CN113592157A - Method and device for predicting bus travel time under sparse data - Google Patents
Method and device for predicting bus travel time under sparse data Download PDFInfo
- Publication number
- CN113592157A CN113592157A CN202110787531.XA CN202110787531A CN113592157A CN 113592157 A CN113592157 A CN 113592157A CN 202110787531 A CN202110787531 A CN 202110787531A CN 113592157 A CN113592157 A CN 113592157A
- Authority
- CN
- China
- Prior art keywords
- travel time
- shift
- predicting
- planned
- time
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q10/10—Office automation; Time management
- G06Q10/109—Time management, e.g. calendars, reminders, meetings or time accounting
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention discloses a method for predicting bus travel time under sparse data, which comprises the steps of determining a planned shift, calculating a proportionality coefficient between an actual operation shift and the planned shift travel time, and predicting travel time information of a bus arriving at a terminal station in real time based on the proportionality coefficient.
Description
Technical Field
The invention relates to the technical field of intelligent transportation, in particular to a method and a device for predicting bus travel time under sparse data.
Background
The prediction of the bus travel time is the basis of bus intelligent scheduling and bus electronic stop board application. The accurate time prediction has important significance for improving the dynamic dispatching effect of the public transport, reducing the waiting time of passengers, improving the public transport service quality, realizing the priority of public transport signals and embodying the superiority of intelligent public transport. According to the difference of prediction methods, the following 5 categories can be basically classified: a space-time model, a regression model, an artificial neural network model, a Kalman filtering model, a support vector machine model and the like. At present, the models cannot be applied to practice, and two problems exist. On one hand, because the traffic conditions are greatly influenced by time (such as rush hour, signal light, holiday and the like) and space (road emergency) factors, the fluctuation is frequent, the prediction results of the methods and the models are often greatly different from the actual values, and the models are complex to construct and operate and are not easy to realize. On the other hand, due to the fact that the existing bus-mounted GPS is not perfect, the predicted travel time data quality of the intelligent bus system is poor, partial data are missing or abnormal, and under sparse data, the prediction performance of the methods is greatly degraded, and the standards of actual requirements cannot be met.
Disclosure of Invention
The invention solves the technical problem of providing a travel time prediction method and a prediction device for intelligent buses under sparse data, wherein the method determines the optimal planned shift by analyzing factors influencing travel times of different shifts, such as weather, time periods, temperature and the like, and then realizes more accurate travel time prediction according to the proportional relationship of the travel time between the planned shift and the actual operation shift, particularly the travel time prediction under the sparse data.
The technical scheme adopted by the invention specifically comprises the following contents:
a travel time prediction method for an intelligent bus comprises the following steps:
s1, analyzing influence factors causing the differences of travel time of different shifts;
s2, determining the optimal planned shift;
s3, predicting the proportional relation between the actual operation shift and the planned shift based on the planned shift;
and S4, completing the real-time prediction of the bus travel time based on the proportional relation.
Preferably, the determination of the optimal planned shift is based on influencing factors influencing the travel time differences between different shifts.
Preferably, the factors causing the difference of travel times of different shifts include time periods, weather and temperature.
Preferably, the travel time of the planned shift is an average of travel times of historical shifts which are within the same time period as the departure time of the actual running shift and have similar weather conditions and temperatures.
Preferably, the ratio relationship between the actual operation shift and the planned shift is updated in real time.
Preferably, the bus travel time information is updated in real time along with the update of the proportionality coefficient.
The invention also provides a device for predicting the bus travel time, which comprises a prediction unit, a first determination unit, a second determination unit and a generation unit, wherein the prediction unit is used for predicting influence factors causing the difference of travel times of different shifts; the first determining unit is used for determining an optimal planned shift, and the second determining unit is used for determining a proportional relation between an actual operation shift and the planned shift; the generating unit is used for predicting real-time bus travel time information.
