13 Summary

The following is a brief summary of important facts dealt with in the report, the main issues, the views of the Commission and its recommendations. We would like to point out that in summarizing the main points in this manner, certain shades of meaning may be lost. As far as the recommendations are concerned, only the individual recommendation itself has been included, without the text explaining its basis. The Commission has based its recommendation on the situation as it was on 4 January 2000.

13.1 Appointment of the Commission and its work

The day after the accident, 5 January 2000, it was decided to appoint a commission of inquiry that was independent of the Norwegian National Railway Administration and the Norwegian State Railway (NSB BA).

The following were appointed members of the Commission:

1. Judge Vibecke Groth, Borgarting Court of Appeals, chair

2. Øystein Skogstad, chartered engineer, SINTEF (Foundation for Scientific and Industrial Research at the Norwegian Institute of Technology)

3. Finn Mørch Andersen, chartered engineer, Directorate for Fire and Explosion Prevention

4. Ingemar Pålsson, chartered engineer, Det norske Veritas, Gothenburg, Sweden

5. Marika Kolbenstvedt, sociologist, Institute of Transport Economics.

The above were appointed by Royal Decree of 7 January 2000. The Commission was expanded to include another member, Joakim Böcher, engineer, Det norske Veritas, Denmark, on 26 July 2000. Secretary to the Commission was Jacob Ferdinand Bull, associate, of the law firm Arntzen, Underland & Co.

The Commission’s mandate was to examine the facts of the accident in order to establish its cause and propose action that in the view of the Commission should be taken to prevent similar accidents in the future.

A few days after the Commission was appointed, it commissioned SINTEF to undertake a technical review of the signalling system to reveal any physical malfunction. This work was assessed by Railcert, The Netherlands.

The Commission also requested an expert opinion from the Swedish National Rail Administration on the condition of points no. 2 at Rudstad station on the day of the accident. In addition, the Commission has received and evaluated the following reports from the police, the Commission of inquiry appointed by the Norwegian National Railway Administration and the NSB BA commission of inquiry:

• Provisional report from the Norwegian National Railway Administration accident commission.

• Norwegian National Railway Administration accident commission interim reports 1, 2 and 3.

• Draft report from the NSB BA accident commission.

• Report from the police investigation into the cause of the accident.

The Commission has interviewed a total of 96 witnesses. All statements made to the police and any documents or other information that might be of interest from the police investigation conducted in parallel with the Commission’s own investigations have been made available to the Commission. In addition, the Commission has sought important documentation and other material of importance to its investigations from the National Railway Administration, NSB BA, the Norwegian Railway Inspectorate, the Norwegian Labour Inspection Authority and the Ministry of Transport and Communications.

13.2 The Accident

On the day of the accident, 4 January 2000, the southbound train left Trondheim at 07.45. The train was on schedule and consisted of a diesel-powered locomotive and three carriages. Its destination was Hamar. A new driver boarded the train at Røros and the train was probably 21 minutes late leaving Røros station. By the time the train had arrived at Rena station and departed again, the delay had been reduced to about 7 minutes. There were 75 persons on board, including the driver and conductor, when the train left Rena at 13.07. According to a witness, the line signal at Rena was green and the log taken from the Hamar rail traffic control centre after the accident also indicates that the exit signal was green. A witness working at Rena station saw that the southbound train had received a green exit signal on the local control panel.

The northbound train left Hamar on schedule at 12.30. The train was a BM 92 engine set consisting of an engine and a steering car. The train was scheduled to run to Rena and then back to Hamar. It arrived at Rudstad station on schedule at 13.06, stopped and picked up a passenger. According to the timetable, the train was supposed to wait at Rudstad station from 13.06 until 13.10 for the southbound train to pass. However, the train left Rudstad at 13.07. There were at this point 11 persons on board, including the engine driver and the conductor. The log indicates that the exit signal was not green and that the set of points on the line exiting Rudstad in the direction of Rena had been forced open.

