TWI524303B - Forecasting Device and Method of Vehicle Trend Forecasting Based on Large Cloud Data Processing - Google Patents
Forecasting Device and Method of Vehicle Trend Forecasting Based on Large Cloud Data Processing Download PDFInfo
- Publication number
- TWI524303B TWI524303B TW103111378A TW103111378A TWI524303B TW I524303 B TWI524303 B TW I524303B TW 103111378 A TW103111378 A TW 103111378A TW 103111378 A TW103111378 A TW 103111378A TW I524303 B TWI524303 B TW I524303B
- Authority
- TW
- Taiwan
- Prior art keywords
- vehicle
- information
- dispatch
- module
- service area
- Prior art date
Links
Landscapes
- Traffic Control Systems (AREA)
Description
本發明係一種乘車趨勢預測裝置及其方法，尤指一種運用雲端大資料處理之乘車趨勢預測裝置及其方法。 The invention relates to a riding tendency prediction device and a method thereof, in particular to a riding tendency prediction device using a cloud large data processing and a method thereof.
目前計程車隊為了在服務區域範圍內提供良好的載客服務，多透過叫車中心進行車輛之派遣。叫車中心係在收到乘客之叫車需求時，透過定位系統找尋最近之車輛，並傳送指派令至該車輛，以提供載客服務。 In order to provide good passenger services within the service area, the current taxi fleet will dispatch vehicles through the call center. The calling center collects the nearest vehicle through the positioning system when receiving the passenger's demand for the car, and transmits an assignment order to the vehicle to provide the passenger service.
由於目前車隊多透過統計過去叫車記錄以及司機之歷史經驗來決定派遣車輛之分佈位置，此種方式由於無法有效的進行量化，因此派遣車輛在尋找乘客以及到達乘客指定搭乘位置時，無形中會浪費許多油資和時間。 Since the current fleet has decided to distribute the location of the dispatched vehicles by counting past car records and the driver's historical experience, this method cannot be effectively quantified, so the dispatched vehicle will be invisible when searching for passengers and arriving at the passenger's designated riding position. Waste a lot of oil and time.
而當舉行特定活動時，例如：跨年、路跑、嘉年華等活動時，由於會吸引大量的人潮，而若無專門人員緊盯特定之活動資訊並進行車輛調度，常會使該區域之派遣車輛無法取得供需平衡。 When a specific event is held, such as a New Year's Eve, a road run, a carnival, etc., it will attract a large number of people, and if there is no special personnel to keep a close eye on specific activity information and carry out vehicle dispatch, it will often send vehicles in the area. Unable to balance supply and demand.
綜上所述，習知之車輛派遣系統由於無法透過量化的方式來預測特定服務區域之派遣車輛數，使得派遣車輛在調度上常會有不必要油資以及時間成本損耗，因此，提供一種可預測特定派遣服務區域之派遣數量之裝置及其方法乃本領域亟需解決之技術問題。 In summary, the conventional vehicle dispatch system can not predict the number of dispatched vehicles in a specific service area through quantitative methods, so that dispatch vehicles often have unnecessary oil and time cost loss in dispatching, thus providing a predictable specific The device and method of dispatching the number of dispatched service areas are technical problems that need to be solved in the field.
為解決前揭習知技術之技術問題，本發明之一目的係提供一種用來預測特定服務區域乘車需求之裝置及其方法，以能有效率的進行車輛派遣作業。 In order to solve the technical problems of the prior art, it is an object of the present invention to provide an apparatus and method for predicting a ride demand in a specific service area, so as to efficiently perform a vehicle dispatch operation.
