Traffic plan generation method and system based on regional state evaluation

Abstract

The invention discloses a traffic plan generation method based on regional state evaluation, and the method comprises the steps: obtaining a historical traffic data set of a regional object, extracting an influence factor index, and building a state evaluation model of a corresponding region; the state evaluation model comprises a corresponding relation between influence factor indexes and initial evaluation values and a mapping relation between the initial evaluation values and final evaluation values; acquiring a real-time traffic data set of the regional object, extracting an influence factor index, and calculating through a state evaluation model to obtain a final evaluation value of a corresponding region; and calling a corresponding plan library based on a comparison result of the final evaluation value and an evaluation threshold value, and carrying out matching and screening according to the influence factor indexes to obtain a traffic plan. According to the invention, the corresponding evaluation result and the generation of the traffic plan can be output according to the evaluation standard of the management department or the organization party of the regional object, and the differentiated state evaluation requirements and the plan generation of different traffic regions can be met.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a method and system for generating a transportation plan based on regional status assessment. Background Art

[0002] With the economic growth and the rising living standards, traveling or participating in various activities has become the first choice for people to relax and live. The huge potential and booming development trend of the tourism industry not only provides different cultural experiences for the masses, but also promotes local employment, overdrafts, infrastructure construction and economic development. For example, people go to various scenic spots for sightseeing during holidays or after get off work; go to stadiums or exhibition halls to attend concerts, music festivals and various exhibitions; or go shopping in various shopping malls during holidays or at night. However, the large number of people gathering in these specific areas in a short period of time will cause great difficulties in management, including traffic congestion around scenic spots or venues and shopping malls. When a large number of people gather in these specific areas, private cars and other means of transportation for transporting people will inevitably gather around these areas, greatly affecting the traffic status of these specific areas. Therefore, it is necessary to evaluate the traffic status and trigger the corresponding traffic control according to the specific status to ensure smooth traffic around the specific area. However, for the management departments or organizers of different scenic spots, venues or shopping malls, their judgment and evaluation standards for the traffic status of the surrounding areas are not the same. The existing regional traffic status evaluation methods are mostly fixed results obtained based on preset indicators, and cannot be evaluated according to the evaluation score standards of different management departments. It cannot meet the differentiated status evaluation needs of different traffic areas.

[0003] The "A method for predicting the traffic situation in the landside area of ​​an airport based on multi-source heterogeneous data" disclosed in Chinese patent documents has a publication number of CN113837446A and a publication date of 2021.12.24. It collects traffic flow data from traffic flow detection points or sections to be predicted in the landside area of ​​the airport, and simultaneously obtains other multi-source heterogeneous data in the landside area of ​​the airport, and uses a feature selection algorithm to sort the feature importance; based on the screened multi-feature data set, a deep learning algorithm is used to predict the traffic flow status in the landside area of ​​the airport; an evaluation index for judging the traffic situation in the landside area of ​​the airport is established, and the influence weight of each evaluation index and its traffic situation membership function are determined; based on the traffic flow state prediction results, the evaluation index is calculated to determine the membership of each evaluation index to each traffic situation; finally, a fuzzy comprehensive evaluation is performed based on the membership and weight of each evaluation index, and the traffic situation level result is determined according to the maximum membership principle. However, this technology is a targeted traffic status assessment method for specific areas around airports. It is not suitable for status assessment in other areas such as scenic spots, shopping malls, and venues, and it is impossible to perform differentiated status assessments based on the assessment score standards of different regional management departments. Summary of the invention

[0004] The present invention aims to overcome the problem that the traffic status assessment methods in the prior art are mostly fixed results obtained based on preset indicators, and cannot be assessed according to the respective assessment score standards of different management departments, and cannot meet the differentiated status assessment needs of different traffic areas. A traffic plan generation method and system based on regional status assessment is provided.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: A method for generating a traffic plan based on regional status assessment, comprising: Obtain the historical traffic data set of the regional object, extract the influencing factor index, and establish a state assessment model for the corresponding region; the state assessment model includes the corresponding relationship between the influencing factor index and the initial assessment value, and The mapping relationship between the initial evaluation value and the final evaluation value; Obtain the real-time traffic data set of the regional object, extract the influencing factor indicators, and calculate the final evaluation value of the corresponding area through the state evaluation model; Based on the comparison result between the final evaluation value and the evaluation threshold, the corresponding plan library is retrieved, and the traffic plan is obtained by matching and screening the influencing factor indicators.

[0006] In the present invention, firstly, a universal evaluation is performed according to the actual traffic conditions of the regional object, and then the mapping relationship between the initial evaluation value and the final evaluation value is determined according to the specific evaluation criteria of different regions, and the result of the universal evaluation is converted into a final evaluation result specifically for a certain regional object. The management department or organizer of the regional object can screen out the corresponding traffic plan according to the final evaluation result and the evaluation threshold set by itself and the corresponding plan library to control the current traffic state. In the traffic plan generation method provided by the present invention, the management department or organizer only needs to input its own evaluation criteria (including historical traffic data sets, evaluation thresholds of the final evaluation value, and plan libraries corresponding to different evaluation thresholds), and can obtain the final evaluation results that meet its own requirements and generate the corresponding traffic plan according to the actual traffic state, which can meet the differentiated state evaluation needs and plan generation of different traffic areas, and there is no need to create a new evaluation model from scratch for each regional object.

[0007] Preferably, in the state assessment model, a corresponding weight coefficient is assigned to each influencing factor indicator; The influencing factor indicators are weighted and summed based on their corresponding weight coefficients to obtain the initial evaluation value; The final evaluation value interval is divided according to the evaluation threshold. For any final evaluation value interval, the corresponding initial evaluation value interval is determined based on the same traffic state, and a mapping relationship is established.