The rail traffic controller responsible for this section of the line was also responsible for the Hamar-Eidsvoll line, where there was heavy traffic. Consequently, for a certain period of time he did not check the screens showing what was happening on the Røros line. An audible alarm to warn of a train on a collision course had not been installed at Hamar rail traffic control centre. Even though a warning displayed in the form of red text 16 millimetres high at the bottom of the screen indicated that points no 2 had been open since 13.08, the rail traffic controller did not realize this until about 13.12.

Neither ATC (Automatic Train Control) nor train radios had been installed on the Røros line. The trains were equipped with mobile telephones. The only way for the rail traffic controller to contact the trains when they are en route between stations was therefore by mobile telephone. Both trains had reported in their phone numbers to the rail traffic controller at Hamar, who went off duty before the accident took place. He had not added these numbers to the list that had been agreed on at the control centre. When the duty rail traffic controller realized that a collision was imminent, he could not find the mobile telephone numbers and could not contact the two trains.

The trains collided at Åsta station between Rudstad and Rena at 13.12.35. The engine car of the northbound train was completely wrecked, while the steering car received minor damage and remained upright on the rails. The locomotive of the southbound train was severely damaged and toppled over onto its side. The front carriage buckled and derailed. The next carriage also derailed, but remained upright. The rear carriage remained on the rails. A major fire broke out immediately in the area around the locomotive and the rest of the engine car, and a few minutes later fire broke out in the front carriage. The fire eventually spread to the remaining two carriages.

19 people were killed in the collision and the subsequent fire. 67 persons survived the accident. None of the survivors was fatally injured.

13.3 Causes of the accident

13.3.1 Possible direct causes

In the view of the Commission, the direct cause of the accident must either be a malfunction in the signalling system or human error.

An overall evaluation of the technical documentation indicates that it is unlikely that a technical malfunction could have affected the signalling and safety system functions on the day of the accident. None of the investigations or tests that have been conducted has revealed physical faults in the system that could have had a bearing on the accident. Nor is there any indication of any specific, functional weakness that can with certainty be said to have been the direct cause of the accident. However, because of the design of the safety system, and because of inadequate logging of operational status in the safety system, the Commission cannot exclude the possibility of a technical malfunction in connection with the accident. Neither the level of safety nor the quality of the safety system is satisfactory. The possibility of short-term operational malfunctions occurring cannot therefore be excluded.

In the light of the above, the Commission cannot state with certainty what signals were showing on the northbound line at Rudstad station on 4 January 2000. From a technical point of view, it would seem highly likely that a red exit signal was showing. At the same time, the design of the safety system makes the potential for error so great that the Commission cannot with certainty exclude malfunction situations that may have produced a different signal aspect.

An examination of the «black box» data recorder shows that the driver of the BM 92 engine set drove smoothly and normally from departure at Hamar until the collision occurred. There is no deviation from the normal except for the stop at Rudstad station. The position taken for the stop may indicate that the driver of the locomotive did not expect another train to cross here. A stop in this position would make it impossible for a southbound train to cross over. The short duration of the stop would indicate that the driver of the locomotive believed that the crossing would take place at Rena instead. The train departed from Rudstad three minutes ahead of schedule. This would indicate that the driver was in a hurry to reach Rena so as not to delay the southbound train that would be waiting there. The early departure would also indicate that the conductor may have believed time was short and that he was expecting the crossing to take place at Rena. He was responsible for ensuring that the departure signal was only given at the scheduled time.

In all events, the driver is required to obey the signals. The fact that the driver of the northbound train drove out from Rudstad station is therefore an indication that the exit signal and any advance signal were showing green. Neither the Commission nor the police have been able to find any indication that the driver or the conductor received information that moving the crossing to Rena was being considered, or that the southbound train was delayed so that a change in location for the crossing could be expected.

