Air Assistant

To develop a smart application to monitor air pollution


My bachelor thesis project explores how to create an engaging long term experience with a smart application. The result is a concept of an adaptable, Smart Air assistant focused on air pollution. It mainly includes carbon dioxide and carbon monoxide. The assistant supports users with the topics in monitoring the ventilation and heating to create healthier, more comfortable, and energy-efficient air in the home. 

The project was sponsored by Municipal Corporation to avail Air Assistant app in nearby areas. 

Role & Duration

Product Designer

Product thinking, Design System, tools, User survey, Logo design, Prototyping and testing

Team- 2 designers, 2 software engineers

Jun 2016 - Mar 2017


Why design a smart app for Air Pollution?

The Times of India has reported 1.4 lakhs death in Maharashtra, India due to Air pollution. During an Industrial visit to nearby areas, we observed the release of Carbon-dioxide due to industries and vehicles. Yet even on days when the air quality degraded to higher,  we rarely saw people take adequate precautions. Fueled by those terrifying facts, we sought to build something to raise awareness and motivate the haze-shrouded country.

How might we develop an app for monitoring air through additional features?

Smart home devices are installed permanently in the home still often have a hard time engaging their users in the long run. With IoT products becoming more service-centered, creating a sustainable interaction with the user is a significant challenge for design since revenues depending on users drawing continuous value from a service instead of just an initial purchase. About 90% of our life is spent inside. Besides temperature, other components like air quality, temperature, and noise levels are essential for staying healthy and comfortable. Nevertheless, a harmful indoor environment is a problem that is overlooked. How can Air Assistant provide valuable support in this field to create further touchpoints with users?

The Problem

To design an application for a system that will monitor air level in environment and update the user from time to time. 


Through the background research, we followed below the design process to develop a smart app.

UX Research

UX Design

Design flow

Air Quality Installation

We followed the design by implementing various sensors, required for monitoring the air. This proposal involved sensors and databases connected to the main cloud storage.

Engaging users outside the proposed system

I thought about how an engaging object in addition to my app concept could look to better communicate my project at the university exhibition. To get people's interest but remain simple at the same time the object should visually evoke associations to the general topic and app concept. Content-wise it should communicate the current status of the indoor climate.

I moved from prototyping and testing small ideas to creating a functional prototype with Arduino. As an exemplary feature, the prototype measures humidity and temperature.


To better understand the current scene and discover opportunities where an app might help, we interviewed residents of Sangli district, asking about their views and feelings on air pollution, and their daily routines.

Our hypothesis

People are suffering from air pollution, but still do not understand the health impacts of air pollution and the effectiveness of installing a system to monitor the Carbon-dioxide levels. 


Through our interviews, we observed following problems and views of people.


Limited solutions available to monitor the carbon-dioxide level in near by surroundings.


Air Pollution management issues is one complaint, and the number one reason that people cannot balance out the usage of other household chores.


People are willing to invest in an mobile application that would help them monitor and balance out CO2 level in industries. house, oil factories, etc


Structures available in the market are expensive and installed on large scale 


The app will guide and keep users in safe and better place. It will also help in reducing the air pollution.

As we moved into defining the problems to solve, our learnings and insights gave clarity about what to build. We learned a lot of useful information, but there were two prominent themes we extracted from my learnings.

Define/ Synthesize

Needs and motivation

Pain points

How might we use motivation to make use of the proposed system?

How might we create a user-friendly interface to enable maximum use?

Observations were collected from the recordings and notes of the interviews, clustered into topics, and translated into needs, pain points, and key insights. There are different motivations of why people care about improving components of their indoor environment which Air Assistant can address. For many users, the topic is connected to underlying goals like well-being and an optimized lifestyle.  While many users have a pragmatic view on the topic, a potential solution also needs to address people's need for reassurance and peace of mind.

We synthesized the research insights into opportunity areas in the form of user needs and distilled those into two guiding questions that correlated to the themes-


The survey helped us get an idea of how we can further define our scope and concepts by focusing on what specific household issues are dealing with, and how common they occur.