[0008] In the present invention, obtaining the initial evaluation value by weighted summing up the acquired influencing factor indicators is a method for obtaining the evaluation result relatively quickly and conveniently. At the same time, as long as a relatively reasonable weight coefficient is set, the final initial evaluation value can be more consistent with the actual traffic status, which is accurate, convenient and has a small amount of calculation.

[0009] In the mapping relationship between the initial evaluation value and the final evaluation value, the actual situation of the traffic status itself is fixed and unchanged. The only difference is the standard of the initial evaluation value during the universal evaluation and the evaluation standard of the specific traffic status by the management departments or organizers in different regions. The final evaluation value interval can be delineated according to its own standards and the evaluation threshold. The final evaluation value in each final evaluation value interval corresponds to its specific traffic status. The initial evaluation value combination of the same traffic status obtains the corresponding initial evaluation value interval, which is mutually associated with the corresponding final evaluation value interval to establish a mapping relationship.

[0010] Preferably, in the correspondence between the influencing factor index and the initial evaluation value: Calculate the coefficient of variation of each influencing factor indicator and arrange them in descending order; Assign the corresponding initial weights in descending order of the influencing factor indicators; The evaluation value set is calculated based on the initial weights and influencing factor indicators; The initial weights are modified until the data in the evaluation value set meets the distribution criteria, and the weight coefficients corresponding to the influencing factor indicators are obtained.

[0011] In the present invention, when the management departments and organizers of the regional objects first use the plan generation method, there may be a situation where the historical traffic data set is not fully preserved and the historical evaluation data set is empty. In this case, it is necessary to collect traffic data for a period of time first, and it is impossible to train the corresponding weights of the influencing factor indicators based on the historical evaluation data set; therefore, it is necessary to first determine the influence of each influencing factor indicator in the specific evaluation process based on the coefficient of variation of the influencing factor indicator, so as to determine the corresponding initial weight; and then adjust and correct the initial weight to obtain a weight coefficient that meets the requirements, so as to establish a corresponding relationship between the influencing factor indicator and the initial evaluation value.

[0012] Preferably, in the correspondence between the influencing factor index and the initial evaluation value: Obtain historical evaluation data sets, including historical influencing factor indicators and corresponding historical initial evaluation values; The weighted sum of the historical influencing factor indicators and the corresponding weight coefficients is the corresponding historical initial evaluation value. The historical evaluation data set is used for fitting training to obtain the weight coefficients corresponding to the various influencing factor indicators.

[0013] In the present invention, after the management departments and organizers of the regional objects have used the plan generation method for a period of time, a large number of historical traffic data sets and historical evaluation data sets have been saved. In this case, fitting training can be directly performed based on the historical evaluation data sets to obtain the weight coefficients of various influencing factor indicators. Compared with the weight coefficients obtained by assigning values ​​in descending order according to the coefficient of variation, they are more accurate and meet the data distribution requirements. And because the data of historical influencing factor indicators and historical initial evaluation values ​​in the historical evaluation data sets will be continuously updated and increased over time, the weight coefficients obtained after fitting training will also be more in line with the actual traffic status.

[0014] Preferably, assigning corresponding initial weights in descending order of the influencing factor indicators includes: Calculate the ratio of the standard deviation to the mean of any influencing factor indicator to obtain the coefficient of variation of the influencing factor indicator; Normalize the coefficient of variation of all influencing factor indicators; The normalized results are used as the initial weights corresponding to the influencing factor indicators.

[0015] Preferably, the traffic plan obtained by matching and screening the influencing factor indicators includes: Arrange in descending order according to the weight coefficient corresponding to each influencing factor indicator; Selecting a number of plans matching the first influencing factor indicator from the plan library to form a plan set; The plan set matched with the next influencing factor index is selected from the plan set obtained by matching the previous influencing factor index, until the number of plans in the plan set is one, and the final traffic plan is obtained.

[0016] Preferably, the step of modifying the initial weight until the data in the evaluation value set meets the distribution standard comprises: The minimum value in the evaluation value set is less than or equal to the preset minimum value threshold; and The proportion of the number of data in the evaluation value set that is less than the low score threshold to the total number is greater than or equal to the low score ratio threshold; The initial evaluation value interval divided by the low score threshold corresponds to the final evaluation value interval divided by the evaluation threshold.

[0017] Preferably, the influencing factor indicators include: Parking lot saturation, which calculates the average ratio of the number of occupied parking lots to the total number of parking lots in the area; Passenger flow saturation, which calculates the average value of the ratio of real-time passenger flow to passenger flow capacity in each sub-area within the area; Traffic saturation, calculate the ratio of the sum of the real-time traffic volume in each sub-area in the area to the traffic capacity.

[0018] Preferably, the influencing factor indicators also include: Road congestion mileage ratio: divide the area object from the inside to the outside into core circle, control circle and diversion circle, and calculate the weighted average of the road congestion ratio in the core circle, the road congestion ratio in the control circle and the road congestion ratio in the diversion circle; The road congestion ratio is the ratio of the length of roads in congestion and oversaturation states to the total length of roads.

[0019] A traffic plan generation system based on regional status assessment, comprising: The index extraction module extracts the corresponding influencing factor index from the traffic data set of the regional object; An evaluation model module stores a state evaluation model and updates the model according to an updated historical traffic data set; The plan library module stores the plan libraries corresponding to different evaluation thresholds.

