612 / 2019-04-25 20:32:18

Predicting plans’ popularity to better manage the rural-urban fringe

plan prediction, decision support, Planning Support Systems, applied artificial intelligence, rural planning, periurban planning

全体主题 > Planning Support Systems

Our self-improving app forecasts what desirability scores different categories of people will assign to alternative plans in any situation. We used it here to better plan the rural-urban fringe of Melbourne, Australia, to eventually conclude, with 95% confidence, that everybody will regard the ‘Do nothing’ plan as significantly worse than the three action plans - regulated suburban Sprawl, a Green Belt and Horticulture. Our app also came close, but did not statistically split these three action plans in terms of predicted desirability. Nevertheless, a spin-off benefit was its myriad of non-significant suggestions about why different types of people rate each plan in the way that they do. Such suggestions prompted the authors to consider several plan improvements that they probably would not have otherwise thought of. That is, our app generated a tentative yet consciousness-expanding form of planning support. After it has ‘learned’ from a larger number of users in the future, it will become more statistically precise and, therefore, more useful.