The impact of time aggregation and travel time models on time-dependent routing solutions

Abstract

Traffic and congestion have a big impact on the performance of transportation systems. Travel time models are required to calculate trip durations and arrival times when traffic information is available. These models rely on the availability of detailed information about the traffic state. With the growing availability of onboard devices, we can now capture very precise data with a very high frequency. The challenge is now to efficiently use these data to solve routing problems and evaluate routing solutions. A key question that emerges is how to determine the best compromise between a huge amount of very precise data and a smaller volume of aggregated data. In this article, we analyze the impact of time aggregation on the performance of the main travel time models, namely the link travel model (LTM), the flow speed model (FSM) and the smoothed travel time model (STTM). We also analyze the impact of different time aggregation levels on these models. Our results show that all models share similar performance, particularly with large intervals. Finally, we show that the LTM largely respects the FIFO property, which is an important hypothesis for routing algorithms.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 2 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.