為達上述之目的，本發明提供一種乘車趨勢預測裝置。乘車趨勢預測裝置包含一狀態擷取模組、一派車資訊模組、一服務區需求預測模組、一主動推播模組以及一雲端大資料平行運算處理模組。雲端大資料平行運算處理模組係連接前述之狀態擷取模組、派車資訊模組、服務區需求預測模組，以及主動推播模組，並提供連接之模組運算服務。狀態擷取模組係自外部網際網路擷取活動資訊以及氣象資訊。活動資訊為記錄位於一服務區域之活動，而氣象資訊，則記錄位於服務區域之氣象狀態。派車資訊模組係提供一描述服務區域之派遣車輛之分佈情形派遣車輛分佈資訊。而服務區需求預測模組係連接狀態擷取模組以及派車資訊模組，服務區需求預測模組更依據活動資訊、氣象資訊或派遣車輛分佈資訊來設置一預測需求車輛數。主動推播模組係連接服務區需求預測模組，主動推播模組更分析預測需求車輛數以設置一推播資訊，推播資訊為提供派遣車輛是否前往服務區域之建議，主動推播模組更選擇的透過外部一車輛派遣系統與至少一車載裝置通訊連接，或者直接與車載裝置通訊連接，以傳送推播資訊，前述之車載裝置係設置於派遣車輛內。雲端大資料平行運算處理模組為連接狀態擷取模組、服務區需求預測模組以及主動推播模組，提供連接之模組運算服務。 To achieve the above object, the present invention provides a ride tendency prediction device. The ride trend prediction device comprises a state capture module, a dispatch information module, a service area demand prediction module, an active push broadcast module and a cloud large data parallel operation processing module. The cloud large data parallel computing processing module is connected to the foregoing state capturing module, the dispatching information module, the service area demand forecasting module, and the active pushing module, and provides a connected module computing service. The state capture module captures activity information and weather information from the external internet. Activity information is to record activities located in a service area, while weather information records the weather status in the service area. The dispatch information module provides information on the distribution of vehicles dispatched to describe the distribution of dispatch vehicles in the service area. The service area demand forecasting module is a connection state capturing module and a dispatching information module. The service area demand forecasting module further sets a predicted demanding vehicle number according to activity information, weather information or dispatch vehicle distribution information. The active push module is connected to the service area demand forecasting module, and the active push broadcast module analyzes and predicts the number of required vehicles to set up a push broadcast information. The push broadcast information is to provide advice on whether the dispatched vehicle goes to the service area, and actively pushes the broadcast mode. The group is further selected to communicate with at least one in-vehicle device through an external vehicle dispatching system, or directly communicates with the in-vehicle device to transmit the push information, and the aforementioned in-vehicle device is disposed in the dispatch vehicle. The cloud large data parallel computing processing module provides a connected module computing service for the connection state capture module, the service area demand prediction module, and the active push broadcast module.
為達上述之目的，本發明提供一種運用雲端大資料處理之乘 車趨勢預測方法。乘車趨勢預測方法係應用於一乘車趨勢預測裝置，並包含下列之步驟：首先，自外部之網際網路擷取用來記錄位於一服務區域之一活動之活動資訊以及用來記錄位於服務區域之氣象狀態的氣象資訊。接著，自外部一車輛派遣系統存取一車輛派遣分佈資訊。再者，依據活動資訊、氣象資訊或派遣車輛分佈資訊來設置一預測需求車輛數。最後，分析預測需求車輛數以設置一個用來提供派遣車輛是否前往服務區域之建議之推播資訊，乘車趨勢預測裝置更選擇的透過外部一車輛派遣系統與至少一車載裝置連接，或者直接與車載裝置連接，以傳送前述之推播資訊。 In order to achieve the above purpose, the present invention provides a multiplication using cloud large data processing Vehicle trend forecasting method. The ride trend prediction method is applied to a ride trend prediction device, and includes the following steps: First, the activity information used to record the activity located in one service area is recorded from the external Internet and used to record the service. Meteorological information on the meteorological status of the area. Next, a vehicle dispatch distribution information is accessed from an external vehicle dispatch system. Furthermore, the number of predicted demand vehicles is set based on event information, weather information, or dispatch vehicle distribution information. Finally, the analysis predicts the number of vehicles required to set up a recommendation information for providing a dispatch vehicle to the service area. The ride trend predicting device further selects an external vehicle dispatch system to connect with at least one vehicle device, or directly The in-vehicle device is connected to transmit the aforementioned push information.
由於傳統之派遣車輛系統皆只能透過被動的分析車輛之派遣狀況，來進行調度，而本發明之乘車趨勢預測裝置及其方法則透過雲端運算來分析特定服務區域之氣象狀況以及事件狀況，並再判斷此服務區域之車輛是否能滿足叫車之需求來進行後續之車輛調度。 Since the conventional dispatch vehicle system can only perform scheduling by passively analyzing the dispatch status of the vehicle, the ride trend prediction device and the method of the present invention analyze the weather conditions and event conditions of a specific service area through cloud computing. And then determine whether the vehicle in this service area can meet the demand of the car for subsequent vehicle scheduling.
