University of Twente Student Theses


Comparison of an aggregated macroscopic trip based 4 step model with a disaggregated tour based microsimulator: A backcasting approach

Mouw, A.V. (2022) Comparison of an aggregated macroscopic trip based 4 step model with a disaggregated tour based microsimulator: A backcasting approach.

[img] PDF
Abstract:The emergence of new mobility services as a result of the societal transition from owning to using goods and services, requires travel demand models to include the decision maker’s context. The traditional aggregated, macroscopic, trip-based 4-step model is not able to fulfil this requirement. Therefore, a generation of more advanced models has arrived with clear theoretical advantages. These models are disaggregated, microscopic, and tour-based, and take the decision maker’s context into account. However, empirical comparisons between these two approaches are scarce and do not address the prediction quality of both models, while the primary aim of travel demand models is to forecast future situations. The presented study attempts to fill this knowledge gap by conducting a backcast: predicting a historical year for which observed travel behaviour is available and compare this with the base year performance. The city of Almere in the Netherlands between 2010 and 2017 is used as case. The model performance is evaluated using three aggregated Key Performance Indicators (KPIs) that can be used for both models: trip frequency, modal split, and trip length distribution. The results demonstrate that the advanced model, called Octavius, is more capable of describing the base year compared to the 4-step model when evaluated against travel survey data. However, the backcast results show that both models perform similarly in terms of prediction quality. Hence, the longitudinal stability of both models is similar. In addition to the three KPIs, the disaggregated behaviour is analysed using two scenarios. The findings show that Octavius is able to include factors that affect travel behaviour in a manner that is consistent with literature. The 4-step model on the other hand exhibits counterintuitive results with increasing car use when car ownership decreases, which demonstrates its lack of behavioural realism. Moreover, the possibility of Octavius to model certain subgroups in society is a valuable tool to evaluate transport policies and other developments, such as the development of new neighbourhoods; analyses that the 4-step is not able to provide. Finally, the main limitation of this study is the limited differences between 2017 and 2010 in travel behaviour and socioeconomic development. Consequently, it is not possible to accurately evaluate how the models respond to changes, and in which situations one model performs better than the other. Future research could clarify this by using a longer backcast period, while ensuring that input and evaluation data are of sufficient quality.
Item Type:Essay (Master)
DAT.mobility, Deventer, The Netherlands
Faculty:ET: Engineering Technology
Subject:56 civil engineering
Programme:Civil Engineering and Management MSc (60026)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page