Citizen sensing system optimization for infrastructure monitoring: developing community currency model which incentivizes behavior changes of "citizen sensors"

Abstract

Infrastructure monitoring has been a big burden for local authorities in Japan. Change detection of aged infrastructure such as roads and ports by using satellite remote sensing has been introduced to judge the needs for their maintenance, however, it has the challenges of rapidity, accuracy and certainty of the observed data. To cover the weakness of remote sensing, reporting from citizens (so called "citizen sensors") which detect the local or micro changes/damages of the infrastructure could work. By matching the remote and citizen sensing data, we could verify and improve the certainty of the observation results. However, there is a challenges of citizen sensing system: incentives and quality control. Upon this problem recognition, this study aims to develop the model to provide incentives to citizens to collect and report the quality data by using the "community currency" system. "Community currency" is a tool to covert the unseen values in the community into the monetary values (such as discounts points or rewards that could be used in the local shopping) to change the behavior of citizens to take such unvalued actions. This paper shows the results of developed model to calculate the reasonable and workable monetary value (rewards) to motivate citizens to collect and report the changes of the community infrastructure (e.g., micro damages of the roads surface) with required quality. The main research question was optimization of the calculation/conversion model to value the citizen sensors. For this, the model has taken the approach to value the expected cost reduction and extra benefits (i.e., damage avoidance) that could not be obtained without citizen sensing (assuming the existing surveying works by local authorities). The result also shows the community currency system with the developed model could enhance the quantity and quality of the citizen sensing data.