Please use this identifier to cite or link to this item:
|Title:||Chapter 2. Data collection techniques|
|Authors:||AUBERLET Jean-Michel; BHASKAR Ashish; CIUFFO BIAGIO; FARAH Hanin; HOOGENDOORN Raymond; LEONHARDT Axel|
|Type:||Articles in periodicals and books|
|Abstract:||The objective of this chapter is to provide an overview of traffic data collection that can and should be used for the calibration and validation of traffic simulation models. There are big differences in availability of data from different sources. Some types of data such as loop detector data are widely available and used. Some can be measured with additional effort, for example, travel time data from GPS probe vehicles. Some types such as trajectory data are available only in rare situations such as research projects. This means that a simulation study carried out as part of a traffic engineering project, having a restricted budget, typically must rely on existing loop data or can at most utilize some GPS probe drives. The objective of calibration and validation in a traffic engineering project is mainly to check whether a model of a specific area replicates—at a desired level of detail— the macroscopic traffic conditions (flow, speed, travel time) for a certain traffic demand. Consequently, data for calibration and validation in traffic engineering projects typically need not to be microscopic. Conversely, data generated with much more effort (e.g., trajectory data) are typically used by researchers to investigate driver behavior in general. Analysis of driving behavior such as car following and lane changing requires highly detailed data to generate adequate insight into the traffic features to be modeled. These data are typically very expensive and/or laborious to acquire. Sections 2.1 through 2.7 briefly describe the technical backgrounds of various data types and detection techniques and discuss typical availability and application areas. Section 2.8 draws conclusions about what data to use for specific purposes. An overview table included in Section 2.8.4 may be useful to get a quick view on the various sorts of data that may be used for the calibration of microscopic traffic simulation models. In accordance with the primary focus of this book, this chapter provides only an overview of data collection. Extensive literature covering the techniques and their performance is available to the public through the World Wide Web. An interesting point is the expected quality of the data. However, there is some ambiguity in existing studies because “performance of a data collection system” is a result of several factors (hardware and software used, sensor configuration, and environmental and traffic conditions). Therefore, this chapter will not answer questions like “What is the expected accuracy?” and in “What sensor is best to be used?”. Specific studies describing detector features and boundary conditions are cited. Errors in data exert impacts on the calibration of a simulation model and hence, on its results. This impact is twofold. First, a calibration step is needed before a simulation can be performed. In Chapter 4, we show that errors in measuring the variables that are compared with the simulation results impact the optimal parameters set for the calibration process. Second, any simulation tool uses measured (or enhanced or estimated; see Chapter 3) variables as inputs. Therefore, data measurement errors must be kept in mind when performing simulation studies. The reader is invited to consult the available documentation to gain knowledge of limits and error bounds of each type of detector.|
|JRC Directorate:||Energy, Transport and Climate|
Files in This Item:
There are no files associated with this item.
Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.