1‧‧‧乘車趨勢預測裝置 1‧‧‧Ride trend prediction device
11‧‧‧狀態擷取模組 11‧‧‧ State acquisition module
12‧‧‧派車資訊模組 12‧‧‧Car Information Module
13‧‧‧服務區需求預測模組 13‧‧‧Service Area Demand Forecasting Module
14‧‧‧主動推播模組 14‧‧‧Active push module
15‧‧‧預測參數調整模組 15‧‧‧Predictive parameter adjustment module
16‧‧‧雲端大資料平行運算處理模組 16‧‧‧Cloud large data parallel processing module
2‧‧‧車輛派遣系統 2‧‧‧Vehicle Dispatching System
3‧‧‧派遣車輛 3‧‧‧ dispatching vehicles
31‧‧‧車載裝置 31‧‧‧ Vehicle-mounted devices
4‧‧‧使用者 4‧‧‧Users
41‧‧‧使用端裝置 41‧‧‧Using end devices
第1圖係為本發明之系統圖。 Figure 1 is a system diagram of the present invention.
第2圖係為本發明之乘車趨勢預測裝置之方塊圖。 Fig. 2 is a block diagram of the ride tendency prediction device of the present invention.
第3圖係為本發明之車輛派遣系統處理使用者叫車之流程圖。 Figure 3 is a flow chart of the vehicle dispatching system of the present invention for handling a user's calling.
第4圖係為本發明之車輛派遣系統與車載裝置之傳輸流程圖。 Fig. 4 is a flow chart showing the transmission of the vehicle dispatching system and the in-vehicle device of the present invention.
第5圖係為本發明之預測參數調整模組之運作流程圖。 Figure 5 is a flow chart showing the operation of the predictive parameter adjustment module of the present invention.
第6圖係為本發明之乘車趨勢預測方法流程圖。 Figure 6 is a flow chart of the method for predicting the riding tendency of the present invention.
第7圖係為本發明之推播資訊設置流程圖。 Figure 7 is a flow chart for setting up the push information of the present invention.
以下將描述具體之實施例以說明本發明之實施態樣，惟其並非用以限制本發明所欲保護之範疇。 The specific embodiments are described below to illustrate the embodiments of the invention, but are not intended to limit the scope of the invention.
請參閱第1圖，其為本發明之系統圖。乘車趨勢預測裝置1係與車輛派遣系統2通訊連接，而位於派遣車輛3上的車載裝置31則是被選擇的與車輛派遣系統2或乘車趨勢預測裝置1通訊連接，使用者4則是透過了使用端裝置41與車輛派遣系統2通訊連接，以傳送叫車資訊。 Please refer to Fig. 1, which is a system diagram of the present invention. The ride tendency prediction device 1 is in communication connection with the vehicle dispatch system 2, and the in-vehicle device 31 located on the dispatch vehicle 3 is selectively connected to the vehicle dispatch system 2 or the ride tendency prediction device 1, and the user 4 is The vehicle end system 41 is used to communicate with the vehicle dispatch system 2 to transmit the car information.
請參閱第2圖，其為本發明之運用雲端大資料處理之乘車趨勢預測裝置1。乘車趨勢預測裝置1包含一狀態擷取模組11、一派車資訊模組12、一服務區需求預測模組13、一主動推播模組14以及一雲端大資料平行運算處理模組16。 Please refer to FIG. 2 , which is a ride trend prediction device 1 using cloud big data processing according to the present invention. The ride trend prediction device 1 includes a state capture module 11, a dispatch information module 12, a service area demand prediction module 13, an active push module 14, and a cloud large data parallel operation processing module 16.
雲端大資料平行運算處理模組16係連接前述之狀態擷取模組11、派車資訊模組12、服務區需求預測模組13，以及主動推播模組14，並提供連接之模組運算服務。前述之雲端大資料運算處理模組16係為可提供平行處理之電腦裝置，以執行大量資訊之運算處理。 The cloud large data parallel computing processing module 16 is connected to the foregoing state capturing module 11, the dispatching information module 12, the service area demand forecasting module 13, and the active pushing module 14, and provides a connected module operation. service. The cloud data processing module 16 described above is a computer device that can provide parallel processing to perform a large amount of information processing.
