Data-Powered Circular Economy - From Unstructured Data to Positive Outcomes

Amid geopolitical turmoil and the global impacts of climate change, leaders around the world are increasingly turning to the circular economy as a conceptual framework to align human activities with planetary boundaries. 

In a circular economy, the value of products and materials is maintained for as long as possible and waste and resource use are minimised. This definition underlines the general principle of a system that decouples economic activities from finite primary resource consumption. 

More than ever, leaders are seeking opportunities to harness the power of data and create analytical systems that enable informed decision making for coordinated global action. Several data driven tools have been developed to assess material and waste flows. One of them developed by GIZ, University of Leeds, eawag and wasteaware, the waste flow diagram is increasingly used in development projects to visualise waste flows and quantify plastic leakage into the environment. After all, data and robust metrics play a critical role in developing and monitoring progress toward the circular economy. 

Together with the GIZ Data Lab and the GIZ Global Programme Go Circular, we consolidated various data sets and explored their potential to accelerate the circular economy transition. 

Until today, traditional data capturing techniques such as conducting household surveys, gathering census data etc. are most prevalently used in international development projects. However, because of their potential to reduce labour-costs and improve scalability, non-traditional data capturing techniques (i.e., data automatically generated by technology, like satellite imagery) are increasingly entering the limelight. 

As one of many benefits, the use of non-traditional data can help practitioners gain an initial understanding of outliers before mobilising resources locally. For example, when investigating litter prevention measures, non-traditional data analysis can help with the identification of communities that employ particularly effective practices. The behaviour of these “positively deviant communities” can then be studied in more detail to eventually replicate their resilience-building practices in other regions. 

The following table presents a summary of the opportunities and limitations of using non-traditional data for circular economy research:

 
Table 1: Summary of the advantages and disadvantages of non-traditional data use 

The transition to a circular economy requires more transparency to make a big leap forward. Currently, a particular challenge is data collection and access, as supply chain metrics such as material use and waste generation are either not collected at all or often not readily available when needed. This hinders the ability of governments, international development organisations or private sector actors to adopt, use, and report on circular metrics, and to implement the circular solutions we need. 

Practitioners and researchers interested in studying the transition to a circular economy also face the challenge of a lack of readily available data sets and harmonised categorization. As the need for more seamless data access and collaboration becomes more apparent, now is the time to consolidate all readily available information and lower the barriers to entry for circular economy analyses. 

We have created a data catalogue (see Figure 1) that summarises some of the most important datasets related to the circular economy transition. The data catalogue contains 34 resources that cover different geographic areas and harness a variety of technology-based data collection techniques. It allows users to either skim the available data or find specific information relevant to their individual circular economy project - particularly suitable for universities, think tanks, researchers and students. 

Figure 1: Circular Economy Data Catalogue - access for free here

In many research projects, data collection is a key challenge to overcome. Once an initial data set is found, practitioners often face the additional hurdle of filling data gaps and removing biases from the collected data. One reason for this is that each data collection technique comes with its unique advantages and limitations.  

In the example developed for this project, we focused on monitoring plastic waste in different regions of the world. We mocked the data captured by different technology sources such as CCTV cameras, drones, satellite imagery and social media channels.  

The advantages of our “multi-data source approach” (see Figure 2) are that litter item records captured from one source can be complemented with additional data points, verified with unbiased reference data, and analysed in more detail to continuously improve the research model. 

Figure 2: Harnessing non-traditional data for litter monitoring – access data story here

In many development projects, statistical methods are insufficiently used to inform strategic and tactical decision-making processes. For example, while many projects collect some sort of data, they often stop at traditional data collection techniques and descriptive statistics.  

However, to truly measure outlier capacity or the potential to achieve positive impact beyond an existing baseline, it is essential to apply inferential statistics and test hypotheses about the significance of certain behaviours over others.  

The novel methodology of “Data-Powered Positive Deviance (DPPD)” allows for a structured end-to-end approach from data collection to project implementation. DPPD assumes that in any population, there are individuals or communities that perform better than their peers despite similar challenges and constraints.  

For a pilot project application using our data catalogue and DPPD, have a look at this data story here.  

Figure 3: Access the DPPD Handbook here

In conclusion, our research confirmed that capturing data with digital technologies is merely an enabler rather than the solution itself - especially in the field of international development. Even in today’s digital society, data cannot solve all our problems and thus in many cases relies on human ingenuity as a complementary resource.  

Nevertheless, we hope that the insights and tools we shared in this blog post will support your everyday life with new inspiration and guidance.  

Thank you for your time. We look forward to accelerating the development of a data-powered circular economy with you. 

 Dominic Santschi is a Co-Founder of Ampliphi, a data-driven plastic action and circular economy platform for consumer goods companies. In his current role, he also leads a working group on plastic data and accounting within the PREVENT Waste Alliance. Dominic holds a MSc in Climate Change, Management & Finance from Imperial College London. He grew up in the Swiss Alps and has been connected with nature ever since. 

 Steffen Blume works as a project manager in GIZ GoCircular and has specialized on plastics and marine litter prevention. Currently, he coordinates GIZ circular economy projects and is contact person for the PREVENT Plastics working group. He worked in Germany, Serbia and Zambia and has a passion for environment protection and data driven approaches.

 Robin Nowok is a specialist at the GIZ Data Lab who has a background in geosciences and energy system optimization. He has expertise in using digital methods like analyzing earth observation data with geographic information systems. Currently, he is focused on using innovative digital methods (e.g. harmonized databases, Big Data, NLP, etc.) to adress the severe impacts of the climate crisis.