CN105719019A - Public bicycle peak time demand prediction method considering user reservation data - Google Patents
Public bicycle peak time demand prediction method considering user reservation data Download PDFInfo
- Publication number
- CN105719019A CN105719019A CN201610039916.7A CN201610039916A CN105719019A CN 105719019 A CN105719019 A CN 105719019A CN 201610039916 A CN201610039916 A CN 201610039916A CN 105719019 A CN105719019 A CN 105719019A
- Authority
- CN
- China
- Prior art keywords
- site
- data
- period
- reservation
- 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 20
- 239000000284 extract Substances 0.000 claims description 20
- 210000003127 knee Anatomy 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 10
- 238000013277 forecasting method Methods 0.000 claims description 9
- 238000011160 research Methods 0.000 claims description 7
- 238000000714 time series forecasting Methods 0.000 claims description 5
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000012546 transfer Methods 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 230000004069 differentiation Effects 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 3
- 238000006467 substitution reaction Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000000556 factor analysis Methods 0.000 claims description 2
- 238000010276 construction Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
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
- 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)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (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)
Abstract
The invention discloses a public bicycle peak time demand prediction method considering user reservation data. The public bicycle peak time demand prediction method is characterized in that 1) user peak time reservation data can be acquired and extracted; 2) the use data of the public bicycle historical peak time can be acquired; 3) the reservation data can be processed, and the reservation demand quantities of different rental stations at different times in the next day can be acquired; 4) the historical data can be processed, and the demand prediction values of different rental stations at different times in the next day can be acquired; 5) the prediction values can be integrated according to the reservation data and the historical data, and the final borrowing and returning demand quantities of different rental stations at different times can be determined; 6) the scheduling demand table having the time window can be drawn. The public bicycle peak time demand prediction method is advantageous in that the accuracy of predicting the bicycle and locking pile demands of different public bicycle rental stations at peak times can be effectively improved, and the data having the actual reference value can be provided for the peak time scheduling.
Description
Technical field
The present invention relates to the technical field of domestic city public bicycles system Construction and operation management, refer in particular to a kind of public bicycles peak period needing forecasting method considering user's reservation data.
Background technology
In current city public bicycle system Construction at home and operation management, requirement forecasting all plays very important effect.In the outlet planning link of system Construction, generally need to consider the factors such as regional function type, regional land use character, service population scale and bicycle working strength, adopt the different Forecasting Methodology such as Logit model, transfer amount reckoning that region public bicycles demand is predicted, so that it is determined that the layout of region public bicycles and the scale of input.Though the result predicting out by these methods can not be said very accurate, but substantially can meet the demand that in planning year survey region, resident's public bicycles is daily gone on a journey.But, from the actual operational effect of public bicycles system of domestic different cities, still quite severe by means of car difficulty difficult problem of returning the car, particularly in peak period morning and evening, the specific node real-time requirement prediction essentially consisting in Relative Fuzzy of tracing it to its cause.
Public bicycles system is as the end subsidiary traffic means solving " last one kilometer ", and in one sky, the demand characteristic of different periods is not quite similar, and this main and urbanite trip custom is closely bound up.Experience have shown that, part is leased a little, as being distributed near residential quarter, tidal phenomena is obvious, intrinsic quantity in stock is difficult to meet the demand of peak period of sooner or later travelling frequently, and is therefore scheduled to for the key link of public bicycles system normal operation, and dispatches it is necessary to have requirement forecasting data accurately do basis efficiently, but there is bigger difference in current domestic existing public bicycles system, seriously constrain the spread of public bicycles in the requirement forecasting of site.
The public bicycles system of domestic several big typical urban (Hangzhou, Guangzhou, Zhuzhou, Beijing) is all monitor the real-time bicycle parking quantity of each website according to the website appearance of vehicle amount early warning system preset, when the bicycle quantity of site is above or below certain threshold value, command centre will allocate in time, in addition, at some special points (returning the car frequency height as borrowed) or in particular cases, the difference scheduling mode such as emergency scheduling, artificial allotment, tide allotment, transregional adjustment is also set up.It is found that these scheduling are not based on requirement forecasting, there is serious hysteresis quality, cause that peak period, ubiquity borrowed car difficulty to return the car difficult problem.Therefore, need a kind of solid needing forecasting method badly, make the efficiently scheduling of public bicycles system be possibly realized.
Summary of the invention
It is an object of the invention to overcome the shortcoming and defect of prior art, a kind of public bicycles peak period needing forecasting method considering user's reservation data is provided, public bicycles each site peak period vehicle and car locking stake requirement forecasting accuracy can be effectively improved, conscientiously provide the data with reference value for peak period scheduling.
