Demand Model in the Agglomeration using Sim Cards

A., Brzeziński; T., Dybicz; Ł., Szymański · 2019 · DOAJ

DOI: 10.2478/ace-2019-0010

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the limitations of current traffic demand models for urban agglomerations, which often rely on simplified assumptions due to a lack of data for suburban areas. Traditional models frequently treat suburban trip generation and attraction using generalized parameters, failing to capture the heterogeneous travel behaviors of individual municipalities. The authors argue that this simplification leads to inaccuracies in forecasting traffic, particularly regarding the conflict between long-distance transit needs and local access requirements. To resolve this, the study utilizes Big Data from mobile operator SIM cards to refine trip generation models within the Warsaw agglomeration, aiming to differentiate between trips within the suburban area, trips between the suburbs and the main city, and trips within the city itself. The research was conducted as part of the INMOP 3 project, analyzing 0.47 million trips recorded between November 15–17, 2016. Data were sourced from the SS7 signaling network of T-Mobile Poland, covering Warsaw districts and 60 suburban municipalities. The analysis defined trips based on movement between base stations, assuming a maximum stop time of three hours to distinguish separate journeys. Due to anonymization requirements, trips with fewer than five observations were aggregated. The authors used these SIM card movements to identify specific trip generators and spatial distributions, revealing significant variations in travel patterns. For instance, while Pruszków was the largest overall trip generator, municipalities like Ząbki showed a dominant connection to Warsaw (73% of external trips), whereas others like Błonie had stronger local connections. Using these insights, the authors developed a refined trip generation model incorporating 11 explanatory variables, such as population, workplace counts, and school places. They introduced specific calibration coefficients for trip generation (Gen-S, Gen-W) and absorption (Abs-S, Abs-W) for each municipality, allowing for distinct modeling of suburban and Warsaw-related trips. The updated model was validated against comprehensive traffic surveys and SIM card data. Results showed high compatibility, with R² values of 0.99 for trip distance distributions within the suburban area and 0.96 for the share of trips related to Warsaw. The model accurately replicated the spatial distribution and volume of trips, significantly outperforming the base model, which had an R² of only 0.37 for Warsaw-related trip shares. The study concludes that simplified agglomeration models are insufficient for accurate traffic forecasting. The strong differentiation in travel behavior across municipalities necessitates individualized modeling approaches. The integration of mobile operator Big Data provides a viable method for obtaining detailed, granular data on suburban mobility, enabling the calibration of specific coefficients for each zone. This approach enhances the accuracy of traffic generation models, improving their utility for planning road networks and analyzing traffic flows in complex agglomeration systems. The findings advocate for the broader adoption of unconventional data sources to overcome traditional data scarcity in suburban transport research.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• urban rural setting