4. Selection, mutation and crossover: Select individuals through the roulette wheel selection method, randomly change the individual solutions through the mutation operation and retain the better solutions, and generate a new generation of population through the crossover operation, and finally form a new population containing 2N individuals;
[0111] S5. Elitist retention: Integrate the new population of the t-th generation with the parent generation into a population R of size 2N t , after non-dominated sorting and crowding degree calculation, first put the non-dominated set individuals into the new parent population. If the number is less than N, continue to supplement the next-level non-dominated set until it exceeds N, and then adjust it to N with the crowding degree operator. Finally, generate a new offspring population through the genetic operator;
[0112] S6. Terminate iteration: Terminate the iteration when the total number of iterations reaches the set value, and generate the final location selection of the shared car network points.
[0113] The present invention further includes a multi-objective optimal location selection system for shared car outlets considering environmental protection factors, which is used to implement the above-mentioned multi-objective optimal location selection method for shared car outlets considering environmental protection factors. The system includes:
[0114] A data acquisition module, which is used to acquire residents' travel information and shared car travel information;
[0115] A network demand prediction module, which is used to predict the demand scale of shared car outlets based on residents' travel information and shared car travel information;
[0116] A multi-objective network location selection module, which is used to construct a first objective function with the shortest total distance as the goal, a second objective function with the largest demand satisfaction as the goal, and a third objective function with the smallest pollution congestion coefficient as the goal based on the alternative shared car outlet locations and the demand scale of shared car outlets, and optimize the layout of shared car outlets;
[0117] A solution module, which is used to solve the multi-objective network location selection module based on the second-generation non-dominated sorting genetic algorithm and generate the final shared car outlet location.
[0118] Embodiment
[0119] In this example, 20 alternative outlets are randomly generated by Python. It is defined that the population in this area is 1 million, the average number of daily trips per resident is 2.2 times, the ratio of shared car trips is 2%, the average driving distance is 6 km per trip, the average number of passengers per trip is 2.75, the average driving time is 0.5 h, the average driving speed is 28.5 km / h, and the percentage of normal operation of shared cars is 90%. The following can be obtained:
[0120] (1) Total daily trips: 2200000.0
[0121] (2) Daily trips of shared cars: 44000.0
[0122] (3) Number of trips that each car can complete per day: 43.2
[0123] (4) Predicted number of shared cars required: 1019 vehicles
[0124] (5) Predicted number of outlets required: 11
[0125] Table 3 Attributes of 20 alternative outlets
[0126]
[0127]
[0128] Before conducting the site selection analysis and solution, relevant parameters are set first, and the parameter settings are as follows: Let the lower limit N of the number of shared car network points min = 11, the upper limit N of the number of shared car network points max = 13, the lower limit d of the distance between any two alternative network points min = 2 km, the service range of each alternative network point is 1.5 km, the weight coefficient θ1 of the pollution index of the shared car alternative network points is 0.33, and the weight coefficient θ2 of the congestion index of the shared car alternative network points is 1. NSGA-II is implemented using python, with the population size N = 100, the population iteration times gen = 100, the number of tournament pairing selections tour = 2, the crossover probability pc = 0.9, the crossover operator mu = 20, and the mutation operator mum = 10. The results obtained are as Figure 4 、 Figure 5 、 Figure 6 and shown in Table 4:
[0129] Table 4 Comparison of Iterative Running Results
[0130]
[0131] In summary, the present invention optimizes the site selection of shared car network points through the NSGA-II algorithm, not only maintaining the original number of network points, but also achieving the following goals through scientific and reasonable adjustment of the network point positions:
[0132] (1) The pollution index drops from 196.02 to 168.94, a decrease of about 14.3%. This indicates that the optimized network layout better considers environmental protection factors and reduces the impact on the environment. This is an important manifestation of the present invention emphasizing green sensitivity design.
[0133] (2) The coverage rate increases from 0.22 to 0.23, an increase of 4.5%. This means that the optimized solution can cover more potential user demand areas, improving the accessibility and convenience of the service.
[0134] (3) The total driving distance is shortened from 624.38 km to 600.48 km, a decrease of 3.8%. This not only helps to reduce operating costs (such as fuel consumption and maintenance costs), but also reduces carbon emissions during vehicle driving, further supporting the environmental protection goal.
[0135] (4) The demand for cars slightly increases from 611 to 614, a growth of 0.5%. This shows that the optimized solution may better meet the travel needs of users and improve the service quality.
[0136] These improvements demonstrate the effectiveness and practical application value of the present invention, especially showing a positive role in dealing with urban traffic congestion, reducing environmental pollution, and improving resource utilization efficiency. Through this innovative method, more convenient and environmentally friendly travel options can be provided for urban residents, and at the same time, it promotes the healthy development of the sharing economy.
[0137] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A multi-objective optimal site selection method for shared car outlets considering environmental protection factors, characterized in that the steps Including: Obtaining residents' travel information and shared car travel information; Generating alternative shared car network site selections based on residents' travel information and shared car travel information; Constructing a network demand prediction module, which predicts the demand scale of shared car networks based on residents' travel information and shared car travel information through the supply-demand balance method; Constructing a multi-objective network site selection module, which optimizes the layout of shared car network site selections based on alternative shared car network site selections and the demand scale of shared car networks. The multi-objective network site selection module includes a first objective function with the shortest total distance as the goal, a second objective function with the largest demand satisfaction as the goal, and a third objective function with the smallest pollution congestion coefficient as the goal; Constructing a solution module, which uses the second-generation non-dominated sorting genetic algorithm to solve the network site selection module and generate the final shared car network site selection.
