Your browser has javascript turned off or blocked. This will lead to some parts of our website to not work properly or at all. Turn on javascript for best performance.

The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here: https://www.microsoft.com/en-us/microsoft-365/windows/end-of-ie-support).

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

Researchers discussed modelling as decision support

The third workshop linked to the Strategic Research Areas' work on the Sustainable Development Goals took place on June 10. The workshop was the second within the theme Future-oriented methodologies and focused on how researchers can use modelling as decision support.

The workshop was divided into two parts - the first two hours were dedicated to three different presentations and the last hour was devoted to group discussions and a summary. The first presentation focused on prognostic health modelling and was given by Anna Jöud, Associate Professor of epidemiology at Lund University. Since the outbreak of the Covid epidemic, she has worked to estimate the number of infected people who need care. Her modelling has served as a basis for organising Region Skåne's care for people with Covid-19. The work has included communicating forecasts to both researchers and the general public. She sums up the past year by saying that it has been a challenge, but that it has been an exciting year for an epidemiologist and that it seems that the number of infections in Skåne is declining.

The second presentation was about modelling of industrial development and was given by Daniel Johansson, cutting data specialist at Seco Tools. The company offers software for managing data-cutting processing to provide recommendations and optimisation of customers' tool selection and process optimisation. He spoke, among other things, about how these services can be used in technical planning and decisions and shared his future visions for Seco's digitalisation in terms of cutting data modelling and what a collaboration with the academic world can look like.

The third presentation was given by Mark Brady, Associate Professor at the Centre for Environmental and Climate Science, CEC, and focused on modelling land use, biodiversity and ecosystem services. He describes himself as an agricultural and environmental economist and uses economic models in his research to find out how farmers are expected to react to various changes in the EU's common agricultural policy. An extremely important issue, as there will be just over nine billion people on earth in 2050, is how food and biofuel production can be increased without aggravating environmental destruction and climate change. 

The problem, according to Mark Brady, is that if the focus is only on one goal, such as increased biomass production, another goal may be left behind, such as the conservation of biodiversity. Today, EU farmers are allocated € 50 billion a year, but the money is not earmarked enough to support environmental goals, and instead largely supports conventional agriculture. Using the AgriPolis model, Mark Brady has investigated how likely it is that farmers will adapt their activities to the SDG´s depending on the current agricultural policy. His research shows that there is a lot of room for raising policy efficiency and in that way steer the sector towards economically and environmentally sustainable production.

After the presentations, the participants were divided into groups to discuss, among other things, risks and opportunities with the models that are used within the SRA´s today, and how models can be developed jointly. The participants agreed that it is important to have in-depth knowledge of the models you use, otherwise things can go very wrong and results can be misinterpreted. The reliability of a model also depends on how much and what data you have and how it has been collected.

Unfortunately, modelling is associated with large costs, which limits its use and makes it difficult to obtain financing. On the one hand, it takes time to validate a model, and on the other hand, it takes time to apply for grants. Since most funders, according to one of the participants, usually want to try something new, it is difficult to apply for money for something that already exists and needs to be tested. Another obstacle, according to the participants, is that it is seldom possible to evaluate the models with empirical data, i.e. researchers who develop models rarely get feedback from end users on how they worked. Thus, it is hard to correct any errors.

The discussion was also about how important it is to have an open dialogue with the users of model results, such as journalists, so that they understand what the research is about and do not spread an incorrect impression. It is important to convey to the user what type of model is being used to understand how the results can be applied and interpreted. For example, how do you communicate that there is no one to blame when forecasts go wrong? Models are tools for dealing with a complex reality, but results will always reflect the assumptions made.

The workshop ended with a concluding presentation of each group discussion. An outcome of the workshop was discovering many similarities between models that are used in different research area, as well as in the challenges researchers face. A forum may therefore be set up to continue the discussions about how models are used as decision support if there is sufficient interest from participants. Interested parties are welcome to contact Christina Windmark, Christel Nielsen or Deniz Koca.

christel [dot] nielsen [at] med [dot] lu [dot] se
christina [dot] windmark [at] iprod [dot] lth [dot] se

The workshop was arranged by the Strategic Research Areas SPI, EpiHealth and BECC in collaboration with Sustainability Forum. 

SPI - lth.se (in Swedish)
EpiHealth - epihealth.lu.se
BECC - becc.lu.se