Crowdsourcing Air Quality Mapping


We are developing a product to address an emerging paradox in urban lifestyle and environmental sustainability.  Canadians are increasingly concerned about how their lifestyles can harm the environment.  Many are opting to reduce their carbon footprints by living in an urban core, close to their workplace and other amenities, thereby lessening air pollution caused by commuting.  Unfortunately, in choosing a lifestyle that is less harmful to the environment, individuals may end up living in a more polluted area themselves, since air pollution levels tend to be highest in the urban core.

This program seeks to address this conflict by providing accurate information about air quality to the government, as well as to commercial and individual stakeholders.  It is presently impossible to eliminate pollution from urban cores completely; however, it should be managed as much as possible.  Awareness is the first step, and our product will provide key stakeholders with the information they need to examine the areas most affected by poor air quality.  This information will also enable policymakers to examine factors that cause poor air quality such as traffic flow.  As a result, long-term strategies to mitigate the effects of air pollution will be better informed.


The product is a data collection system that includes a smartphone-based “sensor” to gather environmental air quality information in dense urban environments, analytics software to transform this data into a usable form (i.e., a map of how air quality changes in a city over the course of the day), and a dashboard application to display this data for easy accessibility.  Data collected will reflect changes to air quality due to traffic congestion, commercial and industrial emissions and other activities.  This program could be used to inform macro-level users (i.e., government), micro-level consumers who would use the information to make lifestyle changes (i.e., runners, bikers, commuters), as well as companies that wish to advertise to health-conscious urban consumers via smartphone application.

Product Development

A private company will develop the technology and own exclusive rights to the data collected.  Government policymakers, environmental organizations, activists and individuals could all benefit from this information.  Marco-level consumers could purchase the data for research, planning and policymaking purposes, while micro-level consumers could access the information relevant to their lifestyles and neighbourhoods via smartphones or computer applications.  Companies that target health-conscious consumers could purchase advertising within these applications.

Substitutes and Existing Products

The University of Southern California developed a smartphone-based application called “Visibility” that measures air quality by analyzing photographs that individuals take using their smartphones.  Individuals submit photographs wirelessly for analysis and then receive an air quality report.  As this application measures pollution via a photograph, it is affected by the colour of the photo and camera quality, and is therefore not very accurate.  In contrast, our product would measure air quality with precision via a dedicated sensor.  Furthermore, the “Visibility” application reports air quality for a given region, where as our product provides information for specific neighbourhoods.

Another existing smartphone application, developed by the company aMobile Future, provides users with information on air, water and soil pollutants for a given location.  This product does not measure pollutants based on real-time feedback, but rather provides static information that does not reflect how air quality levels change at the micro-level each day.  Similarly, many weather channels provide daily reports on factors that indirectly measure air quality, such as visibility ratings or smog alerts.  They do not, however, provide precise measurements for specific neighbourhoods, nor do they reflect how human activities affect air quality at the micro-level.

Performance Story

As illustrated in our logic model (Appendix A), we intend to develop a device that combines a carbon dioxide (CO2) sensor and a smartphone application in order to measure the air pollution within city cores via crowdsourcing.  Using a business intelligence process, we will chart these results on dynamic maps to display the air quality in different areas of the city’s core as CO2 levels change throughout the day.

Logic Model


In order to reach our desired outcomes, we must start by investing resources into product development.  Our product is a remote sensor device that will be used to measure the varying levels of CO2 across the city.  The actual price-per-unit of the remote sensor devices must first be calculated and compared to the budgeted cost.   We must then recruit activists who will use our product to collect CO2 data.  We will measure activists’ participation by how many activists we recruit, what percentage of city coverage they represent, and the fixed and potential variable costs of recruitment and retention—all compared to our forecasts.  Lastly, we must create a data collection centre and website in order to make the data accessible.  The data collection centre will be measured based on the cost of service, and the website will be measured based on the cost of development and ongoing maintenance—all measured as a percentage of our ongoing marketing costs and expenses.  At this level of the model, we must assume that there are sufficient activists in the area whom we may recruit in order to support our business model.


Using these inputs, CO2 data will be collected from across the city on a daily basis.  To determine city coverage, we will measure the percentage of the city from which we collect data based on the GPS locations of data transmissions, and compare it to our target coverage rate.  For example, collecting data from 90% of the city indicates higher performance than collecting data from 20% of the city, compared to a 99% target rate of coverage.  We will then extract this data, transform it and load it into a structure that is more easily accessible.  The percentage of useable data out of all of the data we have collected will indicate the process’ effectiveness compared to our forecasts as some overlap in data is probable.  Thus, although we have ample data, the quality of the data must also be measured as an indicator of high performance.  The data must be both valid, and reliable, which in this case means accurately measured air quality that may be transformed into “user-friendly” information.


