CN112085349A - Demand response bus dispatching method based on passenger travel time window constraint - Google Patents

Demand response bus dispatching method based on passenger travel time window constraint Download PDF

Info

Publication number
CN112085349A
CN112085349A CN202010838030.5A CN202010838030A CN112085349A CN 112085349 A CN112085349 A CN 112085349A CN 202010838030 A CN202010838030 A CN 202010838030A CN 112085349 A CN112085349 A CN 112085349A
Authority
CN
China
Prior art keywords
bus
demand
point
passenger
candidate
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010838030.5A
Other languages
Chinese (zh)
Other versions
CN112085349B (en
Inventor
李欣
王天奇
李炎皓
罗越
徐伟汉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing Eryu Technology Co ltd
Dalian Maritime University
Original Assignee
Chongqing Eryu Technology Co ltd
Dalian Maritime University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing Eryu Technology Co ltd, Dalian Maritime University filed Critical Chongqing Eryu Technology Co ltd
Priority to CN202010838030.5A priority Critical patent/CN112085349B/en
Publication of CN112085349A publication Critical patent/CN112085349A/en
Application granted granted Critical
Publication of CN112085349B publication Critical patent/CN112085349B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses a demand response bus dispatching method based on passenger trip time window constraint, which comprises the steps of firstly obtaining passenger demand and bus running lines and initializing parameters, adopting an ant colony algorithm to sequence and serially connect demand points of passenger trip according to time windows, distributing the demand points to corresponding bus running lines, selecting bus stop candidate points based on the passenger demand, calculating a predicted service time window, taking the minimum error value between the predicted service time window and the time window of the passenger trip demand, solving a final target, and finally obtaining a more practical bus transfer timetable. And finishing the real-time scheduling of the demand response connection bus system.