Compared with the prior art, the invention has the beneficial effects that:
a method for predicting bus travel time under sparse data is analyzed from the principle, influence factors suffered by a bus in the running process can be divided into two types, one is a directional factor, and the directional factor is gradually accumulated along with time, such as the influence of driving behaviors and the influence of background traffic flow, and even comprises filtering control; the second is a stochastic factor, where no cumulative effect occurs, which counteracts or is bounded as the vehicle travels. Therefore, as the vehicle travels, the cumulative effect of the directional factors increases and exceeds the influence of the stochastic factor, so that the travel time exhibits a stable proportional characteristic, and the prediction accuracy is excellent when the directional factors are sufficiently accumulated. From the analysis of operation angle, on the one hand, the model can satisfy the travel time prediction under public transit operational data sparse condition, and on the other hand, the model is built simply relatively, need not spend a large amount of time, perhaps uses a large amount of data to look for the optimal model structure, alright in order to realize higher prediction accuracy, is favorable to improving public transit system service quality, promotes public transit traveler's service experience.
Drawings
FIG. 1 is a schematic flow chart of a method for predicting bus travel time according to the present invention;
FIG. 2 is a schematic diagram of the ratio of actual operating shifts to scheduled shift travel times;
FIG. 3 is a graph illustrating the scaling factor stability relationship based on a planned shift in hours and days;
FIG. 4 is a graphical illustration of accuracy of travel time predictions based on a planned shift in hours and days;
FIG. 5 is a schematic diagram of a travel time prediction apparatus according to the present invention;
FIG. 6 is a schematic diagram of a travel time prediction error analysis based on a proportional characteristic under sparse data.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description of the embodiments, structures, features and effects according to the present invention with reference to the accompanying drawings and preferred embodiments is as follows:
as shown in FIG. 1, the present invention discloses a method for predicting a user portrait, comprising:
predicting influence factors causing travel time differences of different shifts;
determining an optimal planned shift;
predicting the proportional relation between the actual operation shift and the planned shift based on the planned shift;
and (4) completing the real-time prediction of the bus travel time based on the proportional relation.
As shown in fig. 5, the invention further provides a prediction device of bus travel time, which comprises a prediction unit, a first determination unit, a second determination unit and a generation unit, wherein the prediction unit is used for predicting influence factors causing differences of travel times of different shifts; the first determining unit is used for determining an optimal planned shift, and the second determining unit is used for determining a proportional relation between an actual operation shift and the planned shift; the generating unit is used for predicting real-time bus travel time information.
Specific examples are as follows.
A method for predicting bus travel time comprises the following steps:
(1) and predicting influence factors causing the difference of travel time of different shifts, constructing corresponding planned shifts according to the influence factors, and analyzing the change rule of the proportionality coefficients under the different planned shifts. Specifically, the main influence factors causing the difference of travel time of different shifts are obtained by analyzing a certain amount of data of the travel time of the buses, including time periods, weather and temperature, and then the change rule of the proportionality coefficients under the planned shifts constructed according to the influence factors is analyzed, for example, the planned shifts in hours (obtained by calculating the mean value of the travel time of the shift in one time period (1h) at the departure time) and the planned shifts without considering the difference of the time periods (obtained by directly calculating the mean value of the historical travel time) are constructed by using historical data of the travel time, and as can be seen from fig. 2 (data of K19 buses in jiangyin city), the proportionality coefficients of the travel time under the two planned shifts are rapidly converged to a certain stable value.
(2) Determining the optimal planned shift, in particular, by comparing the stability of the scaling factors under different planned shifts and the accuracy for the travel time prediction, in combination with the laws found in (1), giving the optimal planned shift.
Wherein the stability of the scaling factor is determined by the deviation (DeltaGamma) of the scaling factor of the previous station from the scaling factor of the terminal stationn(k)＝γn(k)-γN(k) Is) expressed, i.e.:
Δγn(k)＝γn(k)-γN(k)
in the formula, gamman(k) Is the travel time proportionality coefficient, gamma, corresponding to the station nN(k) Is the travel time scaling factor for the destination station (i.e., station N).
As a result, as shown in fig. 3 (data of K19 buses in jiangyin city), the scaling factors of both planned shifts exhibit better stability, and specifically, the stability of the scaling factor based on the planned shift in hours is negligible because the stability is based on the planned shift in days, but the difference is not obvious.
