Modelling twofold uncertainty in the condition assessment of residential buildings using interval valued fuzzy signatures Ádám Bukovics Department of Structural and Geotechnical Engineering Széchenyi István University Győr, Hungary bukovics@sze.hu István Á. Harmati Department of Mathematics and Computational Sciences Széchenyi István University Győr, Hungary harmati@sze.hu László T. Kóczy Department of Automation Széchenyi István University Győr, Hungary koczy@sze.hu Department of Telecommunications and Media Informatics Budapest University of Technology and Economics Budapest, Hungary koczy@tmit.bme.hu Abstract—In this study we will describe an intervention decision-support method related to residential buildings, which is modelling the condition of buildings based on linguistic expert opinions so that it is capable of taking into consideration the uncertainties included in the expert opinions. Fuzzy signature- based model is used, wherein the uncertainties are integrated into the system by applying interval valued fuzzy sets, as well as linguistic hedges. The result will be a fuzzy set for each building, describing the condition of the whole building, specify the bourdary values between which the condition of the building may fall. Keywords—interval valued fuzzy signature; type-2 method; membership function; residential building; assessment I. INTRODUCTION In order to provide for the continous maintenance of residential buildings and keep up the level of the condition of structures there must be completed status retaining and structure renewing works on buildings in case of justified building structures at the right time and in the reasonable extent. To support it there are often expert opinions available, which are analysing the structural system and condition of the buildings (the structural deteriorations and occasionally the reasons triggering those) in details. Processing the expert opinions there can be a database created based on which we have developed a status evaluating and ranking model, which provides an intervention decision-support. Since partly due to subjective and objective reasons the expert opinions may contain many uncertainties and incorrectness, such uncertainties must also be taken into account when preparing the model. In the course of our study we have developed a fuzzy signature based model wherein modelling of uncertainties were completed by using interval valued fuzzy set as well as linguistic hedges. II. SHORT DISCRIPTION OF FUZZY SIGNATURES AND TYPE-2 FUZZY SIGNATURES A. Fuzzy signatures In many fields of science and technology arise problems of modelling complex systems, but usually the exact mathematical model is not known or too difficult to deal with, due to lack of detailed knowledge of the parameters or the behavior of the system. Moreover, lack of reproducibility and not well-defined interdependencies between the variables are also common features. Fuzzy signatures are useful tools in modelling such complex systems and objects. In this kind of approach, the complex systems is described by a set of qualitative measures, which are also arranged into a hierarchical framework expressing interconnections and dependencies, and modelling the human approach to the problem. In the literature there is a wide variety of applications, for example in economy, in the medical field, and in several fields of engineering and informatics, for example robotics, data mining and civil engineering. The complex system is described by a set of qualitative measures, which are also arranged into a hierarchical