Description

Demand response bus dispatching method based on passenger travel time window constraint
Technical Field
The invention relates to the technical field of demand response type bus scheduling, in particular to a demand response bus scheduling method based on passenger travel time window constraint.
Background
At present, demand response type connection buses are used as a personalized and flexible bus operation form and mainly serve between a passenger trip starting point and an urban main transportation line of a fixed line. The passengers give travel preference including travel time windows and getting-on and getting-off places through a mode of making a telephone reservation in advance or initiating a bus taking application by a network terminal, the dispatching center carries out optimal operation line design according to the travel preference, the expected arrival time of the vehicles is informed to the passengers, and the transfer bus is dispatched to send the passengers to a main line traffic transfer station from a boarding point, so that transfer service is completed.
However, in the prior art, the complex real-time bus scheduling design method considering the travel preference of the passengers has the following defects:
1. the existing public transportation system design method is only limited to a macroscopic level and is not focused on a microscopic scheduling level; based on the assumed passenger distribution condition, the system design steps are simplified by homogenizing the passenger requirements, so that the individual passenger requirements are often ignored, and the individual trip time windows of the passengers cannot be met;
2. the existing public transportation network design considering passenger time windows is usually based on the assumption that a given vehicle visits stations and road sections, and because the passenger is not reasonably guided, the demand response public transportation cannot select station positions, all stations need to be traversed, and the process from an initial station to a terminal station consumes long time;
3. the existing demand response connection bus dispatching algorithm cannot achieve real-time efficient solution due to the limitation of scale, the traditional method for distributing the buses to the shortest distance or the shortest travel time often breaks up the access sequence of the demand points of the passengers, the bus traveling repetition rate is high, and the resource distribution is unreasonable.
Therefore, how to provide a demand response type bus dispatching method which can meet the individual travel time window demand of passengers and is efficient and reliable is a problem which needs to be solved urgently by technical personnel in the field.
Disclosure of Invention
In view of the above, the invention provides a demand response bus scheduling method based on passenger trip time window constraints, which efficiently and reliably realizes the scheduling task of a demand response bus under the condition of meeting the passenger time window constraints, and solves the problems that the existing demand response bus scheduling method cannot meet the individual trip time window requirements of passengers, the time consumption of the process from an origin station to a destination station is long, the bus traveling repetition rate is high, and the resource allocation is unreasonable.
In order to achieve the purpose, the invention adopts the following technical scheme:
a demand response bus dispatching method based on passenger travel time window constraint comprises the following steps:
step 1: acquiring passenger trip demand data and position information of a demand response type bus starting station and a transfer pivot point, constructing a total objective function, and initializing parameters in the total objective function;
step 2: sequencing passenger travel demand points in passenger travel demand data in sequence according to a time window and connecting the passenger travel demand points in series by taking a demand response type bus starting station as a starting point and taking a transfer pivot point as a terminal point to obtain a plurality of candidate paths;
and step 3: acquiring a bus stop candidate set corresponding to a passenger travel demand point on each candidate path, selecting an optimal bus stop from the bus stop candidate set, and distributing the passenger travel demand point to the optimal bus stop to generate a bus operation line;
and 4, step 4: and acquiring an optimal riding time window corresponding to each passenger trip demand point on the bus operation line, and generating a demand response type connection bus schedule.
Further, the total objective function is specifically:
Figure BDA0002640404100000031
in the formula, U is a total objective function, wherein D is a set of bus stations, namely a set of starting points of demand response connection bus routes, and M is a set of candidate bus stop stations; MS is a set of transfer pivot points, namely path end points;
Figure BDA0002640404100000032
and the variable is 0 or 1, the value of the variable is 1 when the candidate stop point j of the demand response bus passes through the stop point m on the path k, otherwise, the value of the variable is 0 if the candidate stop point m does not pass through. t is tjmThe driving time from the candidate stop point j of the demand response bus to the candidate stop point m is taken for the vehicle;
Figure BDA0002640404100000033
for the time on the path k expected to travel away from the bus stop candidate point j,
Figure BDA0002640404100000034
the time predicted to reach the bus stop candidate point j on the path k is obtained; diNumber of passengers at demand point i, dijIs the walking distance, x, from the passenger from the demand point i to the candidate bus stop jijAnd the value is 0 or 1 variable, when the passenger i is distributed to the bus stop candidate point j, the value is 1, otherwise, the value is 0. WsAverage walking speed for the passenger;
Figure BDA0002640404100000035
Figure BDA0002640404100000036
and
Figure BDA0002640404100000037
the upper and lower boundaries of the deviation value of the passenger preference time window and the bus running schedule are respectively.
The sum of the first two items in the total objective function expression is the total operation time of the demand response bus, and comprises two parts, namely the driving time and the stop time, the third item is the total walking time of passengers, and the fourth item is the total deviation value of the predicted service time window and the passenger preference time window at each bus stop.
In step 1, the initialized parameters include the total number of ants N in the ant colony, the total path amount R, and the maximum iteration number itermaxPath pheromone factor tau between any pair of demand points i and demand response bus stop candidate points jijPheromone variation amount delta tauij(ii) a Separately initializing path pheromone factors tauij(0) And corresponding pheromone volatilization factor delta tauij(0) And is 0, the current iteration number is set to 0.
Further, in step 2, obtaining a plurality of candidate paths by using an ant colony algorithm specifically includes:
step 2.1: and sequencing the passenger travel demand points according to the intermediate values of the upper bound and the lower bound of the passenger travel time window in the passenger travel demand data, and placing the N ant individuals at the initial sequence point of the passenger travel time window.
Step 2.2: generating an ant colony candidate path set allowed specifically as follows:
Figure BDA0002640404100000041
