| The first part of the dissertation consists of the first four chapters, i.e. Chapter 1: Introduction, Chapter 2: Deterioration processes and loss of function, Chapter 3: Life of building constructions and Chapter 4: The history of service life prediction. The purpose of these chapters is to give the reader an introduction to the subjects of deterioration mechanisms, loss of function for building envelope components and prediction of service life for building envelope components. The chapters are mostly of a descriptive nature, and no new knowledge is introduced into the domain of service life prediction and building envelope component design.
The second part of the dissertation consists of the next three chapters, i.e. Chapter 5: Standards, guides and methods for assessing service life, Chapter 6: Assessing reference service life and Chapter 7: Discussion of standards and guidelines. The purpose of these chapters is to describe, compare and criticise the standards, guides and methods dealing with service life prediction which are currently available. Most of the standards which are treated in chapter 5 are based on the Japanese Guide for Service Life Planning which was published in 1993. The general approach, which is described both in the Japanese guide and in a number of national and international standards, is that a reference service life for a component is determined using accelerated testing etc. To account for differences in usage patterns, exposure rates etc. a number of modifying factors are determined, making it possible to calculate the estimated service life for a component at a specific location under specific conditions. The standards are not the only available information regarding service life prediction, as a number of researchers have developed methods which are based on either a structural engineering approach, a probabilistic approach or methods t hat are further developments of t he deterministic approaches which the national and international standards are based on. Although the relation between the structural engineering and service life prediction has been recognised, none of the described structural engineering approaches reveal a potential for further development.
Two probabilistic approaches are described, one approach using a mathematical function (Weibull) t o describe the performance of a component over time and one approach using discrete Markov chains. The latter of these two have successfully been used to predict the performance through time of road pavement and bridges. However, one major disadvantage with using Markov chains is that the method requires a large number of similar building envelope components (Chapter 7) which are subjected to the same climatic influences, a demand which cannot be met very often in t he building sector as almost every building is a prototype, being different from the other buildings.
Finally, two variations of the method specified in the Japanese guide and a later ISO standard have been described and examined. Instead of using modifying factors, the two variations introduce statistical functions to describe the influence of the indoor/outdoor climate, quality of materials/work etc., and as such combine the deterministic and probabilistic approach. Based on an investigation of the data-requirement, user-friendliness and supposed accuracy (the accuracy of the different methods has not been evaluated due to the absence of field data) the method which combines the deterministic factor method with statistical distributions for the factors is recommended as the preferred method. The method therefore forms the basis of the method-development which is found in the third part of the dissertation.
The third part of the dissertation consists of the next two chapters, i.e. Chapter 8: Suggestions for improvement of standards and guidelines and Chapter 9: Integrating durability in future building design. Of these two chapters, the aim of the first one (Chapter 8) is to describe how durability assessment can be included in the design process and in the following chapter (Chapter 9) a holistic design process for the building envelope is described which takes into account all relevant aspects, e.g. thermal performance, cost, durability, aesthetics etc.
The basis of the proposed method of chapter 8 is that the properties of a building envelope component can be divided into two subsets, one containing the properties which cannot be optimised during the design process and one subset containing the properties which can be optimised during t he design process. Examples of properties belonging to the first subset are aesthetics, fire safety and the resulting indoor climate of t he building; t hese are properties where a compromise cannot be tolerated. Examples of properties belonging to the second subset are heat loss coefficient, i.e. insulation thickness, and investment; these are properties between which tradeoffs can be made. The choice of the preferred building envelope component is made by calculating the net present value of the total cost of investment, operation, maintenance, repair and replacement during a specified period of time (e.g. the life span of the entire building or a period of 30 years due to governmental regulation etc.). The frequency (and thereby cost) of repair/replacement of the building envelope components is determined using a Monte Carlo simulation which is based on information regarding mean-value and scattering of the service life of a component. The total cost calculation should only be performed for the components which fulfil the requirements of the non-optimisable aspects (e.g. fire safety). Based on a comparison of the net present value of the total cost for the building envelope components, a building designer may choose the component which has the lowest total cost as this represents the best solution.
In chapter 9 t he scope of t he optimisation process is changed from t he component level t o t he building envelope level. Two methods are currently under development, one being developed by the International Energy Agency (IEA) Annex 32 IBEPA and one (called BELCAM) being developed by researchers at the National Research Council Canada. Work in IEA is concentrated on development of a method which can be used when buildings are to be designed, whereas the current focus of BELCAM is on assessing the performance of existing components through time.
The fourth part of the dissertation consists of Chapter 10: Examples on integrating durability in the design. In this chapter the use of the method, which was developed in chapter 8, is illustrated by evaluating two innovative building envelope components that were designed to be prepared for repair and maintenance. Both of these components are insulation systems for flat roofs and low slope roofs; components where repair or replacement is very expensive if the roofing material fails in its function. The principle of both roofing insulation systems is that the insulation can be dried, a task which is impossible in traditional flat roof insulation systems. The drying is made possible by ventilation with outside air through or below the insulation layer for a short period of time once water has been detected. The properties and performance of the systems have been evaluated by experiments in the laboratory and by using t he method from chapter 8 of t his dissertation. Both roofing insulation systems showed good performance in the experiments and in the performance assessment.
The final section of the dissertation consists of Chapter 11: Conclusion where the conclusions of the dissertation and the recommendations for further work are given. The conclusions are divided into three smaller parts regarding the description of the current methods, description of a developed method and use of the findings to evaluate the performance of two innovative building envelope components.
The recommendations for further work points at a number of issues which are specified below:
* Further development of methods for designing building envelope components prepared for repair and maintenance, and ways of tracking and predicting performance through time once the components have been designed, implemented in a building design and built.
* Development of a uniform description format for building envelope components where a designer can easily compare certain or all relevant aspects of building envelope components during the design process. This is in contrast to the current situation where some of the information may be available but organised in a different way for each component.
* Further development of building envelope components that through their design are prepared for repair and maintenance so that expensive repair and replacement may be avoided. Examples of components in need of such a design process could be internal insulation systems where condensation can be removed if detected (including an easy method for detection of moisture) and wall systems where extra insulation can easily be inserted later on if demanded. | 