狀態擷取模組11係提供一活動資訊以及一氣象資訊。活動資訊係記錄位於服務區域之一活動。而氣象資訊係記錄位於服務區域之一氣象狀態。派車資訊模組12提供一派遣車輛3分佈資訊，車輛派遣分佈資訊係描述服務區域之派遣車輛3之分佈情形。服務區需求預測模組13則連接狀態擷取模組以及派車資訊模組12，服務區需求預測模組13更依據活動資訊、氣象資訊或派遣車輛3分佈資訊來設置一預測需求車輛數。主動推播模組14係連接服務區需求預測模組13並分析預測需求車輛數以設置一推播資訊，推播資訊係提供派遣車輛3是否前往服務區域之建議，主動推播模組14更選 擇的透過外部一車輛派遣系統2與至少一車載裝置31連接，或者直接與車載裝置31連接，以傳送推播資訊。雲端大資料平行運算處理模組16係連接狀態擷取模組11、服務區需求預測模組13以及主動推播模組14，雲端大資料平行運算處理模組16係提供連接之模組運算服務。 The state capturing module 11 provides an activity information and a weather information. The activity information is recorded in one of the activities in the service area. The meteorological information record is located in one of the service areas. The dispatch information module 12 provides a dispatch vehicle 3 distribution information, and the vehicle dispatch distribution information describes the distribution of the dispatch vehicle 3 in the service area. The service area demand forecasting module 13 is connected to the state capturing module and the dispatching information module 12, and the service area demand forecasting module 13 further sets a predicted demanding vehicle number according to the activity information, the weather information or the dispatching vehicle 3 distribution information. The active push module 14 is connected to the service area demand forecasting module 13 and analyzes the predicted demanded vehicle number to set a push broadcast information. The push broadcast information provides suggestions for whether the dispatched vehicle 3 goes to the service area, and the active push broadcast module 14 further selected Alternatively, the external vehicle dispatching system 2 is connected to at least one in-vehicle device 31 or directly connected to the in-vehicle device 31 to transmit the push information. The cloud large data parallel computing processing module 16 is a connection state capturing module 11, a service area demand forecasting module 13 and an active pushing module 14, and the cloud large data parallel computing processing module 16 provides a connected module computing service. .
前述之活動更再分成一般事件活動以及週期性活動。一般活動如演唱會、研討會、或體育活動等。而週期性活動則如國慶日、元宵燈會、情人節或跨年活動等。狀態擷取模組11透過網路網路自舉辦活動之網站，如唱片公司之網頁、體委會、人事行政局等網頁來擷取並分析所需之活動資訊。而狀態擷取模組11又經由氣象局或公正第三方氣象單元之網站來擷取特定服務區域之降雨機率、溫度、颱風、豪雨等氣象資訊進行後續之分析。 The aforementioned activities are further divided into general event activities and periodic activities. General activities such as concerts, seminars, or sports events. The recurring activities are like National Day, Lantern Festival, Valentine's Day or New Year's Eve. The state capture module 11 retrieves and analyzes the required activity information through a web site from a website such as a website of a record company, a sports committee, and a personnel administration bureau. The state capture module 11 obtains weather information such as rainfall probability, temperature, typhoon, and heavy rain in a specific service area through the website of the weather bureau or the fair third party weather unit for subsequent analysis.
派車資訊模組12係連接車輛派遣系統2，而前述之車輛派遣系統2更與外部至少一使用端裝置41通訊連接，派車資訊模組12係透過分析下列資訊來設置派遣車輛3分佈資訊： The dispatching information module 12 is connected to the vehicle dispatching system 2, and the vehicle dispatching system 2 is further connected to at least one external terminal device 41. The dispatching information module 12 is configured to analyze the following information to set the dispatching vehicle 3 distribution information. :
(1)使用端裝置41之叫車位置。 (1) The calling position of the end device 41 is used.
(2)車載裝置31記錄之載客位置。 (2) The passenger position recorded by the in-vehicle device 31.
(3)車輛派遣系統2之歷史叫車分佈區域以及數量。 (3) The history of the vehicle dispatch system 2 and the number of vehicles.
(4)位於服務區域之車載裝置31之統計數量。 (4) The statistical quantity of the in-vehicle device 31 located in the service area.