wherein i is a passenger demand point of a selective demand response type connection bus trip, K is a set of all paths, psi (K) is a next candidate station adjacent to the demand point i on the path K, and eψ(k)And lψ(k)Upper and lower bounds, t, of the time window corresponding to the candidate site, respectivelyψ(k)iRepresenting travel time between the candidate site and the demand point i, eiAnd liThe upper bound and the lower bound of the time window corresponding to the demand point i are respectively.
Step 2.3: and determining the transfer path of the ant individual based on the candidate path set allowed.
Step 2.4: and returning to the step 2.2 until all the passenger travel demand points are distributed on a route from the bus starting station to the transfer pivot point to obtain a plurality of candidate routes.
Further, in the step 2.3, the transfer path of the ant individual is determined by adopting a pseudo-random proportion rule, which specifically comprises:
taking pseudo-random number as q and preset parameter as q0；
If q is less than or equal to q0Then the ant individual selects the order parameter tauiψ(k)(t)[ηiψ(k)]βThe kth path with the maximum value is used as the transfer path of the next step;
if q > q0Then calculate a probability value
Figure BDA0002640404100000042
The ant individual selects the kth path with the maximum probability value in the path set as a transfer path of the next step;
wherein m belongs to MS; q. q.s0Is a preset parameter, q0∈(0，1]；ηiψ(k)A heuristic factor between a demand point i and a next point psi (k) on the k-th path; tau isiψ(k)And (t) is a path pheromone factor between a next station next to the demand point i on the kth path at the moment t and the demand point i, alpha is a pheromone relative influence degree factor, and beta is a heuristic factor relative influence degree factor.
Further, the heuristic factor and the path pheromone factor have the following relationship:
ηiψ(k)＝1/(tψ(k)i+0.01)
further, the step 3 specifically includes:
step 3.1: selecting a transfer pivot point m as an end point of a candidate path k by adopting a backtracking method, and constructing a time cost objective function f according to the sum of the bus travel time between two stops and the walking time of passengers walking to the stop candidate point from the end pointk(i)；
Figure BDA0002640404100000051
Wherein the content of the first and second substances,
Figure BDA0002640404100000052
represents the vehicle running time s from the current bus stop candidate point j to the bus stop candidate point corresponding to the next demand point i +1ijRepresents the distance from the demand point i to the candidate point j of the bus stop, WsRefers to the average walking speed of the passengers.
Step 3.2: setting the initial value of the objective function corresponding to the trip demand point i of the passenger to be solved as 0, namely fk(i) Generating a candidate point set J of the bus stop station corresponding to the trip demand point i of the passenger to be solved on the candidate path k as 0k(i)。
Step 3.3: based on the candidate point set J of the bus stopk(i)And selecting the site of the bus stop and selecting the best bus stop.
Step 3.4: and distributing all the passenger travel demand points to the corresponding optimal bus stop points to generate a bus operation line.
In the step 3, the candidate point set J of the bus stop corresponding to the current demand point ik(i)And finding a station which can make the travel time shortest as an optimal bus stop site selection result on the k-th path until the k-th path is traced back to the starting point, generating an optimal vehicle driving route k, and replacing the previous path.
Further, the step 4 specifically includes:
step 4.1: calculating the predicted arrival time of the bus at the destination transfer pivot point through one of the bus lines, e.g. calculating the predicted arrival time of the k-th bus line at the transfer pivot point m
Figure BDA0002640404100000053
The calculation formula is as follows:
Figure BDA0002640404100000054
wherein the content of the first and second substances,
Figure BDA0002640404100000061
for estimated time of arrival, PconFor transferring the time of departure of the main traffic at pivot point m, TconThe minimum time difference allowed between the arrival of the bus at the transfer pivot point m and the departure of the bus at the transfer pivot point m is responded to by the demand.
Step 4.2: calculating the predicted departure time of the bus at the candidate point j of the bus stop, wherein the time is
Figure BDA0002640404100000062
Wherein, tjmAnd (4) selecting the bus running time from the optimal bus stop to the transfer pivot point m in the step (3).
Step 4.3: generating a vehicle predicted arrival time interval [ l ] of each optimal bus stop according to the predicted arrival time and the predicted departure timei-TDmax，li+TDmax]；
Wherein liPassenger travel time window upper bound, TD, for passenger travel demand point i served by optimal bus stopmaxThe maximum difference allowed between the time expected to leave the bus stop candidate point j for the bus and the upper bound of the passenger travel time window.
Step 4.4: and traversing the predicted arrival time intervals of all the buses at the optimal bus stop, and taking the result closest to the time window preferred by the passenger as the final distribution result.
Step 4.5: and if the passenger preference time windows of all the passenger trip demand points are met, forming a demand response type connection bus schedule, and otherwise, repeating the step 4.3.
Because the demand response type bus transfer schedule, passenger route selection and bus stop site selection scheme obtained according to the steps can achieve ideal bus scheduling work, but cannot ensure optimal scheduling result, so that multiple iterative computations are required to find the optimal scheme, the step 4 can further comprise:
step 4.6: and determining the volatilization amount of the pheromone on each bus operation line.
Step 4.7: and updating the value of the total objective function and the pheromone factor on the local path based on the pheromone volatilization amount.
Step 4.8: and judging whether the value of the current total objective function is the minimum value of all the values of the total objective function, if not, returning to the step 2.2 in the step 2, and if so, carrying out the next step.
Step 4.9: and updating the pheromone factor on the global path, judging whether the pheromone factor is the optimal solution in all the iteration times, if not, adding 1 to the iteration times, returning to the step 2.1 to perform a new iteration, and if so, taking the current solution as the global optimal solution to optimize the driving path of the demand response type bus.
According to the technical scheme, compared with the prior art, the demand response bus dispatching method based on the passenger travel time window constraint is characterized in that firstly, after the passenger demand and the bus running line are obtained and parameters are initialized, the ant colony algorithm is adopted to sequence and serially connect the demand points of the passenger travel according to the time windows and distribute the demand points to the corresponding bus running line, the bus stop candidate points are selected based on the passenger demand, the predicted service time window is calculated, the minimum error value between the predicted service time window and the time window of the passenger travel demand is taken to solve the final target, and the more practical bus connection timetable is finally obtained. The system can be coordinated with a trunk bus line, can adapt to various complex scheduling scenes of the demand response bus to be plugged in reality, and completes the real-time scheduling of the demand response bus system to be plugged in.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic flow chart of a demand response bus dispatching method based on passenger travel time window constraints according to the present invention;
