A traffic distribution prediction method for large-scale event activities based on interest degree

Abstract

The present invention discloses a prediction method for traffic distribution of large-scale event activities based on interest degree. The traditional gravity model has relatively weak pertinence for predicting the traffic distribution of large-scale event activities, which may lead to relatively large prediction errors of the model and even mislead the formulation of relevant traffic organization and control plans. Based on the traditional gravity model, this prediction method changes the travel distribution between communities to the travel distribution between communities and competition areas; sets the attraction parameter of the activity according to the scale, number of projects, etc. of the large-scale event; sets the travel volume parameter according to the interest degree of community residents in the event; when the difference in the intention intensity of groups between communities to participate in the activity is obvious, the interest degree can be used to adjust the distance value and set the impedance function to correct the model. The present invention effectively improves the accuracy and pertinence of predicting the traffic distribution of large-scale event activities, and provides a method for predicting the traffic distribution of future large-scale event activities.

Description

Technical Field

[0001] The present invention relates to a method for predicting traffic distribution of large-scale event activities based on interest degree. Background Art

[0002] In recent years, with the rapid development of social economy and the vigorous development of national sports and culture undertakings, the frequency of large-scale event activities held in major cities has been increasing, and the country and all sectors of society have paid more and more attention to various sports events. During the holding of large-scale event activities, the number of travelers is huge and the requirements for traffic services are high, which puts great traffic pressure on the local and surrounding road networks, and the traffic guarantee work faces huge challenges. In order to reasonably respond to the challenges brought by large-scale event activities and ensure the smooth progress of the activities, predicting the traffic volume of large-scale event activities in advance is one of the bases and keys for formulating effective traffic organization and control plans.

[0003] In the prediction of traffic volume, the prediction of traffic distribution is a key link. The traditional gravity model only considers the attraction intensity of traffic zones and the resistance between them, and believes that the trip distribution between two traffic zones is proportional to the traffic generation volume and attraction volume of the two zones, and inversely proportional to the resistance between traffic zones. In previous patents related to traffic distribution prediction, a traffic distribution prediction method combining the gravity model and the Fratar model not only makes full use of the current distribution information, but also considers the changes in the road network and the impact of land use on people's travel, improving the accuracy and applicability of the prediction results. A combined model of traffic distribution and traffic flow assignment considering travelers' destination preferences considers the factor of travelers' destination preferences, and verifies the important impact of travelers' destination preferences on system balance through comparison. The above two methods make up for the deficiencies of the traditional gravity model.

[0004] Since the traditional gravity model has relatively weak pertinence for traffic prediction of large-scale events, it may lead to large model prediction errors, and even mislead the formulation of relevant traffic organization and control plans. The applicability of the two patents on improving the gravity model to large-scale event activities is relatively low. Therefore, the present invention proposes a method for predicting traffic distribution of large-scale event activities based on interest degree.

[0005] The characteristics of the method of the present invention are as follows: on the basis of the traditional gravity model, the trip distribution between zones is changed to the trip distribution between a zone and the venue of the event; and the attraction volume parameter of the event is set according to the scale, type, etc. of the large-scale event activity; the trip volume parameter is set according to the interest degree of the residents of the zone in the event; when the difference in the interest degree of groups between zones in participating in the activity is obvious, the interest degree can be used to adjust the distance value and an impedance function is set to correct the model. Summary of the Invention

[0006] The present invention provides a method for predicting the traffic distribution of large-scale event activities based on interest degree. That is, on the basis of the traditional gravity model, the gravity model is improved according to the traffic characteristics of large-scale event activities based on interest degree, so as to achieve the accuracy and pertinence of predicting the traffic distribution of large-scale event activities.

[0007] The technical solution adopted by the present invention is a method for predicting the traffic distribution of large-scale event activities based on interest degree, including the following steps:

[0008] Step 1: Collect the current traffic volume, current OD distribution and other basic data of each community for large-scale event activities. Among them, the current generation volume of each community is O i , the attraction volume of the sub-venues of large-scale event activities is D j , i = 1, 2... n, j = 1, 2... m, n represents the number of communities with traffic generation for large-scale event activities, and m represents the number of sub-venues of the large-scale event activity.

