Targeted Advertising in the Public Transit Network Using Smart Card Data

A great number of urban residents uses public transit network to travel and reach their destination. While the public transit network could perform as a valuable medium for advertising purposes, the share of transit advertising in annual advertising spending is low due to the lack of passengers’ profiles. This paper proposes a targeted advertising model in the public transit network regarding the extracted passengers’ profiles from smart card data. The model exposes advertisements to groups of passengers in the public transit network regarding their activities and trips. A targeted group includes passengers with similar activities (considering type, location, and time of the activity) and trips (considering spatial and temporal dimensions of the trip). An agglomerative hierarchical clustering method is used to discover activity-trip groups of passengers according to the defined activity and trip similarity measures. An optimization problem is formulated to allocate advertisements to all activity-trip groups aiming at maximizing the coverage and minimizing the cost of the advertisements. Non-Dominated Sorting Genetic-II (NSGA-II) algorithm is used to solve the optimization problem. One-day smart card dataset from Brisbane, Australia is used to implement the model and examine the outcomes. Results show that at different cost intervals, solutions with high coverage can be applied to the network targeting all the activity-trip groups of passengers.