An integrated cloud model for measurement errors and fuzziness

Two kinds of uncertainties - measurement errors and concept (or classification) fuzziness, can be differentiated in GIS data. There are many tools to handle them separately. However, an integrated model is needed to assess their combined effect in GIS analysis (such as classification and overlay) and to assess the plausible effects on subsequent decision-making. The cloud model sheds lights on integrated modeling of uncertainties of fuzziness and randomness. But how to adopt the cloud model to GIS uncertainties needs to be investigated. Indeed, this paper proposes an integrated formal model for measurement errors and fuzziness based upon the cloud model. It addresses physical meaning of the parameters for the cloud model and provides the guideline of setting these values. Using this new model, via multi-criteria reasoning, the combined effect of uncertainty in data and classification on subsequent decision-making can be assessed through statistical indicators, which can be used for quality assurance.