For achieving the above object, technical scheme provided by the present invention is: a kind of public bicycles peak period needing forecasting method considering user's reservation data, comprises the following steps:
1) gather and extract user's reservation data peak period: gathering the plan of travel data of public bicycles user's peak period next day period according to mobile phone A PP or wechat public number, extract the departure time section of reservation, starting point and point of destination these about data；
2) gather public bicycles history and use data peak period: swipe the card record according to each site history, extract borrowing car data, going back the time point of car data and correspondence of history public bicycles peak period；
3) reservation data is processed, draw the reservation demand of difference next day site each time period: the first step is according to the starting point in reservation data and point of destination coordinate, in conjunction with the average overall travel speed of urban road non-linear coefficient and public bicycles, the time period that the actual range of estimation origin and destination and client arrive at；Second step is on the basis of data with existing, the time period in units of ten minutes, it is determined that the demand of site each time period；
4) processing historical data, draw the requirement forecasting value of difference next day site each time period: use data according to the history above extracted, adopt time series forecasting, to the next day same time period, the demand of each site is predicted；
5) predictive value drawn according to reservation data and historical data is integrated, determine that the final of different site each time period borrows the amount of there is also a need for: by step 3) the reservation amount and the step 4 that obtain) predictive value that obtains compares, if reservation amount is more than the predictive value after historical data analysis, then adopt reservation amount as requirement forecasting result；If reservation amount is less than historical data predictive value, take the numerical value between reservation amount and predictive value as requirement forecasting result；
6) draw the dispatching requirement table of free window: according to step 5) predict the outcome, public bicycles and car locking stake quantity situation in conjunction with each site, the dispatching requirement amount of this peak period, site unit interval section it is converted into by formula, and consider user's acceptable waiting time, draw the dispatching requirement table of free window, provide foundation for follow-up traffic control.
In step 1) in gather and extract user's reservation data peak period key step have:
1.1) user logs in mobile phone A PP or wechat public number, and the departure place that selects to use peak period next day public bicycles to go on a journey, set out period and place of arrival；The origin and destination that system selects according to user, transfer public bicycles dot data around, and recommend nearest site to select for user；
1.2) system background is according to reservation data, extracts set out accordingly period, set out site and purpose site；
In step 2) in gather public bicycles history and use data mainly through transferring system background historical record data peak period, extract borrowing car data, going back the time point of car data and correspondence of user's corresponding period；
In step 3) middle process reservation data, show that the key step of the reservation demand of difference next day site each time period has:
3.1) coordinate according to set out in reservation data site and purpose site, in conjunction with urban road non-linear coefficient, calculates the distance of riding between two sites；
R is road non-linear coefficient, (xn,yn) for site coordinate
3.2) utilize average riding speed value, calculate the time period arriving purpose site；
3.3) time period in units of ten minutes, determining day part time interval successively, screen the trip period of each site and arrive the period, the period of setting out belonging to this time interval illustrates that there is public bicycles vehicle demand this site once, remembers+1；The time of advent belonging to this period illustrates that there is city bicycle lock knee demand this site once, remembers-1；Statistics site i is at vehicle/knee demand R (i, Δ t) of Δ t time period；
In step 4) middle process historical data, show that the key step of the requirement forecasting value of difference next day site each time period has:
4.1) with step 3) time period consistent, the historical data collected is carried out classified statistic, caution area divides the situation on working day and nonworkdays, and rejects part abnormal data；
4.2) data that statistics obtains are weaved into time series, and be depicted as broken line graph according to time series；
4.3) comprehensive long-term trend and Season Factor Analysis, uses ARIMA model in SPSS that time series is analyzed prediction, obtains Demand Forecast value H (i, Δ t) of site i in peak period Δ t period next day；
If site is more in research range, then advise that the land character type of differentiation public bicycles site and region carry out unifying prediction；
In step 5) in integrate the predictive value that draws according to reservation data and historical data, it is determined that the final demand amount of different site each time periods, realize mainly through in the following manner:
Comparison step 3) the reservation demand R (i, Δ t) and step 4) that obtains obtain historical data predictive value H (i, Δ t),
If R is (i, Δ t) >=H (i, Δ t)
O (i, Δ t)=R (i, Δ t) × α
If R is (i, Δ t) < H (i, Δ t)
In formula: (i, the public bicycles vehicle predicted required amount of site i in Δ t) the Δ t period, negative value represents knee demand to O；
The non-subscriber of α, β reserves coefficient, and initial value takes 1.1～1.2, revises with historical data is accumulative；
The use initial stage, R (i, Δ t) < H (i, during Δ t), and O (i, Δ t) takes average, with the popularization of the accumulative of historical data and subscription services, it is weighted meansigma methods and is adjusted, near reservation demand value；
In step 6) in draw the key step of dispatching requirement table of free window and have:
6.1) gather peak period and start the bicycle vehicle number of front each site, by step 5) the substitution dispatching requirement transformation model that predicts the outcome, try to achieve the dispatching requirement amount of each site different time sections；
V (i, Δ t)=O (i, Δ t)-Pi
In formula:
(in the public bicycles vehicle scheduling demand of Δ t period, negative value represents knee demand to i, Δ t) site i to V；
PiPeak period starts the public bicycles quantity of front site i；
6.2) scheduling time window is determined: choose the 3rd minute of the time period that abovementioned steps is determined as the moment (t+3) arriving site i of dispatching buses, wait in conjunction with user's acceptable and picking up the car the time, front and back increase by 5 minutes surpluses altogether as scheduling time window interval (t-2, t+8)；
6.3) the dispatching requirement table of foundation number node, dispatching requirement amount, the free window of these data organizations of scheduling time window, offer foundation is dispatched in the peak period for public bicycles.
