Moving beyond averages: uncertainty and variability analysis in industrial ecology

Jonathan Cullen

University of Cambri

Jonathan Cullen is the Lecturer in Energy, Transport and Urban Infrastructure, at Cambridge University Engineering Department (CUED) and a Fellow at Fitzwilliam College, Cambridge. He leads the newly formed Resource Efficiency Collective, exploring options to reduce energy use and emissions through resource efficient solutions. Jonathan holds an undergraduate degree from the University of Canterbury, and MPhil and PhD degrees from the University of Cambridge, all in engineering. He is the co-author of Sustainable Materials: with both eyes open, described by New Scientist as “clear, well-supported, and vital” and recently listed in Bill Gate’s top six reads for 2015.

David Laner

Technische Universitaet Wien

David Laner is a Senior Researcher at the Institute for Water Quality, Resource and Waste Management at the TU Wien. He holds a Master of Science in land and water engineering (2005) from the University of Natural Resources and Life Sciences in Vienna and a PhD in civil engineering (2011) from the Vienna University of Technology. His major research interest is the analysis and design of waste and resource systems, in particular by using material flow analysis and life cycle assessment.

Rick Lupton

University of Cambridge

Rick is a Research Associate in the Use Less Group at the University of Cambridge. He studied engineering at the University of Cambridge, before working for several years on the design and modelling of wind turbines, first at a renewable energy consultancy, and later back in the Department of Engineering at Cambridge. Since completing his PhD in 2014, he has been working with Professor Julian Allwood on modelling and visualisation of resource use.Rick's research interests include dealing with uncertainty in modelling, using analytical and stochastic methods, and the subsequent analysis and visualisation of model results. He is a supporter of open research and is the author of several open-source software packages related to visualisation of resource flows with Sankey diagrams, amongst other things.

Leonardo Paoli

University of Cambridge

A second-year PhD student at the University of Cambridge, Department of Engineering. Leonardo holds an MEng (Hons) degree from Imperial College in mechanical engineering. Leonardo’s PhD supervisor is Dr Jonathan M Cullen. His research focuses on the characterisation of energy conversion efficiency from primary to useful energy.

Abstract

The scientific tools of industrial ecology have proved immensely useful for tracking the flows of resources through society, estimating the environmental impacts of such resource flows, and for providing evidence for decision makers about the effects of human activities on the planet’s natural resources and eco-systems. Our industrial ecology toolkit has developed rapidly in recent years, in both the breadth of tools available and the specificity of individual tools; we have moved well beyond the simple categories of MFA, LCA and IO to a myriad of sub-, hybrid- and support- tools. The advancement of tools stems from the community’s concern for a fragile planet and commitment to increasing knowledge of our natural and human systems.

Yet, our efforts to develop rigorous evidence come at a time in history when we face an increasing breakdown of trust between scientists and the public. In his recent address to the California Institute of Technology entitled “The Mistrust of Science”, Atul Gawande states that “even where the knowledge provided by science is overwhelming, people often resist it—sometimes outright deny it." (Gawande A, 2016). One reason for this mistrust is that scientific knowledge is rarely complete and scientific models are successive approximations. Therefore, a key part of our role as scientists and industrial ecologists is to communicate more effectively the variability and uncertainty in our work.

All industry ecology tools involving data collection, manipulation and interpretation are confronted with data variability and uncertainty. Therefore, understanding how variability and uncertainty affect results and finding strategies to deal with it are crucial for a wide range of tools. In this Special Session we propose to explore the latest methodical advances in: (1) data quality assessment and uncertainty characterisation; (2) uncertainty propagation and inference; (3) the evaluation and communication of uncertain results. The use of various methods such as information theory, frequentist and Bayesian statistics or fuzzy set theory and relevant tools such as analytical error propagation, Monte Carlo simulation, Markov Chain or regression analysis will be explored. An important focus of the session will be the effective communication of uncertainty and variability with decision makers and the public, including new approaches for visualising uncertainty.

It is hoped that by holding a Special Session at the Chicago conference we can draw attention to the importance of uncertainty and variability analysis and bring together a group of academics who are keen to explore these issues across the expanding array of industrial ecology modelling tools.

Gawande A (2016) The mistrust of science. The New Yorker, 10 June 2016.

Authors

Jonathan Cullen (University of Cambri)
David Laner (Technische Universitaet Wien)
Rick Lupton (University of Cambridge)
Leonardo Paoli (University of Cambridge)

Topic Areas

• Complexity, resilience and sustainability , • Decision support methods and tools

Session

ThS-21 » Special session: "Moving beyond averages: distributional analysis in industrial ecology" (13:45 - Thursday, 29th June, Room F)

Presentation Files

The presenter has not uploaded any presentation files.

Email Support • Blog • Privacy Policy • Cancellation Policy