前述之派遣車輛3分佈資訊更包含一派遣量資訊，服務區需求預測模組13更依據下述之手段進行預測： The distribution information of the dispatch vehicle 3 described above further includes a dispatch amount information, and the service area demand forecasting module 13 further predicts according to the following means:
(1)服務區需求預測模組13透過預測模型設置派遣建議資訊，其預測模型為F t =F t-1+α(F t-2-F t-1)+α 2(F t-3-F t-1)+β(F e-1-F t-1)+γ(F w-1-F t-1) 公式(1) (1) The service area demand prediction module 13 sets the dispatch recommendation information through the prediction model, and the prediction model is F t = F t -1 + α ( F t -2 - F t -1 ) + α 2 ( F t -3 - F t -1 )+ β ( F e -1 - F t -1 )+ γ ( F w -1 - F t -1 ) Formula (1)
公式(1)之各參數說明如下：F t 為服務區域之一預測需求車輛數；F t-1為服務區域之上一周之同時段派遣量資訊；F t-2為服務區域之上一個月之同時段派遣量資訊；F t-3為服務區域之上一年之同時段派遣量資訊；F e-1為前一次同類型之活動，服務區域之派輛量資訊；F w-1為前一次同類型之氣象狀態，服務區域之派遣量資訊；α為一歷史需求參數；β為一天氣變異影響參數；γ為一特殊節日與同類型活動影響參數。 The parameters of formula (1) are as follows: F t is the predicted number of vehicles required for one of the service areas; F t- 1 is the information of the same segment dispatch amount for one week above the service area; F t- 2 is one month above the service area At the same time, the dispatch amount information; F t -3 is the information of the same period of dispatch in the service area; F e -1 is the previous activity of the same type, the information of the service area; F w -1 is The previous meteorological state of the same type, the dispatch quantity information of the service area; α is a historical demand parameter; β is a weather variation influence parameter; γ is a special holiday and the same type of activity influence parameter.
服務區需求預測模組13更自車輛派遣系統2接收一實際車輛派遣數。實際派遣車輛3數係為車輛派遣系統2實際派遣出去之車量數，乘車趨勢預測裝置1更包含一預測參數調整模組15。預測參數調整模組15係連接服務區需求預測模組13，預測參數調整模組15更依據實際派遣車輛3數以及預測需求車輛數之差異來修正為歷史需求參數、天氣變異影響參數、特殊節日與同類型活動影響參數中至少一個參數，以減少預測需求車輛數與實際車輛派遣數之差異，並將調整後之參數記錄與儲存。據此，服務區需求預測模組13能過透過分析氣象資訊、活動資訊、以及動態調整後之歷史需求參數、天氣變異影響參數以及特殊節日與同類型活體影響參數，來計算每個服務區域之預測需求車輛數，再將預測需求車輛數記錄至雲端大資料平行運算處理模組16，作為下一次預測之參考值。 The service area demand prediction module 13 receives an actual number of vehicle dispatches from the vehicle dispatch system 2. The actual number of dispatched vehicles is the number of vehicles actually dispatched by the vehicle dispatching system 2, and the riding tendency prediction device 1 further includes a predictive parameter adjusting module 15. The prediction parameter adjustment module 15 is connected to the service area demand prediction module 13, and the prediction parameter adjustment module 15 is further modified into historical demand parameters, weather variation influence parameters, special festivals according to the difference between the actual number of dispatched vehicles 3 and the number of predicted demand vehicles. At least one of the parameters of the same type of activity influence parameter to reduce the difference between the predicted demand vehicle number and the actual vehicle dispatch number, and record and store the adjusted parameters. Accordingly, the service area demand prediction module 13 can calculate each service area by analyzing weather information, activity information, and dynamically adjusted historical demand parameters, weather variation influence parameters, and special holiday and same type living influence parameters. The number of required vehicles is predicted, and the number of predicted demand vehicles is recorded to the cloud large data parallel operation processing module 16 as a reference value for the next prediction.
(2)服務區需求預測模組13更判斷當下是否為特殊節日、週期性活動或事件 等，若有，則搜尋歷史紀錄中最接近當前同類型之活動之最近一筆同時段派遣數量F e-1。 (2) The service area demand prediction module 13 further determines whether the current special holiday, periodic activity or event, etc., if any, searches for the latest simultaneous dispatch quantity of the activity in the history record that is closest to the current same type of activity F e - 1 .
(3)服務區需求模組更從乘車趨勢預測裝置1所連結之資料庫進行搜尋，並找出歷史紀錄中最接近當前天氣之最近一筆同時段派遣數量F w-1。 (3) The service area demand module searches from the database linked to the ride trend forecasting device 1 and finds the latest simultaneous dispatch quantity F w -1 in the history record which is closest to the current weather.
當乘車趨勢預測裝置1由派遣車輛3分佈資訊中得知此服務區域內的派遣車輛3無法滿足預測需求車輛數時，乘車趨勢預測裝置1會發送推播資訊以對派遣車輛3進行調度，調度之說明如下： When the riding tendency prediction device 1 knows from the dispatch vehicle 3 distribution information that the dispatch vehicle 3 in the service area cannot satisfy the predicted demanded vehicle number, the ride tendency prediction device 1 transmits the push information to schedule the dispatch vehicle 3. The description of the schedule is as follows:
(1)當判斷此服務區域車輛短缺時，乘車趨勢預測裝置1會尋找其他區域是否有閒置派遣車輛3，並發送推播資訊給前述之閒置派遣車輛3。 (1) When it is judged that there is a shortage of vehicles in the service area, the riding tendency prediction device 1 searches for whether there are idle dispatch vehicles 3 in other areas, and transmits the push information to the aforementioned idle dispatch vehicle 3.
