2026-11-11 –, Video Room
Public Transport Network Analysis in the City of Torreón, Coahuila, Mexico
Claudia Guerrero Sepúlveda
The city of Torreón, Coahuila, is experiencing a chronic public transportation crisis that has led residents to rely increasingly on private automobiles and motorcycles to meet their mobility needs, particularly in the areas where the city has expanded. According to the Vehicle Registration Program (VMRC, Vehículos de Motor Registrados en Circulación) maintained by the General Directorate of Statistics of INEGI, the number of passenger buses increased from 3,988 in 2015 to 6,522 in 2024. However, their share of the total vehicle fleet declined from 2.79% to 2.39%. Furthermore, within this category, vehicles dedicated to public transportation decreased from 48.23% to 38.88% during the same period, showing a downward trend despite continued population growth and the expansion of the urban territorial index—calculated by INEGI to define urban and rural areas in Mexico. These trends suggest that the problem is not only the insufficient number of vehicles but also the limited territorial coverage of the public transport network.
This study proposes a spatial analysis of the public transportation system using route data provided by IMPLAN (Torreón Carto API) together with INEGI data, including population information from the 2020 Population Census and road network data from the 2025 Geostatistical Framework. The analysis includes the following components:
1) Route segmentation and overlap
The 55 routes available through the IMPLAN Torreón database are segmented, and the number of unique routes sharing the same roadway is calculated using a 5-meter tolerance. Segments are classified into the following categories: 1 route, 2–5 routes, 6–10, 11–15, 16–20, 21–25, 26–30, and more than 31 routes. This analysis identifies areas with a high concentration of public transport services as well as areas where no routes operate.
2) Route intersections and redundancy
The percentage and total length of each route that overlaps with other routes are calculated. Shared lengths are summarized in a pairwise comparison matrix, and the intensity of overlap is represented through a heat map showing the 20 routes with the greatest shared distance. The complementary measure, namely the unique distance traveled by each route, is also calculated.
3) Network topology and structure (Sienkiewicz & Hołyst, 2005 metrics)
The analysis calculates the number of routes sharing segments with each route (degree), the total shared distance with neighboring routes (strength), the average distance to all other routes (closeness), transfer nodes (locations where more than two routes converge), and betweenness centrality, which identifies the routes that are critical for network connectivity. It also identifies the most interconnected routes, as well as isolated or redundant routes that could potentially be removed without significantly affecting the overall network.
4) Accessibility and territorial coverage (population by census block, INEGI)
By combining the spatial coverage of public transport routes with census block population data from the 2020 Population Census, the study estimates the population located within five- and ten-minute walking distances from the network, as well as the percentage of the population with access to public transportation. Although equitable access to the network is desirable, some neighborhoods are served by numerous routes while others have none. Therefore, the Gini coefficient and the Lorenz curve are calculated to measure inequality in network coverage.
5) Oversupply and territorial efficiency
For each route segment, the population of census blocks whose centroid falls within a 400-meter buffer is calculated. The oversupply index is defined as the relationship between the number of routes traversing each segment and the population served, together with the number of network meters per inhabitant as an indicator of territorial efficiency. This analysis identifies areas lacking service, as well as sparsely populated zones such as industrial areas, parks, and the airport.
6) Route geometry and spatial analysis
The sinuosity index is calculated as the ratio between the actual route length and the straight-line distance between its origin and destination, together with the percentage by which the route exceeds the direct path. Another indicator is the circularity index, which identifies routes returning close to their starting point, as well as route deviations, detected when the endpoint of a segment is located more than 200 meters farther from the destination than its starting point. These indicators are calculated for each segment and then expressed as a percentage of the total route length.
The Historic Downtown polygon is intersected with each route to calculate the percentage of its length located within this area, allowing routes to be classified as central trunk, central, or peripheral. In addition, the predominant orientation of each route is determined using the circular mean weighted by segment length. Finally, natural mobility corridors are identified by grouping segments shared by more than three routes according to their orientation, revealing the roadways where multiple routes converge in the same direction.
The identification and measurement of these indicators, such as route sinuosity, directly affect the number of buses required to maintain service frequency and travel times. Network centrality and peripherality reveal the concentration of routes in the city center, forcing passengers traveling between peripheral areas to transfer through downtown, where numerous routes converge.
The analysis of the indicators derived from these approaches highlights the structural problems of the public transportation network and demonstrates the need for further studies involving transportation specialists to identify major trip generators and attractors, such as wholesale markets, shopping centers, hospitals, and educational institutions. Such studies would improve network coverage through the design of new routes and strategically located transfer points, promoting greater use of public transportation while reducing traffic congestion, traffic accidents, and CO₂ emissions.
Use of Artificial Intelligence Tools
Claude (Anthropic) was used as a programming assistant during the development of the spatial analysis workflow. Specifically, it assisted in generating, debugging, and optimizing code in the R programming language, as well as providing suggestions for implementing spatial algorithms and network analysis procedures. All analytical decisions, validation of results, interpretation of findings, and conclusions were performed and verified by the author.
Korompay, M. & Földes, D. (2025). Where to change from public transportation to car-sharing? Developing a transfer point optimization method. Transactions on Transport Sciences, 16(SI SCSP), 4-9. https://doi.org/10.5507/tots.2025.005
Sienkiewicz, J. & Holyst, J.A. (2005). Statistical analysis of 22 public transport networks in Poland. Physical Review E, 72, 046127. https://doi.org/10.1103/PhysRevE.72.046127
Geurs, K.T. & van Wee, B. (2004). Accessibility evaluation of land-use and transport strategies. Journal of Transport Geography, 12(2), 127-140. https://doi.org/10.1016/j.jtrangeo.2003.10.005
Currie, G. (2010). Quantifying spatial gaps in public transport supply based on social needs. Journal of Transport Geography, 18(1), 31-41. https://doi.org/10.1016/j.jtrangeo.2009.01.002
Levine, J. & Karner, A. (2019). Route directness and transit ridership. Transportation Research Record, 2673(5), 425-435.
Alonso, W. (1964). Location and Land Use: Toward a General Theory of Land Rent. Harvard University Press.
Gutiérrez, J. & García-Palomares, J.C. (2008). Distance-measure impacts on the calculation of transport service areas using GIS. Environment and Planning B: Planning and Design, 35(3), 480-503.
Mardia, K.V. & Jupp, P.E. (2000). Directional Statistics. John Wiley & Sons.
Boeing, G. (2019). Urban spatial order: street network orientation, configuration, and entropy. Applied Network Science, 4, 67. https://doi.org/10.1007/s41109-019-0189-1
Derrible, S. & Kennedy, C. (2010). Characterizing metro networks: state, form, and structure. Transportation, 37(2), 275-297.