Determinação de matriz origem/destino utilizando dados do sistema de bilhetagem eletrônica

Abstract

The characterization of demand in terms of spatial and temporal distribution is critical in transportation networks development. There are numerous methods for analysis and estimates of transportation demand, however conventional methods have high cost and execution time, which restrict its application frequency. Transportation systems are adopting smart technologies for specific purposes, which generate a wealth of information, but are rarely used to improve relevant studies and analysis, as it occurs with data from Automated Fare Collection - AFC. In order to mitigate the shortcomings of conventional transportation research methods, particularly implementation time and costs, this paper proposes a methodology to determinate a Transit Origin-Destination Matrix (OD) that use AFC data based on international experience. The methodology is divided into two stages. The first is given a sample matrix that includes the users of electronic card, called seed matrix. In the nextstep, the seed matrix is expanded through a process that uses counts of passenger flows. The main advantages associated with thist mhodology are: low cost and time for obtain the trix, large sample size and capacity to rebuild matrix with high frequency. This work also presents the application of the methodology for the municipal transport system of Maceió city, capital of Alagoas state, Brazil. The results were consistent, however, to measure more accurately the quality of the final matrix is suggested that in future work, are performed comparative tests between the matrix obtained by the proposed methodology and a matrix produced by a new methodology already consolidated.