(2)主動推播模組14會透過無線傳輸機制，諸如簡訊、3G網路、LTE等形式來傳送推播資訊。 (2) The active push module 14 transmits the push information through a wireless transmission mechanism such as a short message, a 3G network, or an LTE.
(3)主動推播模更會於一預定時段內，根據車輛短缺或過多之情形來決定是否發送推播資訊給車載裝置31。 (3) The active push mode will decide whether to send the push information to the in-vehicle device 31 according to the shortage or excessive condition of the vehicle within a predetermined period of time.
請參閱第3圖，當車輛派遣系統2接收到使用者4之叫車請求時，其車輛派遣系統2之處理流程圖下： Referring to FIG. 3, when the vehicle dispatching system 2 receives the request for the vehicle of the user 4, the processing flow chart of the vehicle dispatching system 2 is as follows:
S101：接收使用者4之叫車資訊，使用者係透過使用端裝置發送叫車資訊，使用端裝置包含電話、手持電話、超商kiosk、數位互動電視或等。 S101: Receive the information of the calling vehicle of the user 4, and the user sends the calling information through the using device, and the using device includes a telephone, a handheld phone, a super kiosk, a digital interactive television or the like.
S102：進入車輛派遣系統2之資料庫找出符合叫車資訊之派遣車輛3。 S102: Enter the database of the vehicle dispatch system 2 to find the dispatch vehicle 3 that meets the vehicle information.
S103：判斷是否讓派遣車輛3前往？ S103: Determine whether to send the dispatch vehicle 3 to go?
S104：若S103之判斷為否，則再通知乘客當前無合適之派遣車輛3，並詢問是否再呼叫一次，若乘客要再呼叫一次，則回到S102；若乘客拒絕呼叫，則到S106。 S104: If the judgment of S103 is negative, the passenger is notified that there is no suitable dispatch vehicle 3 at present, and asks whether to call again. If the passenger wants to call again, the process returns to S102; if the passenger rejects the call, then the process goes to S106.
S105：若S103判斷為是，則提供派遣服務 S105: If the determination in S103 is YES, the dispatch service is provided.
S106：將此次叫車服務之記錄傳送至派車資訊模組12。 S106: Transfer the record of the car service to the dispatch information module 12.
請參閱第4圖，其為車輛派遣系統2與車載裝置31之傳輸流程圖。 Please refer to FIG. 4, which is a transmission flow chart of the vehicle dispatching system 2 and the in-vehicle device 31.
S201：與車載裝置31連線。 S201: Connected to the in-vehicle device 31.
S202，判斷是否有符合此車載裝置31之叫車資訊，若S202之判斷為是，則執行S203；若S202判斷為否，則執行S205。 S202, it is determined whether there is any vehicle information conforming to the in-vehicle device 31. If the determination in S202 is YES, then S203 is performed; if the determination in S202 is NO, then S205 is executed.
S203：提供叫車資訊給車載裝置31，以讓派遣車輛3前開往叫車資訊內指定之地點，並接著執行S204。 S203: Providing the calling information to the in-vehicle device 31 to cause the dispatching vehicle 3 to drive to the designated place in the calling information, and then executing S204.
S204：將此筆叫車資訊傳送至派車資訊模組12或服務區需求預測模組13，回到S202。 S204: The call information is transmitted to the dispatch information module 12 or the service area demand prediction module 13, and the process returns to S202.
S205：建議車載裝置31是否前往特定之服務區域。 S205: It is recommended whether the in-vehicle device 31 goes to a specific service area.
S206：發送建議前開之推播資訊。 S206: Send the push information that is opened before the suggestion.
請參閱第5圖，其為本發明之預測參數調整模組15之運算流程圖。 Please refer to FIG. 5 , which is a flowchart of the operation of the prediction parameter adjustment module 15 of the present invention.
S301：計算一時間區間內，實際車輛派遣數與預測需求車輛數之差異。 S301: Calculate the difference between the actual number of vehicles dispatched and the number of predicted demand vehicles in a time interval.
S302：調整歷史需求參數α。 S302: Adjust the historical demand parameter α .
S303：調整天氣變異影響參數β。 S303: Adjusting the weather variation affects the parameter β .
S304：調整特殊節日與同類型活動影響參數γ。 S304: Adjust the special holiday and the same type of activity influence parameter γ .
