Abstract

Ellona provides solution for indoor environment quality monitoring, including air quality, comfort and odor nuisance measurements. These data can be difficult to interpret for non-expert stakeholders (e.g., Facility Managers, Health and Safety Managers...). In collaboration with customer success and product development teams, we designed an automated analytical report pipeline aimed at non-technical customers. Expert rules were combined with Gaussian modeling to define three KPIs relevant for health and confort surveillance on any indoor site. We implemented the business and scientific logic in Python and set out on a static yet responsive user interface built with a mix of HTML/CSS/Google Charts. A beta release was first tested on a selected pannel of customers. Their feedback was used for minor UI adjustments and release the first version of the product. A Python API was designed to interface the module with backoffice parameters and equipment fleet database (MongoDB). This endpoint was called periodically to produce batch of reports for subscribing customers.