Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Basic engineering errors led to Indonesia bridge collapse

The official investigation into November’s catastrophic bridge collapse in Indonesia has found that a catalogue of engineering errors triggered the failure.

The report, commissioned by Indonesia’s Ministry of Public Works, said a lack of engineering knowledge and understanding of the suspension bridge structure was the primary cause of shear-failure in the connection between the steel hanger and the bridge truss near its deck.

Repair failings

One of the key investigators, Bandung Institute of Technology structural engineering research group professor Bambang Budiono told NCE that the failings occurred during repairs to the 10 year old structure.

“The main reasons why the bridge collapsed was lack of knowledge [about the forces running through the bridge], an over simplified approach [to repairing the bridge] and lack of maintenance and repair,” he said.

The three-span, 710m long Kutai Kertanegara suspension bridge collapsed into the River Mahkam in East Kaliminta province last December leaving only the bridge cables and tower in place. At least 20 people were killed and many more injured (NCE 1 December 2011).

A team of nine engineers from three Indonesian universities − Bandung Institute of Technology, University of Indonesia in Jakarta and Yogyakarta’s Gadjah Mada University - along with an official from the Ministry of Public Works investigated the collapse.

Extra stresses

Failure occurred when engineers where jacking underneath one side of the bridge deck at mid span. The jacking procedure caused extra stresses on the hanger connection triggering the shear failure of the bolt connection between the steel hanger and the bridge deck.

Once one connection failed, the increased stress on the other connections caused a chain-reaction resulting in all the other bolts failing with 20 seconds, according to the report.

Budiono said that a witness statement from the only surviving bridge worker - four of his colleagues died in the collapse -said that a maintenance firm was in the process of jacking the bridge to restore the bridge’s lateral camber of 3.7m after it had dropped to just 3m. The work was commissioned by the bridge’s owner, the Kutai local authority.

Workers first began jacking on the upstream side of the bridge two days before the collapse on 24 November, according to the witness. They raised it by 150mm, before beginning on the downstream side. It was at this point that the collapse occurred.

Failure to jack both sides

Budiono said the contractors should have jacked-up the bridge deck at both sides, and closed the bridge to traffic.

However, they shut just one lane, and jacked each span individually - which added to the death toll on the bridge.

But there were numerous errors prior to the repair work being carried out on site, according to the report.

Engineers working on the repairs failed to follow numerous standard practices, says the report.

“They did not follow any guidelines or good engineering principles,” said Budiono.

Before carrying out the maintenance, the company failed to make structural checks on the bridge to analyse what impact the jacking procedure would have on the bridge. The bridge had also not been checked during the first three years after its opening in 2001 due to lack of funds, alleges the report.

“Engineers did not follow any guidelines”

Prof Bambang Budiono

However, because of the dramatic drop in camber, engineers checked the bridge and discovered the foundation of its western tower had moved horizontally 200mm, which changed the forces within the bridge structure, but no further investigation was carried out to explore the effects on the forces of this or the proposed jacking.

The other major contributing factor in the collapse was the use of poor quality bolts connecting the hangers to the bridge deck and suspension cable, says the report.

“They used a very brittle material manufactured locally which was not covered by the codes,” added Budiono.

Brittle bolt

Tests on the connections following the collapse revealed the bolt was made of using a cast-iron grade that was brittle and sensitive to impact loading. “This was the wrong choice of material,” said Budiono. “There should have been redundancy in the structure.”

But the choice of material was approved during design, he said.

Indonesia does not have any suspension bridge design codes and engineers used the Japanese equivalent, but it is unclear how accurately these were adhered to.
Indonesia’s Department for Public Works has accepted the conclusions of the report and is reviewing its procedures.

NCE was unable to contact the consultants and contractors involved in the bridge’s construction and maintenance at the time of going to press.

Readers' comments (3)

  • Michael Paul

    I find the description somewhat incomplete and hard to follow. for example, a movement of one tower by 200mm would not in my opinion fully explain a 700 mm reduction in the camber. Also the description of jacking "underneath the bridge" (how? - from a barge in the river?) seems strange - or were the jacks being applied to shorten the hangers and thus restore the camber, which would fit better to the mode of failure. More details please!
    Mike Paul, Stuttgart, Germany

    Unsuitable or offensive? Report this comment

  • I'm sorry, but I can't say I 'm much wiser as to the cause of failure after having read this article.
    Another, unofficial, paper states the bridge has spans of 100, 270 and 100m, which tie in with measurements made on Google map. Not quite the 710m O/A stated. The very basic drawing of a typical suspension bridge really doesn't help.
    The paper also states it was the end anchor supports which had moved laterally by 10cm each (due, purportedly, to the lack of raker piles). That would make more sense than the west tower moving horizontally and also explains the reduction of longitudinal (not lateral) camber.
    Could we have a more detailed description of the actual jacking procedures please? We have no idea of how many hangers were being jacked at any one time and how the loads were shared.
    Finally, I find it difficult to believe that the initial failure was at the connection between the hanger/deck. If that were the case, why are all the hangers missing from the cables? Are there any photos/ details of the connections?
    Sorry NCE, but only 4/10 for technical content.

    Unsuitable or offensive? Report this comment

  • Clarifications to your comments:<br/>According to the report, the 200mm horizontal movement of the western tower caused the slackening of the main cable which resulted in the deck moving vertically downwards and reducing the camber.<br/>To rectify the problem, the engineers used jacks (the report did not state which type but liner acuaters jacks have been suggested) to shorten the hangers to lift the deck back into the initial or the design position. <br/>As the article states, engineers working on the bridge did not perform an engineering investigation prior to restoring the camber. Furthermore, there is no details on how the jacking procedure would be carried out, or how that would affect the structure.<br/>To clarify, the steel clamps holding the steel hangers to the suspension cable snapped causing the collapse.<br/>In reference to the bridge length, the total span including viaducts was 710m according to the official reports from the Indonesian Public Works.<br/>

  • Thanks for the reply Declan. However, I would be rather surprised if the tower had moved horizontally unless there was a serious failure of its footings. The cables are tied to the tops of the towers and so horizontal movement is controlled by the cables. Even if the cables could slide over the towers, there would be no discernable change in cable profile. Something doesn't add up here.
    As for the jacking, yes it is only possible to use linear actuators, or hydraulic rams, as they are more commonly known. They would be connected such that they 'drag' the deck up a hanger. It's a pity there isn't any further info on the sequencing.
    WRT the hanger connections, there are quite a few photos and other bits and pieces on the web. These clearly show the failures being at the cable connections and yes, the O/A length of the bridge is 710m, but this includes approach spans of 132.5m each.
    So, at the end of the day, are we any the wiser about any specific reasons, apart from a lack of engineering perception on the part of the maintaining authority?

    Unsuitable or offensive? Report this comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Please note comments made online may also be published in the print edition of New Civil Engineer. Links may be included in your comments but HTML is not permitted.