S305：在α、β、γ中取調整後使實際車輛派遣數與預測需求車輛數之差異最小之參數作為下一次預測之參數，其餘二參數不調整。 S305: The parameter that minimizes the difference between the actual vehicle dispatch number and the predicted demand vehicle number after adjusting α , β , and γ is used as the parameter of the next prediction, and the other two parameters are not adjusted.
請參閱第6圖，其為本發明之乘車趨勢預測方法，應用於一 種乘車趨勢預測裝置1，此方法包含下列步驟： Please refer to FIG. 6 , which is a method for predicting the riding tendency of the present invention, which is applied to A ride trend prediction device 1, the method comprising the following steps:
S401：平行處理步驟S402~S406。 S401: Parallel processing steps S402 to S406.
S402：自外部之網際網路擷取氣象資訊以及活動資訊。接著，循環執行S402。 S402: Extract weather information and activity information from the external Internet. Next, S402 is executed in a loop.
S403：當預測需求車輛數與實際派遣之車輛數有落差時，則調整公式(1)之α、β、γ參數其中之一。接著，循環執行S403。 S403: When there is a difference between the number of predicted demand vehicles and the number of actually dispatched vehicles, one of the α , β , and γ parameters of formula (1) is adjusted. Next, S403 is executed in a loop.
S404：自車輛派遣系統2存取派遣車輛3分佈資訊，並循環執行S404。 S404: The vehicle dispatching system 2 accesses the dispatch vehicle 3 to distribute information, and executes S404 in a loop.
S405：預測需求車輛數之設置，並將預測需求資料庫存入資料庫，前述之預測需求車輛數之設置係依據活動資訊、氣象資訊或派遣車輛3分佈資訊來進行設置。接著，循環執行S405。 S405: Predicting the number of required vehicles and storing the predicted demand data into the database. The foregoing setting of the predicted number of required vehicles is set according to the activity information, the weather information, or the dispatching vehicle 3 distribution information. Next, S405 is executed in a loop.
S406：分析派遣車輛3分佈資訊以及預測需求車輛數以設置一推播資訊，推播資訊係提供派遣車輛3是否前往服務區域之建議。接著，循環執行S406。 S406: Analyze the distribution information of the dispatched vehicle 3 and predict the number of required vehicles to set a push information, and the push information provides a suggestion as to whether the dispatched vehicle 3 is going to the service area. Next, S406 is executed in a loop.
前述之步驟S406係包含下列之步驟： The aforementioned step S406 includes the following steps:
S501：週期性分析各服務區域之預測需求車輛數與對應之預測需求車輛數，以判斷此服務區域是否有缺車，若判斷為是，則執行S502；若判斷為否，則循環執行S501。 S501: Periodically analyze the number of predicted demand vehicles in each service area and the corresponding predicted demand vehicle number to determine whether there is a lack of vehicle in the service area. If the determination is yes, execute S502; if the determination is no, loop S501.
S502：尋找服務區域周圍之其他服務區域是否有可提供載客之派遣車輛3。 S502: Looking for other service areas around the service area, whether there is a dispatch vehicle 3 that can provide passengers.
S503：傳送建議前往該服務區域之推播資訊至可提供載客之派遣車輛3之車載裝置31。 S503: transmitting the push information to the service area to the in-vehicle device 31 of the dispatch vehicle 3 that can provide the passenger.
S504：判斷缺車之服務區域其派遣車輛3之需求項是否足夠，若判斷為是，則執行S501，若判斷為否，則執行S505。 S504: It is determined whether the demand item of the dispatched vehicle 3 is sufficient in the service area of the vehicle outage. If the determination is yes, the process proceeds to S501, and if the determination is no, the process proceeds to S505.
S505：判斷預設定時段其時間是否已到，若判斷為是，則執行S501；若判 斷為否，則執行S503。 S505: determining whether the time has elapsed in the preset time period, if the determination is yes, executing S501; If it is not, execute S503.
在步驟S406中，若多個服務區域皆有缺車，則開啟多個執行緒來分別執行S502~505。 In step S406, if there are any missing vehicles in the plurality of service areas, a plurality of threads are turned on to execute S502-505 respectively.
上列詳細說明係針對本發明之一可行實施例之具體說明，惟該實施例並非用以限制本發明之專利範圍，凡未脫離本發明技藝精神所為之等效實施或變更，均應包含於本案之專利範圍中。 The detailed description of the preferred embodiments of the present invention is intended to be limited to the scope of the invention, and is not intended to limit the scope of the invention. The patent scope of this case.