fig. 2 is a schematic diagram of an implementation flow of a three-stage heuristic algorithm of the ant colony algorithm in the embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to the attached figure 1, the embodiment of the invention discloses a demand response bus dispatching method based on passenger travel time window constraint, which comprises the following steps:
s1: acquiring passenger demands and position information of a demand response bus starting station and a transfer pivot point, constructing a model total objective function, and initializing parameters; defining the model overall objective function as:
Figure BDA0002640404100000081
d is a set of bus stations, namely a set of starting points of demand response transfer bus routes, and M is a set of candidate bus stop stations; MS is a set of transfer pivot points, namely path end points;
Figure BDA0002640404100000082
variable of 0 or 1 when stopping from bus on route kAnd when the site candidate point j passes through the site m, the variable takes the value of 1, otherwise, if the non-passing site m takes the value of 0. t is tjmThe driving time of the vehicle from the candidate point j to the point m of the bus stop is obtained;
Figure BDA0002640404100000083
for the time on the path k expected to travel away from the bus stop candidate point j,
Figure BDA0002640404100000084
predicting the time for reaching the candidate point j of the demand response bus stop on the path k; diNumber of passengers at demand point i, dijIs the walking distance, x, from the passenger demand point i to the bus stop candidate point jijAnd the variable is 0 or 1, the value is 1 when the passenger i is allocated to the bus stop candidate point j, and otherwise, the value is 0. WsAverage walking speed for the passenger;
Figure BDA0002640404100000085
and
Figure BDA0002640404100000086
the upper and lower boundaries of the deviation value of the passenger preference time window and the bus running schedule are respectively.
The sum of the first two items in the total objective function expression is the total operation time of the demand response bus, and comprises two parts, namely the driving time and the stop time, the third item is the total walking time of passengers, and the fourth item is the total deviation value of the predicted service time window and the passenger preference time window at each bus stop.
In the above step S1, the parameters to be initialized include:
the total number N of ants in the ant colony, the total path quantity R and the maximum iteration number itermaxPath pheromone factor tau between any pair of demand points i and demand response bus stop candidate points jijPheromone variation amount delta tauij(ii) a Separately initializing path pheromone factors tauij(0) And corresponding pheromone volatilization factor delta tauij(0) And is 0, the current iteration number is set to 0.
S2: the ant colony algorithm is adopted, the demand points of the passengers going out are sequentially sequenced according to the time window and are connected in series by taking the demand response bus starting station as the starting point and taking the transfer pivot point as the end point.
The step S2 specifically includes the following steps:
step 2.1: sequencing passenger travel demand points according to the intermediate values of the upper bound and the lower bound of the passenger travel time window, and placing N ant individuals at the initial point of the sequence of the time window;
step 2.2: generating an ant colony candidate path set allowed specifically as follows:
Figure BDA0002640404100000091
wherein: i is the passenger demand point for selecting demand response for the plug-in bus trip, K is the set of all paths, psi (K) is the next station on path K next to demand point i, eψ(k)And lψ(k)Upper and lower bounds, t, of the time window corresponding to this site, respectivelyψ(k)iRepresenting travel time between the candidate site and the demand point i, eiAnd liThe upper bound and the lower bound of the time window corresponding to the demand point i are respectively.
Step 2.3: determining an ant individual transfer path based on the candidate path set allowed;
in the step 2.3, a pseudo-random proportion rule is adopted, and a pseudo-random number q is taken;
if q is less than or equal to q0Then ant selects order parameter tauiψ(k)(t)[ηiψ(k)]βTaking the kth path with the maximum value as a transfer path of the next step;
if q > q0Then calculate a probability value
Figure BDA0002640404100000092
The method specifically comprises the following steps:
Figure BDA0002640404100000101
the ant takes the kth path with the maximum probability value in the selected path set allowed as the transfer path of the next step;
where m is MS, which is the set of transfer pivot points, i.e. path end points, q0Is a preset parameter, q0∈(0，1]；ηiψ(k)Is a heuristic factor, eta, between the required point i and the next point psi (k) on the kth pathiψ(k)＝1/(tψ(k)i+0.01)；τiψ(k)And (t) is a path pheromone factor between a station next to the demand point i on the kth path at the time t and the demand point i, alpha is a pheromone relative influence degree factor, and beta is a heuristic factor relative influence degree factor.
Step 2.4: and returning to the step 2.2 until all demand points are distributed on the line from the bus starting station to the transfer pivot point.
In the embodiment of the invention, whether a certain ant individual can select the kth path as the travel route or not is judged through an allowed generation rule, and the rule shows that the ant can select the kth path to transfer to the next position only when the travel time is saved.
In the presence of pseudo-random numbers q and predetermined parameters q0When k is such that the parameter τ isiψ(k)(t)[ηiψ(k)]βWhen the value reaches the maximum path label, the ant will select the kth path, etaikThe heuristic factor is used for expressing the attraction degree of the kth path to the ants; otherwise, by probability value
Figure BDA0002640404100000102
And selecting the kth path with the maximum probability value in the path set allowed as the next transfer path.
S3: and selecting the best bus stop from the candidate set of bus stops corresponding to the demand points, and distributing the demand to the bus stops so as to generate a bus operation line.
The above step S3 includes the following steps:
step 3.1: the backtracking method is adopted, the transfer pivot point m is taken as the end point of the path k, and the transfer pivot point m is sent out from the end point to represent two phasesThe sum of the bus running time between adjacent stops and the walking time of passengers walking to the bus stop candidate point is the objective function (i.e. time cost objective function) in the step 3, and the objective function f corresponding to the demand point ik(i) Is set to 0.
Step 3.2: generating a corresponding bus stop candidate point set J for a demand point i on the kth pathk(i)。
Step 3.3: candidate point set J based on bus stopk(i)Carrying out bus stop site selection;
specifically, in step 3.3, the state transition equation (i.e., the time cost objective function) is set to:
Figure BDA0002640404100000111
calculating the time cost for meeting the demand point i through the path k; wherein the content of the first and second substances,
Figure BDA0002640404100000112
means the vehicle running time from the current bus stop candidate point j to the bus stop candidate point corresponding to the next demand point i +1, dijRefers to the distance from the demand point i to the candidate point j of the bus stop, WsRefers to the average walking speed.
Bus stop candidate point set J corresponding to current demand point ik(i)And finding a station which can make the travel time shortest as an optimal bus stop site selection result on the k-th path until the k-th path is traced back to the starting point, generating an optimal vehicle driving route k, and replacing the previous path.