24 Target Users

Survey chart

User Entactments

Our experience prototyping study involved conducting 3 user enactments with each of our 3 participants for a total of 9 user enactments to answer our four research questions. Through the user enactments, we wanted to understand users’ perception and the strengths and weaknesses of our product in varying scenarios.

Use cases for the assistants


A two-pronged framework

Drawing from the two HMW’s, we generated ideas for solutions. Some of these included the ability to, getting sense the temperature in the surrounding area, notifications when users are experiencing dangerous levels of air pollution, and gamification of in-app actions.  We bundled these ideas in various ways to get a feel for how the most promising ones could fit within an educate-and-motivate framework.

Rough framework for initiating the ideas

Findings and Design Implications

As we conducted research about user’s general pain points, it was in this stage of our research that we were able to directly observe users interacting with our product in the way they would in their home. As a result, we were able to perceive the user's authentic reaction to our suggested features, rather than our abstract ideas or hypotheticals. It was these reactions that helped us gain a more accurate understanding of if and how our ideas were addressing user needs and the pain points we had previously identified.

Users should have more control over the system

People chose to rely more on the app rather than the automation part. Set up should include the feature, voice assistant, and notification bar. 

Being unable to react is the worst

Notificiations should be service-oriented

Data is a king

Display information in an easy to read visualizations, and show progress.

The system should be responsive

Users should be able to log in and connect data with various devices.


A simple blue color logo for carbon dioxide to indicate the richness of application and enhance the meaning of theme.

Logo Making


Paper Prototyping

We made use of low fidelity prototyping to understand whether the design solution I had envisioned was working or not. It helped us understand how the users would use the application. We conducted usability tests using the paper prototype. During the usability tests, we got to know what solution was missing and whether users were facing problems looking at the mobile application.


The final concept is adaptable to changing needs during different stages of usage. A conversational layer connected to all features gives the assistant the possibility to reach out to the user and provide in-context help to increase engagement. The assistant will over time get to know the user and learn to show relevant content and send notifications at the right moment.

Customize list

On the initial page, a user can choose up the location and select the main issue accordingly. 


An adaptable home screen can suit changing needs over time: during a busy day a simple status is often enough - but if there is a problem the user can access detailed information. A pie chart that visualizes the amount of carbon dioxide in the given particular area. Users will able to follow the trend of ongoing carbon dioxide level


The trend will enable us to track the amount of CO2, temperature, and humidity occupying the surrounding area. A track that features with a given specific unit of the sensor. 


Make annoying little tasks, like improving air, opening your windows regularly, and creating healthy habits easier through motivation and tangible goals.

Engaging dashboard

With trending stuff, the dashboard allows navigating through tips and notification. Tips are one of the primary features to guide users about significant suggestions. 

Users get an overview of the total carbon dioxide pressure in the surrounding area. This interaction explores the how users can quickly look at the data from their mobile by connecting their device via Bluetooth.

Device Prototype

After researching different IoT microcontrollers, our group decided on utilizing a Raspberry Pi 4 Model B+ instead of an Arduino due to limited functions. The Raspberry Pi's core processor matched our processing designs and inbuilt WiFi function provided inside the device. It helped us connect the device to the computer and monitor the data uploaded on the cloud. The Raspberry Pi's extensive community helped guide our software implementation that involved running open-source "Motion" software, which offers to sense the movement or pressure. The Raspberry new model helped to connect the data to the Twitter page, which availed the users to monitor the Carbon dioxide levels. Twitter was our secondary platform. 

CO and CO2 sensors

Onboarding of sensors via connecting to ThingSpeak platform


How did it go?

It was challenging to take on a problem and create a solution for a mobile application that daily users implement on daily basis.

Implementing a system

I believe that this designed platform would help the users in case this problem statement would implement in real-time. This information could be further used in professional work or as a base to design small interfaces for Internet of Things apps. I look at my solution as an icebreaker for people who want to get introduced to a new IoT and Interaction design field.

Further user research and testing

What other features would people in the surrounding benefit from? One consideration would be introducing user customizations such as setting a custom range of elevation that the user manages. I would like to improve upon the UX and app features based on further user research and testing

Are you a design geek, a pretzel lover, a movie buff, or a foodie who likes trying out new cuisines? If yay, do get in touch! ✨

© 2021 with 🥨 Savani Shrotri