The present invention compared with prior art, has the advantage that and beneficial effect:
The present invention is by by mobile terminal, utilize Internet technology, gather and extract the reservation data that user is gone on a journey by public bicycles, secondly segment processing, determine time window, then gather the history arranging the same period and use data, finally by input backstage integrated treatment, obtain the real-time requirement amount of peak period, each site, thus knowing follow-up traffic control, so can be effectively improved public bicycles each site peak period vehicle and car locking stake requirement forecasting accuracy, conscientiously provide the data with reference value for peak period scheduling.
Accompanying drawing explanation
Fig. 1 is the public bicycles peak period needing forecasting method flow chart of the present invention.
Fig. 2 preengages flow chart for user.
Fig. 3 is reservation data process chart.
Fig. 4 a is the China scape new city amount of hiring a car time series forecasting figure.
Fig. 4 b is the China scape new city amount of returning the car time series forecasting figure.
Detailed description of the invention
Below in conjunction with specific embodiment, the invention will be further described.
As it is shown in figure 1, public bicycles peak period of the present invention needing forecasting method, comprise the following steps:
1) gather and extract user's reservation data peak period.Gather the plan of travel data of public bicycles user's peak period next day period according to modes such as mobile phone A PP or wechat public number, extract the relevant data such as the departure time section of reservation, starting point and point of destination.
2) gather public bicycles history and use data peak period.Swipe the card record according to each site history, extract borrowing car data, going back the time point of car data and correspondence of history public bicycles peak period.
3) process reservation data, draw the reservation demand of difference next day site each time period.The first step is according to the starting point in reservation data and point of destination coordinate, in conjunction with the average overall travel speed of urban road non-linear coefficient and public bicycles, the time period that the actual range of estimation origin and destination and client arrive at；Second step be data with existing (starting point, point of destination, departure time section, the time of advent section) basis on, the time period in units of ten minutes, it is determined that the demand of site each time period.
4) process historical data, draw the requirement forecasting value of difference next day site each time period.Using data according to the history above extracted, adopt time series forecasting, to the next day same time period, the demand of each site is predicted.
5) predictive value drawn according to reservation data and historical data is integrated, it is determined that the final of different site each time periods borrows the amount of there is also a need for.By step 3) the reservation amount that obtains and step 4) predictive value that obtains compares, if reservation amount is more than the predictive value after historical data analysis, then adopts reservation amount as requirement forecasting result；If reservation amount is less than historical data predictive value, for lowering the impact on the data precision of the software utilization rate, take the numerical value between reservation amount and predictive value as requirement forecasting result.
6) the dispatching requirement table of free window is drawn.According to step 5) predict the outcome, public bicycles and car locking stake quantity situation in conjunction with each site, the dispatching requirement amount of this peak period, site unit interval section it is converted into by formula, and consider user's acceptable waiting time, draw the dispatching requirement table of free window, provide foundation for follow-up traffic control.
As in figure 2 it is shown, in step 1) in gather and extract the key step of user's reservation data peak period and have:
1.1) user logs in the platform such as mobile phone A PP or wechat public number, and the departure place that selects to use peak period next day public bicycles to go on a journey, set out period and place of arrival.The origin and destination that system selects according to user, transfer public bicycles dot data around, and recommend nearest site to select for user.
1.2) system background is according to reservation data, extracts set out accordingly period, set out site and purpose site.
Step 2) in gather public bicycles history and use data mainly through transferring system background historical record data peak period, extract borrowing car data, going back the time point of car data and correspondence of user's corresponding period.
As it is shown on figure 3, step 3) middle process reservation data, show that the key step of the reservation demand of difference next day site each time period has:
3.1) coordinate according to set out in reservation data site and purpose site, in conjunction with urban road non-linear coefficient, calculates the distance of riding between two sites；
R is road non-linear coefficient, (xn,yn) for site coordinate
3.2) utilize average riding speed value, calculate the time period arriving purpose site；
3.3) time period in units of ten minutes, determining day part time interval successively, screen the trip period of each site and arrive the period, the period of setting out belonging to this time interval illustrates that there is public bicycles vehicle demand this site once, remembers+1；The time of advent belonging to this period illustrates that there is city bicycle lock knee demand this site once, remembers-1.Statistics site i is at vehicle/knee demand R (i, Δ t) of Δ t time period.
In step 4) middle process historical data, show that the key step of the requirement forecasting value of difference next day site each time period has:
4.1) with step 3) time period consistent, the historical data collected is carried out classified statistic, caution area divides the situation on working day and nonworkdays, and rejects part abnormal data；
4.2) data that statistics obtains are weaved into time series, and be depicted as broken line graph according to time series；
4.3) factor such as comprehensive long-term trend and seasonal move, uses ARIMA model in SPSS that time series is analyzed prediction, obtains Demand Forecast value H (i, Δ t) of site i in peak period Δ t period next day.
If site is more in research range, then advise that the land character type of differentiation public bicycles site and region carry out unifying prediction.
In step 5) in integrate the predictive value that draws according to reservation data and historical data, it is determined that the final demand amount of different site each time periods.Realize mainly through in the following manner:
Comparison step 3) the reservation demand R (i, Δ t) and step 4) that obtains obtain historical data predictive value H (i, Δ t),
If R is (i, Δ t) >=H (i, Δ t)
O (i, Δ t)=R (i, Δ t) × α
If R is (i, Δ t) < H (i, Δ t)
In formula: O (i, the public bicycles vehicle predicted required amount (negative value represents knee demand) of site i in Δ t) the Δ t period；
The non-subscriber of α, β reserves coefficient, initial value desirable 1.1～1.2, revises with historical data is accumulative；
Initial stage, R (i, Δ t) < H (i is used in scheme, during Δ t), O (the desirable average of i, Δ t), with the popularization of the accumulative of historical data and subscription services, meansigma methods can be weighted and be adjusted, suitably near reservation demand value.
