Estimating Base Matrices: the Combined Calibration Method
GUNN H, MIJJER P H, LINDVELD K, Hague Consulting Group and HOFMAN F, Ministry of Transport, The Netherlands
This paper describes recent and ongoing work in connection with the estimation of current or recent past travel patterns, summarised as origin to destination matrices distinguished by mode and purpose. It is intended as an short introduction to a research
This paper describes recent and ongoing work in connection with the estimation of current or recent past travel patterns, summarised as origin to destination matrices distinguished by mode and purpose. It is intended as an short introduction to a research initiative, rather than a comprehensive report of conclusions.
The context is the continuing development and refinement of the Dutch National Model (DNM), a predictive system for multi-modal transport demand which uses disaggregate analysis and forecasting techniques to generate forecasts of likely change in travel patterns in conjunction with estimates of base year travel patterns which set the levels from which these changes will apply. This model was developed from research in the 1970's and 1980's, and has been in use since 1985. It currently uses base matrices which are approximations to travel patterns in 1990, based on updates of earlier matrices built for the year 1983. Only the car-driver and train passenger are identified in the original base matrices (although the disaggregate models can and do generate separate estimates of matrices for all modes and purposes). The current project will generate estimates for car-driver and car-passenger matrices, consistent with matrices for train travel which have been developed earlier.
The focus of this paper is on the theoretical underpinnings of the method being used. It is intended to provide an overview of the approach, and of the relationship to previous research. The paper is set out in four sections, following this introduction; in turn we deal with the background to the work, the general aims of the project in respect of making some advance on the state-of-the-art, the technical approach selected to attempt this and a technical appendix recording the mathematical nature of the problem being solved. At the time of writing, the work is ongoing; our only conclusions are that early results confn-m the practical feasibility of the approach.
Association for European Transport