Step 3.4: if all the passenger travel demand points are distributed to the optimal bus stop, the step S4 is carried out, otherwise, the step 3.2 is repeated.
In this embodiment, in step S3, each demand point is allocated to an address selection point of a bus stop, so as to determine a bus driving route of the demand response type connection bus system; the method realizes the allocation of the site selection and demand points of the bus stop candidate points through a dynamic planning method, backtracks from the terminal transfer station by taking the shortest travel time of passengers as a target, and further finds the optimal bus stop site selection point.
And 4, step 4: and calculating a service time window of the bus station, taking the minimum error value between the predicted response time of the bus and the travel demand time window of the passengers, and solving by taking the minimum total error value as the objective function of the step S4 to obtain a transfer bus schedule.
The step S4 specifically includes:
step 4.1: for the kth bus route, the predicted arrival time of the kth bus route to the transfer pivot point m through the route is calculated
Figure BDA0002640404100000113
The method specifically comprises the following steps:
Figure BDA0002640404100000114
wherein, PconThe time for transferring the main line traffic departure at the pivot point m is designated; t isconThe minimum time difference allowed between the time when the demand response bus arrives at the transfer pivot node m and the time when the demand response bus departs from the transfer pivot node m;
step 4.2: the predicted departure time of the bus at the bus stop candidate point j is calculated as
Figure BDA0002640404100000121
Figure BDA0002640404100000122
Wherein, tjmThe bus travel time from the candidate point j of the bus stop station to the transfer pivot point m of the demand response bus selected in the step S3;
step 4.3: generating a predicted arrival time interval [ l ] for each demand response bus stopi-TDmaxli+TDmax]Wherein l isiPassenger travel time window upper bound, TD, for passenger travel demand point i served by optimal bus stopmaxForecasting for public transportThe maximum difference allowed between the time of leaving the bus stop candidate point j and the upper bound of the passenger travel time window; (ii) a
Step 4.4: traversing the vehicle arrival time of all bus stop candidate points j, and taking the result closest to the passenger preference time window as a final algorithm distribution result; the step adopts a mode of traversing all the demand points, so that the method has a polynomial time complexity characteristic;
step 4.5: if the passenger preference time windows of all the passenger trip demand points are met, a demand response connection bus schedule is formed; otherwise, repeating the step 4.3.
After the demand response connection bus schedule is formed, in order to further optimize the scheme, the following steps can be further set:
step 4.6: determining the volatilization amount Delta tau of pheromone on each pathijThe method specifically comprises the following steps:
Figure BDA0002640404100000123
wherein the content of the first and second substances,
Figure BDA0002640404100000124
the pheromone variation quantity of the s-th ant left on the side (i, j) in the iteration is represented and is a constant;
step 4.7: updating the total objective function value and the pheromone factor tau on the local path based on the obtained pheromone volatilization amountij(t + Δ t), specifically:
τij(t+Δt)＝(1-ρ)τij(t)+Δτij
wherein rho is the evaporation coefficient of pheromone on the path, and rho is more than 0 and less than 1;
step 4.8: storing the current total objective function value and the allocation scheme, judging whether the current objective function value is the optimal solution in the current ant colony, namely the minimum value in all objective function values, if not, returning to the step 2.2 to calculate the next ant, otherwise, performing the next step;
step 4.9: and updating the pheromone factor on the global path, judging whether the optimal solution is the optimal solution in all the iteration times, if not, adding 1 to the iteration times, returning to the step 2.1 to carry out a new iteration, and if not, judging that the optimal solution is the global optimal solution.
The demand response bus dispatching method based on the passenger trip time window constraint disclosed by the embodiment of the invention is solved by a three-stage heuristic algorithm based on an ant colony algorithm, and the method is shown in the attached figure 2 and specifically comprises the following steps:
the first stage, an ant colony algorithm is adopted, ant colony generation and transfer rules are designed, random passenger travel demands are distributed to all lines from a bus starting station to a trunk line transfer terminal station, and candidate lines are generated;
in the second stage, aiming at each candidate bus line, a dynamic planning algorithm is adopted to distribute the demand of passengers belonging to the line to each bus stop on the line, and the stop site selection is completed for each demand point;
and in the third stage, designing an optimal riding time window for each demand point on each line through a polynomial algorithm to finish scheduling.
In the embodiment of the invention, after one round of three-stage calculation is completed, the current objective function value and the allocation scheme are stored, the pheromones on each path are updated, and meanwhile, when the iteration times are smaller than the maximum iteration times, the current iteration times are added by one to enter the next round of calculation.
In this embodiment, the passenger travel demand points and the bus candidate stations are distributed discretely and randomly, and only the starting point and the ending point of the route are fixed and are the bus starting station and the transfer pivot point respectively. The public transportation starting station and the transfer pivot point are urban trunk public transportation stations including but not limited to subway stations, BRT stations and trunk common public transportation stations.
It is easy to find that the method provided by this embodiment efficiently solves the NP-hard problem, and under the condition of satisfying the constraint of the passenger time window, by minimizing the passenger walking connection distance and the total system trip time, the demand response connection bus route and departure time are designed and scheduled and coordinated with the trunk bus route, the selection process of the bus stop is increased to reflect the actual demand response connection bus operation state, and the three-stage heuristic algorithm based on the ant colony algorithm frame is embedded into the dynamic programming and the polynomial algorithm to solve the approximate optimal solution, and finally the complex scheduling scenes of various demand response connection buses in reality are realized, and the real-time scheduling of the demand response connection bus system is completed.
In summary, compared with the prior art, the demand response type connection bus scheduling method provided by the embodiment of the invention has the following advantages:
1. three independent stages of demand response public traffic network in the existing research are integrated in the same algorithm frame, so that the method is suitable for the real scheduling condition with complex change;
2. the traditional ant colony algorithm is improved, and a dynamic programming algorithm and a polynomial algorithm are fused, so that the algorithm can more accurately find the optimal solution in limited memory and time resources;
3. the method has low requirement on hardware configuration and high solving efficiency, and the algorithm shows good convergence in different test scenes.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (10)

