Standard municipal waste systems are failing because they rely on fixed schedules, ignoring the reality of overflow in high-traffic urban zones. We developed Seiketsu to solve this. Unlike IoT-heavy solutions that cost thousands to install, our framework uses the smartphones already in citizens' pockets. The system develops a weighted graph of the city where each edge (path between two vertices) has a weight determined by one or more of the following criteria: cleanliness, air quality, accessibility, safety, and cost of travel. A multi-objective approach is used to combine these criteria to calculate the weight of each edge which can be submitted to routing algorithms such as A* for routing contextually aware paths between vertices. In contrast to traditional ``shortest path'' routing, which focuses solely on distance and time, this method allows users to identify routes that are both cleaner and more environmentally safe. The Seiketsu platform has been designed to be scalable for integrating user-generated data and open spatial data, forming the basis for providing adaptive, sustainable urban navigation. In addition, a Logistic Regression model performs analysis on user engagement data to classify users as being in either a high or low engagement group. Seiketsu provides a solution for developing scalable and cost-effective smart city applications by utilizing a software-driven and participatory approach, rather than relying on expensive sensor infrastructure.
Partha Pratim Mahanta (Thu,) studied this question.