HUMANE D3.4 - Typology-driven modelling and validation of design options

Abstract

The HUMANE project has developed a typology and method for characterising and analysing Human-Machine Networks (HMNs) in order to help the design process when new networks are being developed or existing networks are evolved. However, evaluating design options is a non-trivial task as networks can be complex and emergent behaviour can be difficult to predict. As an alternative to building and testing prototypes, we propose a simulation modelling approach that not only has a potential cost saving, but may also allow evaluation of scenarios that may otherwise be infeasible or difficult to test empirically, e.g., due to potential dangers involved. Grounded in the HUMANE method, we propose a modelling approach for network simulation using the agent-based modelling paradigm. We propose a Core HMN Model for describing networks that can be readily extended and used for simulation purposes of specific HMNs. We demonstrate the approach via two case studies: Wikipedia and Truly Media. The former provides a case study for introducing design-changes to a well-established HMN, while the latter provides a case study for evaluating design options while the HMN is being developed. As part of the design and evaluation phases of the HUMANE method, we pose some example design-oriented what-if scenarios for simulation modelling, demonstrate how the Core HMN Model can be used and extended, and discuss results from simulations.

More information

Identifiers

Catalogue record

Export record