[0020] In the present invention, the index extraction module and the evaluation model module are the parts of the system that have been built when they are provided to the management department or the organizer, and the plan library module is the part provided by the management department or the organizer itself and associated with the system; in addition, it also includes a traffic data module, which is provided by the relevant departments. The historical traffic data and the real-time traffic data are obtained, and are associated with the system for the index extraction module to obtain data; after the association is completed, the index extraction module extracts the influencing factor index, and the evaluation model module obtains the final evaluation value result according to the influencing factor index, and the corresponding plan library is retrieved in the plan library module for screening and matching to obtain the corresponding traffic plan. In the case that the management department or the organizer does not have an independent evaluation standard, the system can be used directly for evaluation and plan generation. In the case of having its own evaluation standards, it also includes a standard input module, which inputs its own evaluation standards into the system, updates the mapping relationship between the initial evaluation value and the final evaluation value in the evaluation model, and thus performs differentiated status evaluation for different traffic areas.

[0021] The present invention has the following beneficial effects: a state evaluation model is set, including the corresponding relationship between the influencing factor index and the initial evaluation value, and the mapping relationship between the initial evaluation value and the final evaluation value. The corresponding evaluation results and the generation of traffic plans can be output according to the evaluation standards of the management department or organizer of the regional object, which can meet the differentiated state evaluation needs and plan generation of different traffic areas, and there is no need to create a new evaluation model from scratch for each regional object. Different weight coefficient acquisition methods are set in the state evaluation model. When it is just used or the amount of data in the historical evaluation data set is insufficient, the weight coefficient can be obtained by calculating the coefficient of variation of the influencing factor index to assign values ​​in descending order and make corrections; when enough data of the historical evaluation data set is saved, the weight coefficient can be obtained by fitting training, and more accurate evaluation results can be obtained in different situations. For the generation of traffic plans, by matching with the influencing factor index and screening from the plan library, a traffic plan that is more in line with the actual traffic state needs can be obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of the traffic plan generating method in the present invention.

[0023] Figure 2 It is a flow chart of the traffic plan generating method in the first embodiment of the present invention.

[0024] Figure 3 It is a flow chart of the traffic plan generating method in the second embodiment of the present invention. DETAILED DESCRIPTION

[0025] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0026] like Figure 1 As shown, a method for generating a traffic plan based on regional status assessment includes: Obtain the historical traffic data set of the regional object, extract the influencing factor indicators, and establish the state assessment model of the corresponding area; the state assessment model includes the corresponding relationship between the influencing factor indicators and the initial assessment value, as well as The mapping relationship between the initial evaluation value and the final evaluation value; Obtain the real-time traffic data set of the regional object, extract the influencing factor indicators, and calculate the final evaluation value of the corresponding area through the state evaluation model; Based on the comparison result between the final evaluation value and the evaluation threshold, the corresponding plan library is retrieved, and the traffic plan is obtained by matching and screening the influencing factor indicators.

[0027] It should be noted that in the present invention, first, a universal evaluation is performed based on the actual traffic conditions of the regional object, and then the mapping relationship between the initial evaluation value and the final evaluation value is determined based on the specific evaluation criteria of different regions, and the result of the universal evaluation is converted into a final evaluation result specifically for a certain regional object. The management department or organizer of the regional object can screen out the corresponding traffic plan based on the final evaluation result and the evaluation threshold set by itself and the corresponding plan library to control the current traffic state. In the traffic plan generation method provided by the present invention, the management department or organizer only needs to input its own evaluation criteria (including historical traffic data sets, evaluation thresholds of the final evaluation value, and plan libraries corresponding to different evaluation thresholds), and can obtain the final evaluation results that meet its own requirements and generate the corresponding traffic plan according to the actual traffic state, which can meet the differentiated state evaluation needs and plan generation of different traffic areas, and there is no need to create a new evaluation model from scratch for each regional object.

[0028] As a specific embodiment, in the state assessment model, a corresponding weight coefficient is assigned to each influencing factor indicator; The influencing factor indicators are weighted and summed based on their corresponding weight coefficients to obtain the initial evaluation value; The final evaluation value interval is divided according to the evaluation threshold. For any final evaluation value interval, the corresponding initial evaluation value interval is determined based on the same traffic state, and a mapping relationship is established.

[0029] It should be noted that in the present invention, obtaining the initial evaluation value by weighted summation based on the acquired influencing factor indicators is a method for obtaining the evaluation result relatively quickly and conveniently. At the same time, as long as a relatively reasonable weight coefficient is set, the final initial evaluation value can be more in line with the actual traffic conditions, which is accurate, convenient and requires little calculation.

[0030] It is worth explaining that in the mapping relationship between the initial evaluation value and the final evaluation value, the actual situation of the traffic status itself is fixed and unchanged. The only difference is the standard of the initial evaluation value during the universal evaluation and the evaluation standard of the specific traffic status by the management departments or organizers in different regions. The final evaluation value interval can be defined according to its own standards and the evaluation threshold. The final evaluation value in each final evaluation value interval corresponds to its specific traffic status. The initial evaluation values ​​of the same traffic status are combined to obtain the corresponding initial evaluation value interval, which is mutually associated with the corresponding final evaluation value interval to establish a mapping relationship.

[0031] Optionally, when the management department or organizer of the regional object does not have its own specific evaluation criteria, the mapping relationship between the initial evaluation value and the final evaluation value in the status evaluation model may not be performed, and the initial evaluation value obtained by the evaluation can be directly used as the final evaluation value to match and filter in the corresponding plan library to generate the corresponding traffic plan.

[0032] Optionally, when the management department or organizer of the regional object has its own specific evaluation requirements or standards, it is necessary to first input specific evaluation standards, such as the evaluation threshold of the final evaluation value, the plan library corresponding to each evaluation threshold, and the final evaluation value of several actual traffic conditions, and then generate a mapping relationship between the initial evaluation value and the final evaluation value to complete the setting of the traffic plan generation method. Subsequently, after extracting the influencing factor indicators based on real-time traffic data, the initial evaluation value and the corresponding final evaluation value can be obtained through the state evaluation model, and the corresponding traffic plan can be obtained by matching and screening in the corresponding plan library according to the final evaluation value.