1. A demand response bus dispatching method based on passenger travel time window constraint is characterized by comprising the following steps:
step 1: acquiring passenger trip demand data and position information of a demand response type bus starting station and a transfer pivot point, constructing a total objective function, and initializing parameters in the total objective function;
step 2: sequencing passenger travel demand points in passenger travel demand data in sequence according to a time window and connecting the passenger travel demand points in series by taking a demand response type bus starting station as a starting point and taking a transfer pivot point as a terminal point to obtain a plurality of candidate paths;
and step 3: acquiring a bus stop candidate set corresponding to a passenger travel demand point on each candidate path, selecting an optimal bus stop from the bus stop candidate set, and distributing the passenger travel demand point to the optimal bus stop to generate a bus operation line;
and 4, step 4: and acquiring an optimal riding time window corresponding to each passenger trip demand point on the bus operation line, and generating a demand response type connection bus schedule.
2. The demand response bus dispatching method based on passenger travel time window constraint according to claim 1, characterized in that the total objective function is specifically:
Figure FDA0002640404090000011
in the formula, U is a total objective function, D is a set of starting points of demand response type transfer bus routes, M is a set of candidate bus stop points, MS is a transfer pivot point, namely a set of route end points,
Figure FDA0002640404090000012
is a variable with a value of 0 or 1, tjmFor the travel time of the vehicle from the bus stop candidate point j to the point m,
Figure FDA0002640404090000013
for the time on the path k expected to travel away from the bus stop candidate point j,
Figure FDA0002640404090000014
is the predicted time to reach the bus stop candidate point j on the path k, DiNumber of passengers at demand point i, dijFor the walking distance, x, of the passenger from the demand point i to the candidate point j of the bus stopijFor variables taking the value 0 or 1, WsFor the average walking speed of the passengers,
Figure FDA0002640404090000021
and
Figure FDA0002640404090000022
the upper and lower boundaries of the deviation value of the passenger preference time window and the bus running schedule are respectively.
3. The demand response bus dispatching method based on passenger travel time window constraint according to claim 1, wherein in the step 2, a plurality of candidate paths are obtained by adopting an ant colony algorithm, and the method specifically comprises the following steps:
step 2.1: sequencing the passenger travel demand points according to the intermediate values of the upper bound and the lower bound of the passenger travel time window in the passenger travel demand data, and placing N ant individuals at the initial sequence point of the passenger travel time window;
step 2.2: generating an ant colony candidate path set according to the sorted passenger travel demand points and the adjacent candidate stations;
step 2.3: determining the transfer path of the ant individual based on the candidate path set;
step 2.4: and distributing all the passenger travel demand points to a route from the bus starting station to the transfer pivot point to obtain a plurality of candidate routes.
4. The demand response bus dispatching method based on passenger travel time window constraint, according to claim 3, characterized in that the ant colony candidate path set specifically is:
Figure FDA0002640404090000023
wherein allowed is an ant colony candidate path set, i is a passenger demand point for selecting a demand response type connection bus trip, K is a set of all paths, psi (K) is a next candidate station adjacent to the demand point i on the path K, and eψ(k)And lψ(k)Upper and lower bounds, t, of the time window corresponding to the candidate site, respectivelyψ(k)iRepresenting travel time between the candidate site and the demand point i, eiAnd liThe upper bound and the lower bound of the time window corresponding to the demand point i are respectively.
5. The demand response bus dispatching method based on passenger travel time window constraint, as recited in claim 4, wherein in step 2.3, the transfer path of the ant individual is determined by adopting a pseudo-random proportion rule, specifically:
taking pseudo-random number as q and preset parameter as q0；
If q is less than or equal to q0Then the ant individual selects the order parameter tauiψ(k)(t)[ηiψ(k)]βThe kth path with the maximum value is used as the transfer path of the next step;
if q > q0Then calculate a probability value
Figure FDA0002640404090000031
Figure FDA0002640404090000032
The ant individual selects the kth path with the maximum probability value in the path set as a transfer path of the next step;
wherein m belongs to MS; q. q.s0Is a preset parameter, q0∈(0，1]；ηiψ(k)A heuristic factor between a demand point i and a next point psi (k) on the k-th path; tau isiψ(k)(t) is the next station on the kth path immediately adjacent to the demand point i at time tPath pheromone factors between the points and the demand points i; alpha is a relative pheromone influence degree factor; beta is a heuristic factor relative influence degree factor.
6. The demand response bus dispatching method based on passenger travel time window constraint according to claim 1, wherein the step 3 specifically comprises:
step 3.1: selecting a transfer pivot point as a destination of a candidate route by adopting a backtracking method, and constructing a time cost objective function according to the sum of the bus running time between two stops and the walking time of passengers walking to the candidate point of the stop from the destination;
step 3.2: setting an initial value of a time cost objective function corresponding to the trip demand point of the passenger to be solved as 0, and generating a candidate point set of the bus stop corresponding to the trip demand point of the passenger to be solved on the candidate path;
step 3.3: based on the candidate point set of the bus stop, selecting the address of the bus stop and selecting the best bus stop;
step 3.4: and distributing all the passenger travel demand points to the corresponding optimal bus stop points to generate a bus operation line.
7. The demand response bus dispatching method based on passenger travel time window constraint according to claim 6, characterized in that the time cost objective function is:
Figure FDA0002640404090000033
wherein the content of the first and second substances,
Figure FDA0002640404090000041
representing the vehicle running time from the current bus stop candidate point j to the bus stop candidate point corresponding to the next demand point i +1, dijRepresents the distance from the demand point i to the candidate point j of the bus stop, WsFinger passengerAverage walking speed.
8. The demand response bus dispatching method based on passenger travel time window constraint according to claim 1, wherein the step 4 specifically comprises:
step 4.1: calculating the predicted arrival time of the bus to a target transfer pivot point through a certain line in the bus operation line;
step 4.2: calculating the predicted departure time of the bus at the candidate point of the bus stop;
step 4.3: generating a vehicle predicted arrival time interval [ l ] of each bus stop candidate point according to the predicted arrival time and the predicted departure timei-TDmax，li+TDmax]；
Wherein liPassenger travel time window upper bound, TD, for passenger travel demand point i served by optimal bus stopmaxThe maximum difference allowed between the time of the bus expected to leave the bus stop candidate point j and the upper bound of the passenger travel time window is calculated;
step 4.4: traversing the predicted arrival time intervals of the buses of all the bus stop candidate points, and taking the result closest to the time window preferred by the passenger as a final distribution result;
step 4.5: and if the passenger preference time windows of all the passenger trip demand points are met, forming a demand response type connection bus schedule.
9. The passenger travel time window constraint-based demand response bus dispatching method according to claim 8, wherein the calculation formula of the predicted arrival time at the target transfer pivot point is as follows:
Figure FDA0002640404090000042
wherein the content of the first and second substances,
Figure FDA0002640404090000043
for estimated time of arrival, PconFor transferring the time of departure of the main traffic at pivot point m, TconThe minimum time difference allowed between the arrival of the bus at the transfer pivot point m and the departure of the bus at the transfer pivot point m is responded to by the demand.
10. The method of claim 8, wherein the step 4 further comprises:
step 4.6: determining the volatilization amount of pheromones on each bus operation line;
step 4.7: updating the value of the total objective function and the pheromone factor on the local path based on the pheromone volatilization amount;
step 4.8: judging whether the value of the current total objective function is the minimum value of all the values of the total objective function, if not, returning to the step 2, and if so, carrying out the next step;
step 4.9: and updating the pheromone factor on the global path, judging whether the pheromone factor is the optimal solution in all the iteration times, if not, adding 1 to the iteration times, returning to the step 2, and if so, taking the current solution as the global optimal solution to optimize the driving path of the demand response type bus.
CN202010838030.5A 2020-08-19 2020-08-19 Demand response bus dispatching method based on passenger travel time window constraint Expired - Fee Related CN112085349B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010838030.5A CN112085349B (en) 2020-08-19 2020-08-19 Demand response bus dispatching method based on passenger travel time window constraint