In step 6) in draw the key step of dispatching requirement table of free window and have:
6.1) gather peak period and start the bicycle vehicle number of front each site, by step 5) the substitution dispatching requirement transformation model that predicts the outcome, try to achieve the dispatching requirement amount of each site different time sections.
V (i, Δ t)=O (i, Δ t)-Pi
In formula:
(i, Δ t) site i is in the public bicycles vehicle scheduling demand (negative value represents knee demand) of Δ t period for V；
PiPeak period starts the public bicycles quantity (monitoring in 30 minutes in advance obtains data or adopts historical average evidence) of front site i.
6.2) scheduling time window is determined.Choose the 3rd minute of the time period that abovementioned steps is determined as the moment (t+3) arriving site i of dispatching buses, pick up the car the time in conjunction with the acceptable wait of user, 5 minutes surpluses can be increased altogether as scheduling time window interval (t-2, t+8) in front and back.
6.3) according to the dispatching requirement table of the free windows of data organization such as number node, dispatching requirement amount, scheduling time window, offer foundation is dispatched in the peak period for public bicycles.
The dispatching requirement table of the free window prepared is as shown in table 1 below.
The dispatching requirement table of the free window of table 1
Analysis of cases
Choose the magnificent scape new city of Tianhe District, happy and carefree flower garden, Shang She, province's postal school, Tianhe Park north gate, east sunshine garden, mansion, Huajian, Tianhe District government, Tianhe Park west gate, ten public bicycles lease points of cyber port as object of study, the demand Forecasting Methodology is carried out analysis of cases.(note: to research site from 1-10 number consecutively)
1) gather and extract reservation data；Owing to condition limits, the reservation data of this research is by Excel stochastic generation.
2) gather history and use data；The historical data of this research, from extracting with public bicycles system, amounts to trimestral historical record data.
3) processing reservation data, show that the reservation demand of difference next day site each time period is as shown in table 2 below, wherein T represents the time period, and B represents reservation amount of hiring a car, and R represents the amount of returning the car.
The reservation amount of table 2 different site each time period
4) process historical data, draw the requirement forecasting value of difference next day site each time period.The main ARIMA model used in SPSS of this research is predicted, and as shown in figures 4 a and 4b, is the scape new city to China schematic diagram that carries out time series analysis, in like manner obtains the predictive value of other site, as shown in table 3 below.
The predictive value of table 3 different site each time period
5) predictive value drawn according to reservation data and historical data is integrated, it is determined that the final of different site each time periods borrows the amount of there is also a need for, and as shown in table 4 below, α, β take 1.2.
What table 4 integrated reservation amount and predictive value borrows the amount of there is also a need for
6) the dispatching requirement table of free window is drawn.
Assume that the public bicycles quantity of first three ten minutes each site, peak period is followed successively by (5,7,10,8,5,9,8,5,8,5), then dispatching requirement table is as shown in table 5 below, and wherein TW represents scheduling time window, and V represents dispatching requirement amount.
The dispatching requirement table of the free window of table 5
The examples of implementation of the above are only the preferred embodiments of the invention, not limit the practical range of the present invention with this, therefore all changes made according to the shape of the present invention, principle, all should be encompassed in protection scope of the present invention.
Claims (2)
1. the public bicycles peak period needing forecasting method considering user's reservation data, it is characterised in that comprise the following steps:
1) gather and extract user's reservation data peak period: gathering the plan of travel data of public bicycles user's peak period next day period according to mobile phone A PP or wechat public number, extract the departure time section of reservation, starting point and point of destination these about data；
2) gather public bicycles history and use data peak period: swipe the card record according to each site history, extract borrowing car data, going back the time point of car data and correspondence of history public bicycles peak period；
3) reservation data is processed, draw the reservation demand of difference next day site each time period: the first step is according to the starting point in reservation data and point of destination coordinate, in conjunction with the average overall travel speed of urban road non-linear coefficient and public bicycles, the time period that the actual range of estimation origin and destination and client arrive at；Second step is on the basis of data with existing, the time period in units of ten minutes, it is determined that the demand of site each time period；
4) processing historical data, draw the requirement forecasting value of difference next day site each time period: use data according to the history above extracted, adopt time series forecasting, to the next day same time period, the demand of each site is predicted；
5) predictive value drawn according to reservation data and historical data is integrated, determine that the final of different site each time period borrows the amount of there is also a need for: by step 3) the reservation amount and the step 4 that obtain) predictive value that obtains compares, if reservation amount is more than the predictive value after historical data analysis, then adopt reservation amount as requirement forecasting result；If reservation amount is less than historical data predictive value, take the numerical value between reservation amount and predictive value as requirement forecasting result；
6) draw the dispatching requirement table of free window: according to step 5) predict the outcome, public bicycles and car locking stake quantity situation in conjunction with each site, the dispatching requirement amount of this peak period, site unit interval section it is converted into by formula, and consider user's acceptable waiting time, draw the dispatching requirement table of free window, provide foundation for follow-up traffic control.