[0033] Specifically, for example, a first evaluation threshold and a second evaluation threshold are set for the final evaluation value, and it is considered that there is no need to activate the traffic plan between the first evaluation threshold and the full score (the first interval); if it is less than or equal to the first evaluation threshold and greater than the second evaluation threshold (the second interval), it is necessary to screen the corresponding traffic plan in the first plan library; if it is less than or equal to the second evaluation threshold (the third interval), it is necessary to screen the corresponding traffic plan in the second plan library.

[0034] According to the evaluation standards of the management department or the organizer itself (there is no specific quantification, but more of a subjective judgment, to evaluate and score different traffic conditions), each final evaluation value interval corresponds to a large number of specific traffic conditions in each interval. The data information of the specific traffic conditions in the first interval is used to calculate the first range of initial evaluation values ​​[a, b], the data information of the specific traffic conditions in the second interval is used to calculate the second range of initial evaluation values ​​[c, d], and the data information of the specific traffic conditions in the third interval is used to calculate the third range of initial evaluation values ​​[e, f]. b as the upper limit of the initial evaluation value can be corrected to full marks, and e as the lower limit of the initial evaluation value can be corrected to the lowest score of zero.

[0035] If two adjacent ranges are just connected, such as a=d, f=c, then the interval of the initial evaluation value can be modified to obtain the first interval corresponding to d to full marks, the second interval corresponding to c to d, and the third interval corresponding to zero to c. If two adjacent ranges are intersecting or non-overlapping, the interval of the initial evaluation value is redefined with the average value of the boundary, the first interval corresponds to (a+d) / 2 to full marks, the second interval corresponds to (f+c) / 2 to (a+d) / 2, and the third interval corresponds to zero to (f+c) / 2. Since the first interval does not require a traffic plan, and the second and third intervals both correspond to the need for a traffic plan, the intervals of the initial evaluation value can also be merged into two parts, (a+d) / 2 to full marks and zero to (a+d) / 2, and are respectively associated with the corresponding final evaluation value intervals to obtain a mapping relationship.

[0036] As a specific embodiment, the traffic plan obtained by matching and screening the influencing factor index includes: Arrange in descending order according to the weight coefficient corresponding to each influencing factor indicator; Selecting a number of plans matching the first influencing factor indicator from the plan library to form a plan set; The plan set matched with the next influencing factor index is selected from the plan set obtained by matching the previous influencing factor index, until the number of plans in the plan set is one, and the final traffic plan is obtained.

[0037] It should be noted that, first of all, the weight coefficient corresponding to the influencing factor indicator can be used to represent the influence of the influencing factor indicator on the evaluation result. The larger the weight coefficient of the influencing factor indicator, the greater the corresponding degree of influence. Therefore, when screening and matching the plan set, priority should be given to the influencing factor indicators with large weight coefficients, and priority should be given to traffic plans that focus on controlling the corresponding influencing factor indicators, so as to generate traffic plans that meet the actual traffic status requirements.

[0038] Specifically, the influencing factor indicators A, B, and C are arranged in descending order of the weight coefficient. Different meta-plans in all the plans in the plan set focus on different directions. Therefore, when screening and matching, all plans in the plan set that give priority to influencing factor indicator A are first screened out to form a new plan set. In the new plan set, the plans that give priority to influencing factor indicator B are screened to obtain a plan set, and the plans that give priority to influencing factor indicator C are screened from the plan set until the number of plans is one. If the number of plans is not one after all indicators are screened, the influencing factor indicator results of the actual traffic status corresponding to the plan are calculated among these plans, and the plan with the smallest difference from the current influencing factor indicator is selected as the final traffic plan.

[0039] Based on the traffic plan generation method, a traffic plan generation system based on regional status assessment is also provided, including: The index extraction module extracts the corresponding influencing factor index from the traffic data set of the regional object; An evaluation model module stores a state evaluation model and updates the model according to an updated historical traffic data set; The plan library module stores the plan libraries corresponding to different evaluation thresholds.

[0040] It should be noted that the index extraction module and the evaluation model module in the present invention are the parts of the system that have been built when they are provided to the management department or the organizer, and the plan library module is the part provided by the management department or the organizer itself and associated with the system; in addition, it also includes a traffic data module, which is provided by the relevant departments. The historical traffic data and real-time traffic data are obtained, and are associated with the system for the index extraction module to obtain data; after the association is completed, the index extraction module extracts the influencing factor index, and the evaluation model module obtains the final evaluation value result according to the influencing factor index, and the corresponding plan library is retrieved in the plan library module to screen and match to obtain the corresponding traffic plan. In the case that the management department or the organizer does not have an independent evaluation standard, the system can be used directly for evaluation and plan generation. In the case of having its own evaluation standards, it also includes a standard input module, which inputs its own evaluation standards into the system, updates the mapping relationship between the initial evaluation value and the final evaluation value in the evaluation model, and thus performs differentiated status evaluation for different traffic areas.

[0041] Embodiment 1: In this embodiment, the management department or organizer of the area object has just started to use the plan generation method or has just used it. At this time, the historical traffic data set of the corresponding area is obtained from the traffic management bureau or relevant departments. The data may include traffic data on different dates such as historical working days, weekends, and major holidays such as May 1st and National Day. The data can be collected at fixed intervals such as three minutes or one minute, and the historical traffic data set is obtained by arranging them in the order of collection time. Figure 2 As shown in the figure, after extracting the influencing factor indicators from the historical traffic data set, a status assessment model for the corresponding area is established.

