The elicitation and processing of relevant information is the core of any policy decision-making process. Modelling is about making sense of the available information. Models are able to incorporate the contextual influences on policy making (e.g. political and economic environments, community sentiment…etc). Systems analysis allows quantitative, empirical testing of models that exist in the study of public policy. Simulation and visualisation techniques can help policy makers to reduce uncertainties on the possible impacts of policies.

In an effort to enable adoption of the systems thinking approach to address the central problem of empirical political study, this thesis presents a framework for prescriptive policy analysis that provides decision support to: the problem definition, ex-ante impact assessment and evaluation activities carried out at the policy formulation stage of the policymaking process.

We contribute a new tool for systemic modelling and simulation of public policy decision situations. It aims to facilitate the cognitive activity of representing complex mental models using system dynamics simulation modelling. Using the ’labelled causal mapping’ method, a policy-oriented problem structuring method introduced in this research, the tool bridges the gap between the user’s mental model and the explicit graphical representation in order to enable knowledge representation and system analysis. The method provides a basis for further computational decision analysis using a common policy appraisal format, a multi-criteria model with main evaluation criteria (effectiveness, efficiency, relevance, coherence and added value), linked to a set of measurable, context dependent attributes (targeted impact variables from the policy model).

A web-based tool prototype has been implemented in a Node.js environment and is accessible both from a web-based graphical user interface as well as a hosted API.  Multiple demonstration and test cases, from various policy areas and different EU policymaking levels, were used in several iterations of the build-evaluate cycle. This approach lead to the different studies that make up this research.

The ultimate aim of support for public policy decision making is to develop ways of facilitating policymaking that can create policies that are consistent with the preferences of policymakers and stakeholders (such as an increase in economic growth, the reduction of social inequalities, and improvements to the environment), and that are at the same time based on the available knowledge and evidential information.

Using the design science research methodology, an iterative design process was followed to build and evaluate a research artefact in the form of an analytical method that is also operationalised as a decision support system (DSS) – in order to facilitate the problem analysis, the impact assessment and the decision evaluation activities carried out at the policy formulation stage of the policymaking process. The DSS provides a web-based, user-friendly interface for two main software modules: (i) a tool for modelling and simulation of policy scenarios; and (ii) a tool for multi-criteria evaluation of policy decisions. The target end-users of the DSS tools are policymakers, the support staff of politicians, policy analysts and researchers within governmental departments and parliaments at the various institutional levels of the European Union.

The proposed model-based decision support approach integrates systems thinking, problem structuring methods and multi-criteria decision analysis, in what can be described as a ‘sense-making’ approach. A new policy-oriented quantitative problem structuring method is introduced in this research, the ‘labelled causal mapping’ method, which aims to reduce the cognitive overload involved in representing complex mental models using system dynamics simulation modelling in order to facilitate knowledge representation and system analysis. One contribution of this work is an object-oriented implementation of a prototype tool for systems modelling and simulation of policy decision situations based on the labelled causal mapping method. The method provides a basis for further computational decision analysis. We proposed criteria models and data formats for common (generic) policy appraisal, and a preference elicitation method for in-depth decision evaluation based on the results of scenario simulation and the preferences of decision makers and stakeholder groups.

The artefact evaluation clarifies how well the proposed approach and the DSS tool prototype support a solution to the problem and the extent to which the outcomes in two policy analysis use cases are useful in terms of output analysis and knowledge synthesis.

The contributions of this research to theory and practice were articulated based on the design knowledge obtained through an iterative design process, notably the emergence of the concepts of transparency and intelligibility in policymodelling, (i.e., the need for explicit and interpretable models that can provide justification of a specific decision).