The Commission cannot with certainty identify the direct cause of the accident that took place on 4 January. Neither a signal malfunction nor human error can be excluded. The signal that technically speaking is most likely to have been shown, a red exit signal, is the signal that a driver of a locomotive is least likely to have driven through. Similarly, the least likely signal technically speaking is the green exit signal, the all-clear signal for the driver of a locomotive.

However, it has been established that although the trains involved were on a collision course for four and a half minutes, the collision was not prevented. How this could happen is in the view of the Commission just as important to establish as the direct cause of the accident.

13.3.2 Indirect causes

We know that technical systems can malfunction. We also know that people make mistakes. Consequently, there must be a safety system to ensure that individual faults do not result in accidents. This has been the guiding principle of safety work for years. Nonetheless, in the Røros incident, a signal failure or a mistaken observation by an engine driver led to a serious accident. ATC had not been installed on the Røros line in spite of the fact that installation had been planned and funding had been allocated. Even though ATC had not been installed, changes in the departure procedure were also introduced on the Røros line. In addition, stations were no longer manned and crossing plans were removed. These changes were made without performing risk analyses for the individual change or for the Røros line. If the Norwegian National Railway Administration had done so, it would and should have been possible to see that an individual fault could lead to an accident.

13.3.2.1 Damage limitation measures

With no barriers to prevent an emergency from arising, there should at least have been measures designed to avert it. The Røros line was not and is not electrified. The trains travelling on the line are therefore diesel-powered. This means that the rail traffic controller does not have the same possibility of stopping the trains as he or she has on an electrified line, where the electricity can be cut in an emergency.

In an emergency, it is at all events vital that the rail traffic controller is aware of the situation. On the Røros line, he would also have to make contact with the trains to stop the situation before the accident happened. At Hamar rail traffic control centre no audible alarm had been installed to warn the rail traffic controller that a dangerous situation had arisen. Nor were there any rules stipulating how often the rail traffic controller should monitor each individual section of the line. Between three and a half to four minutes passed from the time the dangerous situation arose before the rail traffic controller became aware of it.

Train radios allow a rail traffic controller to make contact with the trains when they are travelling between stations. Communication is always possible. However, train radios had not been installed on the Røros line. When the rail traffic controller became aware of the situation, he could not find the correct mobile telephone numbers for the two trains. There were no regulations laid down by a central authority that train personnel should report their mobile telephone numbers to the rail traffic control centre, and there were no regulations for the logging of numbers if they were reported in. At Hamar rail traffic control centre a list was made and it had been agreed that mobile telephone numbers that were reported in would be written down on this list. However, no safety grounds had been given for keeping a list of mobile telephone numbers in spite of the requirement in the regulations of 22 July 1994 that rapid two-way contact between train and control centre should be possible in an emergency. With clear rules and procedures and a focus on the importance of having mobile telephone numbers available in an emergency, it is possible that the rail traffic controller could have made contact with the two trains in time. He would undoubtedly have done so had an audible alarm been installed and he had had a good method of communication with the two trains.

13.3.2.2 Risk and safety management

The Norwegian National Rail Administration should have conducted more risk analyses over the last few years in the light of the changes introduced that affected safety. Furthermore, a risk analysis should have been conducted of the safety level on the individual section of a line, including the Røros line. A risk analysis would have shown that the safety level on the Røros line was far from adequate. Whatever the direct cause of northbound train 2369 incorrectly passing the exit signal at Rudstad station on 4 January 2000, our examination of the reasons why it happened at all, and why the situation was not discovered and stopped at an earlier stage, has revealed a basic lack of a systematic approach to safety issues, particularly within the Norwegian National Rail Administration, whose responsibility it is to ensure that the overall safety of a section of a line is acceptable.

