Inferences on trip matrices from observations on link volumes: A Bayesian statistical approach |
| |
Authors: | M. J. Maher |
| |
Affiliation: | Department of Probability and Statistics, Sheffield University, Sheffield, England |
| |
Abstract: | Previous methods for estimating a trip matrix from traffic volume counts have used the principles of maximum entropy and minimum information. These techniques implicitly give as little weight to prior information on the trip matrix as possible. The new method proposed here is based on Bayesian statistical inference and has several advantages over these earlier approaches. It allows complete flexibility in the degree of belief placed on the prior estimate of the trip matrix and also allows for different degrees of belief in diffeent parts of the prior estimate. Furthermore under certain assumptions the method reduces to a simple updating scheme in which observations on the link flows successively modify the trip matrix. At the end of the scheme confidence intervals are available for the estimates of the trip matrix elements. |
| |
Keywords: | |
本文献已被 ScienceDirect 等数据库收录! |