Participatory multi-modelling: boundary object ecology perspective

Decision making required for dealing with wicked socio-technical problems like the energy transition or climate change requires a diversity of perspectives. These different perspectives are hard, if not impossible, to capture within a single model or simulation, created from a single perspective. Participatory multi-modelling offers a way to use participatory modelling methods to connect different models, datasets and stakeholder questions into a coherent boundary object (BO) which can be used for decision support. While constructing a single model within a participatory process is already difficult, multi-modelling brings several new challenges. First, there are multiple models that need to be created, and they may require vastly different domain expertise and modelling tools. Second, the interfaces between models need to be developed both technically and conceptually, thus requiring many additional BOs on which consensus must be reached. Third, the multi-model is a BO that dynamically emerges from interactions between sub-models and whose structure may change while being used. Participatory modelling literature insufficiently addresses the specific challenges of multi-modelling. We explore the participatory multi-modelling process from the lens of many BOs interacting and co-evolving during the process: a boundary object ecology. This perspective provides us with the concepts and vocabulary to explore the interactions in a structured manner. For example, notion of temporal path dependency can reveal (lack of) design choices, as a series of BOs are being constructed and the notion of BO co-evolution within a coupled fitness landscape allows us to explicitly design parallel participatory modelling processes, and their interaction over time. We apply this perspective on a case study and derive several specific design principles for structuring the modelling process.