[0042] The state assessment model includes the corresponding relationship between the influencing factor index and the initial assessment value, and the mapping relationship between the initial assessment value and the final assessment value. The influencing factor index includes at least five factors: key area index L1, parking lot saturation L2, passenger flow saturation L3, vehicle flow saturation L4, and road congestion mileage ratio L5; the corresponding relationship between these five influencing factor indicators and the initial assessment value is sc=100-100*(L1*c1+L2*c2+L3*c3+L4*c4+L5*c5). In this embodiment, the full score is 100 points, and the assessment is performed by deducting points according to the influencing factor index; the larger the value of the influencing factor index, the lower the initial assessment value, and the worse the corresponding traffic state.

[0043] After the initial evaluation value sc is calculated based on the influencing factor index, the final evaluation value score is obtained based on the mapping relationship between the initial evaluation value and the final evaluation value. Before obtaining the corresponding mapping relationship, the management department or organizer of the regional object inputs its own evaluation criteria, including the historical traffic data set, the evaluation threshold of the final evaluation value and the plan library corresponding to different evaluation thresholds, and the final evaluation value under the actual traffic conditions in history (which can be evaluated manually according to one's own judgment criteria); and calculates the initial evaluation value under the same actual traffic conditions, so as to determine the one-to-one corresponding initial evaluation value interval and final evaluation value interval.

[0044] In this embodiment, the management department of the area object believes that when the final evaluation value is the first evaluation threshold score1=70 points, it is necessary to trigger the first plan library to filter out the corresponding traffic plan for control. If the final evaluation value is greater than score1=70 points, no control is required; when the final evaluation value is the second evaluation threshold score2=40 points, it is necessary to trigger the second plan library to filter out the corresponding traffic plan for control. According to the final evaluation value interval division method of the present invention, the final evaluation value interval can be divided into a first interval of [70, 100], a second interval of [40, 70] and a third interval of [0, 40].

[0045] The corresponding initial evaluation value range is calculated based on the actual traffic status in the final evaluation value interval, and the corresponding initial evaluation value intervals [sc1, 100] and [0, sc1] are determined according to the correction method of the present invention. That is, the traffic status with an initial evaluation value greater than sc1 does not require emergency plan control in the standards of the relevant management department, and the traffic status with an initial evaluation value less than sc1 selects the corresponding emergency plan for control according to the evaluation threshold corresponding to the mapped final evaluation value.

[0046] Specifically, the mapping between [sc1, 100] corresponding to the initial evaluation value sc and [score1, 100] of the final evaluation value score is: score = (sc-sc1) / (100-sc1)*(100-score1)+score1.

[0047] The mapping between the initial evaluation value sc corresponding to [0, sc1] and the final evaluation value score is: score=(sc-(sc1-thr)) / (sc1-(sc1-thr))*(score1-score2)+score2; Wherein thr is determined by the value sc2 in the initial evaluation value corresponding to the final evaluation value score2, thr = sc1-sc2, that is, when the initial evaluation value is sc2, the corresponding final evaluation value is score2, when the initial evaluation value is less than sc2, the final evaluation value is less than score2.

[0048] Taking score1=70, score2=40 as an example, after correction calculation, the initial evaluation value interval is [30, 100] and [0, 30], that is, sc1 is 30. In the evaluation standard of the management department, the initial evaluation value of 30 points needs to be screened in the first plan library for plan control. At the same time, according to the initial evaluation value of 10 points corresponding to the final evaluation value of 40 points, the traffic plan needs to be screened in the second plan library for plan control, that is, sc2=10. Based on this, thr=20 can be obtained, thereby obtaining the mapping relationship between the initial evaluation value and the final evaluation value: score = (sc-30) / (100-30)*(100-70)+70, sc belongs to [30, 100]; score=(sc-(30-20)) / (30-(30-20))*(70-40)+40, sc belongs to [0,30].

[0049] After the determination of each parameter in the state assessment model is completed, the real-time traffic data set of the regional object is obtained, and the influencing factor indicators are extracted, and the final assessment value of the corresponding area is calculated through the state assessment model; Based on the comparison result between the final evaluation value and the evaluation threshold, the corresponding plan library is retrieved, and the traffic plan is obtained by matching and screening the influencing factor indicators.

[0050] As an optional embodiment, the influencing factor indicators include: Parking lot saturation L2, calculate the average value of the ratio of the number of occupied parking lots to the total number of parking lots in the area; The passenger flow saturation L3 is calculated by calculating the average value of the ratio of real-time passenger flow to passenger flow capacity in each sub-area in the area; Traffic saturation L4 calculates the ratio of the sum of the real-time traffic volume in each sub-area within the area to the traffic capacity.

[0051] Specifically, parking lot saturation Where Np ark Indicates the number of parking lots in the area, N i停车 and N i总车位 They represent the number of occupied parking spaces and the total number of parking spaces in the i-th parking lot in the area respectively.

[0052] For the number of parking lots in the area, the relevant management department can obtain specific results based on the regional map or planning map or the parking lot's filing records or through manual visits and surveys. For the total number of parking lots in the area, the number of parking lots in the area can be obtained while the number of parking lots in the area is obtained through investigation. For the number of occupied parking lots in each parking lot, the number of occupied parking lots can be determined by counting the cameras installed at the entrance and exit of the parking lot, or the number of parking lots in the parking lot can be determined by linking the parking lot's intelligent charging system.

[0053] Specifically, traffic saturation Where N area Indicates the number of sub-areas in the area, N i客流 and N i客流容 They represent the real-time passenger flow and passenger flow capacity of the i-th sub-area in the area respectively.

[0054] The number of sub-areas divided in an area is determined by the relevant management department or organizer based on the shape or actual situation of the area. For example, different sub-areas can be divided according to different attractions in a scenic area; near stadiums or exhibition halls, different sub-areas can be divided according to the distribution of surrounding buildings or streets. The passenger flow capacity of each sub-area is pre-set after the sub-area division is completed, indicating the passenger flow carrying capacity of the sub-area under ideal conditions. Since it is difficult to count the real-time passenger flow in the sub-area one by one, it can be calculated by sampling and counting the distribution of cameras in the sub-area; for the situation where there is isolation between different sub-areas (such as a building as a sub-area, or a closed scenic spot as a sub-area), it can be counted according to the population flow in and out.

[0055] Specifically, traffic saturation Where N area Indicates the number of sub-areas in the area, N i客流 represents the real-time traffic flow in the ith sub-area of ​​the region, N 车流容量 Represents the traffic capacity in the area.

[0056] The number of sub-areas divided in an area is determined by the relevant management department or organizer based on the shape or actual situation of the area. For example, in a scenic area, different sub-areas can be divided according to different attractions; near stadiums or exhibition halls, different sub-areas can be divided according to the distribution of surrounding buildings or streets. For the real-time traffic flow in each sub-area, it can be collected and counted based on the cameras on the road. In addition, when calculating the traffic capacity, because some sub-areas, such as inside the venues and scenic spots, cannot park or drive vehicles, it is necessary to consider the traffic capacity of the entire area from the overall perspective of the area.

[0057] The specific traffic capacity can be selected from the planned carrying capacity during regional planning, or it can be calculated through historical traffic data and the historical maximum value of real-time vehicles on the road in the area can be selected as the traffic capacity.

[0058] As a specific embodiment, the influencing factor index also includes: Road congestion mileage ratio L5, divide the regional object from the inside to the outside into a core circle, a control circle and a diversion circle, and calculate the weighted average of the road congestion ratio in the core circle, the road congestion ratio in the control circle and the road congestion ratio in the diversion circle; The road congestion ratio is the ratio of the length of roads in congested and oversaturated states to the total length of roads.

[0059] Specifically, the road congestion mileage ratio Among them, α1, α2 and α3 are the coefficients of the road congestion ratio in the core circle, the road congestion ratio in the control circle and the road congestion ratio in the diversion circle respectively; L 1 and are the total length of roads in the core circle and the length of roads in congestion or above; L 2 and are the total length of roads in the control circle and the length of roads in congestion or above; L 3 and are the total length of roads within the diversion circle and the length of roads in congestion or above.

[0060] For regional objects, the core circle, control circle and diversion circle are divided from the inside to the outside. The more inward the circle, the greater the impact of road congestion. The traffic status of the road can be divided into free, non-congested, congested and oversaturated from free to busy. The specific division standard can set the corresponding traffic flow threshold, and make a judgment based on the comparison between the real-time traffic flow and the traffic flow threshold; the road congestion ratio is calculated by dividing the sum of the road lengths in the congested state and the oversaturated state by the total road length.

[0061] Furthermore, the influencing factor index also includes a key area index L1. In this embodiment, the key area index is selectable, and the saturation of key parking lots or the congested mileage ratio of key roads, or other evaluation indicators required by the management department, ie, the organizer, can be selected.

[0062] When the key area indicator L1 is the saturation of the key parking lot in Indicates the number of key parking lots, and They represent the number of occupied parking lots and the total number of parking lots in the ith key parking lot. The selection of key parking lots can be determined manually by the management department or the organizer, or the first several parking lots with the largest data fluctuation in the historical parking lot saturation data can be selected as key parking lots.

[0063] When the key area indicator L1 is the ratio of congested mileage of key roads Where L core and They are the total length of key roads and the length of roads in congestion or above. The selection of key roads can be determined by the management department or organizer according to actual needs, or the first several roads with the largest data fluctuation in the historical road congestion mileage ratio can be selected as key roads.

[0064] In this embodiment, after extracting a series of data of each influencing factor index, each influencing factor index is normalized to facilitate subsequent calculations. The specific normalization method can be: extracting the maximum and minimum values ​​of each influencing factor index, and the normalized value = (influencing factor index-minimum value) / (maximum value-minimum value).

[0065] After completing the extraction and normalization of the influencing factor indicators, the corresponding relationship between the influencing factor indicators and the initial evaluation values ​​is established: Calculate the coefficient of variation of each influencing factor indicator and arrange them in descending order; Assign the corresponding initial weights in descending order of the influencing factor indicators; The evaluation value set is calculated based on the initial weights and influencing factor indicators; The initial weights are modified until the data in the evaluation value set meets the distribution criteria, and the weight coefficients corresponding to the influencing factor indicators are obtained.

[0066] It should be noted that in the present invention, when the management departments and organizers of the regional objects first use the plan generation method, there may be situations where the historical traffic data set is not fully preserved and the historical evaluation data set is empty. In this case, it is necessary to collect traffic data for a period of time first, and it is impossible to train the corresponding weights of the influencing factor indicators based on the historical evaluation data set; therefore, it is necessary to first determine the influence of each influencing factor indicator in the specific evaluation process based on the coefficient of variation of the influencing factor indicator, so as to determine the corresponding initial weight; and then adjust and correct the initial weight to obtain a weight coefficient that meets the requirements, thereby establishing a corresponding relationship between the influencing factor indicator and the initial evaluation value.

[0067] Specifically, the corresponding initial weights are assigned in descending order according to the order of the influencing factor indicators, including: Calculate the ratio of the standard deviation to the mean of any influencing factor indicator to obtain the coefficient of variation of the influencing factor indicator; Coefficient of variation of influencing factor indicators Where σ(L i ) represents the standard deviation of the influencing factor index, L i Corresponding to one of the five influencing factor indicators; It represents the average value of the influencing factor index.

[0068] Optionally, after calculating the coefficient of variation of each influencing factor indicator, each influencing factor indicator is assigned an initial weight in descending order of the coefficient of variation. For example, if the influencing factor indicators in descending order are L1, L2, L3, L4, and L5, the corresponding weights c1=0.4, c2=0.3, c3=0.2, c4=0.09, and c5=0.01 can be assigned, so that the sum of the initial weights is 1; the specific initial weight values ​​can be changed according to actual conditions. However, it is believed that there are too many subjective factors when assigning initial weights, and it is not easy to obtain a more accurate evaluation result. Therefore, the initial weights can be processed according to the obtained coefficient of variation.

[0069] Optionally, normalize the coefficients of variation of all influencing factor indicators; The normalized results are used as the initial weights corresponding to the influencing factor indicators: Where ci is the initial weight corresponding to the i-th influencing factor index, cv i is the coefficient of variation of the i-th influencing factor index; cv min and cv min The larger the coefficient of variation of the influencing factor index is, the higher the impact of the index on the traffic status is, and the greater the weight should be given to the impact.

[0070] After the initial weights are assigned, the evaluation value set is calculated based on the initial weights and influencing factor indicators. The evaluation value set contains all evaluation values ​​calculated based on the influencing factor indicators: sc=100-100*(L1*c1+L2*c2+L3*c3+L4*c4+L5*c5) The initial weight is modified according to the data distribution in the evaluation value set, and the evaluation value set is recalculated, and the calculation is iterated continuously until the data in the evaluation value set meets the distribution standard. Finally, the weight coefficient that meets the requirements is obtained.

[0071] Further, modifying the initial weights until the data in the evaluation value set meets the distribution criteria includes: The minimum value in the evaluation value set is less than or equal to the preset minimum value threshold; and The proportion of the number of data in the evaluation value set that is less than the low score threshold to the total number is greater than or equal to the low score ratio threshold; The initial evaluation value interval divided by the low score threshold corresponds to the final evaluation value interval divided by the evaluation threshold.

[0072] Specifically, the minimum threshold is set to 5 as needed, and the low score threshold sc1 is determined according to the boundary of the initial evaluation value interval determined by the evaluation threshold. In this embodiment, sc1=30; at the same time, the low score ratio threshold is set according to specific needs.

[0073] View sc data minimum sc min , count the proportion of the number of data below sc1 in the total data occ; when the minimum sc value sc min >5, it means that the initial weight is unreasonable and the distribution is not satisfied, so adjust the initial weight; When c1<0.95, c1=c1+n, c2=c2-n / 4, c3=c3-n / 4, c4=c4-n / 4, c5=c5-n / 4; n is the adjustment step size, which defaults to 0.05 in this embodiment, and then sc is recalculated; When c1>=0.95, c2=c2+n, c1=c1-n / 4, c3=c3-n / 4, c4=c4-n / 4, c5=c5-n / 4; n is the adjustment step size, which defaults to 0.05 in this embodiment, and then sc is recalculated; When sc is the minimum value sc min <=5, and occ>=low score ratio threshold, the iteration of the initial weight is completed to obtain the weight coefficient.

[0074] After completing the determination of various parameters in the status assessment model (the weight coefficient of the influencing factor index and the mapping relationship between the initial assessment value and the final assessment value), the real-time traffic data set of the regional object is obtained, and the influencing factor index is extracted. The final assessment value of the corresponding area is calculated through the status assessment model; based on the comparison result between the final assessment value and the assessment threshold, the corresponding plan library is retrieved, and the traffic plan is obtained according to the matching and screening of the influencing factor index.

[0075] Example 2: In this example, the management department or organizer of the area object has used the plan generation method for a period of time and has collected a historical evaluation data set containing historical initial evaluation values ​​and historical influencing factor indicators; and has obtained the historical traffic data set of the corresponding area from the traffic management bureau or relevant departments. The data may include traffic data on different dates such as historical working days, weekends, and major holidays such as May 1st and National Day. The data can be collected at fixed intervals such as three minutes or one minute, and the historical traffic data set is obtained by arranging them in the order of collection time. Figure 3 As shown in the figure, after extracting the influencing factor indicators from the historical traffic data set, a status assessment model for the corresponding area is established.

[0076] The other method steps in this embodiment are the same as those in the first embodiment, so they are not described in detail. The specific difference lies in the way of obtaining the weight coefficient corresponding to the influencing factor index. Specifically, the corresponding relationship between the influencing factor index and the initial evaluation value is: Obtain historical evaluation data sets, including historical influencing factor indicators and corresponding historical initial evaluation values; The weighted sum of the historical influencing factor index and the corresponding weight coefficient is the corresponding historical initial evaluation value: sc h =100-100*(L1 h *c1+L2 h *c2+L3 h *c3+L4 h *c4+L5 h *5) where sc h is the historical initial evaluation value, L1 h 、L2 h 、L3 h 、L4 h and L5 h The historical influencing factor index corresponding to the historical initial evaluation value; The historical evaluation data set is used for fitting training to obtain the weight coefficients corresponding to each influencing factor indicator.

[0077] It should be noted that in the present invention, after the management departments and organizers of the regional objects have used the plan generation method for a period of time, a large number of historical traffic data sets and historical evaluation data sets have been saved. In this case, fitting training can be directly performed based on the historical evaluation data sets to obtain the weight coefficients of various influencing factor indicators. Compared with the weight coefficients obtained by assigning values ​​in descending order according to the coefficient of variation, they are more accurate and meet the data distribution requirements. And because the data of historical influencing factor indicators and historical initial evaluation values ​​in the historical evaluation data sets will continue to be updated and increased over time, the weight coefficients obtained after fitting training will also be more in line with the actual traffic conditions. The specific method of obtaining weight coefficients based on historical evaluation data sets and calculation relationship training belongs to the prior art, so it will not be described in detail. The sum of the weight coefficients obtained by fitting may not be 1. In this case, the weight coefficients can be normalized first, and the processed results are used as new weight coefficients.

[0078] It is worth noting that the weight coefficients obtained through fitting training can also be corrected in order to obtain more accurate results. The specific correction process is the same as that in Example 1. The evaluation value set is obtained by weighted calculation based on the fitted weight coefficients and influencing factor indicators, and the correction is performed until the data in the evaluation value set meets the distribution standard, thereby obtaining the final weight coefficient.

[0079] The above embodiments are further elaborations and illustrations of the present invention for ease of understanding, and are not limitations of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for generating a traffic plan based on regional status assessment, characterized in that: include: Obtain the historical traffic data set of the regional object, extract the influencing factor indicators, and establish the status assessment model of the corresponding area; The state assessment model includes the corresponding relationship between the influencing factor index and the initial assessment value, and The mapping relationship between the initial evaluation value and the final evaluation value; Obtain the real-time traffic data set of the regional object, extract the influencing factor indicators, and calculate the final evaluation value of the corresponding area through the state evaluation model; Based on the comparison result between the final evaluation value and the evaluation threshold, the corresponding plan library is retrieved, and the traffic plan is obtained by matching and screening the influencing factor indicators.

2. A method for generating a traffic plan based on regional status assessment according to claim 1, characterized in that: In the state assessment model, a corresponding weight coefficient is assigned to each influencing factor indicator; The influencing factor indicators are weighted and summed based on their corresponding weight coefficients to obtain the initial evaluation value; The final evaluation value interval is divided according to the evaluation threshold. For any final evaluation value interval, the corresponding initial evaluation value interval is determined based on the same traffic state, and a mapping relationship is established.

3. A method for generating a traffic plan based on regional status assessment according to claim 1 or 2, characterized in that: In the corresponding relationship between the influencing factor index and the initial evaluation value: Calculate the coefficient of variation of each influencing factor indicator and arrange them in descending order; Assign the corresponding initial weights in descending order of the influencing factor indicators; The evaluation value set is calculated based on the initial weights and influencing factor indicators; The initial weights are modified until the data in the evaluation value set meets the distribution criteria, and the weight coefficients corresponding to the influencing factor indicators are obtained.

4. A method for generating a traffic plan based on regional status assessment according to claim 1 or 2, characterized in that: In the corresponding relationship between the influencing factor index and the initial evaluation value: Obtain historical evaluation data sets, including historical influencing factor indicators and corresponding historical initial evaluation values; The weighted sum of the historical influencing factor indicators and the corresponding weight coefficients is the corresponding historical initial evaluation value. The historical evaluation data set is used for fitting training to obtain the weight coefficients corresponding to the various influencing factor indicators.

5. A method for generating a traffic plan based on regional status assessment according to claim 3, characterized in that: The step of assigning corresponding initial weights in descending order of the influencing factor indicators includes: Calculate the ratio of the standard deviation to the mean of any influencing factor indicator to obtain the coefficient of variation of the influencing factor indicator; Normalize the coefficient of variation of all influencing factor indicators; The normalized results are used as the initial weights corresponding to the influencing factor indicators.

6. A method for generating a traffic plan based on regional status assessment according to claim 1, 2 or 5, characterized in that: The traffic plan obtained by matching and screening the influencing factor indicators includes: Arrange in descending order according to the weight coefficient corresponding to each influencing factor indicator; Selecting a number of plans matching the first influencing factor indicator from the plan library to form a plan set; The plan set matched with the next influencing factor index is selected from the plan set obtained by matching the previous influencing factor index, until the number of plans in the plan set is one, and the final traffic plan is obtained.

7. A method for generating a traffic plan based on regional status assessment according to claim 3, characterized in that: The step of modifying the initial weight until the data in the evaluation value set meets the distribution standard includes: The minimum value in the evaluation value set is less than or equal to the preset minimum value threshold; and The proportion of the number of data in the evaluation value set that is less than the low score threshold to the total number is greater than or equal to the low score ratio threshold; The initial evaluation value interval divided by the low score threshold corresponds to the final evaluation value interval divided by the evaluation threshold.

8. A method for generating a traffic plan based on regional status assessment according to claim 1, 2, 5 or 7, characterized in that: The influencing factor indicators include: Parking lot saturation, which calculates the average ratio of the number of occupied parking lots to the total number of parking lots in the area; Passenger flow saturation, which calculates the average value of the ratio of real-time passenger flow to passenger flow capacity in each sub-area within the area; Traffic saturation, calculate the ratio of the sum of the real-time traffic volume in each sub-area in the area to the traffic capacity.

9. A method for generating a traffic plan based on regional status assessment according to claim 1, 2, 5 or 7, characterized in that: The influencing factor indicators also include: Road congestion mileage ratio: divide the area object from the inside to the outside into core circle, control circle and diversion circle, and calculate the weighted average of the road congestion ratio in the core circle, the road congestion ratio in the control circle and the road congestion ratio in the diversion circle; The road congestion ratio is the ratio of the length of roads in congestion and oversaturation states to the total length of roads.

10. A traffic plan generation system based on regional status assessment, applicable to the traffic plan generation method according to any one of claims 1 to 9, characterized in that: include: The index extraction module extracts the corresponding influencing factor index from the traffic data set of the regional object; An evaluation model module stores a state evaluation model and updates the model according to an updated historical traffic data set; The plan library module stores the plan libraries corresponding to different evaluation thresholds.

Images