2. a kind of public bicycles peak period needing forecasting method considering user's reservation data according to claim 1, it is characterised in that: in step 1) in gather and extract user's reservation data peak period key step have:
1.1) user logs in mobile phone A PP or wechat public number, and the departure place that selects to use peak period next day public bicycles to go on a journey, set out period and place of arrival；The origin and destination that system selects according to user, transfer public bicycles dot data around, and recommend nearest site to select for user；
1.2) system background is according to reservation data, extracts set out accordingly period, set out site and purpose site；
In step 2) in gather public bicycles history and use data mainly through transferring system background historical record data peak period, extract borrowing car data, going back the time point of car data and correspondence of user's corresponding period；
In step 3) middle process reservation data, show that the key step of the reservation demand of difference next day site each time period has:
3.1) coordinate according to set out in reservation data site and purpose site, in conjunction with urban road non-linear coefficient, calculates the distance of riding between two sites；
R is road non-linear coefficient, (xn,yn) for site coordinate
3.2) utilize average riding speed value, calculate the time period arriving purpose site；
3.3) time period in units of ten minutes, determining day part time interval successively, screen the trip period of each site and arrive the period, the period of setting out belonging to this time interval illustrates that there is public bicycles vehicle demand this site once, remembers+1；The time of advent belonging to this period illustrates that there is city bicycle lock knee demand this site once, remembers-1；Statistics site i is at vehicle/knee demand R (i, Δ t) of Δ t time period；
In step 4) middle process historical data, show that the key step of the requirement forecasting value of difference next day site each time period has:
4.1) with step 3) time period consistent, the historical data collected is carried out classified statistic, caution area divides the situation on working day and nonworkdays, and rejects part abnormal data；
4.2) data that statistics obtains are weaved into time series, and be depicted as broken line graph according to time series；
4.3) comprehensive long-term trend and Season Factor Analysis, uses ARIMA model in SPSS that time series is analyzed prediction, obtains Demand Forecast value H (i, Δ t) of site i in peak period Δ t period next day；
If site is more in research range, then advise that the land character type of differentiation public bicycles site and region carry out unifying prediction；
In step 5) in integrate the predictive value that draws according to reservation data and historical data, it is determined that the final demand amount of different site each time periods, realize mainly through in the following manner:
Comparison step 3) the reservation demand R (i, Δ t) and step 4) that obtains obtain historical data predictive value H (i, Δ t),
If R is (i, Δ t) >=H (i, Δ t)
O (i, Δ t)=R (i, Δ t) × α
If R is (i, Δ t) < H (i, Δ t)
In formula: (i, the public bicycles vehicle predicted required amount of site i in Δ t) the Δ t period, negative value represents knee demand to O；
The non-subscriber of α, β reserves coefficient, and initial value takes 1.1～1.2, revises with historical data is accumulative；
The use initial stage, R (i, Δ t) < H (i, during Δ t), and O (i, Δ t) takes average, with the popularization of the accumulative of historical data and subscription services, it is weighted meansigma methods and is adjusted, near reservation demand value；
In step 6) in draw the key step of dispatching requirement table of free window and have:
6.1) gather peak period and start the bicycle vehicle number of front each site, by step 5) the substitution dispatching requirement transformation model that predicts the outcome, try to achieve the dispatching requirement amount of each site different time sections；
V (i, Δ t)=O (i, Δ t)-Pi
In formula:
(in the public bicycles vehicle scheduling demand of Δ t period, negative value represents knee demand to i, Δ t) site i to V；
PiPeak period starts the public bicycles quantity of front site i；
6.2) scheduling time window is determined: choose the 3rd minute of the time period that abovementioned steps is determined as the moment (t+3) arriving site i of dispatching buses, wait in conjunction with user's acceptable and picking up the car the time, front and back increase by 5 minutes surpluses altogether as scheduling time window interval (t-2, t+8)；
6.3) the dispatching requirement table of foundation number node, dispatching requirement amount, the free window of these data organizations of scheduling time window, offer foundation is dispatched in the peak period for public bicycles.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610039916.7A CN105719019A (en) | 2016-01-21 | 2016-01-21 | Public bicycle peak time demand prediction method considering user reservation data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610039916.7A CN105719019A (en) | 2016-01-21 | 2016-01-21 | Public bicycle peak time demand prediction method considering user reservation data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105719019A true CN105719019A (en) | 2016-06-29 |
