Socio-technical systems integration and design: a multi-objective optimisation method based on integrative preference maximisation

TL;DR

A new Open Design Systems (Odesys) methodology and a new Integrative Maximisation of Aggregated Preferences (IMAP) method are introduced, implemented in the Preferendus tool.

Abstract

Current systems design optimisation methodologies are one-sided, as these ignore the socio-technical integration between stakeholder preferences ('what a human wants') and the capability of technical assets ('what a system can deliver'). Moreover, classical multi-objective optimisation methods contain fundamental mathematical flaws. Also, the often-used classical Pareto front does not provide a single best-fit design configuration, but rather a set of design alternatives. This leaves designers without a unique solution to their problems. Finally, current multi-objective optimisation processes are not well aligned with design practices, because they do not sufficiently involve decision makers and do not translate their interests into a single common preference domain to find an overall group optimum. This paper introduces a new Open Design Systems (Odesys) methodology and a new Integrative Maximisation of Aggregated Preferences (IMAP) method, implemented in the Preferendus tool. Its added value and use are exemplified in two infrastructure design applications, which show how to achieve the pure best-fit for common-purpose design results.

Figures (8)

• Figure 1: Socio-technical interplay between (un)desirability and (in)capability.
• Figure 2: Conceptual threefold framework of the Odesys mathematical statement, where subject desirability (preference functions) and the object capability (design performance functions) are integrated subject-object (objective functions). Note: the shapes of the curves are arbitrary.
• Figure 3: The workflow of the Preferendus, presented as a concept diagram.
• Figure 4: Conceptual threefold diagram, describing the systems design integration for the rail level-crossing design application. Note: the aim of this figure is to illustrate the relationship between the different functions and some curves may not represent the actual function.
• Figure 5: The three stakeholder preference functions ($P_{1..3,1..3}$) for different objectives ($O_{1..3}$) for the level-crossing design application, including the results of the different optimisations. The numerical results can be found in \ref{['tab:support_pref_curves_rail']}.