Objective
A.BACKGROUND
Forest products in the European building industry: Europe produces and consumes the roundwood equivalent of approximately 400 million m3 of wood products 1. Estimate for the short term future indicate that the demand for wood products will increase much more than the wood harvest from European forests.
Due to significant efforts at national and European level, the use of wood in the building industry should follow similar patterns. This means that this material must be used in a more efficient manner. This can be achieved through a better design of materials and systems, leading to an increased reliability of structures. Such an approach should also strengthen the position of timber in building and facilitate maintenance policies for existing building, including historical buildings. Reliability based design (RBD): computers and new mathematical methods have made it possible to determine the force or stress distribution with high accuracy in even very complicated structures, provided the material properties are known. The knowledge concerning the actions on structures has increased due to better measuring techniques and international cooperation, such as the drafting of Eurocodes. Further methods (the so called level 2 methods for instance) have been developed. These enable experts to take into account uncertainties in material properties and actions. For some materials this has led to an increasingly more precise evaluation of the safety or reliability, i.e. the probability that a structure will not become unfit for its intended use due to collapse or unacceptable deformations.
This development can be seen in the format of design codes, which may be based on crude allowable stresses to advanced limit state design and partial coefficient methods. This is known as level 1 reliability method and takes into account some aspects of the randomness of actions and material properties. For steel and concrete even more advanced fully probabilistic level 2 methods have been brought into practice and are used to:
- Calibrate level 1 codes to insure a more uniform and consistent safety level,
- Improve design of mass produced standard elements,
- Design extraordinary structures,
- Improve maintenance policies for existing buildings by the use of post critical analysis.
For timber the development has been less impressive for several reasons:
-The variability of the material properties is much higher than for other building materials (this poses problems but also means that the advantages of introducing level 2 methods may be higher);
-The material properties, and therefore the probability of failure, depend on the whole load and moisture history of the structure: the mechanical properties in a design situation depend on the load duration and moisture content. These two effects interact in a complicated manner for the wooden materials used in timber structures, structural timber, glued laminated timber, panel products together with the joints between them.
Links with current activities, including other COST activities: an attempt was made some years ago in CIB 1 W18 to draft a probabilistic code as part of a project headed by JCSS 2 aiming at having international model codes for the reliability design of all building materials. It was, however, concluded that the knowledge basis was insufficient.
Since then many national projects, reinforced by European cooperation, have been started to get a better knowledge base:
The ongoing COST Action E8 "Mechanical performance of wood and wood products" runs from 1996 to 2000. The field of this Action consists of basic level material science on the mechanical behaviour of wood, with the objective of demonstrating that wood is a modern engineering material. Much of the research undertaken within COST Action E8 can be utilised in the present programme. For example, many of the results on material properties can be used as input data in reliability analysis of timber structures. Also gaps in knowledge of the mechanical properties have been identified. A significant topic linked to both Actions concerns the long-term strength of wood. In COST Action E8, material models are developed to account for the degradation of strength under load and moisture changes. Such models will be utilised in the present COST proposal for reliability purposes, considering the stochastic nature of long term loads and air humidity as well as the variability of the material.
- Within the framework of COST Action C1 - which ended in 1998 - the behaviour of semi-rigid joints has been studied. One of the main objectives of the COST Action C1 sub-group "Timber joints" was to study design principles for new jointing techniques and to improve the structural performance of timber connections with regard to stiffness, strength and ductility, and to use this knowledge to improve the design of structures.
- In COST Action E5, "Timber frame building systems", emphasis is placed on system performance, with the aim of demonstrating that timber can be used as a competitive structural material for medium-rise buildings in European countries.
European cooperation in the areas mentioned above provides a justifiable basis for the implementation of a new Action involved in reliability based design methods for timber structures. International cooperation with countries with longer experience in this field (North America, Australia) may also be important for success.
B.OBJECTIVES AND BENEFITS
The main objective is to extend established statistical methods and structural mechanics models to take into account the specific properties of wood, wood based materials and joints, making it possible to analyse the reliability of timber structures and structural systems and thus improve the competitiveness of timber structures and the economy of the forest sector.
Specific objectives:
Specific objectives of the Action are:
- to improve fundamental understanding of the reliability aspect of timber structures,
- to gain an overview of ongoing and state of the art research,
- to define goals and coordinate ongoing and new research at a national and European level,
- to identify fundamental research needs and areas where basic data are missing,
- to collect background information for possible future design codes,
- to bring European experts to work together towards a common goal.
Longer term objectives:
- to improve design methods, and maintenance policies, for timber structures;
- to provide more uniform reliability levels of structures and improve the competitiveness of wood as a safe structural material;
- to contribute to a more efficient use of forest resources in Europe.
These objectives will not be achieved during the Action; however the Action will contribute to their achievement.
Benefits
The results of the Action will serve as a basis for drafting a unified RBD-code for timber structures.
One consequence of this could be an increased competitiveness of timber in building with respect to other materials, especially with the development of well designed wooden products, such as glulam, LVL, OSB, and connections. This, in turn, will lead to a better economy for the important forest sector in the European countries.
The Action will also provide fundamental understanding on wood heterogeneity, statistical size effects, and on the behaviour of timber structures taking into account the strong variability of this material (load sharing effects). Finally the proposed Action will facilitate scientific exchange and cooperation between experts in the field of reliability, between wood scientists, designers and engineers.
C.SCIENTIFIC PROGRAMME
Treating any material or system in a probabilistic sense requires specific data format and numerical methods, and we need to collect information in that way, considering the most recent knowledge about wood (E8) and timber structures (E5), including connections (C1).