1‧‧‧乘車趨勢預測裝置 1‧‧‧Ride trend prediction device
2‧‧‧車輛派遣系統 2‧‧‧Vehicle Dispatching System
3‧‧‧派遣車輛 3‧‧‧ dispatching vehicles
31‧‧‧車載裝置 31‧‧‧ Vehicle-mounted devices
4‧‧‧使用者 4‧‧‧Users
41‧‧‧使用端裝置 41‧‧‧Using end devices
Claims (16)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW103111378A TWI524303B (en) | 2014-03-27 | 2014-03-27 | Forecasting Device and Method of Vehicle Trend Forecasting Based on Large Cloud Data Processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW103111378A TWI524303B (en) | 2014-03-27 | 2014-03-27 | Forecasting Device and Method of Vehicle Trend Forecasting Based on Large Cloud Data Processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW201537509A TW201537509A (en) | 2015-10-01 |
| TWI524303B true TWI524303B (en) | 2016-03-01 |
Family
ID=54850940
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW103111378A TWI524303B (en) | 2014-03-27 | 2014-03-27 | Forecasting Device and Method of Vehicle Trend Forecasting Based on Large Cloud Data Processing |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI524303B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI640955B (en) * | 2016-12-15 | 2018-11-11 | 中華電信股份有限公司 | System and method for estimating the number of vehicles |
| TWI635447B (en) * | 2017-06-16 | 2018-09-11 | 宏碁股份有限公司 | Method and System of Predicting Passengers' Demand |
| CN109284880B (en) * | 2017-07-20 | 2021-09-07 | 北京嘀嘀无限科技发展有限公司 | Data processing method, device, server, mobile terminal and readable storage medium |
| CN111737632B (en) * | 2017-08-16 | 2024-05-31 | 北京嘀嘀无限科技发展有限公司 | Queuing time determining method, queuing time determining device, server and computer readable storage medium |
-
2014
- 2014-03-27 TW TW103111378A patent/TWI524303B/en not_active IP Right Cessation
Also Published As
| Publication number | Publication date |
|---|---|
| TW201537509A (en) | 2015-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10192448B2 (en) | Method to control vehicle fleets to deliver on-demand transportation services | |
| CN104077915B (en) | Riding trend prediction device and method | |
| Cats et al. | Real-time bus arrival information system: an empirical evaluation | |
| CN104346921B (en) | Taxi information communication services (PCS) system, terminal and method based on location information | |
| US20190057480A1 (en) | Method and apparatus for providing transportation service information | |
| US20170193625A1 (en) | Driver supply control | |
| TWI524303B (en) | Forecasting Device and Method of Vehicle Trend Forecasting Based on Large Cloud Data Processing | |
| JP2022514134A (en) | A method for managing a transportation service provider, a computer program containing instructions for implementing the method, a non-temporary storage medium for storing instructions for executing the method, and a device for managing the transportation service provider. | |
| CN106373387A (en) | Vehicle scheduling, apparatus and system | |
| US20200342418A1 (en) | Vehicle service center dispatch system | |
| Cats et al. | Evaluating the added-value of online bus arrival prediction schemes | |
| CN111062629B (en) | Vehicle scheduling method and device, computer equipment and storage medium | |
| GB2540817A (en) | Improvements in or relating to distributed vehicular data management systems | |
| CN103218769A (en) | Taxi order allocation method | |
| TW201433480A (en) | Vehicle maintenance reminding method and device | |
| CN105139505A (en) | Off-time pre-appointment remote queuing method for bank business handling, and system thereof | |
| CN106530677B (en) | A kind of school bus dispatching method and system based on real time information | |
| Pi et al. | Understanding transit system performance using avl-apc data: An analytics platform with case studies for the pittsburgh region | |
| CN109816128A (en) | The net about processing method of vehicle order, device, equipment and readable storage medium storing program for executing | |
| Tirachini et al. | Headway variability in public transport: A review of metrics, determinants, effects for quality of service and control strategies | |
| CN114298559A (en) | Battery swapping method of battery swapping station, battery swapping management platform and storage medium | |
| CN114971135A (en) | Intelligent charging station power equipment management system based on 5G Internet of things and scheduling method | |
| CN116051052A (en) | Digital visual intelligent management platform for automobile road rescue | |
| Hua et al. | Effect of information contagion during train service disruption for an integrated rail-bus transit system | |
| TW202034277A (en) | Transportation method and apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MM4A | Annulment or lapse of patent due to non-payment of fees |