Gary Watmough

Image

What is your innovation idea? 

My research commercialisation idea is to develop estimates of socioeconomic conditions at fine spatial scales on an annual basis using available data and satellite imagery. It will require frameworks to be created that allow context specific machine learning models that can be updated and trusted by companies to produce accurate estimations of socioeconomic conditions. 

Why does this matter? 

There is an increasing need for companies, banks, investors to know how their investments are impacting the communities in which they are based. Supply chain analysis is required to ensure that companies are having no negative impacts on the areas where they are investing. To produce these estimates we need to have information on environmental conditions and local populations, which is a requirement of ESG reporting. Data on environmental conditions and changes in these conditions over time are well established and are an increasing area of growth in the industry. However population data, that is data on how many people there are as well as the socioeconomic levels/conditions in these areas, is often lacking. Traditional survey-based measures of poverty and livelihoods are often very high quality. However, they are expensive and time consuming and often have gaps of years between them. Thus, ESG reporting is difficult, often relying on national level records and out of date information. 

What problem will it solve? 

Current methods are too general and therefore lack accuracy in local areas. Since most companies invest in regions of a country they need regional specific estimates of their impacts. This isn’t possible with the large generalised global models that currently exist. My research commercialisation will provide this more local output. Once this is available it is more likely that we can then develop estimates in countries with little data that are emerging from periods of unrest. Thus, providing investment focused companies an opportunity to understand the conditions in new regions earlier. 

What is the future of your research? 

My concept is focused on product innovation in taking the research on poverty estimations from satellite data and turning them into products that can be used to inform ESG reporting as well as allowing organisations to consider longer term projections of their impacts. Additional products such as frameworks/pipelines for estimating conditions in countries with little or no data due to conflict could also be developed which would potentially be framed around process innovation as we take our product and develop additional processes for applying it elsewhere. 

What motivated you to apply to the Bayes Innovation Fellows programme?

One of the main attractions of the Bayes Innovation Fellowship is the opportunity to use my experience of co-production and co-creation of research projects to work with funders, companies, and investors to tweak the products and processes to work for their needs.