2. The multi-objective optimal location selection method for shared car network points considering environmental protection factors according to claim 1, wherein, The network demand prediction module predicts the demand scale of shared car networks based on residents' travel information and shared car travel information through the supply-demand balance method, including: Based on residents' travel information and shared car travel information, predicting the total number of shared cars through the supply-demand balance method. The calculation formula for predicting the total number of shared cars is: where N is the predicted total number of shared cars, R is the total resident population, A is the average daily travel frequency of residents, P is the travel ratio occupied by shared cars, D is the average driving distance for one shared car trip, S is the average passenger capacity for one shared car trip, t is the average driving time for one shared car trip, v is the average driving speed for one shared car trip, and δ is the normal operation percentage of shared cars.
3. The multi-objective optimal site selection method for shared car outlets considering environmental protection factors according to claim 1, characterized in that The specific form of the first objective function with the shortest total distance as the goal is: Among them, min f1 is the first objective function, and y j is the first decision variable indicating whether the j-th alternative network point is selected as a shared car network point, and d ij is the distance between the shared car demand point i and the alternative network point j, j is a single shared car alternative network point, i is a single shared car demand point, I is the set of shared car demand points, and J is the set of shared car alternative network points The specific form of the second objective function with the largest demand satisfaction as the goal is: where minf2 is the second objective function, y j is the first decision variable indicating whether the j-th alternative network point is selected as a shared car network point, x ij is the second decision variable indicating whether the shared car demand point i can be covered by the alternative network point j, ρ i is the population density of the shared car demand point i, j is a single shared car alternative network point, i is a single shared car demand point, I is the set of shared car demand points, J is the set of shared car alternative network points The specific form of the third objective function with the smallest pollution congestion coefficient as the goal is: Among them, min f3 is the third objective function, y j is the first decision variable indicating whether the j-th alternative network point is selected as a shared car network point. θ1 is the weight coefficient of the pollution index of the shared car alternative network point i, θ2 is the weight coefficient of the congestion index of the shared car alternative network point i, c j is the pollution index of the shared car alternative network point i, t j is the congestion index of the shared car alternative network point i, j is a single shared car alternative network point, i is a single shared car demand point, I is the set of shared car demand points, and J is the set of shared car alternative network points.
4. The multi-objective optimization site selection method for shared car network points considering environmental protection factors according to claim 3, characterized in that, The constraint conditions of the first objective function, the second objective function, and the third objective function include: Among them, N min is the lower limit of the number of shared car network points, y j is the first decision variable indicating whether the j-th alternative network point is selected as a shared car network point, N max is the upper limit of the number of shared car network points, y m is the third decision variable indicating whether the m-th alternative network point is selected as a shared car network point, y n is the fourth decision variable indicating whether the n-th alternative network point is selected as a shared car network point, d min is the minimum allowable distance between any two selected shared car network points, d mn is the actual distance between alternative network point m and alternative network point n, x ij is the second decision variable indicating whether the shared car demand point i can be covered by the alternative network point j, d ij is the distance between the shared car demand point i and the alternative network point j, d0 is the service radius of the shared car network point, j is a single alternative shared car network point, i is a single shared car demand point, I is the set of shared car demand points, and J is the set of alternative shared car network points.
5. The multi-objective optimal location selection method for shared car network points considering environmental protection factors according to claim 1, characterized in that The working process of the solution module is: S1. Population initialization: According to the number of objective functions M and the number of decision variables V, using the real number coding method, initialize an N×(M + V) matrix by setting the population size N to generate the initial population; S2. Non-dominated sorting: Use the non-dominated sorting algorithm to stratify the population of size N. By continuously comparing and marking non-dominated individuals, remove them layer by layer and execute in a loop until the population is fully stratified; S3. Crowding degree calculation: Determine the crowding degree value by calculating the average distance between a solution in the population and its adjacent left and right points, so as to ensure the population diversity, and use sorting and weighting methods to determine the weights of the objective functions in the total weighted average crowding degree to select the final solution set; S4. Selection, mutation, and crossover: Select individuals through the roulette wheel selection method, randomly change the individual solutions through the mutation operation and retain the better solutions, and generate a new generation of populations through the crossover operation, finally forming a new population containing 2N individuals; S5. Elite retention: The new population of the t-th generation is merged with the parental generation to form a population R of size 2N t , after non-dominated sorting and crowding degree calculation, the individuals in the non-dominated set are first placed into the new parental population. If the number is less than N, the next-level non-dominated set is continuously supplemented until it exceeds N, and then it is adjusted to N using the crowding degree operator. Finally, a new offspring population is generated through genetic operators; S6. Terminate iteration: Terminate the iteration when the total number of iterations reaches the set value, and generate the final shared car network site selection.
6. A multi-objective optimal location system for shared car outlets considering environmental protection factors, which is used to implement any one of the multi-objective optimal location methods for shared car outlets considering environmental protection factors described in claims 1-5, characterized in that, Including: A data acquisition module for acquiring residents' travel information and shared car travel information; A network point demand prediction module for predicting the demand scale of shared car network points based on residents' travel information and shared car travel information; A multi-objective network point location selection module for constructing a first objective function with the shortest total distance as the goal, a second objective function with the largest demand satisfaction as the goal, and a third objective function with the smallest pollution and congestion coefficient as the goal based on the alternative shared car network point locations and the demand scale of shared car network points, and optimizing the layout of shared car network points; A solution module for solving the multi-objective network point location selection module based on the second-generation non-dominated sorting genetic algorithm to generate the final shared car network point location.