"Comb" And "Solution" Of Traffic Volume Prediction

The accuracy and reliability of expressway traffic volume prediction will have an important impact on the scale of project construction, the selection of technical standards and the evaluation of economic benefits.

Therefore, it is necessary to combine the feasibility study of Expressway and the reliability of highway construction project prediction, and analyze the main reasons for the deviation of expressway traffic volume prediction at the research stage, and put forward a practical solution method, which is of practical significance to optimize and improve the early stage demonstration work of Expressway Construction project.

Current situation analysis

The research on traffic volume in our country is relatively late. In the 80s of last century, the index method was used to forecast highway traffic volume. Since 1990, the "four stage" method used by European and American urban traffic planning and the total control method put forward by Xi'an Highway Communication University are mainly adopted.

On the whole, the prediction methods of existing expressways mainly include: single prediction (time series method, regression analysis method), and the "four stage" method based on the origin and destination survey. Among them, the "four stage" method is most commonly used.

Judging from the regulations of the industry, there is no definite provision for the method and model of traffic volume prediction at present stage in China.

As China's expressways have been completed and opened to traffic, some post highway construction projects post evaluation analysis shows that there is a certain discrepancy between the predicted traffic volume and the measured traffic volume, no matter what kind of prediction methods are inadequate.

Therefore, we choose the actual traffic volume and feasibility study stages of different highways in the East, middle and West.

　　從表1可以看出，不同地區高速公路可研中預測的交通量與實際交通量差距較大：東部福泉、泉廈高速公路運營前五年預測交通量較實際高百分之30-百分之70，特別是前三年交通量差距最大，主要是由于預期對高速公路通車后分擔通道交通比例估計過高;中部宜黃高速連接了湖北省東西經濟最發達地區，形成湖北長江經濟帶的運輸大動脈，運營前五年工可預測值均低于實際值百分之15左右，主要由于預測階段過低考慮了通道交通增長趨勢以及高速公路在其中承擔的比例，而哈大高速公路主要功能為連接哈爾濱和大慶兩個較為成熟的城市，通車初期交通誘增影響考慮偏大，實際交通量較預測偏低;西部桂柳高速交通量預測與實際值的誤差在百分之35以上，預測階段對區域經濟發展趨勢評估較為樂觀，且通道旅游誘增影響考慮偏大。

The high error of expressway traffic volume prediction is not only the above individual items, but also exists in the traffic volume forecast of some expressway projects.

On the trend, most of the Expressways

Main problems

In terms of method, the method of combining qualitative analysis with quantitative analysis is generally adopted in the prediction of highway traffic volume.

In addition to referring to the conclusion of the model, we should grasp the rationality of the qualitative analysis. For example, the determination of the proportion of new old roads should not only take into account the pportation conditions and travel costs of each route, but also depend on the traffic volume of the present investigation. Short haul vehicles require less pportation conditions and are more sensitive to travel expenses in actual travel, so they can not be assigned to the expressway by fixed models only.

The selection of elastic coefficient is not only based on the calculation results of the model established by historical data, but also should be qualitatively grasped according to the actual situation of the future economic and traffic development in the region.

From the point of view of quantitative prediction, the "four stage" method is generally adopted in China's traffic volume prediction. The method is based on the premise that the relationship between traffic volume and environmental factors (economic society, population, etc.) does not change, the relationship between the two is analyzed, and the future traffic volume is calculated according to the development of the environment.

This method is quite suitable for the relatively stable social environment and stable industrial and product structure in foreign countries. At present, in the actual operation of China's traffic volume forecast, the growth rate of growth rate from high to low, slow and fast growth is generally predicted, and the number of traffic elasticity is basically selected according to this rule.

Under the environment of rapid economic development and rapid adjustment of industrial structure in China, the development of economic and traffic growth will show the characteristics of leaping or wave band. It is still questionable to predict the relatively stable correlation between economy and traffic.

{page_break}

In terms of models and parameters, due to the lack of standard restrictions, the selection of models in the prediction stage is more casual, and the requirements of the specific methods for data are not clear enough, which makes the data analysis of the early stage or the traffic survey data not closely linked with the traffic volume forecast.

The use of the model is too simple for some projects, and has not been adjusted according to the actual characteristics of the specific projects. For example, the Fletcher method, commonly used in traffic volume distribution, is not applicable to the areas where the traffic distribution changes greatly in the future.

At the same time, the determination method of parameters and coefficients in the model is not clear enough, and the experience and formula operation are excessive.

The derived traffic demand arises from the original demand such as economic and social activities. Traffic volume prediction is based on the trend of economic and social development.

Therefore, the reasonable prediction of the future economic and social development will affect the traffic demand forecast of the whole road network and affect the prediction of the traffic volume of the project.

The economic forecast of existing freeway projects is mainly based on the mathematical analysis of the statistical data of regional economic and social history, and referring to the forecast data of regional economic development plans, such as national economic development planning, so as to grasp the trend of gradual decreasing trend in the future.

In practical work, we often focus on mathematical analysis of historical statistical data, ignoring the analysis of relevant policies and plans at the national, provincial and regional levels, resulting in an optimistic or pessimistic outlook for future economic and social development based on statistical data analysis, which directly results in a high or low traffic volume forecast value.

For highway traffic volume forecasting, it is usually based on the survey of the origin and destination of the regional traffic trip, the observation of the relevant highway traffic volume and the investigation of the relevant pport parameters.

At the present stage, the traffic volume observation in the provincial road of our country is based on intermittent statistics, and the accuracy of some statistical data is poor, and the collection of traffic statistics of most expressways is still difficult.