[0009] Step 2: Determine the form of the traffic distribution model of large-scale event activities based on interest degree as follows:

[0010]

[0011] Among them, X ij is the predicted distribution value from community i to the sub-venue j of the large-scale event activity; k is the model parameter; O i is the traffic generation volume of community i; α i is the generation volume parameter of community i, which is set according to the interest degree of the residents of community i in participating in and watching similar event activities; D j is the attraction volume of the sub-venue j of the event activity. This model is only applicable to large-scale event activities with multiple sub-venues located in different regions; β j is the attraction volume parameter of the sub-venue j of the event activity; f(c ij ) is the impedance function; c ij is the impedance function parameter. Considering that there are significant differences between participating in event activities and daily travel behaviors, in addition to considering the distance, time and cost between community i and the sub-venue j of the event activity, the interest degree of community residents in the event is introduced to reflect the travel willingness of residents.

[0012] This model satisfies the constraint That is, the generation volume of traffic communities is equal to the attraction volume of large-scale event activities.

[0013] Step 3: Determine the expressions of the parameters α i , β j and f(c ij ) of the traffic distribution model of large-scale event activities based on interest degree

[0014] Generation volume parameter α of traffic zone i i , that is, the attention index of residents in zone i to this large-scale sports event, is related to the interest of community residents in this sports event. Therefore, the calculation formula is as follows:

[0015]

[0016] Among them, α i is the generation volume parameter; g i is the number of sports facilities related to this sports event in zone i; b i is the number of times of retrieving relevant entries of this sports event through the web page in zone i.

[0017] Attraction parameter β of sub-venue j of large-scale sports event j , which is related to the scale of the sports event, the number of events and the scope involved. When the scale of the sports event is larger, the number of events is more, and the scope involved is wider, the value of the attraction parameter is larger to adjust the attraction of the activity site. The calculation formula is as follows:

[0018]

[0019] Among them, β j is the attraction parameter of sub-venue j of the sports event; n j is the maximum capacity of sub-venue j of this sports event; is the average value of the number of people that can be accommodated in the same type of sports events; cj is the number of countries or regions participating in sub-venue j of this sports event. Here, if this sports event is a national-level event, the number of participating regions is taken; if it is an international sports event, the number of participating countries is taken; is the average value of the number of countries or regions participating in the same type of sports events; p j is the number of events in sub-venue j of this sports event; is the average number of events in the same type of sports events.

[0020] Impedance function f(c ij ) of the traffic distribution model of large-scale sports events based on interest degree, uses the exponential function to be proportional to the travel distance or travel time; in addition, considering the difference between the travel during large-scale sports events and the normal travel of residents, the impedance function is also inversely proportional to the economic level of the community, the population, the interest of community residents in this sports event, etc. The calculation formula is as follows:

[0021]

[0022] Among them, f(c i ) is the impedance function; R ij is the distance from zone i to sub-venue j of the large-scale sports event; N iis the total population of community i; is the average population of each community in the planned area; G i is the total output value of community i; is the average total output value of each community in the planned area; S i is the total commodity sales amount of community i; is the average total commodity sales amount of each community in the planned area; gi is the number of sports facilities related to the event in community i; b i is the number of times of retrieving event-related entries through the web page in community i.

[0023] Step 4: Calibrate the model parameters k and λ

[0024] Determine the sample data set through the generated volume of the current traffic community and the predicted attraction volume of the large-scale event, as well as the OD distribution of the current large-scale event. Calibrate the model using the least squares method to determine the model parameters k and λ.

