Introduction
The importance of effective road planning is greater than ever, since the traffic volume have increased significant over the last years. The need for implementation of an effective road planning that ensures efficient traffic flow and minimizes and concentrate the negative effects in the traffic is therefore high. One of the main road planning tools that the road authorities uses is road classification, where the individual roads and road sections in the road network are nested hierarchy according to their function and significance in the overall road network.
The modern principles of road planning used today can be traced back to the urban-planning paradigm "functionalism", which among other Raymond Unwin and Clarence S. Stein was inspired by in the English Garden Cities and Radburn in the U.S. The real breakthrough attributed to Le Corbusier, who in 1933 in connection with a congress released the “Charte d'Athènes”. This statement contained 95 ideas for how the growing traffic challenges could be reduced. In the Danish context, the modern principles of road planning were first officially implemented in vejregel; "Bytrafik, projektgruppe 0" from 1980 and since then only minor adjustments has been done.
Problem statement
An interesting reflection, which still only few inquiries have been about, is the correlation between the road users route choice and the road classification performed by the road authorities.
If the road users do not select the individual trips in accordance to the road classification, then the road planning will lose much of its purpose. This is the basis for this study, which will identify and analyze the extent the road classification is respected in the drivers' route choice. Traffic planners want road users to use the main road when they transport themselves between the individual local areas.
Study's problem statement reads:
“Is it possible to develop a method based on FCD, which can describe the extent to which the motorists follow the mission for the road classification as the road authorities perform?
It was chosen to use the Floating Car Data as data source for this study. FCD is chosen because it offers some unique opportunities in monitoring the motorists' traffic patterns and route choice over both time and place. Meanwhile, FCD is in a format that makes it well suited for further processing and analysis.
It was decided to conduct the study as a Proof of Concept Study, where much of the focus will be on developing and quality assuring methods since FCD has not yet been used in studies with this problem area.
Methodology and data
The FCD is from the project; “Spar På Farten”, and the data used is from 1. – 30 of September 2007. The analyses are based on approx. 1.6 million GPS-points, which were subsequently linked to 1,749 trips. The trips are driven by 65 vehicles.
It was through the study developed two methods which could either be used individually or in combination; the methods are:
* A quantitative analysis method
* A qualitative analysis method
The two methods have their own approach to identifying the correlation between drivers' route choice and road classification. The quantitative analysis is based on the identification and counting of the number of shifts in roads classes for each of the individual trips. By comparing the number of shifts with each trips origo and destinations street class, it is possible to identify whether the trip was conducted in accordance with the functional category of roads. The qualitative analysis method was developed to handle large data sets and dynamically identify the correlation between the road classification and route selection.
For the qualitative analysis is the starting point a more thorough analysis of the route choice between an OD zone or trips from the qualitative analysis. This method is based on GIS and has a more manual approach. Therefore the number of trips which is analysed is smaller than for the quantitative analysis.
Results and Conclusion
In the quantitative method it was unfortunately found that the method was not able to demonstrate classification to the extent requested. It was found that the numbers of shifts in the category of roads were overestimated mainly due to uncertainties and discrepancies in the online map matching, as the data base was going through the SPF project. Furthermore, the executed trip segregation also weak. It was not possible to account for these sources of error through adjustment of the method, and therefore the results are not assumed to be completely accurate.
The results from the qualitative analysis are not similarly affected by the error sources as the quantitative. The studied route choices between Aalborg City and City Syd are analyzed and mapped. There was completed a total of 41 tours of 10 different vehicles between the two zones. It was found that they spent a total of 7 alternative routes, where 3 of them contained a choice of route / routing, which was not consistent with the functional road classification. The extent of the inappropriate choice of route was limited, as it was only short stretches of the 3 routes that have run in inappropriate areas.
The conclusion of the study is that it is possible to develop a method that can describe the degree the motorists follow the road classifications. However, it is found that in order to implement this dynamically, the quality of FCD must be improved through better trip segregation and map matching.