The accuracy of the travel time prediction is represented by the Mean Absolute Error (MAE), i.e.:
wherein K is the number of test shifts,
is the predicted path travel time, t, from station n to the end stationn,N(k) Is the path travel time, pt, from station n to the destinationN(j) The path cruise time for the planned shift constructed.
The precision of the travel time prediction is expressed by the Mean Absolute Percentage Error (MAPE), i.e.:
in the formula (I), the compound is shown in the specification,
is the predicted path travel time, t, from station n to the end stationn,N(k) Is the path travel time from station n to the end station.
As a result, as shown in fig. 4 (data of K19 buses in jiangyin city), the travel time prediction models constructed based on the two planned shifts have high accuracy, specifically, the prediction model based on the planned shift in hours has higher accuracy than the prediction model constructed based on the planned shift in an undifferentiated period, and therefore, the following model accuracy degradation analysis under sparse data is performed to select a prediction model based on the planned shift in hours as a reference shift for analysis.
(3) Calculating the proportionality coefficient according to the planned shift given in (2) and combining the real-time operation data (travel time) of the actual operation shift, wherein the calculation formula is as follows:
in the formula pti,n(k) For the path travel time, pt, from station i to destination station n for shift ki,n(B) The path travel time for the planned shift from station i to target station n.
(4) And (4) giving the travel time information of the bus in real time according to the proportional relation given in the step (3). The specific calculation formula is as follows:
in the formula, pti,j(k) The bus route travel time from the station i to the station j can be obtained through bus AVL data. pt isi,n(k) Is the predicted path travel time from station i to destination station n, which can be obtained by the following equation:
according to the existing data (K19 buses in Jiangyun city), the analysis finds that under sparse data (the actual data set only stores 90%, 80%, 70%, 60% and 50% of the complete data set), the prediction accuracy of the travel time prediction method provided by the invention has no obvious degradation phenomenon in the data loss environment, and still shows better prediction performance (as shown in figure 6).
The above embodiments are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby, and any insubstantial changes and substitutions made by those skilled in the art based on the present invention are within the protection scope of the present invention.
Claims (7)
1. A method for predicting bus travel time under sparse data is characterized by comprising the following steps:
s1, predicting influence factors causing travel time differences of different shifts;
s2, determining the optimal planned shift by analyzing influence factors;
s3, predicting the proportional relation between the actual operation shift and the planned shift based on the planned shift;
and S4, completing the real-time prediction of the bus travel time based on the proportional relation.
2. The method for predicting bus travel time under sparse data according to claim 1, wherein in step S1, a certain amount of bus travel time data is analyzed to obtain influence factors causing differences in travel time of different shifts, and then a change rule of a proportionality coefficient under a planned shift constructed based on the influence factors is analyzed to construct the planned shift in hours and the planned shift without considering time interval differences from historical travel time data.
3. The method for predicting bus travel time under sparse data as recited in claim 2, wherein the planned shift in hours is obtained by calculating an average of departure times over shift travel time over a time period.
4. The method for predicting bus travel time under sparse data as claimed in claim 2, wherein planned shifts without consideration of time period differences are obtained by calculating historical travel time means.
5. The method for predicting bus travel time under sparse data as claimed in claim 1, wherein the factors causing travel time differences of different shifts comprise time periods, weather and temperature.
6. The method for predicting the bus travel time under the sparse data as claimed in claim 1, wherein the travel time of the planned shift is an average value of the travel times of historical shifts which are in the same time period as the departure time of the actual running shift and have similar weather conditions and temperatures.
