A platform for research: civil engineering, architecture and urbanism
Survival analysis methodology for service live prediction and building maintenance
This paper deals with, on one hand, the introduction of survival analysis techniques for being used in building maintenance and, on the other hand, the application of this methodology for analyzing a large building stock in order to obtain information for maintenance strategies and/or prevention policies. In particular, in this contribution the description of the time to the event when the event of interest is some damage (or some level of degradation or extent) on the building façade is the main goal to reach. For the time being, building follow-up is based on inspections. However, data coming from building inspections are always not completed, but censored, due to the fact that, at each inspection time, the event of interest is already happened, or not yet. In order to solve this problematic, the existing methodology for fields like medicine, biology, industrial engineering or event history analysis is adapted, and routines in S-PLUS for a numerical and graphical systematic analysis are implemented. Estimates for non-parametric durability and hazard functions are derived. The possibilities of the proposed methodology will be illustrated with its application to the building façades in Hospitalet de Llobregat, the second most important city in population in Catalonia (Spain), where more than 14.000 buildings have been inspected. The analysis of the results allows technicians to detect different zones and levels of intervention to be applied in the city. ; Postprint (published version)
Survival analysis methodology for service live prediction and building maintenance
This paper deals with, on one hand, the introduction of survival analysis techniques for being used in building maintenance and, on the other hand, the application of this methodology for analyzing a large building stock in order to obtain information for maintenance strategies and/or prevention policies. In particular, in this contribution the description of the time to the event when the event of interest is some damage (or some level of degradation or extent) on the building façade is the main goal to reach. For the time being, building follow-up is based on inspections. However, data coming from building inspections are always not completed, but censored, due to the fact that, at each inspection time, the event of interest is already happened, or not yet. In order to solve this problematic, the existing methodology for fields like medicine, biology, industrial engineering or event history analysis is adapted, and routines in S-PLUS for a numerical and graphical systematic analysis are implemented. Estimates for non-parametric durability and hazard functions are derived. The possibilities of the proposed methodology will be illustrated with its application to the building façades in Hospitalet de Llobregat, the second most important city in population in Catalonia (Spain), where more than 14.000 buildings have been inspected. The analysis of the results allows technicians to detect different zones and levels of intervention to be applied in the city. ; Postprint (published version)
Survival analysis methodology for service live prediction and building maintenance
Serrat Piè, Carles (author) / Gibert Armengol, Vicente (author) / Universitat Politècnica de Catalunya. Departament de Matemàtica Aplicada I / Universitat Politècnica de Catalunya. Departament de Construccions Arquitectòniques II / Universitat Politècnica de Catalunya. GREiP - Grup de Recerca d'Edificació i Patrimoni / Universitat Politècnica de Catalunya. GREMA - Grup de Recerca en Estadística Matemàtica i les seves Aplicacions
2011-01-01
Conference paper
Electronic Resource
English
DDC:
720
Service life prediction and maintenance of building
British Library Conference Proceedings | 1994
|Methodology for the survival analysis of urban building stocks
British Library Online Contents | 2007
|Methodology for the survival analysis of urban building stocks
Online Contents | 2007
|Systematic methodology for service life prediction of building materials and components
Springer Verlag | 1989
|Systematic methodology for service life prediction of building materials and components
Online Contents | 1989
|