Origin-destination (O-D) data

Origin-destination data records where trips begin and end, whether by car, transit, or on foot, without necessarily tracking the specific path taken between the two points. Aggregated across many trips it is usually organized into an O-D matrix, a table showing how many trips moved from each origin zone to each destination zone over a given period.

Sources range from household travel surveys and transit ticketing to anonymized, aggregated mobile-location data and connected-vehicle records, each with different coverage, cost, and privacy tradeoffs. This differs from traffic flow analysis, which measures conditions on the road network itself; O-D data instead describes underlying demand, the trips people actually want to make, independent of which route they took to make them.

Planners use O-D data to design transit routes, forecast the effect of new development through land-use modeling, and build evacuation modeling scenarios that need realistic assumptions about where people will actually try to go. A common pitfall is treating an O-D matrix built from one data source, such as a single mobile carrier's users, as representative of the whole population, when coverage can skew heavily by demographic and device type.