Family
ID=56153682
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610039916.7A Pending CN105719019A (en) | 2016-01-21 | 2016-01-21 | Public bicycle peak time demand prediction method considering user reservation data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105719019A (en) |
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106327871A (en) * | 2016-09-06 | 2017-01-11 | 华南理工大学 | Highway congestion forecasting method based on historical data and reservation data |
| CN106529703A (en) * | 2016-10-31 | 2017-03-22 | 浙江大学 | Executive right distribution-based urban traffic peak shifting method |
| CN106705981A (en) * | 2017-01-03 | 2017-05-24 | 上海量明科技发展有限公司 | Method for sharing communication media data statistics, terminal and system |
| CN107092973A (en) * | 2016-11-25 | 2017-08-25 | 口碑控股有限公司 | The Forecasting Methodology and device of a kind of portfolio |
| CN107103506A (en) * | 2017-03-31 | 2017-08-29 | 芜湖泰领信息科技有限公司 | The decision method of the heavy duty zone of public bicycles |
| CN107103388A (en) * | 2017-04-26 | 2017-08-29 | 杨毅 | Robot scheduling System and method for based on requirement forecasting |
| CN107145714A (en) * | 2017-04-07 | 2017-09-08 | 浙江大学城市学院 | Based on multifactor public bicycles usage amount Forecasting Methodology |
| CN107633680A (en) * | 2016-07-12 | 2018-01-26 | 阿里巴巴集团控股有限公司 | Acquisition methods, device, equipment and the system of trip data |
| CN107798446A (en) * | 2016-09-07 | 2018-03-13 | 北京嘀嘀无限科技发展有限公司 | The evaluation process method and device of rich ore route |
| CN108053053A (en) * | 2017-11-21 | 2018-05-18 | 东南大学 | The non-commuting travel forecasting approaches of electric bicycle based on space autoregression model |
| CN108062855A (en) * | 2017-12-18 | 2018-05-22 | 北京物资学院 | The method and system of peak time of using cars section vehicle scheduling |
| CN108182170A (en) * | 2017-11-21 | 2018-06-19 | 东南大学 | Electric bicycle commuter Forecasting Methodology based on space autoregression preference pattern |
| WO2018126367A1 (en) * | 2017-01-04 | 2018-07-12 | 上海温尔信息科技有限公司 | Data cleaning method and device |
| CN108346010A (en) * | 2018-04-23 | 2018-07-31 | 徐漫洋 | Shared Truck dispartching method based on user requirements analysis |
| CN108460537A (en) * | 2018-03-23 | 2018-08-28 | 华南理工大学 | A method of bicycle demand is shared based on big data and user psychology prediction |
| CN108564225A (en) * | 2018-04-24 | 2018-09-21 | 徐漫洋 | Shared automobile scheduling system and dispatching method |
| CN108596484A (en) * | 2018-04-24 | 2018-09-28 | 徐漫洋 | A kind of shared Truck dispartching method and system based on site attributive analysis |
| CN108960590A (en) * | 2018-06-15 | 2018-12-07 | 平安科技（深圳）有限公司 | Vehicle leasing method, apparatus, computer equipment and storage medium |
| CN110033161A (en) * | 2019-02-27 | 2019-07-19 | 山东开创云软件有限公司 | A kind of dispatching method and system of shared automobile |
| CN110059839A (en) * | 2019-04-25 | 2019-07-26 | 北京首汽智行科技有限公司 | A kind of method that need-based extracts unmet demand |
| CN110191184A (en) * | 2019-05-31 | 2019-08-30 | 国信优易数据有限公司 | A kind of duration information method for pushing, device, equipment and medium |
| CN110807921A (en) * | 2019-10-24 | 2020-02-18 | 上海钧正网络科技有限公司 | Vehicle scheduling method, device, equipment and storage medium |
| CN110969834A (en) * | 2019-11-19 | 2020-04-07 | 上海钧正网络科技有限公司 | Processing method, processing device, computer equipment, storage medium and vehicle dispatching system |
| CN111178948A (en) * | 2019-12-18 | 2020-05-19 | 同济大学 | Method for realizing sharing of dynamic automobile borrowing |
| CN111383055A (en) * | 2020-03-12 | 2020-07-07 | 北京嘀嘀无限科技发展有限公司 | Method, device and equipment for selecting products in center cabin of taxi and computer readable storage medium |
| CN111625375A (en) * | 2020-05-22 | 2020-09-04 | 腾讯科技（深圳）有限公司 | Account reservation method and device, storage medium and electronic equipment |
| CN111882171A (en) * | 2020-07-06 | 2020-11-03 | 湖北合纵科创能源科技有限公司 | Intelligent scheduling system and method for scenic spot sightseeing vehicle based on demand prediction |
| CN111882158A (en) * | 2020-06-24 | 2020-11-03 | 东南大学 | Mixed public bicycle scheduling demand prediction method based on Voronoi diagram |
| US10859387B2 (en) | 2016-04-27 | 2020-12-08 | Beijing Didi Infinity Technology And Development Co., Ltd. | System and method for determining routes of transportation service |
| CN113065915A (en) * | 2021-03-17 | 2021-07-02 | 北京首汽智行科技有限公司 | Car-arranging method for sharing cars |
| CN113095636A (en) * | 2021-03-25 | 2021-07-09 | 深圳前海联动云软件科技有限公司 | Intelligent scheduling system and method for fuel sharing automobile |
| CN114595953A (en) * | 2022-03-01 | 2022-06-07 | 杭州骑迹科技有限公司 | Method and system for dispatching shared electric bicycle |