Safety-consciousness and safety management, which in other comparable sectors have been basic principles for many years, have not been implemented in the former NSB and later in the Norwegian National Railway Administration. When the incident-based form of safety management on which safety on the railways has supposedly been based has not been followed either, the result is a system that will only discover that there are basic inadequacies in the safety of a section of line when an accident happens on that particular line. This was allowed to happen on the Røros line on 4 January 2000, but could have been avoided relatively easily if the recommendations that already existed and the plans that had already been made had been implemented.

13.3.3 Conclusion

In the view of the Commission, the Åsta accident occurred because of basic inadequacies in the Norwegian National Rail Administration with regard to safety consciousness and safety management. This means that the effect that serious and in some cases well-known safety deficiencies on the Røros line had on safety was neither analysed nor followed up. These basic deficiencies in safety management apply to all the aspects of the Norwegian National Rail Administration’s activities that the Commission has examined and must therefore be regarded as a serious systems failure.

13.4 The rescue operation

Before the Rena fire service arrived at Åsta, passengers who had survived the crash tried to put the fire out using portable fire extinguishers taken from the southbound train. This had little or no effect.

The accident occurred at 13.12.35. The Rena fire service arrived just after half past one. By that time the fire was very severe. Firefighters began by using a high-pressure hose delivering water from the water tank on the fire engine while larger hoses with greater capacity were being laid out. These were also connected to the fire engine water tank. An attempt was made to extinguish the flames using foam, but this had no noticeable effect. There was a brief interruption in the water supply while the hoses were disconnected from the fire engine and reconnected to the water tank truck. This cannot be regarded as having had any significance in the progress of the fire.

All the efforts of the fire service personnel were focused on saving surviving passengers in the front carriage on the southbound train. Several passengers were trapped here. The Elverum fire service arrived at the scene of the accident at 13.47. More water tank trucks were immediately called in and efforts were intensified. However, the firefighters were unable to extinguish the fire or prevent it from continuing to spread in the front carriage during the phase when this was vital to the surviving passengers who were trapped there.

The resources that were applied and the efforts made were sufficient to extinguish a relatively large fire. The reason why it was not possible to extinguish the fire at Åsta in spite of this involved several factors:

• In addition to carriage furnishings, large amounts of diesel were on fire.

• Parts of the site of the fire in the front carriage of the southbound train (carriage no. 3) were inaccessible to the firefighters because of the position of the carriage and the damage it received in the collision.

• The diesel fire outside carriage no. 3 was inaccessible in many places as diesel had collected underneath pieces of wreckage and the carriage. The diesel fire led to extreme temperatures inside the carriage, which in turn reduced the effect of the efforts to put the fire out here.

• After the collision, carriage no. 3 lay at an angle so that once the fire had taken hold in the lowest sections it was very difficult to prevent smoke and heat from spreading to sections higher up.

With the benefit of hindsight, the question might be raised of whether the Elverum fire service should have been dispatched as soon as the emergency services call centre had received the first report of the accident at 13.16. If so, the Elverum fire service would have arrived at Åsta about 10 minutes earlier. Under the current regulations, assistance from the Eleverum fire service would have to be requested as it belongs under another municipality. In the view of the Commission, the deputy fire chief in Åmot (for the Rena fire service) did not have sufficient knowledge of the scope of the accident to make such a request any earlier than he did. However, the Commission feels it would have been an advantage if assistance from Elverum had arrived sooner.

13.4.1 Priorities set by fire service

It would have been preferable to have a greater supply of water than was actually available since all the water had to be transported to the scene by water tank truck. However, other solutions would have required resources and time that were otherwise spent on efforts focused directly on saving trapped passengers. In the view of the Commission, the right choice was made. To establish a water supply from the river Glomma would have taken so long that by the time it had been set up, it would have been too late to save any lives.

The efforts made to free trapped passengers were quickly and professionally carried out. The reason why more passengers could not be saved was primarily the fact that it was not possible to hold the fire back, and consequently there was too little time to complete the complicated and time-consuming work necessary to free people who were trapped between heavy steel structures.

13.5 The Commission’s recommendations

The Commission has the following recommendations: