Kim Malafant, Director of Complexia, an Australian Company specialising in the design and application of high performance systems, recently gave a Seminar at the Lodge on "Scenario Analysis, Modelling and Planning".
Complexia specialises in a series of applications, such as data analysis, data mining, modelling, decision support systems and integrated framework systems.
Kim has previously worked in CSIRO (Commonwealth Scientific and Industrial Research Organisation) and in private firms before setting up his own company. His background is in Mathematics, Statistics and Computer Sciences.
During the lecture, he discussed how complex systems can be linked together. He referred to the difference between quantitative and qualitative models, plus the importance of mental models, which has not been well understood in the past. Kim also divided the systems into hard and soft ones, the former requiring quantifiable inputs, and the latter being more qualitative and tending to come from socio-economic groups.
A scenario is a description of the current situation as well as of a series of events that could lead from a current to a future scenario. Scenario Analysis Method involved forward projection which allows us to progress from the current situation to a series of alternative, future scenarios. Another method is to look at backward modes or inverse cases, then to progress from a current scenario to a desired future one and find out what we need to do to achieve the desired situation. In addition, to be conceptually different for the first one, this approach is computationally more intensive.
The best scenario approach is to develop a framework by discipline experts and those with local knowledge. This is called the top down approach and is also an interactive development. An important consideration is that it ought to be flexible. In the interactive approach, it is essential to keep reviewing and changing the model as conditions vary.
Kim described the main characteristics of some of the best known toolkits available. The include WHAT IF? from Canada which is an object-oriented scenario analysis and modelling toolkit. Other models are such as FACET, a special spreadsheet providing a complete spatial decision toolkit; MODEL MAKER uses mode diagrams to construct decision frameworks. They are all toolkits, some of them with their own language and mathematical functionality.
Some of these models have been used by Kim for scenario analysis and exploration of irrigation and water management options and policies in Australia. Another use is for Forestry and multiple-land-use databases in the USA, Canada and Australia. A further project was long term socio-economic analysis of policy decisions in Canada and the USA.
The advantage of the tools is that they provide a good overview of the problems and issues; encouraging examination and experimentation; that they are modular, flexible and re-usable; can be effective in communicating ideas and information; can be applied to built flexibility and enhance the transparency of the problems under study.
Kim then described an implementation dealing with the irrigation future framework in Australia. The aim may be to build a system that can look at the interaction between different components, including physical (such as seepage); changes in vegetation characteristics; economic and socio-economic models to predict what will happen in the region. The modules are all in different scales.
Another interesting application is the case of ballast water discharged from ships, which has introduced new species of fish in Australian waters. The aim in this case was to develop a decision support system.
The general idea is to link model components within one system and integrate knowledge across disciplines. This builds credibility and enhances transparency for unfamiliar 'users'. They also act as awareness and educational tools.
The talk was well received by the audience and was followed by a lively discussion.