MEDIA is a powerful computer tool, developed by Pieter W.G. Bots at Delft University of Technology, that supports a multi-actor extension of a method called Analysis of Interconnected Decision Areas (AIDA) developed at the Tavistock Institute of Human Relations. The authors of AIDA – operational researchers – originally aimed at automated search for optimal designs in a multi-dimensional solution space, in analogy to linear programming. The method never became successful, mainly due to the combinatoric explosion that occurs when the degrees of freedom of a design problem become large.
MEDIA uses the AIDA concept not for the purpose of optimisation, but to structure the dialogue on design choices and their consequences for sustainability. The conceptual model of AIDA is basically unchanged: decision problems are described as a set of decision areas, each with at least two mutually excluding decision options. If for each decision area one option is chosen, this constitutes a design.
The combinatoric explosion immediately presents itself here: for N decision areas Ai (i = 1 ... N) each having n(Ai) decision options, the number of possible designs from which to choose is n(A1) x n(A2) x ... x n(AN). This solution space can be reduced to some extent by defining preclusions and exclusions: a preclusion occurs when the choice for one particular option in decision area A makes decision areas B obsolete; an exclusion occurs when choosing option X in decision area A rules out the possibility of choosing option Y in decision area B. But even if many such relations between decision areas and options exist, the design space will still be vast if N ranges beyond 100.
Still, MEDIA intends to support decision makers in finding sustainable designs. Not by systematically searching the solution space, but by providing a conceptual language for design problem formulation with the capacity of impact assessment. The tool has been designed to cope with incomplete models and data sets. The collection of decision areas, options and their expected effects on sustainability indicators can continuously be modified and expanded without loss of consistency; validity is another matter, but that is part of the dialogue between model builder and the model user (and among the model users).
Click here to download the MEDIA software.
Although MEDIA is domain-independent, it has been developed to support decision making on the built environment (urban planning, housing and construction) with emphasis on sustainability. This work has been documented in several publications:
More recently, MEDIA has been used to design a project decision making game that has been used for training purposes in a large energy corporation. You can read about this application in: