CN115907266B - Custom bus route planning method based on passenger flow travel characteristics - Google Patents

Abstract

The invention discloses a customized bus route planning method based on passenger flow travel characteristics, which comprises the following steps: s1, determining the type and coverage area of a customized bus line; s2, selecting a target area and an analysis period; s3, obtaining a high-frequency trip OD pair of the target area; s4, determining passenger flow concentratedly as a bus stop of a customized bus route; s5, determining the custom bus route trend from the starting point region to the end point region. According to the technical flow of selection scene-OD analysis-route planning, aiming at demand analysis, passenger flow research and judgment and point location selection of customized buses, the intelligent planning of customized bus organization modes is realized by combing, designing and developing customized bus high-potential passenger flow mining based on commuting passenger flow OD analysis, current bus track service resource supply evaluation and passenger flow travel service supply and demand matching degree evaluation.

Description

Custom bus route planning method based on passenger flow travel characteristics

Technical Field

The invention relates to the technical field of bus route planning, in particular to a customized bus route planning method based on passenger flow travel characteristics.

Background

Urban traffic precise operation and precise management are the necessary trend of future development, and pushing public transportation system operation management mode change is compliance with new era development requirements driven by data as key elements and cores. The conventional buses are used as capillaries in a public transportation system, the customized buses are used as upgrade versions and beneficial supplements of the conventional buses, and the method is an important means for promoting urban public transportation reform and meeting the ever-increasing beautiful travel demands of people. Custom buses, also known as commercial buses, are one-stop direct buses from cell to cell, from cell to cell. Citizens can put forward own demands through special websites, and public transport groups design public transport lines according to the demands and passenger flow conditions.

However, most of current customized public transport services are actively initiated by passengers, and a public transport group main department collects passenger demands, and a traditional organization mode of 'demand collection-site survey-line demonstration-test operation' of optionally opening lines has the defects of difficulty in obtaining travel demands of vulnerable groups, few demand collection channels, long opening period and the like, and lacks effective means for mining potential passenger flow demands, so that the passenger flow demands cannot be fed back, and customized public transport users are lost.

Of course, with the continuous progress of technology, intelligent bus planning has also been applied to a certain extent, taking a "public bus network planning method" disclosed in chinese patent application No. CN201611209172.5 as an example, this scheme aims at minimizing the travel time and transfer times of passengers and maximizing the demand density of the network, comprehensively considers the benefits of passengers and the operation efficiency of the network, and effectively improves the utilization efficiency of the line by searching the line with the largest passenger flow between OD pairs. The defect that the traditional model is limited to the traveling time of passengers or the direct demand density is avoided, so that the line and the passenger flow are more consistent, and the service level of the public transportation network is improved. However, the expression for the constraint conditions in this technical solution is as follows: in the step S2, the constraint conditions are specifically as follows: 1) The wire nets are communicated; 2) No loop wire; 3) The number of lines in the wire network meets a preset value; 4) The transfer times are not more than n times; 5) The number of the line nodes is smaller than a preset maximum value and larger than a preset minimum value; 6) The number of bus lines arranged on the road section is smaller than a preset value; 7) The wire mesh density is larger than a preset value; 8) The nonlinear coefficient of the line is smaller than a preset value ", obviously, in the technical scheme, all constraint conditions are definite and invariable, and as a constraint condition is always a constant constraint condition, partial high-quality feasible scheme is necessarily sacrificed when influencing factor changes in consideration of the bottom covering property of the constraint condition.

The same situation also occurs in the method of integrated optimization design of pure electric public transportation network disclosed in chinese patent application No. CN202010581155.4, the method of setting constraint conditions is also disclosed in the technical solution, but in such a technical solution, for example, "the constraint conditions include line constraint conditions, charging station constraint conditions, vehicle operation constraint conditions and charging plan constraint conditions", but in the technical solution, the representation that the constraint conditions are variable along with influencing factors does not occur, so that there is still the problem that part of high-quality viable solutions is necessarily sacrificed when influencing factors change due to the bottom of the constraint conditions.

Disclosure of Invention

In order to solve the technical problems mentioned in the background art, a customized bus route planning method based on the passenger flow trip characteristics is provided.

In order to achieve the above purpose, the present invention adopts the following technical scheme:

the customized bus route planning method based on the passenger flow travel characteristics comprises the following steps:

s1, determining a customized bus line type and a coverage area, wherein the customized bus line type comprises a special shopping line, a special commuting line, a special medical line, a special reading line and a special travel line, different line types are configured with different POIs, and different urban travel databases are called to realize classified management and cross calculation of passenger flow sources;

s2, selecting a target area and an analysis period, wherein the target area comprises a starting point area and an end point area which are used as a starting point and an end point of a customized bus route;

s3, aiming at the target area, carrying out OD analysis on at least four influencing factors including transfer times, destination distribution, travel time ratio and walking distance, and sequencing according to OD intensity from high to low to obtain a high-frequency travel OD pair of the target area;

s4, according to the high-frequency trip OD pairs, aiming at a target cell in a target area, carrying out cell OD special analysis on at least three influencing factors including residence/employment population thermodynamic distribution, multidimensional trip thermodynamic distribution and trip characteristic distribution, and determining that passenger flows are intensively used as bus stops of customized bus routes;

s5, determining an optimal bus path between target cells in a target area through a line generation model which minimizes path cost based on the bus stop in the step S4, and finally determining the customized bus line trend from a starting point area to an end point area, wherein at least a plurality of constraint conditions which are mutually related and are suitable for the condition that calculation parameters are variable are set in the line generation model, and the line generation model is as follows:

minZ=∑n+1i=0∑n+1j=0c_ij x_ij ——（1）;

equation (1) represents minimizing path costs, where c represents distance transportation costs between bus stops i, j, x_ij Is a 0-1 variable, x as the vehicle goes from station i to station j_ij =1, otherwise, x_ij =0；

The constraint conditions are as follows:

∑ n+1 j=1，j≠ix_ij =1，"i∈C——（2）；

equation (2) indicates that each station must pass once, wherein, the passenger set C= {1, how much, n };

∑ n+1 i=0，k≠ix_ik =∑ n+1 j=1，k≠ix_kj ，"k∈C——（3）；

equation (3) indicates that each station must be stopped once;

∑n j=1x_0j ≤K——（4）；

equation (4) indicates that the starting station has K arcs at most;

y_i +x_ij +q_i -Q（1-x_ij ）≤y_j ，"i，j∈N——（5）；

q_j ≤y_j ≤Q，"i∈N——（6）;

x_ij ∈{0,1}，"i，j∈N——（7）；

in the formulas (5) and (6), y_i Representing accumulated occupant demand when the vehicle arrives at station i, y_j Indicating accumulated occupant demand when a vehicle arrives at station j, q_j Representing the demand of the occupant j, Q representing the total vehicle capacity, the vertex set n=c {0, n+1}, y_i 、y_j And q_j Are all variable parameters which are adjusted according to actual influencing factors.

In the prior art, for constraint conditions, a constant constraint condition is often adopted, so that partial high-quality feasible schemes are necessarily sacrificed when influencing factor changes in consideration of the trashness of the constraint conditions, therefore, the application is improved according to the weaknesses of the technical schemes, the constraint conditions which are mutually related and are suitable for the condition that the calculation parameters are variable parameters are introduced, and in order to further reduce the excessive influence of the constraint conditions, the constraint conditions also need to have a certain linkage property. The technical scheme overcomes the technical prejudice that constraint conditions are unchanged, greatly improves technical rationality, and can be combined with passenger flow OD time-varying data under different influencing factors to give a practical high-quality feasible scheme.

As a further description of the above technical solution:

the method also comprises the following steps:

s6, predicting the time-varying data of the passenger flow OD in the coverage area of the customized public transportation line in the step S5 based on the Goldmost data, acquiring the passenger flow intensity parameter of each unit time period, and determining the scheduling plan of the customized public transportation;

and S7, in the coverage area of the customized bus, evaluating and comparing traffic modes of the customized bus, the conventional bus and the taxis from a plurality of angles including economy, travel duration, walking distance, travel distance, transfer times and the like.

As a further description of the above technical solution:

in step S3, after the destination distribution is determined as a starting point area, according to a plurality of end point areas obtained by OD analysis, the travel time ratio is a ratio of the longest traffic mode time consumption to the shortest traffic mode time consumption, and the walking distance includes less than 500m, 500m-1km and more than 1 km.

As a further description of the above technical solution:

in step S4, the multidimensional travel thermal distribution includes a full-day travel thermal distribution, an early-peak travel thermal distribution, a late-peak travel thermal distribution, a bus travel thermal distribution and a rail travel thermal distribution, and the travel characteristic distribution includes a travel period duty distribution, a travel mode duty distribution and a travel time characteristic distribution.

As a further description of the above technical solution:

in step S2, the analysis period includes an early peak period, a late peak period, and a full day continuous period.

In summary, due to the adoption of the technical scheme, the beneficial effects of the invention are as follows: the invention takes 'demand guidance and digital driving' as principles, and carries out system function carding, design and development on the customized bus high-potential passenger flow mining on the basis of the commuting passenger flow OD analysis, the existing bus track service resource supply evaluation and the passenger flow travel service supply and demand matching degree evaluation according to the technical flow of 'selection scene-OD analysis-line planning-shift planning-submitting scheme', aiming at the new crossing of the conversion from 'passive response' to 'intelligent planning' of the existing customized bus organization mode, has certain innovativeness and practicability, and can form a new data intelligent driving bus line customization model which can be popularized nationally according to the technical flow of 'selection scene-OD analysis-line planning-shift planning-submitting scheme' for the customized bus.

Detailed Description

The technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Example 1

The embodiment provides a technical scheme: the customized bus route planning method based on the passenger flow travel characteristics comprises the following steps:

s1, selecting a scene: determining the types and coverage areas of customized public transportation lines, wherein the types of customized public transportation lines comprise shopping special lines, commuting special lines, medical special lines, reading special lines and travel special lines, the shopping special lines mainly affect the association analysis range configured with urban POIs, the shopping special lines mainly cover residential communities and large business circles, the commuting special lines mainly cover residential communities and industrial park post concentration areas, the medical special lines mainly cover three hospitals, hubs, residential communities and the like, the reading special lines mainly cover primary schools and residential communities, the travel special lines mainly cover hot spots, business circles, residential communities and hubs, the different types of lines are configured with different POIs, different urban travel databases are called, classification management and cross calculation of the sources of the passenger flows are realized, the specific user self-defines the lines, and the POIs are configured simultaneously, such as: schools, hospitals and the like, through identifying the passenger flow of each area through the internet trip data, the area occupancy data and the like, and distinguishing by combining with POI attributes, such as: taking a living area as a starting point and taking a office building as an ending point, and simultaneously carrying out cross analysis on the passenger flow of each area of the commute special line by combining the working day with the non-working day to obtain the passenger flow of each characteristic period, such as the working day and the non-working day;

s2, selecting a target area and an analysis period, setting the target area as a starting point or an end point, and performing primary locking of a main research area with a traffic area as a target to realize primary frame selection of a city database, so that data analysis efficiency is improved, and simultaneously, anchoring different analysis periods, such as early peak period, late peak period, full-day continuous period and the like, by a self-defined set analysis period to perform highly self-defined basic target selection definition;

s3, aiming at a target area, carrying out OD analysis from four aspects of transfer times, destination distribution, travel time ratio and walking distance, sequencing according to OD intensity from high to low, and obtaining high-frequency travel OD pairs of the target area, specifically, sequencing according to OD intensity from high to low by a list, manually selecting a passenger flow with high passenger flow intensity, and generally, selecting an OD pair of a top 10;

the transfer times are important indexes reflecting the transfer strength of the current city traveler in the travel chain, are important indexes reflecting the current travel level, directly reflect the travel service level and the experience level, are classified according to the non-transfer, one-time, multiple-time and all transfer times by combining with the basic rule of city travel, realize the dynamic synchronous response of the transfer times and the intensity of the passenger flow OD, realize the dynamic response connection between the transfer times and the passenger flow OD, and better assist more targeted customized analysis;

after the destination distribution is a fixed starting point area, a plurality of end point areas are obtained according to the OD analysis, for example, passenger flows from the area A respectively reach five B, C, D, E, F areas, and the destination distribution is the five areas;

the travel time ratio reflects the time degree that a traveler spends adopting different traffic modes at the starting point and the ending point, is the ratio of the longest traffic mode time consumption to the shortest traffic mode time consumption, reflects the path advantage and the channel advantage among different traffic modes, is the centralized representation of the comprehensive travel efficiency of the different traffic modes, is the core index for measuring the advantages and disadvantages of customized public transport and other traffic modes, and is obtained through the aggregation analysis of the whole sample elements;

the walking distance is an important index reflecting the connection level of urban travelers and traffic infrastructures such as parking lots, rail stations and bus stations, is an important index reflecting the current trip level, directly reflects the overall layout balance and supply level of facilities, is the most important ring in construction of travel chains, and is classified and analyzed according to three categories of 500m or less, 500m-1km or more and 1km or more by combining the overall layout distribution of the facilities, so that dynamic response connection is realized between transfer times and passenger flow OD, and more targeted customized analysis can be better assisted;

s4, aiming at a high-frequency trip OD pair, aiming at a target cell in a target area, carrying out cell OD pair special analysis from three aspects of residence/employment population thermodynamic distribution, multidimensional trip thermodynamic distribution and trip characteristic distribution, determining that passenger flows are concentrated to serve as preselected bus stops, specifically, carrying out residence/work population thermodynamic distribution display on the target area based on a gis map, and taking the position with large passenger flows in the concentrated condition of the distribution of the user-integrated bus stops and the passenger flows as a custom bus stop;

aiming at population thermodynamic distribution of living and employment, the population thermodynamic distribution of the cell level of a starting and ending region is mainly researched, quantitative support analysis is provided for determination of bus stops, population thermodynamic refined distribution of living and employment of the starting and ending region is displayed, and the bus route stops are directly guided to be customized to be distributed around a high thermodynamic distribution region;

the multidimensional travel thermodynamic distribution comprises full-time travel thermodynamic distribution, early-peak travel thermodynamic distribution, late-peak travel thermodynamic distribution, public transport travel thermodynamic distribution and rail travel thermodynamic distribution, is mainly characterized in that the district-level travel intensity thermodynamic distribution of a starting and ending region is researched, the living and employment distribution is converted in a mode, and the distribution is matched into a travel chain through a high-German big data algorithm, so that different types of travel thermodynamic distribution of full-time travel, early-late-peak travel, public transport travel and rail travel are realized, population thermodynamic refinement distribution of various travel types of the starting and ending region is displayed, and the customization of bus route stations is more directly guided to be distributed around the high thermodynamic distribution region;

the travel characteristic distribution comprises travel time period proportion distribution, travel mode proportion distribution and travel time characteristic distribution, analysis is carried out by using the accumulated data of the current month, long-term change rule parameters are extracted, the travel time period proportion is used for describing the analogy proportion of travel time consumption, the integral travel time consumption level is reflected, the travel mode proportion is used for reflecting the proportion of different travel modes, the weight of customized bus high-potential passenger flow subdivision categories is reflected, the travel time characteristic distribution displays the accumulated value of the time period of the whole month, and the frequency of time period change is further reflected;

specifically, the OD analysis is a core functional module for mining high-potential passenger flows, and based on a Gao Deduo source big data base, the multi-element unified analysis of urban data and traffic travel data is carried out, the cross comprehensive comparison analysis is carried out around population and travel two levels from four aspects of transfer times, destination distribution, travel time ratio and travel distance, and from various dimensions of living population, employment population, whole-day travel heat, early-peak travel heat, late-peak travel heat, bus travel heat, track travel heat and the like, the urban travel OD distribution rule and space-time two-dimensional characteristics are accurately grasped and comprehensively depicted, wherein the special analysis of cell OD pair is carried out for four aspects of transfer times, destination distribution, travel time ratio and travel distance, and after comprehensive research and judgment are carried out to obtain high-frequency travel OD pairs, the core OD analysis link for customizing bus line planning is carried out;

s5, determining an optimal bus path between target cells in a target area through a line generation model which minimizes path cost based on the bus stop in the step S4, and finally determining a customized bus line trend from a starting point area to an end point area, wherein at least a plurality of constraint conditions which are mutually related and are suitable for the condition that calculation parameters are variable parameters are set in the line generation model, and the line generation model is as follows:

minZ=∑n+1i=0∑n+1j=0c_ij x_ij ——（1）;

equation (1) represents minimizing path costs, where c represents distance transportation costs between bus stops i, j, x_ij Is a 0-1 variable, x as the vehicle goes from station i to station j_ij =1, otherwise, x_ij =0；

The constraint conditions are as follows:

∑ n+1 j=1，j≠ix_ij =1，"i∈C——（2）；

equation (2) indicates that each station must pass once, wherein, the passenger set C= {1, how much, n };

∑ n+1 i=0，k≠ix_ik =∑ n+1 j=1，k≠ix_kj ，"k∈C——（3）；

equation (3) indicates that each station must be stopped once;

∑n j=1x_0j ≤K——（4）；

equation (4) indicates that the starting station has K arcs at most;

y_i +x_ij +q_i -Q（1-x_ij ）≤y_j ，"i，j∈N——（5）；

q_j ≤y_j ≤Q，"i∈N——（6）;

x_ij ∈{0,1}，"i，j∈N——（7）；

in the formulas (5) and (6), y_i Indicating the accumulated passengers when the vehicle arrives at station iMember demand, y_j Indicating accumulated occupant demand when a vehicle arrives at station j, q_j Representing the demand of the occupant j, Q representing the total vehicle capacity, the vertex set n=c {0, n+1}, y_i 、y_j And q_j And (3) carrying out specific route trend layout on the customized bus route layout target cells according to the analysis result of the step (S4), mainly selecting stations covered by high passenger flows at the starting and ending cells for route layout through the station passenger flow coverage ranges of different service radiuses, and automatically determining the customized bus route trend through reasonable route planning by utilizing the Goldmap base map.

Specifically, in the application, the influence factors are not only in the use of OD pairs, for example, at least four influence factors including transfer times, destination distribution, travel time ratio and walking distance, but also influence constraint conditions, so that consideration and use of various influence factors are more comprehensive, and the design of the scheme is more in line with the actual situation.

In this application, when a route generation model is built to minimize the path cost, taking the demands of passengers as an example, part of constraint conditions can be changed, and the model constraint conditions can be used according to the change of parameters to obtain different calculation results, for example, the demands of passengers are adjustable and variable according to travel time, seasons, travel will and the like.

The variable parameters can be parameters which change according to environmental influence factors including time, holidays, seasons and temperature, different route schemes can be formed when the parameters are calculated through different constraints, and according to the different route schemes, the lines which better and more meet actual demands can be obtained by combining manual selection.

As in the present embodiment, a simple example is set as follows: y is_j =（Ka_i T_i + Kb_i S_i + KC_i D_i ）y_ej ，Ka_i +Kb_i + KC_i =1, in the above formula, y_ej In order to calculate parameter values of accumulated passenger demands when arriving at station i, in a general technical scheme, y_ej Often equal to y_j However, in this embodiment, y_j The weighting conversion is performed on the basis of the original data so as to meet the requirement that the constraint condition changes according to the environmental influence transformation, in the embodiment, T_i To calculate the temperature in the influencing factors (when reaching site i), S_i To calculate the value of the season in the influencing factors (when reaching site i), D_i To calculate the holiday value (when reaching site i) of the influencing factors, the influencing factors are converted into the required values through calculation, and the corresponding Ka_i 、Kb_i And KC_i In order to influence the weight value corresponding to the factor, the weight value is set manually and adjusted, and the constraint condition can be more in line with the actual situation through reasonably adjusting the big data result before fitting. The corresponding weighting value may also be increased as influencing factors are added. As for T_i The values can be obtained by selecting the corresponding results by using a conversion function, for example, the calculated temperature value corresponding to 20 ℃ of the daily average air temperature is 0.7, and the calculated temperature value corresponding to 32 ℃ of the daily average air temperature is 0.3, namely, when the calculated temperature is comfortable, the travel requirement of the passengers on the bus is far greater than that of the bus in high-temperature weather, and S is as follows_i The numerical value can also be obtained by adopting a segmented corresponding result mode to correspond, for example, the season calculated value in summer is 0.8, and the season calculated value in winter is 0.6, namely, the time length in summer is longer, the total travel requirement of passengers on the bus is greater than that in winter, and D_i The holiday is not holiday 0-1, which is shown as busy in some areas and opposite in some areas. Through the steps, y_j Compared with the original y_ej With reasonable variation, the source of the related data is the statistical data accumulated over the years.

In order to further reduce the excessive influence of the constraint conditions, each constraint condition needs to have a certain linkage property, for example, the constraint conditions that each station passes once and must stop once can be adjusted according to the demands of passengers, and whether the constraint conditions are applicable or not needs to comprehensively consider various influence factors, but is not invariable.

The method overcomes the technical prejudice that constraint conditions are unchanged, greatly improves technical rationality, and can be combined with the time-varying data of the passenger flow OD under different influencing factors to give a practical high-quality feasible scheme.

According to the above technical solution, it is assumed that the target area A, B is the starting point and the ending point of the customized bus route, respectively, where the target area a has target cells X1, X2, and X3 in sequence, the target area B has target cells Y1, Y2, and Y3 in sequence, and a unique optimal path L exists between the target area a and the target area B_A-B ，L_A-B I.e. the only route between the target cells X3, Y1, further, according to the above-mentioned customized bus route planning method, the preselected bus stops of the target cells X1, X2, X3, Y1, Y2, Y3 are determined as shown in the following table 1, and the optimal path between the adjacent target cells is determined as shown in the following table 2:

TABLE 1

TABLE 2

Then it can be seen from table 2 that the optimal path of the customized bus route is: "L_X1-X2 （x11-x23）——L_X2-X3 ：（x22-x31）——L_A-B （x31-y13）——L_Y1-Y2 （y13-y22）——L_Y1-Y2 （y31-y32）”。

Further, in step S5, the following alternatives are also included: taking the bus station determined in the step S4 as an anchor point, taking the peripheral 300m, 500m and 800m as different coverage radiuses as a passenger flow gathering ring, and obtaining road trend conditions of a starting and ending point area and urban land block distribution conditions of peripheral cells based on a data engine of a Goldmap to carry out line trend arrangement, wherein the cells have corresponding POI points, mainly see whether the positions are concentrated or not, and have the value of setting up the bus station or not, for example, parks are not suitable for setting up commutes, wherein the passenger flow gathering ring is the passenger flow distribution condition of the starting and ending point, namely living/working population distribution and passenger flow distribution condition in the step S4;

s6, scheduling planning is to carry out capacity space-time configuration on a customized bus line, passenger flow OD time-varying data of a primary line coverage area is obtained based on Gooded big data, passenger flow intensity parameters of each unit time period are obtained, the passenger flow planning is an important auxiliary reference for scheduling, multi-dimensional calling inquiry of a historical time period can be carried out through big data accumulation and different characteristic differences such as seasonal variation, so that more comprehensive, fine and accurate scheduling planning is carried out, the passenger flow planning is an important data analysis achievement foundation of actual operation of the line, specifically, passenger flow in each time period is output according to the time period of each hour, customized buses do not operate all day, only planning operation is carried out in the time period of high-frequency travel, and scheduling is carried out by acquiring the passenger flow of each hour and combining actual conditions by a user;

if after the scheme of customizing the bus route is planned, automatically extracting the passenger flow covered by the periphery of the route, performing preliminary fitting according to unit hour, sequencing according to the intensity of the passenger flow, and setting the passenger flow in a high-intensity period to the top, wherein the route is found to be typical commuter passenger flow distribution characteristics, the passenger flow intensity in the early peak period is very concentrated, and the passenger flow intensity in the late peak is also most concentrated from the aspect of reverse analysis, and belongs to barbell type passenger flow distribution with obvious characteristics;

s7, in the custom bus coverage, from the angles of economy, travel duration, walking distance, travel distance and transfer times, the traffic modes of custom buses, conventional buses and taxis are evaluated and compared, the final ring of custom bus line layout is the summarization and display of the overall situation of planning custom bus lines, the centralized expression of relevant basic parameters of the custom bus lines is the comprehensive comparison evaluation of the custom bus lines and other traffic modes, and the important technical means for evaluating the utility and the quality of the custom bus lines are realized, so that the comprehensive comparison evaluation of the custom bus lines in the urban comprehensive traffic and transportation system based on high-grade big data is realized.

Further, as shown in the following table 3, the traffic between the destination target area A, B and the target cells X1, X2, X3, Y1, Y2, Y3 covered by the destination target area A, B is comprehensively evaluated by customizing three traffic modes of buses, conventional buses and taxis:

TABLE 3 Table 3

As can be seen by way of example from table 3, the following conclusions are drawn:

(1) The transit time of the customized bus is about 47.18 minutes, compared with the transit time of the conventional bus, which is 101.17 minutes, the customized bus has certain advantage in trip efficiency;

(2) The customized bus travel distance is about 28.74 km, the conventional bus travel distance is 35.28 km, and the shorter customized bus travel distance can provide faster travel efficiency;

(3) The customized bus is not required to be transferred along the way, and the conventional bus is required to be transferred for 1 time, so that compared with the customized bus travel experience is better;

(4) The whole-course walking distance of the customized bus is about 0.00 meter, the walking distance of the conventional bus is about 2476.00 meters, and the customized bus is more friendly;

(5) The customized bus is expected to cost 12 yuan, 84.00% cheaper than the bus driving, and the customized bus is more economical and preferential.

The method takes 'demand guidance and digital driving' as principles, and carries out system function carding, design and development on customized bus high-potential passenger flow mining on the basis of commuting passenger flow OD analysis, current bus track service resource supply evaluation and passenger flow travel service supply and demand matching degree evaluation by carrying out deep communication and flow restoration with a bus operation department and a bus management department according to the technical flow of 'selection scene-OD analysis-line planning-scheduling planning-submitting scheme', and aiming at the demand analysis, passenger flow research and point selection of customized buses. The research results penetrate through each link of the planning design, the operation management of the customized public transportation system, the customized public transportation operation service level can be effectively improved, the new crossing of the existing customized public transportation organization mode from passive response to intelligent planning is realized, certain innovation and practicability are realized, and a new generalized form of intelligent data-driven public transportation line customization can be formed nationally.

First, unlike traditional buses: the main difference between the custom public transport and the traditional public transport is the accurate response to the passenger demands, and the starting and ending point and the shift time of the line are customized completely according to the passenger flow scale and the travel demands. Therefore, unlike the universality and fixity (fixed line, fixed station and fixed shift moment) of the traditional buses, the customized buses are more flexible and autonomous, and can meet the individual demands of passengers. In the aspect of popularization and operation, compared with the traditional buses, the popularization and operation of customized buses depend on the mouth-to-mouth transmission among users, and the user experience is more concerned, so that higher-quality services are won. Compared with the traditional bus, the customized bus has the following characteristics:

(1) The circuit design is more flexible: the traditional bus route has limited service area in the operation of fare determination, route determination and stop location, and has high requirements on setting of the stops along the way and stopping at the first and last stops. In addition, the number of stop stations of the customized buses is small, namely, direct buses with few or no intermediate stations along the way from a fixed starting point to a fixed ending point are provided, the advantages of good route direct property are achieved, the service coverage area of the public transportation can be effectively increased, and convenient and efficient services can be provided for more citizens;

(2) The operation service is more efficient: by collecting the travel requirements of passengers with similar travel starting and ending points, the customized buses can realize the starting and ending point direct transport service, and the bus priority channels are authorized to be used, so that the interleaving conflict with social vehicles is reduced, and the operation efficiency of the bus is higher than that of the conventional buses, subways and other traffic modes needing to stop for multiple times for getting on and off;

(3) Travel demand is more accurate: customized buses often utilize an Internet platform, meet accurate demands of passengers, collect, arrange and analyze the travel demands of the passengers, and the service scene relates to commute, general school, tour, scenic spot and the like, so that the accurate demands of point-to-point direct arrival of the passengers are met.

Secondly, the conventional mode: if citizens subscribe to purchase tickets in advance for 10 minutes through the Beijing customized public transport APP, the customized public transport platform selects the direction, the area and the time period in the travel concentration according to the subscription conditions on the citizens line in the surrounding area, and the directional travel service is designed. If the line meeting the self demand is not available, line customization can be initiated, and the bus is organized to provide customization service for shifts with the size of the opening person. And the public transport group can collect background data regularly, pinch the contract demand and carry out route evaluation planning. Generally, the number of passengers reaches 50% of the number of seats in the vehicle, and the line is open. Taking the most common commute shift as an example, the commute shift generally uses 54 vehicles, and the number of passengers reaches 27, so that the commute shift can be operated;

while in this embodiment: based on urban traffic travel data, a periodic dynamic accumulated data observation analysis method is adopted, high-frequency travel demands are extracted according to multi-element dimensions affecting travel quality, and the inner and outer traffic travel pain points of an important research area are found out. Aiming at the trip pain point, the method fully combines various comprehensive transportation advantages of intensive, high-quality, rapid transportation and the like of the customized buses, organically matches the customized buses with the urban trip pain point through big data association analysis mainly comprising urban POIs, and finds out a brand new solution. According to the planning trend of the customized public transportation line, the space-time two-dimensional distribution characteristics of urban travel passenger flows are combined, the highly intelligent customized public transportation service of dynamic scheduling, dynamic connection and dynamic response is realized, the comprehensive service capacity of the urban public transportation system is comprehensively improved, a more multi-element, efficient and layered urban public transportation system is created, high-quality travel is served, and high-quality life is boosted;

compared with the traditional mode, the method and the system for the public transportation are capable of changing the problem of passive discovery reported by citizens into the travel area of each area actively excavated by a management department, guiding public transportation investment and improving the willingness of citizens to travel in public. Meanwhile, the demand analysis, passenger flow research and judgment and point location selection of the customized buses are performed through deep communication and flow restoration with a bus operation department and a bus management department, and the method is based on the basis of commuting passenger flow OD analysis, current bus track service resource supply evaluation and passenger flow travel service supply and demand matching degree evaluation; the bus route can be planned and customized more comprehensively.

The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.

Claims (8)

1. The customized bus route planning method based on the passenger flow travel characteristics is characterized by comprising the following steps of:

s1, determining the type and coverage area of a customized bus line;

s2, selecting a target area and an analysis period;

s3, obtaining a high-frequency trip OD pair of the target area;

s4, performing cell OD pair special analysis aiming at a target cell in a target area according to the high-frequency trip OD pair, and determining a bus station of a customized bus route;

s5, determining an optimal bus path among target cells in a target area through a route generation model with minimized path cost, and determining the trend of a customized bus route, wherein at least a plurality of constraint conditions which are mutually related and are suitable for the condition that calculation parameters are variable parameters are set in the route generation model;

the line generation model is as follows:

；

equation (1) represents minimizing path cost, where c_ij Representing the distance transportation cost between bus stops i and j, x_ij Is a 0-1 variable, x as the vehicle goes from station i to station j_ij =1, otherwise, x_ij =0；

The constraint of formula (1) includes:

；

in the formulas (5) and (6), y_i Representing accumulated occupant demand when the vehicle arrives at station i, y_j Indicating accumulated occupant demand when a vehicle arrives at station j, q_j Representing the demand of the occupant j, Q representing the total vehicle capacity, the vertex set n=c {0, n+1}, y_i 、y_j And q_j All are variable parameters which are adjusted according to actual influencing factors;

y_j =（Ka_i T_i + Kb_i S_i + Kc_i D_i ）y_ej ，Ka_i +Kb_i + Kc_i =1；

in the above, y_ej To calculate the parameter value, T, for the accumulated occupant demand when arriving at station i_i To calculate the temperature in the influencing factors when arriving at site i, S_i To calculate the seasonal value, D, in the influencing factors when arriving at site i_i To calculate whether or not holiday in the influence factors when arriving at site i, ka_i 、Kb_i And Kc_i The weighting value is set manually for the weighting value corresponding to the influence factor;

s6, determining a scheduling plan of the customized bus according to the time-varying data of the passenger flow OD in the coverage area of the customized bus line.

2. The customized bus route planning method based on the characteristics of the passenger flow trip according to claim 1, wherein the constraint condition of formula (1) comprises:

；

equation (2) indicates that each station must pass once, wherein, the passenger set C= {1, and n.

3. The customized bus route planning method based on the characteristics of the passenger flow trip according to claim 2, wherein the constraint condition of formula (1) further comprises:

；

equation (3) indicates that each station must be stopped once it passes.

4. A customized bus route planning method based on passenger flow travel characteristics as set forth in claim 3, wherein the constraint of formula (1) further comprises:

；

equation (4) indicates that the starting station has at most K arcs.

5. The customized bus route planning method based on the characteristics of the passenger flow trip according to claim 1, wherein the target area comprises a start area and an end area, and the analysis period comprises an early peak period, a late peak period and a full day continuous period.

6. The customized bus route planning method based on passenger flow travel characteristics according to claim 5, wherein for the target area, OD analysis is performed from at least four influencing factors including transfer times, destination distribution, travel time ratio and walking distance, and the high-frequency travel OD pairs of the target area are obtained by sorting from high to low according to OD intensity.

7. The customized bus route planning method based on passenger flow travel characteristics according to claim 6, wherein the destination distribution is a plurality of destination areas obtained according to an OD analysis after determining a start area, and the travel time ratio is a ratio of the longest traffic mode time consumption to the shortest traffic mode time consumption.

8. The customized bus route planning method based on the passenger flow travel characteristics according to claim 1, wherein the specific analysis of the cell OD is performed from at least three influencing factors including residence/employment population thermodynamic distribution, multidimensional travel thermodynamic distribution and travel characteristic distribution for a target cell in a target area according to the high-frequency travel OD pair, and it is determined that passenger flows are intensively used as bus stops of the customized bus route.