A user-flocksourced bus intelligence system for Dhaka

Abstract

Flocksourcing, or guided crowdsourcing, is an experimental data collection technique where users become the sensors to generate a large amount of information that improves a public service like public transport. In an ideal world, users would self-organize to collect data that would improve their own experience-but in a place like Dhaka, the megacapital of Bangladesh, the technology that enables users to become sensors is not (yet) widely available. In this thesis, we attempt to test the viability of flocksourcing by co-developing and seeding those technologies with a local resident-led flock who targeted a segment of the largely owner-operated bus system in Dhaka. The results from a flock of eight's weeklong data collection efforts demonstrated that flocksourcing can be a viable data collection technique for generating big amounts of data even in some of the most challenging urban environments. While the demonstration helps lay the groundwork showing that the technique can be used, questions still remain on whether and how it will be used in a truly sustainable way.