Although there are relatively mature survey theories and methods for traffic investigation, because of the difficulty of organization investigation and so on, only sampling investigation can be conducted. There are many survey options that need subjective judgement or estimation in the field traffic investigation, which leads to a certain degree of deviation in the reliability of the survey data.

At the same time, whether the data processing is reasonable will directly affect the reliability of the survey data. The forecaster should reasonably expand the sample size according to the survey date and the historical statistics, so as to ensure that the survey data can correctly reflect the present situation of regional traffic and pportation.

Under the background of the gradual improvement of the comprehensive pportation network, the relationship between passenger and freight pportation and competition between highway and railway and water pportation is becoming more and more closely related.

Although the current traffic volume forecasting method has considered the competitive relationship between highway construction projects and other modes of pportation to some extent, it has failed to provide a scientific, concrete and operable method.

From the angle of highway pportation system, prediction is limited to qualitative analysis and prediction based on pfer rate method. The prediction time span is too large, light freight and heavy passenger pport in the forecast process are insufficient, and the understanding of complex competition and cooperation between highway and other modes of pportation is deficient, which affects the reliability of prediction results.

The induced traffic reflects the economic growth in the area near the road after the improvement of traffic conditions, thus inducing new traffic trips.

The induced traffic volume is generally reflected only after three to five years after the project is completed. In actual traffic volume prediction, the induced traffic volume is often added to the highway at the opening stage. Based on this, traffic growth is predicted by a certain traffic growth rate, resulting in a larger forecast result of initial traffic volume.

At the same time, after the completion of the project, the induced effect gradually diminished. If the traffic volume is also taken into consideration in the whole period of project operation, the traffic volume forecast result will be too large.

The change of road network will change the proportion of new old roads in the highway passageway and directly affect the traffic volume of expressway projects.

In some highway works, the old roads are mostly two or three grade roads.

For some reasons, part of the expressway has been postponed than the estimated commencement and opening time, while the old road has undergone large-scale renovation during this period, some of which have reached the first or two level standards.

The increase of relevant road capacity has weakened the attractiveness of the highway and caused partial traffic diversion, especially in the economically developed coastal provinces.

At present, although highway tolls are understood and recognized by people, they are very sensitive to toll costs when they consider vehicle operating costs, due to factors such as income level and economic affordability.

After the improvement of road conditions, the reduction of vehicle operation cost is difficult to quantify relative to tolls, and more truck drivers do not attach importance to the value of time saving. Therefore, the highway toll level has become a "lever" for the traffic assignment of new old roads. The number of highway charges directly affects the road users' route choice behavior.

In addition, after the improvement of the local road conditions along the expressway area, more vehicles will choose to use the local roads and national trunk highways in the passageway under the influence of the low level of local road charges.

According to the previous norms, it is more feasible to determine whether the expressway construction project is feasible from the traffic volume point of view. If the prediction meets the standard requirements, it is feasible, and if it is not achieved, it is not feasible.

This factor often causes the design unit to make the feasibility report to be a batch report in order to make the project submitted for examination and approval, and the traffic forecast value is adjusted to the higher level. This results from the comparison between the predicted traffic volume and the actual traffic volume, and the predicted traffic volume is often higher than the actual traffic volume.

Countermeasures and suggestions

The "four stage" law in China is not exactly the same with that in other countries. Most of the existing methods are empirical qualitative analysis. Although some reports also mention the contents of public rail and public water diversion, there has not yet been a mature method. Although similar investigations have been done in China, it has not been further promoted and verified.

The existing freeway workers can report mainly on the experience to determine the proportion of distribution between new roads and old roads, and generally fail to make up the traffic volume according to the current situation. The impact of the systematic charging on the traffic volume distribution causes the actual traffic volume of the expressway to be lower than the predicted traffic volume.

In the background of the gradual abolition of the two grade road toll in the future, we should calibrate the traffic assignment model according to the highway survey data in our country, and take the traffic composition of the future passageway as the basis, and consider the influence of the highway toll on the traffic volume assignment.

{page_break}

We should attach importance to the improvement of forecasting methods for economic and social development and avoid the impact of economic optimism or pessimism on traffic volume development. At the same time, we should also pay attention to the improvement of forecasting methods for the relationship between pportation and economic society.

It is suggested that different economic development plans should be given to various important factors to forecast the economic forecast under different plans so as to make decision reference.

For example, the forecast of economic development can be divided into three plans: high growth, neutral growth and low growth. The traffic volume of each scheme is predicted, and the possibility of the three options is estimated to obtain the expected value of the traffic volume. Different schemes can be considered for the relationship between the traffic volume and the economic society and whether the fees are adjusted or not. According to the different combinations of economic development, pportation and economic relations, and whether the fees are adjusted, the traffic volume of the project is forecasted. Then the possibility of the occurrence of different traffic projects of the three factors is judged to obtain the expected value of the traffic volume of the project.

We should strengthen the integrity and authenticity of data statistics, grasp the true and reliable data of traffic volume prediction from the source, enhance the check and analysis of the original data, and take it as the basis for evaluating the accuracy of expressway traffic volume prediction.

We should strengthen the function orientation of the project in the entire road network and analyze the role in the pport corridor, pay attention to the coordination and convergence of various pportation modes, and analyze the characteristics and sources of the traffic volume of the project services, so as to deepen our understanding of the project.

The changes of the relevant road network directly affect the change of the traffic volume undertaken by the project. The regional road network planning should be orderly and standardized, and maintain the continuity and stability of the road network planning plan in a certain period, so as to provide a reliable basis for road network traffic assignment.

After carrying out post evaluation of highway construction projects, we can find problems in time and sum up experience, which is very useful for carrying out traffic volume prediction in the future.

Therefore, it is suggested that highway construction project post evaluation be one of the statutory steps of highway construction project management.