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010838030.5A CN112085349B (en) 2020-08-19 2020-08-19 Demand response bus dispatching method based on passenger travel time window constraint

Publications (2)

Publication Number Publication Date
CN112085349A true CN112085349A (en) 2020-12-15
CN112085349B CN112085349B (en) 2021-06-01

Family

ID=73728553

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010838030.5A Expired - Fee Related CN112085349B (en) 2020-08-19 2020-08-19 Demand response bus dispatching method based on passenger travel time window constraint

Country Status (1)

Country Link
CN (1) CN112085349B (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112381472A (en) * 2021-01-15 2021-02-19 深圳市城市交通规划设计研究中心股份有限公司 Subway connection bus route optimization method and device and storage medium
CN113408790A (en) * 2021-05-31 2021-09-17 银江股份有限公司 Modular vehicle operation path and combination strategy collaborative optimization method and system
CN113780808A (en) * 2021-09-10 2021-12-10 西南交通大学 Vehicle service attribute decision optimization method based on flexible bus connection system line
CN114331060A (en) * 2021-12-15 2022-04-12 东南大学 DRT vehicle path generation method for high-speed rail station connection under MaaS background
CN115018211A (en) * 2022-08-08 2022-09-06 北京建筑大学 Method and device for setting transportation scheduling line
CN115049156A (en) * 2022-08-11 2022-09-13 深圳市城市交通规划设计研究中心股份有限公司 Traffic line network operation diagram linkage optimization method, electronic equipment and storage medium
CN115100896A (en) * 2022-06-17 2022-09-23 大连海事大学 Electric demand response bus dispatching method considering opportunity charging strategy
CN116629567A (en) * 2023-06-01 2023-08-22 大连海事大学 Demand response public traffic and shared electric bicycle coupling optimization method considering dynamic demand
CN117252323A (en) * 2023-11-17 2023-12-19 西华大学 User preference-based multi-mode travel path planning method
CN117789955A (en) * 2024-02-28 2024-03-29 济南大学 Medical service distribution and path planning method, system, equipment and medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102169512A (en) * 2010-02-26 2011-08-31 同济大学 Urban mass transit network dynamic accessibility computing method based on time window constraints
CN102737129A (en) * 2012-06-20 2012-10-17 东南大学 Method for compiling bus regional running timetable
CN106096868A (en) * 2016-07-01 2016-11-09 广州地理研究所 A kind of multiple constraint network intensive minibus dispatching method
CN108960539A (en) * 2018-08-31 2018-12-07 东南大学 A kind of demand response formula feeder bus sytem method for optimizing route
CN109344529A (en) * 2018-10-22 2019-02-15 北京航空航天大学 A kind of customization public bus network design method based on two-phase heuristic algorithm
CN109409599A (en) * 2018-10-24 2019-03-01 天津市市政工程设计研究院 Customization public bus network based on real-time requirement opens optimization method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102169512A (en) * 2010-02-26 2011-08-31 同济大学 Urban mass transit network dynamic accessibility computing method based on time window constraints
CN102737129A (en) * 2012-06-20 2012-10-17 东南大学 Method for compiling bus regional running timetable
CN106096868A (en) * 2016-07-01 2016-11-09 广州地理研究所 A kind of multiple constraint network intensive minibus dispatching method
CN108960539A (en) * 2018-08-31 2018-12-07 东南大学 A kind of demand response formula feeder bus sytem method for optimizing route
CN109344529A (en) * 2018-10-22 2019-02-15 北京航空航天大学 A kind of customization public bus network design method based on two-phase heuristic algorithm
CN109409599A (en) * 2018-10-24 2019-03-01 天津市市政工程设计研究院 Customization public bus network based on real-time requirement opens optimization method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
BO SUN 等: "An Optimization Model for Demand-Responsive Feeder Transit Services Based on Ride-Sharing Car", 《INFORMATION》 *
王健 等: "考虑出行时间窗的定制公交线路车辆调度方法", 《中国公路学报》 *