[0025] Advantages of the present invention:

[0026] A traffic distribution prediction method for large-scale events based on interest proposed by the present invention breaks through the weakness of the traditional gravity model for large-scale events, which may lead to large prediction errors in the model and even mislead the formulation of relevant traffic organization and control plans. Considering factors such as the interest of residents in the event and the attraction of large-scale events, it provides a more accurate prediction model for predicting the future traffic volume of large-scale events, improves the accuracy of prediction, so that relevant units can adopt more effective traffic organization methods and provide guarantee for the smooth holding of large-scale events. Specific implementation manner

[0027] The following describes the specific steps of the present invention.

[0028] Step 1: Collect the current generated traffic volume, current OD distribution and other basic data of each community for large-scale events. Among them, the current generated volume of each affiliated community is O i , the attraction volume of the sub-venues of large-scale events is D j , i = 1, 2... n, j = 1, 2... m, n represents the number of communities with traffic generation for large-scale events, and m represents the number of sub-venues of the large-scale event.

[0029] Step 2: Determine the traffic distribution model for large-scale events based on interest as follows:

[0030]

[0031] Among them, X ijis the predicted distribution value from community i to the sub - venue j of a large - scale sports event; k is the model parameter; O i is the traffic generation volume of community i; α i is the generation volume parameter of community i, which is set according to the interest of residents in community i in participating in and watching similar sports events; D j is the attraction volume of the sub - venue j of the sports event. This model only targets large - scale sports events with multiple sub - venues located in different regions; β j is the attraction volume parameter of the sub - venue j of the sports event; f(c ij ) is the impedance function; c ij is the impedance function parameter. Considering that there are significant differences between participating in sports events and daily travel behavior, in addition to considering the distance, time, and cost between community i and the sub - venue j of the sports event, the interest of community residents in the event is introduced to reflect the travel willingness of residents.

[0032] This model satisfies the constraint That is, the generation volume of traffic communities is equal to the attraction volume of large - scale sports events.

[0033] Step 3: Determine the model parameters α i , β j and the expression of f(c ij )

[0034] The generation volume parameter α i of traffic community i, that is, the attention index of residents in community i to this large - scale sports event, is related to the interest of community residents in this sports event. When the interest of community residents in this sports event is higher, there are more venue facilities related to this sports event. Therefore, the calculation formula is as follows:

[0035]

[0036] Among them, α i is the generation volume parameter; g i is the number of sports facilities related to this sports event in community i; b i is the number of times of retrieving relevant entries of this sports event through web pages in community i.

[0037] The attraction volume parameter β j of the sub - venue j of the large - scale sports event is related to the scale of the sports event, the number of event items, and the scope involved. When the scale of the sports event is larger, the number of event items is more, and the scope involved is wider, the value of the attraction volume parameter is larger to adjust the attraction volume of the event location. The calculation formula is as follows:

[0038]

[0039] Among them, βj is the attraction parameter of the competition event sub - venue j; n j is the maximum capacity of the competition sub - venue j; is the average value of the number of people that can be accommodated in the same type of competition; cj is the number of countries or regions participating in the competition sub - venue j. Here, if the competition event is national, the number of participating regions is taken; if it is an international competition, the number of participating countries is taken; is the average value of the number of countries or regions participating in the same type of competition event; p j is the number of events in the competition sub - venue j; is the average number of events in the same type of competition event.

[0040] The impedance function f(c ij ) of the traffic distribution model for large - scale competition events adopts an exponential function which is proportional to the travel distance or travel time; in addition, considering the difference between the travel during large - scale competition events and the normal travel of residents, the impedance function is also inversely proportional to the economic level of the community, the population, the interest of community residents in the competition event, etc. The calculation formula is as follows:

[0041]

[0042] where f(c i ) is the impedance function; R ij is the distance from community i to the competition event sub - venue j of the large - scale competition; N i is the total population of community i; is the average value of the population of each community in the planning area; G i is the total output value of community i; is the average value of the total output value of each community in the planning area; S i is the total sales of goods in community i; is the average value of the total sales of goods of each community in the planning area; gi is the number of sports facilities related to the competition event in community i; b i is the number of times of retrieving relevant entries of the competition on the web page in community i.

[0043] Step 4: Calibrate the model parameters k and λ

[0044] Determine the sample data set through the origin - destination (OD) distribution of the current situation of the traffic community's production volume and the predicted value of the attraction volume of large - scale competition events, and calibrate the model using the least - squares method to determine the model parameters k and λ.

Claims

1. A traffic distribution prediction method for large-scale event activities based on interest degree, characterized in that, it includes the following steps: Step 1: Collect the current traffic generation volumes, current OD distributions, and other basic data of each community for large-scale sports events. Among them, the current traffic generation volume of each community is O i , the attraction volume of the sub-venues of large-scale sports events is D j , i = 1, 2... n, j = 1, 2... m, where n represents the number of communities with traffic generation for large-scale sports events, and m represents the number of sub-venues of the large-scale sports event; Step 2: Determine the form of the traffic distribution model for large-scale event activities based on interest degree as follows: Among them, X ij is the distribution prediction value from community i to large-scale event venue j; k is the model parameter; O i is the current traffic volume of community i; α i is the current traffic volume parameter of community i, which is set according to the interest of residents in community i in participating in and watching similar events; D j is the attraction of sub-venue j of the event. This model is only applicable to large-scale events with multiple sub-venues located in different regions; β j is the attraction parameter of the event venue j; f(c ij ) is the impedance function; c ij is the impedance function parameter. Considering that there is a big difference between participating in events and daily travel behavior, in addition to considering the distance, time and cost between community i and event venue j, the interest of community residents in the event is introduced to reflect the residents' travel willingness. The model satisfies the constraints That is, the current traffic volume in the community is equal to the volume attracted by large-scale events; Step 3: Determine the parameters α i , β j and the expression of f(c ij ) The current traffic generation parameter α of zone i i , which is the attention index of residents in zone i to this large-scale sports event and is related to the interest of residents in this sports event. Therefore, the calculation formula is as follows: Among them, α i is the current traffic generation parameter of community i; g i is the number of sports facilities related to the event in community i; b i is the number of times of retrieving relevant entries of the event through the web page in community i. Attraction parameter β of sub - venue j of large - scale event j , which is related to the scale of the event, the number of events and the scope involved. When the scale of the event is larger, the number of events is more, and the scope involved is wider, the value of the attraction parameter is larger, so as to adjust the attraction of the activity site. The calculation formula is as follows: Among them, β j is the attraction parameter of the sub - venue j of the event; n j is the maximum capacity of the sub - venue j of the event; is the average value of the number of people that can be accommodated in the same type of event; c j is the number of countries or regions participating in the sub - venue j of the event. Here, if the event is a national - level event, the number of participating regions is taken; if it is an international event, the number of participating countries is taken; is the average value of the number of countries or regions participating in the same type of event; p j is the number of events in the sub - venue j of the event; is the average number of events in the same type of event; The impedance function f(c ij ) of the traffic distribution model for large-scale event activities based on interest is an exponential function that is proportional to the travel distance or travel time. In addition, considering the differences between the trips for large-scale event activities and the normal trips of residents, the impedance function is also inversely proportional to the economic level of the community, the population, and the interest of community residents in the event. The calculation formula is as follows: Among them, f(c ij ) is the impedance function; R ij is the distance from cell i to the sub-venue j of the large-scale sports event; N i is the total population of cell i; is the average value of the population of each cell in the planning area; G i is the total output value of cell i; is the average value of the total output value of each cell in the planning area; S i is the total sales of goods in cell i; is the average value of the total sales of goods of each cell in the planning area; g i is the number of sports facilities related to the event in cell i; b i is the number of times of retrieving event-related entries through the web page in cell i; Step 4: Calibrate the model parameters k and λ Determine the sample data set through the current traffic generation volume of the cell and the predicted value of the attraction volume of the large-scale event activity, as well as the OD distribution of the current large-scale event activity, and calibrate the model using the least squares method to determine the model parameters k and λ.

Images