We will create dashboards for to display our collected data to each type of user.  Macro-level users (i.e., government) will pay to access to a more enhanced and detailed dashboard that will be useful for research purposes, while micro-level users (i.e., individuals) will be provided a complimentary dashboard as part of the product offering.  We will measure the dashboard’s effectiveness by comparing the actual user rates to the user rates expected by the macro-level customers.  We assume that by taking the proper steps to ensure adequate city coverage, the data collected will be accurate and representative; for example, the data will not come from a single location in the city, but rather, from many points across the city.  Secondly, we assume that the device is user-friendly, and that the average individual will be able to use it easily.


Our business idea is directed at the government (specifically the Ministry of the Environment), as well as both companies and individuals who may be interested in the data collected.  The key performance indicators for those who participate are the amount of sales that we make to macro-level users and application downloads by micro-level users compared to sales targets, as well as application advertising sold to companies who target health-conscious urban consumers.  We will also measure the short-term outcomes through surveys determining the rate of satisfaction of users compared to target rates.

The data that we provide will reveal which locations in the city have the highest CO2 levels.  In the short-term, municipal governments could use this data to make infrastructure changes; for example, city planners could focus on re-positioning traffic lights to help reduce the number of places where cars idle in the city, or change the location of city paths so that pedestrians, cyclists, joggers, etc., will not have to travel through areas with high CO2 levels.  The ultimate outcome will be for the government to improve macro-level air quality, which we can track using yearly assessments of CO2 concentration in the city (3, 5, 10 years and ongoing).

We anticipate that companies will be interested in advertising to health-conscious urban consumers through the smartphone application.  Some companies may find the data useful as well; for example, real estate companies could attract health-conscious condominium buyers by using the data to select less-polluted neighbourhoods in which to build.

Lastly, we believe that individuals would be interested in accessing this data in order to make informed decisions related to their health.  By being able to see which parts of the city have the highest levels of CO2 at certain times of day, individuals will be able to avoid these areas when they go jogging, etc.  This would be particularly useful for people who suffer from respiratory problems such as asthma; avoiding areas of the city with high CO2 levels could help decrease the number of asthma attacks that the individual suffers since asthma is directly linked to high CO2 levels.  The overall outcome is to improve the lifestyle of individuals in the city, and can be measured based on the overall decrease in the number of cases of respiratory illnesses caused by poor air quality.  We make the assumption that there are adequate medical resources to assess the inherent causes of respiratory illnesses.


Risks to this project’s success were assessed by taking into account the KPIs listed in the logic model (Appendix A) and by quantifying key risks in terms of likelihood and impact (Appendix B).  Using this approach, we have identified major risks and developed mitigation strategies for each.


Risk: Activists may lose the device or reduce their level of commitment to the project, resulting in inadequate data collection.

Mitigation: Design the device to attach to a keychain when not in use.  Design the application to remind the user to plug in the device, and to notify the company if it has not been used for a certain period of time (i.e., 5 days).  The company can then follow up with the activist to determine if the device has been lost, etc.

Risk: Activists’ foot traffic may not align with vehicular traffic (particularly due to weather/seasonal factors) or they may not travel sufficiently on a regular basis to provide thorough city coverage, therefore, data collected may not be representative of true CO2 levels in all areas.

Mitigation: Predict activists’ city coverage based on their daily routine routes (i.e. where they reside and where they work/study).  Recruit additional activists in low-coverage areas or ask activists to take alternate routes.  Restrict coverage to main routes during offseason; recruit committed activists to cover these areas.


Other Top Risks

Risk: Cost may affect both macro- and micro-level customers’ choices, thus hampering penetration into new market.

Mitigation: Offer accurate data that is not available elsewhere to increase demand.  Outsource the device manufacturing to developing countries to lower prices.

Risk: Improving air quality may not be a significant political priority, hampering the government’s willingness to pay for the information.

Mitigation: Promote the importance of clean air and its benefits to individuals via advertisements that encourage the public to do their part for the environment for their own health.  In turn, the public may pressure the government to pass air quality legislation.

Risk: Other pollutants aside from CO2 (i.e., dust, chemical waste) that the sensor does not detect may cause poor air quality.

Mitigation: Remind government users that CO2 concentration is only one component of air pollution.  While CO2 concentration is one of the two main forms of air pollution and traffic is the number one cause of air pollution in urban environments, the government should be advised to closely monitor other pollutants (i.e., dust, chemical waste) on a regular basis.


Project Team: Matthew Farnand, Adam Zaret, Thibaud Clement, Britney Hanlon, Matt Xu, Dana Summers

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>