Article from Newsletter No.16 November 2003

COST SAVING IN BRIDGE ASSESSMENT BY APPLICATION OF ADVANCED METHODS

Many countries are experiencing problems in management of bridge stocks due to the increasing demand on load carrying capacity combined with low budgets for rehabilitation and strengthening of older bridges.

Benefits are obtained from the difference between a general approach and a more thorough individual approach. The general approach for safety evaluation of existing bridges is based on codes and regulations for evaluation of bridges. The fact that codes generalise to be applicable for the design of many types of new bridges is efficient because the load and safety calculations become easy and because the extra cost due to the generalisation is marginal in the budget for a new bridge.

But in the case of rehabilitation or strengthening of an existing bridge, the required safety or load carrying capacity can often not be obtained from the general approach and the result may be an expensive strengthening or replacement project.

The individual approach is based on the concept that a bridge does not necessarily have to fulfil the specific requirements of a general code, as long as the overall level of safety defined by the code is satisfied. In other words the purpose of the individual approach is to cut or reduce the strengthening or rehabilitation cost without compromising on the level of safety. In this method, safety evaluations are based on probabilistic methods. In a probabilistic-based safety evaluation the uncertainties of the specific bridge condition and the local traffic situation can be taken into account consistently. It could be said that this approach establishes a ‘code’ for the individual bridge.

Examples are presented where the application of a probabilistic-based assessment approach has resulted in savings of over €30 million.

Skovdiget West Bridge (Saving ~€20 million)

The Skovdiget identical twin bridges were constructed in 1965-67. The bridges are concrete post-tensioned, combined box-girder and beam-slab bridges. Due to poor workmanship and poor design, both bridges started to deteriorate shortly after construction. The damage was related to un-injected or poorly injected post-tensioned cable ducts, insufficient drainage, bad waterproofing and an uneven bridge deck.

A major repair was performed on the East Bridge in 1978. The repair proved so costly that is was decided to leave the West Bridge without repair. Instead it was decided to monitor the West Bridge closely.

In 1998 RAMBØLL performed a probabilistic based assessment of the structure to optimise maintenance and provide a management strategy for the structure. By use of a probabilistic-based assessment model and stochastic modelling of load (including traffic) and resistance (incorporating a stochastic deterioration model) the safety of the structure was established. Using estimates of future deterioration the remaining service lifetime was found to be 7-8 years. This was extended to 15 years by implementation of a cost optimal plan without compromising the level of safety prescribed by the code. The overall result is a saving of more than €20 million when compared to the result of a traditional deterministic analysis, i.e. immediate replacement of the structure.

Probabilistic and Plasticity Based Assessment of a Concrete Slab Bridge (Saving ~€1 million)

Plastic response analysis and probabilistic-based safety analysis are combined using RAMBØLL’s program PROCON. The intention is to use models, which are closer to the actual structural behaviour in the failure situation and to the safety of the bridge.

The bridge considered is from 1942 and was constructed to under-pass a 4-lane motorway for a small factory railroad with a span of 4.43 m, a width of 24 m and a skew angle of 25.6°.

Motorway bridges in Denmark are required to have a minimum capacity termed Class 100. This requires the structure to be capable of supporting side-by-side a 100t and a 50t truck positioned at the critical location. The result of an elastic FE analysis is that only 29 % of the load from the class 100 truck can be carried together with the dead load and the class 50 truck.

The analysis with PROCON is that 79% of the load from the class 100 truck can be carried.

The result of the probabilistic-based safety analysis is that 116 % of the load from the class 100 truck can be carried. The use of plastic analysis in combination with probabilistic analysis makes a strengthening project redundant with considerable cost savings.

Examples of the savings made on various projects are indicated in Tables 1 and 2.

Table 1 Comparison of Deterministic vs. Probabilistic Assessment of 4 Danish Bridges

Table 2 Comparison of Load Carrying Capacity from Deterministic vs. Probabilistic Assessment of 3 Swedish Bridges

The general approach for assessment of existing bridges by application of standard general codes is quick and efficient, but can be costly due to expensive strengthening or rehabilitation projects. The individual approach incorporating probabilistic-methods is based on the concept that a bridge does not necessarily have to fulfil the specific requirements of a general code, as long as the overall level of safety defined by that code is satisfied. The individual approach is therefore able to cut or reduce the strengthening or rehabilitation cost without compromising on the level of safety.

In Ireland as Eirspan moves to the next phase in 2004, i.e. structural assessment, it is likely that a percentage of the Irish national bridge stock will be deemed 'unsafe' by traditional deterministic analysis. It is therefore important that engineers embrace advanced methods so as to avoid costly and unnecessary bridge strengthening or replacement.

A lecture entitled Cost Savings in Bridge Assessment by Application of Advanced Methods will be presented by the authors at 7pm on Tuesday 18^th November 2003 in The Ussher Theatre, Arts Block, Trinity College Dublin.

Dr. Ib Enevoldsen and Dr. Alan O’Connor RAMBØLL Consulting Engineers, Denmark