A METHOD FOR UPDATING EXISTING DATA BASES OF AN ORIGIN-DESTINATION TRAFFIC MATRIX
Gerd Sammer, K Zelle
Abstract
Gerd Sammer, K Zelle
Abstract
The updating of existing origin-destination traffic surveys represents a big problem. Methods are lacking to follow up these surveys made with great expenditure of time and money by means of minor sample traffic surveys of the travel behaviour and traffic counts. The present report describes such an estimating method. According to the principle of minimum gain of an origin-destination traffic matrix obtained some years ago can be updated and calibrated at small costs with the aid of actual sample traffic surveys. It is possible to incorporate in this estimation the distribution models of the modal split and the traffic assignment. The method is described in detail. All information available from traffic counts, behaviour studies on the mobility, the modal choice and the routing split are considered in the estimation. With the use of the respective distribution models the latter information can be incorporated in the estimation. The estimation itself is not based on empirical data: it is a mathematical-statistical method satisfying the actual information on the travel behaviour as limit conditions and choosing the most probable solution from that limited margin of solutions. The method has been successfully used in the practice. (TRRL)
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The updating of existing origin-destination traffic surveys represents a big problem. Methods are lacking to follow up these surveys made with great expenditure of time and money by means of minor sample traffic surveys of the travel behaviour and traffic counts. The present report describes such an estimating method. According to the principle of minimum gain of an origin-destination traffic matrix obtained some years ago can be updated and calibrated at small costs with the aid of actual sample traffic surveys. It is possible to incorporate in this estimation the distribution models of the modal split and the traffic assignment. The method is described in detail. All information available from traffic counts, behaviour studies on the mobility, the modal choice and the routing split are considered in the estimation. With the use of the respective distribution models the latter information can be incorporated in the estimation. The estimation itself is not based on empirical data: it is a mathematical-statistical method satisfying the actual information on the travel behaviour as limit conditions and choosing the most probable solution from that limited margin of solutions. The method has been successfully used in the practice. (TRRL)
Key concepts: Trip distribution, Computer science, Sample (material), Modal, Traffic count, Estimation, Margin (machine learning), Matrix (chemical analysis)