| CN115879016A (en) * | 2023-02-20 | 2023-03-31 | 中南大学 | Prediction method for shared bicycle trip tide time interval |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104464104A (en) * | 2014-11-03 | 2015-03-25 | 南京邮电大学 | Public bicycle ordering, renting and returning system and realizing method thereof |
| CN104573830A (en) * | 2014-12-17 | 2015-04-29 | 北京石油化工学院 | Intelligent terminal and system for subscribing and renting bicycle |
| CN104636828A (en) * | 2015-02-02 | 2015-05-20 | 西南交通大学 | Public bicycle station supply and demand prediction method based on Markov chain |
-
2016
- 2016-01-21 CN CN201610039916.7A patent/CN105719019A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104464104A (en) * | 2014-11-03 | 2015-03-25 | 南京邮电大学 | Public bicycle ordering, renting and returning system and realizing method thereof |
| CN104573830A (en) * | 2014-12-17 | 2015-04-29 | 北京石油化工学院 | Intelligent terminal and system for subscribing and renting bicycle |
| CN104636828A (en) * | 2015-02-02 | 2015-05-20 | 西南交通大学 | Public bicycle station supply and demand prediction method based on Markov chain |
Non-Patent Citations (1)
| Title |
|---|
| 焦云涛: "公共自行车借还特性分析及调度模型研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (46)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10859387B2 (en) | 2016-04-27 | 2020-12-08 | Beijing Didi Infinity Technology And Development Co., Ltd. | System and method for determining routes of transportation service |
| CN107633680A (en) * | 2016-07-12 | 2018-01-26 | 阿里巴巴集团控股有限公司 | Acquisition methods, device, equipment and the system of trip data |
| CN106327871B (en) * | 2016-09-06 | 2018-09-14 | 华南理工大学 | A kind of crowded prediction technique of highway of fusion historical data and reservation data |
| CN106327871A (en) * | 2016-09-06 | 2017-01-11 | 华南理工大学 | Highway congestion forecasting method based on historical data and reservation data |
| CN107798446A (en) * | 2016-09-07 | 2018-03-13 | 北京嘀嘀无限科技发展有限公司 | The evaluation process method and device of rich ore route |
| CN106529703A (en) * | 2016-10-31 | 2017-03-22 | 浙江大学 | Executive right distribution-based urban traffic peak shifting method |
| CN107092973A (en) * | 2016-11-25 | 2017-08-25 | 口碑控股有限公司 | The Forecasting Methodology and device of a kind of portfolio |
| US11443251B2 (en) | 2016-11-25 | 2022-09-13 | Koubei Holding Limited | Method and device for predicting traffic |
| CN106705981B (en) * | 2017-01-03 | 2023-10-24 | 上海量明科技发展有限公司 | Method, terminal and system for sharing vehicle data statistics |
| CN106705981A (en) * | 2017-01-03 | 2017-05-24 | 上海量明科技发展有限公司 | Method for sharing communication media data statistics, terminal and system |
| WO2018126367A1 (en) * | 2017-01-04 | 2018-07-12 | 上海温尔信息科技有限公司 | Data cleaning method and device |
| CN107103506A (en) * | 2017-03-31 | 2017-08-29 | 芜湖泰领信息科技有限公司 | The decision method of the heavy duty zone of public bicycles |
| CN107145714B (en) * | 2017-04-07 | 2020-05-22 | 浙江大学城市学院 | Multi-factor-based public bicycle usage amount prediction method |
| CN107145714A (en) * | 2017-04-07 | 2017-09-08 | 浙江大学城市学院 | Based on multifactor public bicycles usage amount Forecasting Methodology |
| CN107103388A (en) * | 2017-04-26 | 2017-08-29 | 杨毅 | Robot scheduling System and method for based on requirement forecasting |
| CN108182170A (en) * | 2017-11-21 | 2018-06-19 | 东南大学 | Electric bicycle commuter Forecasting Methodology based on space autoregression preference pattern |
| CN108053053A (en) * | 2017-11-21 | 2018-05-18 | 东南大学 | The non-commuting travel forecasting approaches of electric bicycle based on space autoregression model |
| CN110930687B (en) * | 2017-12-18 | 2020-12-25 | 北京物资学院 | Method and system for dispatching vehicles in peak vehicle time period |
| CN110930687A (en) * | 2017-12-18 | 2020-03-27 | 北京物资学院 | Method and system for dispatching vehicles in peak vehicle time period |
| CN108062855A (en) * | 2017-12-18 | 2018-05-22 | 北京物资学院 | The method and system of peak time of using cars section vehicle scheduling |
| CN108062855B (en) * | 2017-12-18 | 2020-03-10 | 北京物资学院 | Method and system for dispatching vehicles in peak vehicle time period |
| CN108460537A (en) * | 2018-03-23 | 2018-08-28 | 华南理工大学 | A method of bicycle demand is shared based on big data and user psychology prediction |
| CN108346010A (en) * | 2018-04-23 | 2018-07-31 | 徐漫洋 | Shared Truck dispartching method based on user requirements analysis |
| CN108346010B (en) * | 2018-04-23 | 2021-07-13 | 徐漫洋 | Shared automobile scheduling method based on user demand analysis |
| CN108564225A (en) * | 2018-04-24 | 2018-09-21 | 徐漫洋 | Shared automobile scheduling system and dispatching method |
| CN108596484A (en) * | 2018-04-24 | 2018-09-28 | 徐漫洋 | A kind of shared Truck dispartching method and system based on site attributive analysis |
| CN108960590A (en) * | 2018-06-15 | 2018-12-07 | 平安科技（深圳）有限公司 | Vehicle leasing method, apparatus, computer equipment and storage medium |
| CN110033161A (en) * | 2019-02-27 | 2019-07-19 | 山东开创云软件有限公司 | A kind of dispatching method and system of shared automobile |
| CN110059839A (en) * | 2019-04-25 | 2019-07-26 | 北京首汽智行科技有限公司 | A kind of method that need-based extracts unmet demand |
