CUMTA’s decision to map travel demand around Chennai’s 41 operational Metro stations is more than a data exercise. It is an attempt to identify the journeys that the network does not currently capture—and to understand why people who can reach a Metro-served destination may still choose another mode of transport.
The Chennai Unified Metropolitan Transport Authority has invited bids for a two-month consultancy to prepare origin-destination matrices for the catchment areas of all Phase-I stations. The stations together cater to nearly 3.5 lakh passengers daily. The study will use anonymised mobile and GPS data to examine movement around each station and support the planning of first- and last-mile feeder services.
That focus is important because Metro systems do not operate as isolated rail corridors. Their usefulness depends on the ability of passengers to reach stations conveniently, affordably and predictably, and to complete the rest of their journeys after leaving the train. Ticketing data can show who enters and exits a station, but it cannot fully show the travel choices made by people who live, work or travel within the station’s wider catchment but do not use the Metro.
CUMTA special officer I Jeyakumar said the exercise would examine not only how existing Metro passengers travel, but also wider movement that the network is currently not capturing. A person travelling between two locations served by the Metro may not necessarily use it, he said, making anonymised mobility data necessary to understand those movements beyond ticketing records.
The proposed geography reflects this wider approach. CUMTA will define a five-kilometre road-distance influence area around every operational station and divide it into traffic analysis zones of approximately 0.5 square kilometres each. Using road distance rather than a simple circular radius is intended to represent the routes people can actually take through Chennai’s urban street network.
The distinction matters for station access. A location that appears close to a station on a map may be separated from it by major roads, disconnected streets or other barriers. Conversely, a feeder route may link a station to a wider area than a straight-line distance suggests. The source material does not establish which locations will emerge as underserved, but the proposed traffic analysis zones are designed to make such differences visible at a more granular scale.
The consultant will have to procure anonymised GPS traces from smartphone applications or mobile-network data covering at least three months. The tender requires representative coverage of around 10 to 15 per cent of the target population. The data supplied to CUMTA, according to the authority, will contain aggregated travel movements rather than individual movement histories.
This gives the exercise two distinct functions. First, it can identify the movement patterns of current Metro users, including where they begin and end their journeys, how long their trips are and when they travel. Second, it can show travel between locations that the Metro network serves but does not necessarily attract. The second function is potentially more consequential for planning because it can reveal demand that exists independently of current ridership.
The raw location information will not be directly converted into ready-made passenger journeys. Machine learning-based trip-chaining algorithms will be used to convert location pings into trips and dwell periods. The resulting movements will then be classified by origin, destination, peak or off-peak period and trip length.
CUMTA has asked for separate origin-destination matrices covering average daily travel, a three-hour morning peak, a three-hour evening peak and a six-hour off-peak period. This time-based structure can help distinguish different types of access problems. A station may face heavy demand during office commuting hours but a different pattern during the middle of the day. Likewise, a feeder service designed around the morning peak may not respond to off-peak travel needs.
The proposed output is not limited to a report. The consultant will prepare a web-based visualisation platform showing the study area, traffic analysis zones and origin-destination movements through heatmaps and desire-line diagrams. Such visualisations can allow CUMTA to see where movements concentrate and how they relate to station locations.
The institutional significance of the project lies in the authority’s attempt to retain the underlying planning capability. The consultant will be required to hand over raw and final datasets, maps, geographic information system shapefiles and working files to CUMTA. This means the assignment is intended to create an evidence base that remains under the authority’s control rather than producing only a one-time set of recommendations.
The privacy question is central because the proposed method depends on location data collected through smartphones and mobile networks. CUMTA has said that individual-level data will not be supplied to it and that the information it receives will be anonymised and aggregated. Jeyakumar also said the data would remain under the authority’s control. These safeguards describe the intended handling of the information, while the tender’s data requirements will determine how the consultant obtains, processes and transfers the datasets.
The study also highlights a recurring challenge in urban transport planning: the difference between network availability and practical accessibility. A Metro line may connect two parts of a city, but the existence of stations alone does not establish that all potential passengers can use the system conveniently. The missing link may involve the distance to a station, the absence of a feeder service, the timing of connections or a mismatch between travel patterns and the operating design of surrounding services. The supplied material does not identify which of these factors is dominant around Chennai’s stations, but it sets out a method to investigate them.
Chennai’s Phase-I network currently has 41 operational stations and nearly 3.5 lakh daily passengers, according to the report. CUMTA’s decision to study movements within five-kilometre road-distance catchments therefore shifts attention from station-level ridership to the wider urban areas that stations are expected to serve. The data could show whether demand is concentrated close to stations or distributed across larger areas requiring organised feeder connections.
The proposed study is also a test of how transport authorities define demand. If demand is measured only through existing Metro trips, the analysis risks describing current users rather than identifying people who are not using the system. By combining ticketing-independent mobility data with station catchment mapping, CUMTA is seeking a broader picture of movement. That does not by itself guarantee better services, but it can give planners a more detailed basis for deciding where feeder links and other access interventions are needed.
The project’s limitations will need to be read alongside its potential. Smartphone and mobile-network data may provide broad movement patterns, but the source material does not specify how different user groups, modes or travel purposes will be distinguished beyond the classifications listed in the tender. Nor does it establish how the authority will translate the resulting heatmaps and origin-destination matrices into service changes. Those details will become important once the consultant is selected and the study is completed.
For Chennai, the larger urban question is whether Metro expansion can be matched by a planning system capable of understanding the streets and neighbourhoods around every station. CUMTA’s tender recognises that rail infrastructure is only one part of a journey. Its proposed use of aggregated mobility data attempts to measure the rest of that journey at a scale fine enough to guide station-area and feeder planning.
The immediate next step is the appointment of the consultant for the two-month exercise. The study is expected to produce station catchment maps, time-based origin-destination matrices, a web visualisation platform and datasets that CUMTA can retain for further planning. What remains to be seen is how the authority uses that evidence to address the movements the Metro network currently fails to capture.