The scientific field to be covered by the action is directly related to the overall objective to develop the ability to analyse the reliability of timber structures and systems. The relevant research issues follows from the need for knowledge, data and methods necessary to perform such an analysis with sufficient accuracy, considering in a correct manner the special behaviour of wood, wood based materials and joints. The scientific content of the action can be grouped within three main areas:
- Experimental data and models for wood materials and products
- Methodology for reliability analysis when applied to timber structures
- Applications of reliability analysis for timber structures.
1.Experimental data and models for wood materials and products
Any assessment of reliability depends strongly on the availability of material data necessary to predict the load bearing capacity of structures. A significant amount of material data is available for wood and wood based products, but it has not been systematically collected for the present purpose. An important activity within the action will be to collect existing statistical data on relevant mechanical properties for:
- Structural timber
- Glulam, LVL and other engineered wood products
- Wood based panel products
- Joints.
As a part of this process gaps in existing data will be identified.
Mechanical properties include various types of short-term strength and stiffness properties (basic data). Information is also needed about:
- the effect of moisture conditions on mechanical properties;
- the influence of duration of load on mechanical properties;
- the degree of inhomogeneity (variation of strength within members).
This area of cooperation can be seen as a partial continuation of COST E8 Action. However there is a significant difference between the proposed Action and the COST E8 Action: the latter has been involved in fundamental understanding of wood behaviour, whereas the proposed Action will concentrate on probabilistic treatments of this scientific knowledge, the physical background of which has been previously established. In summary, this proposed Action will build upon the knowledge base acquired through E8.
2.Methodology for reliability analysis when applied to timber structures
Methods of reliability analysis need to be developed and applied, in order to consider
- variation of strength with time, depending on load and moisture history;
- spatial variation of mechanical properties (inhomogeneity);
- the performance of structural timber systems.
Existing numerical methods used to assess the limit state of a complete system (such as, the branch and bound method) need to be evaluated for their possible application to timber systems that which exhibit a multi-scale variability and strong environmental interactions. To the best of our knowledge, these aspects have not been previously investigated.
Activities within this area will focus on this methodology, reporting the state of art, research needs and coordinating ongoing research activities. A key element will be to bring together general experts in statistics and reliability analysis on one side and timber engineering specialists on the other side. The intention is to reach a general agreement among European experts about a common methodology.
3.Applications of reliability analysis for timber structures
One important issue in the action is to stimulate and exchange experience from direct application of probability based design of timber structures and systems. Such studies will consist of
- Analyses of typical timber structures to assess the actual level of safety in today's practice and to identify critical details
- Analyses with the aim to assess the reliability of existing structures.
It will be an important task to coordinate such studies, which will form an important basis for development future design rules for timber structures. In addition to this an effort will be made to begin development of a proposal for a reliability based design code which is internationally recognised. The purpose of such a code will be to lay down rules and principles for probabilistic design of timber structures, ensuring a unified approach to analyses of reliability. This work can also be seen as a direct output from the action summarising the state of art.
D.ORGANISATION AND TIMETABLE
The scientific programme will be covered by working groups that will be set up at the beginning of the Action. The Management Committee (MC) will assess the work of the task groups at every meeting and update the work plan as found necessary. The MC will also support short-term scientific missions and other researcher visits within these working groups. One of the groups will especially investigate time dependent models for wood in variable environments. A second group will be involved in probabilistic treatments of standard properties, while a third working group will report on reliability tools and their application to timber structures.
At the common mid-term conference the progress of these working groups will be discussed, and revised groups will be set up, one of them being especially involved in the draft code proposal.
The activities carried out in the COST Action will produce a scientific background consisting of new research results and analysing methods which are used as a basis for a European code for Reliability based design of timber structures. This work is carried out in cooperation with the CIB W18 and JCSS groups as shown in the figure.
Meetings established in cooperation with CIB W18 and JCSS to draft the code
E.ECONOMIC DIMENSION
The following COST countries have actively participated in the preparation of the Action or otherwise indicated their interest:
- AUSTRIA
- CZECH REPUBLIC
- DENMARK
- FINLAND
- FRANCE
- GERMANY
- IRELAND
- ITALY
- NETHERLANDS
- NORWAY
- PORTUGAL
- SWEDEN
- UNITED KINGDOM
On the basis of national estimates provided by representatives of these different countries, and taking into account the coordination costs to be covered by the COST budget of the European commission, the annual cost of the activities to be carried out under the Action has been estimated to be EUR 3 million at 1999 prices. This amounts to approximately EUR 12 million for the duration of the Action.
F.DISSEMINATION OF THE RESULTS
As indicated in the timetable, the management committee will organise workshops at least once a year. These meetings will promote interdisciplinary research in the fields of wood science and reliability based design. The final conference could be linked to an ICASP meeting (International Conference on Application of Statistics and Probabilities in Civil Engineering). Co-authored papers as well as dissemination of research findings in international conferences, will also be encouraged.
Another important vehicle of the dissemination activities is the internet. The construction of a web page devoted to the Action and the applications of research findings will be included in the program of dissemination. Simple case studies will be presented, in order to clarify reliability concepts for the end users, and to show that it is possible to quantify safety even for timber structures.
The Action MC will maintain good contacts with the COST Forestry and Forest Products Technical Committee, the good contact will entail:
- A written annual report to the COST Forestry and Forest Products Technical Committee, normally through the Wood Technology Sector Group,
- Attendance of the chairman or his representative at one of the two annual Wood Technology Sector Group meetings,
- A contribution to the COST Forestry and Forest Products Technical Committee Activity report (usually once every two years),
- Assistance for the Wood Technology Sector Group in their mid-term and final evaluation of the Action.
Cooperation between this COST Action and the CIB W18 and JCSS groups will also be a way for dissemination, especially towards the participation to the CIB W18 meetings.