The Importance of Data Quality for Modelling in Transport
Title: The importance of data quality for modelling in transport
Data is always the predominant part of a certain engineering research activity. In the transport area, it is extremely necessary for scientists, engineers and researchers to set up quantitative mathematical models or to adopt suitable modelling to understand a huge amount of transport events and then to solve transport problems. The initial function of a model is to reflect the reality so that researchers can make use of the operability of the model to achieve some goals. While usually data is collected from the physical circumstances to represent some of the elements that influence the definite transport events. Functionally, a transport model helps to show the framework of a transport issue or problem, in which data has its power to present how the reality would react when conditions change. Therefore, the quality of the enquiry data and the organised data sets directly determines whether a transport model performs well with producing satisfying results. This essay is going to present some discussions around the importance of data quality in respect of transport modelling.
When data quality is emphasised in the scientific and engineering fields, the meaning of "data quality" cannot be just the accuracy that is on the most noticeable layer. It should be noted that being “multi-dimensional” is a vitally important characteristic of data quality (Karr, Sanil and Banks, 2005). Karr et al. (2005) indicated that data quality at least contains not only record-level accuracy, relevance and accessibility, but also timeliness, metadata, documentation, user capabilities and expectations, cost and context-specific domain knowledge. As for transport modelling, data quality is commonly considered the level of the reliability, capability and usefulness of data that fit the purpose of the models or can be used for the planning, operation, and decision making of the transport business or the governmental
Data is always the predominant part of a certain engineering research activity. In the transport area, it is extremely necessary for scientists, engineers and researchers to set up quantitative mathematical models or to adopt suitable modelling to understand a huge amount of transport events and then to solve transport problems. The initial function of a model is to reflect the reality so that researchers can make use of the operability of the model to achieve some goals. While usually data is collected from the physical circumstances to represent some of the elements that influence the definite transport events. Functionally, a transport model helps to show the framework of a transport issue or problem, in which data has its power to present how the reality would react when conditions change. Therefore, the quality of the enquiry data and the organised data sets directly determines whether a transport model performs well with producing satisfying results. This essay is going to present some discussions around the importance of data quality in respect of transport modelling.
When data quality is emphasised in the scientific and engineering fields, the meaning of "data quality" cannot be just the accuracy that is on the most noticeable layer. It should be noted that being “multi-dimensional” is a vitally important characteristic of data quality (Karr, Sanil and Banks, 2005). Karr et al. (2005) indicated that data quality at least contains not only record-level accuracy, relevance and accessibility, but also timeliness, metadata, documentation, user capabilities and expectations, cost and context-specific domain knowledge. As for transport modelling, data quality is commonly considered the level of the reliability, capability and usefulness of data that fit the purpose of the models or can be used for the planning, operation, and decision making of the transport business or the governmental
References: Berglund, S., 2001, GIS in Transport Modelling. Unpublished PhD dissertation. Stockholm: Royal Institute of Technology. Couclelis, H., 1992, Geographic knowledge production through GIS: Towards a model for quality monitoring [online]. Santa Barbara: NCGIA. Available at: [Accessed 10 November 2012] Dia, H. and Panwai, S., 2010, Evaluation of discrete choice and neural network approaches for modelling driver compliance with traffic information. Transportmetrica, 6(4), 249-270. Enoch, M., 2012, Sustainable Transport, Mobility Management and Travel Plans. 1st. ed. Surrey: Ashgate. Karr, A. F., Sanil, A. P. and Banks, D. L., 2005, Data Quality: A Statistical Perspective [online]. North Carolina: NISS. Available at: [Accessed 10 November 2012] Yang T., 2010, A two-stage stochastic model for airline network design with uncertain demand. Transportmetrica, 6(3), 187-213. Zhang, J. and Timmermans, H., 2012, Activity-travel behaviour analysis for universal mobility design. Transportmetrica, 8(3), 149-156.