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112381472B (en) * 2021-01-15 2021-07-30 深圳市城市交通规划设计研究中心股份有限公司 Subway connection bus route optimization method and device and storage medium
CN112381472A (en) * 2021-01-15 2021-02-19 深圳市城市交通规划设计研究中心股份有限公司 Subway connection bus route optimization method and device and storage medium
CN113408790B (en) * 2021-05-31 2022-09-30 银江技术股份有限公司 Modular vehicle operation path and combination strategy collaborative optimization method and system
CN113408790A (en) * 2021-05-31 2021-09-17 银江股份有限公司 Modular vehicle operation path and combination strategy collaborative optimization method and system
CN113780808A (en) * 2021-09-10 2021-12-10 西南交通大学 Vehicle service attribute decision optimization method based on flexible bus connection system line
CN113780808B (en) * 2021-09-10 2023-04-07 西南交通大学 Vehicle service attribute decision optimization method based on flexible bus connection system line
CN114331060A (en) * 2021-12-15 2022-04-12 东南大学 DRT vehicle path generation method for high-speed rail station connection under MaaS background
CN114331060B (en) * 2021-12-15 2023-04-28 东南大学 DRT vehicle path generation method for high-speed rail station connection under MaaS background
CN115100896A (en) * 2022-06-17 2022-09-23 大连海事大学 Electric demand response bus dispatching method considering opportunity charging strategy
CN115018211B (en) * 2022-08-08 2022-11-01 北京建筑大学 Method and device for setting transportation scheduling line
CN115018211A (en) * 2022-08-08 2022-09-06 北京建筑大学 Method and device for setting transportation scheduling line
CN115049156A (en) * 2022-08-11 2022-09-13 深圳市城市交通规划设计研究中心股份有限公司 Traffic line network operation diagram linkage optimization method, electronic equipment and storage medium
CN115049156B (en) * 2022-08-11 2023-02-03 深圳市城市交通规划设计研究中心股份有限公司 Traffic line network operation diagram linkage optimization method, electronic equipment and storage medium
CN116629567A (en) * 2023-06-01 2023-08-22 大连海事大学 Demand response public traffic and shared electric bicycle coupling optimization method considering dynamic demand
CN116629567B (en) * 2023-06-01 2024-01-16 大连海事大学 Demand response public traffic and shared electric bicycle coupling optimization method considering dynamic demand
CN117252323A (en) * 2023-11-17 2023-12-19 西华大学 User preference-based multi-mode travel path planning method
CN117252323B (en) * 2023-11-17 2024-01-30 西华大学 User preference-based multi-mode travel path planning method
CN117789955A (en) * 2024-02-28 2024-03-29 济南大学 Medical service distribution and path planning method, system, equipment and medium
CN117789955B (en) * 2024-02-28 2024-05-03 济南大学 Medical service distribution and path planning method, system, equipment and medium

Also Published As

Publication number Publication date
CN112085349B (en) 2021-06-01

Similar Documents

Publication Publication Date Title
CN112085349B (en) Demand response bus dispatching method based on passenger travel time window constraint
CN109409599B (en) Customized bus line opening optimization method based on real-time requirements
US10639995B2 (en) Methods, circuits, devices, systems and associated computer executable code for driver decision support
CN108417031B (en) Intelligent parking berth reservation strategy optimization method based on Agent simulation
JP7166222B2 (en) System for assigning commuter vehicles to passengers
JP6341352B1 (en) Demand traffic operation system
CN107146068B (en) Train operation daily plan compiling and allocating method based on balanced application
CN104598994A (en) Related logistics transportation optimized dispatching method with time-varying time window
CN112561249B (en) Real-time demand-oriented city customized bus scheduling method
CN112418514A (en) Method for optimizing campus bus route planning by using ant colony system
CN115222156A (en) Automobile charging scheduling method considering user demand response based on time-sharing dual road network
CN111724076A (en) Regional multi-type rail transit passenger flow dynamic distribution method under operation interruption condition
Ning et al. Passenger-centric vehicle routing for first-mile transportation considering request uncertainty
CN104477217A (en) Operation and circulation of trains on city rail with single end station
Bagamanova et al. Extended atm for seamless travel (x-team d2d)
Yousaf et al. Ride-sharing: A multi source-destination path planning approach
CN112465211A (en) Rail transit train full load rate control method and application
El Falou et al. On demand transport system's approach as a multi-agent planning problem
Kim et al. Model and algorithm considering time-varying travel times to solve static multidepot dial-a-ride problem
JP2021015379A (en) Vehicle allocation processing device
Poulhès et al. User assignment in a smart vehicles’ network: Dynamic modelling as an agent-based model
Zhao et al. Online vehicle dispatch: From assignment to scheduling
Lee et al. Optimal Zonal Design for Flexible Bus Service Under Spatial and Temporal Demand Uncertainty
Lees-Miller et al. Sampling of redistribution of empty vehicles for personal rapid transit
Asatryan et al. Ridepooling and public bus services: A comparative case-study

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20210601