7. The device for predicting the bus travel time under the sparse data is characterized by comprising a prediction unit, a first determination unit, a second determination unit and a generation unit, wherein the prediction unit is used for predicting influence factors causing the difference of travel times of different shifts; the first determining unit is used for determining an optimal planned shift, and the second determining unit is used for determining a proportional relation between an actual operation shift and the planned shift; the generating unit is used for predicting real-time bus travel time information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110787531.XA CN113592157A (en) | 2021-07-13 | 2021-07-13 | Method and device for predicting bus travel time under sparse data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110787531.XA CN113592157A (en) | 2021-07-13 | 2021-07-13 | Method and device for predicting bus travel time under sparse data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113592157A true CN113592157A (en) | 2021-11-02 |
Family
ID=78247078
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110787531.XA Pending CN113592157A (en) | 2021-07-13 | 2021-07-13 | Method and device for predicting bus travel time under sparse data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113592157A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102737504A (en) * | 2012-07-13 | 2012-10-17 | 重庆大学 | Method for estimating bus arrival time in real time based on drive characteristics |
| JP2013054771A (en) * | 2012-11-21 | 2013-03-21 | Mitsubishi Electric Information Systems Corp | Operation management support system and program |
| CN109191845A (en) * | 2018-09-28 | 2019-01-11 | 吉林大学 | A kind of public transit vehicle arrival time prediction technique |
| CN109215374A (en) * | 2018-10-26 | 2019-01-15 | 上海城市交通设计院有限公司 | A kind of bus arrival time prediction algorithm |
-
2021
- 2021-07-13 CN CN202110787531.XA patent/CN113592157A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102737504A (en) * | 2012-07-13 | 2012-10-17 | 重庆大学 | Method for estimating bus arrival time in real time based on drive characteristics |
| JP2013054771A (en) * | 2012-11-21 | 2013-03-21 | Mitsubishi Electric Information Systems Corp | Operation management support system and program |
| CN109191845A (en) * | 2018-09-28 | 2019-01-11 | 吉林大学 | A kind of public transit vehicle arrival time prediction technique |
| CN109215374A (en) * | 2018-10-26 | 2019-01-15 | 上海城市交通设计院有限公司 | A kind of bus arrival time prediction algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9159032B1 (en) | Predicting arrival times of vehicles based upon observed schedule adherence | |
| JP5374067B2 (en) | Traffic condition simulation apparatus and program | |
| Roncoli et al. | Hierarchical model predictive control for multi-lane motorways in presence of vehicle automation and communication systems | |
| US9177473B2 (en) | Vehicle arrival prediction using multiple data sources including passenger bus arrival prediction | |
| Gkiotsalitis et al. | An analytic solution for real-time bus holding subject to vehicle capacity limits | |
| Xinghao et al. | Predicting bus real-time travel time basing on both GPS and RFID data | |
| US8874354B2 (en) | Method and system for expansion of real-time data on traffic networks | |
| CN105788270A (en) | Internet of things-based traffic data prediction method and processing server | |
| CN113219933B (en) | Strip mine unmanned truck dispatching system and method based on digital twin prediction | |
| US20140100766A1 (en) | Traffic data prediction device, traffic data prediction method and computer program | |
| CN111688538A (en) | System and method for providing a predictive distance-to-empty for a vehicle | |
| CN111476389A (en) | Method and device for pre-estimating order receiving waiting time | |
| CN109064742A (en) | A kind of adaptive public transport arrival time prediction technique based on SVM | |
| Jenelius | Data-driven bus crowding prediction based on real-time passenger counts and vehicle locations | |
| US20220415170A1 (en) | Method and arrangement for predicting switching times of a signal group of a signal installation for controlling a flow of traffic | |
| CN112258856B (en) | Method for establishing regional traffic signal data drive control model | |
| CN113592157A (en) | Method and device for predicting bus travel time under sparse data | |
| CN113232671A (en) | Computer-implemented method, storage medium, and system | |
| Peng et al. | Network-wide coordinated control based on space-time trajectories | |
| CN105303246A (en) | Multiline arrival time prediction for public transportation | |
| Bazan et al. | Rebalancing and fleet sizing of mobility-on-demand networks with combined simulation, optimization and queueing network analysis | |
| CN115909734A (en) | Vehicle driving data screening and updating system and method based on track big data | |
| CN114781243A (en) | ETA prediction and model training method, device, medium and product | |
| CN112101677B (en) | Public transport travel path planning method, device, equipment and storage medium | |
| Zeng et al. | A transient queuing model for analyzing and optimizing gate congestion of railway container terminals |
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 |