| CN110191184B (en) * | 2019-05-31 | 2022-02-01 | 国信优易数据股份有限公司 | Duration information pushing method, device, equipment and medium |
| CN110191184A (en) * | 2019-05-31 | 2019-08-30 | 国信优易数据有限公司 | A kind of duration information method for pushing, device, equipment and medium |
| CN110807921A (en) * | 2019-10-24 | 2020-02-18 | 上海钧正网络科技有限公司 | Vehicle scheduling method, device, equipment and storage medium |
| CN110969834A (en) * | 2019-11-19 | 2020-04-07 | 上海钧正网络科技有限公司 | Processing method, processing device, computer equipment, storage medium and vehicle dispatching system |
| CN111178948A (en) * | 2019-12-18 | 2020-05-19 | 同济大学 | Method for realizing sharing of dynamic automobile borrowing |
| CN111178948B (en) * | 2019-12-18 | 2023-08-04 | 同济大学 | Method for realizing dynamic borrowing of shared automobile |
| CN111383055A (en) * | 2020-03-12 | 2020-07-07 | 北京嘀嘀无限科技发展有限公司 | Method, device and equipment for selecting products in center cabin of taxi and computer readable storage medium |
| CN111383055B (en) * | 2020-03-12 | 2023-11-17 | 北京嘀嘀无限科技发展有限公司 | Method, device and equipment for selecting car renting center bin and computer readable storage medium |
| CN111625375A (en) * | 2020-05-22 | 2020-09-04 | 腾讯科技（深圳）有限公司 | Account reservation method and device, storage medium and electronic equipment |
| CN111625375B (en) * | 2020-05-22 | 2023-10-24 | 腾讯科技（深圳）有限公司 | Account reservation method and device, storage medium and electronic equipment |
| CN111882158A (en) * | 2020-06-24 | 2020-11-03 | 东南大学 | Mixed public bicycle scheduling demand prediction method based on Voronoi diagram |
| CN111882171A (en) * | 2020-07-06 | 2020-11-03 | 湖北合纵科创能源科技有限公司 | Intelligent scheduling system and method for scenic spot sightseeing vehicle based on demand prediction |
| CN113065915A (en) * | 2021-03-17 | 2021-07-02 | 北京首汽智行科技有限公司 | Car-arranging method for sharing cars |
| CN113095636A (en) * | 2021-03-25 | 2021-07-09 | 深圳前海联动云软件科技有限公司 | Intelligent scheduling system and method for fuel sharing automobile |
| CN113095636B (en) * | 2021-03-25 | 2024-01-23 | 深圳前海联动云软件科技有限公司 | Intelligent scheduling system and method for fuel sharing automobile |
| CN114595953A (en) * | 2022-03-01 | 2022-06-07 | 杭州骑迹科技有限公司 | Method and system for dispatching shared electric bicycle |
| CN115879016A (en) * | 2023-02-20 | 2023-03-31 | 中南大学 | Prediction method for shared bicycle trip tide time interval |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105719019A (en) | Public bicycle peak time demand prediction method considering user reservation data | |
| CN103632212B (en) | System and method for predicating time-varying user dynamic equilibrium network-evolved passenger flow | |
| CN102332210B (en) | Method for extracting real-time urban road traffic flow data based on mobile phone positioning data | |
| CN102157075B (en) | Method for predicting bus arrivals | |
| CN106503869A (en) | A kind of public bicycles dynamic dispatching method that is predicted based on website short-term needs | |
| CN105427594B (en) | A kind of public transport section volume of the flow of passengers acquisition methods and system based on two-way passenger flow of getting on the bus | |
| CN108491682B (en) | Rainfall runoff forecasting system | |
| CN105303882A (en) | Parking management system and method based on space-time dimension dynamic parking charging strategy | |
| CN110390349A (en) | Bus passenger flow volume based on XGBoost model predicts modeling method | |
| CN106127329A (en) | Order forecast method and device | |
| CN102419580B (en) | Flood forecasting and dispatching system | |
| CN104636828A (en) | Public bicycle station supply and demand prediction method based on Markov chain | |
| CN110956807B (en) | Highway flow prediction method based on combination of multi-source data and sliding window | |
| CN105513359A (en) | Method for estimating city expressway traffic states based on mobile detection of smartphones | |
| CN105809962A (en) | Traffic trip mode splitting method based on mobile phone data | |
| CN102087788A (en) | Method for estimating traffic state parameter based on confidence of speed of float car | |
| CN102169606A (en) | Method for predicting influence of heavy passenger flow of urban rail transit network | |
| CN110379198B (en) | Urban intelligent parking guide system and parking guide method thereof | |
| CN105512741A (en) | Bus passenger traffic combined prediction method | |
| CN114638440B (en) | Charging load ultra-short-term prediction method based on charging pile utilization degree | |
| CN104318081A (en) | Method for allocating bicycles at public bicycle rental stations with urgent demand in city | |
| CN109003128A (en) | Based on improved random forest public bicycles website Demand Forecast method | |
| CN109389243A (en) | In conjunction with the freeway traffic condition survey website distribution method of networked fee collection data | |
| CN106503848A (en) | The load forecasting method of many small power station's bulk sale area power grids | |
| CN113821547B (en) | Rapid and efficient short-time prediction method, system and storage medium for occupancy of parking lot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160629 |