 Exposure to poor air quality leads to serious health conditions. This impact can be minimized by reducing our exposure to air pollution. We can do this by leveraging the latest technology which is available to us. One of the best ways of doing that is forecasting air quality in our cities. The accuracy of a forecast is as good as the quality of underlying input data. High data quality and versatile applicability of devices is the central design objective of OISM. An air quality forecasting system essentially requires two data sets. First, the condition of current air pollution, and second, meteorological data. To generate good air quality forecasts, the models require large numbers of input data points. Conventional air quality monitors due to their high cost cannot be deployed on such a large scale. This is the biggest advantage of sensor-based monitors in that they can provide high quality data at an economical cost. As a result, they can be easily deployed to create a dense monitoring network. With the help of modern data assimilation capabilities, additional data layers can be added to generate insightful air quality forecasts. The goal of technology is to help and improve the lives of people. It is also the central goal of smart cities to provide a clean and healthy environment for its citizens. Air quality forecasts can help city administration in issuing health alerts and advisory. It also facilitates data-driven informed decision making.