Swanwick - The latest (10/12/98) ComputerWeekly report
Guest
Posts: n/a
Swanwick - The latest (10/12/98) ComputerWeekly report
Swanwick audit mixes its message.
A government audit of the £217m systems at the Swanwick air traffic control centre has concluded that the technical problems have been overcome. But does this mean the systems will work? Tony Collins reports
The main conclusion of an independent audit into the management of the world's most advanced air traffic control system at Swanwick in Hampshire, is that the project can be completed satisfactorily.
"In view of the good progress on the technical development of the system, we see no reason to abandon the project at this stage," says a report by the state-owned Defence Evaluation Research Agency.
This positive conclusion will be welcomed by airlines, the travelling public and the Government who collectively have paid about £450m so far for the New En Route Centre, including £217m so far for the IT systems.
The Swanwick centre was due to have all but replaced the ageing London Air Traffic Control Centre near Heathrow by 1996 but the date has been deferred four times, and is not expected to go live until 2001/2.
To compensate for the delay, the London air traffic control centre's IBM 4381-based systems have been upgraded. But there is only limited expansion potential.
Under pressure
Meanwhile, the increasing pressures on air traffic controllers and their systems has contributed to longer waits in departure lounges, flight delays, and fears about safety.
On the face of it, the Defence Evaluation Research Agency report assuages many of the concerns. It says there is no reason why the late introduction of the Swanwick centre should compromise safety, though there may be an increase in flight delays. The report's conclusions also suggest that the £217m which had been paid by April 1998 to Swanwick's main IT contractor Lockheed Martin has been money well spent.
"The Dera audit team is strongly of the opinion that the New En Route Centre at Swanwick will meet its design purpose," says the report.
But those who study the agency's report closely might wonder if some of the more positive conclusions were written by one set of people, and the report by a different group. These conclusions are politically sensitive because they are likely to be the only part of the 78-page report which is read by many MPs and national newspaper journalists.
For example the transport minister John Reid, in an interview on Radio Four's Today programme mainly referred to the report's more positive findings. He said that the report would lay to rest the "scare stories" that have been running in the media about Swanwick.
But within the body of the report are long lists of the specific risks that, if they materialise, could jeopardise delivery of the project in 2002 or even lead to its abandonment.
There was praise for Bill Semple, who became chief executive of National Air Traffic Services in 1997. But the agency also criticised his organisation's project management and expressed anxiety that testing of the system might be curtailed which could cause "serious problems" when the system is operational, and referred to the "high" number of bugs which posed a "significant" risk to the project's target date.
The agency also expressed concern that the safety case - which is aimed at convincing regulators of the Swanwick systems' safety in operational use - was "not as complete and mature as we would have expected at this stage". It added, "We judge there to be a significant timescale risk associated with its [the safety case] completion and approval."
Testing problems
The report goes on to disclose the difficulties involved in trying to test all of the sub-systems together. It says the final phase involves parallel running between the old London and new Swanwick air traffic control centres, but, "even here operators will not actually control aircraft in a way representative of the full system load . . . the first time this will really happen is when the centre goes operational".
Another of the agency's "main concerns" is the backlog of bugs - called programme trouble reports - and change requests. There are 1,260 "significant" bugs which are being cleared at a rate of 70 per week but new bugs "continue to be generated". This "high" number of bugs represents a "significant" risk to the target date.
There are also nearly 50 changes that need to be made to the design of the system which has already been subjected to major change. And further changes may be needed. This is because the end-users who originally helped to specify the systems were not "truly representative" of the whole population of air traffic controllers. "The result of this is that the original requirements fall significantly short of those that are now required for the New En Route Centre."
User testing
It adds that "in view of the considerable amount of user testing still to be carried out, we believe that the raising and clearance of change requests remains a considerable risk to the programme timescales".
Also, the new communications network may not be big enough to handle the loads and so might need to be upgraded urgently.
Page 46 of the report reveals that there are "significant risks still to be overcome," - a point which is not made in the report's summary, or in the announcements by the minister or the Department of Transport.
Despite the short time available for the audit, it is at least as penetrative as one would expect, which is an achievement. But it avoids any discussion on how National Air Traffic Services would cope with air traffic should the whole project prove unfeasible.
In addition the agency might be said to have hedged its bets. If the Swanwick systems are a success, the agency can say that it always said the systems would work. Should the project fail, the agency could argue that it had pointed out all the most significant risks in the body of the report.
And should the worst of all possible worlds occur, and there is a mid-air collision where a contributory cause is a failure of procedural and IT systems, the agency could argue that it never guaranteed that the existing systems were safe.
Indeed it could say it had recommended that a further study be carried out to assess the maximum safe capacity of the existing systems.
The transport minister John Reid finds the agency's report reassuring. Not everyone will agree.
Why the audit was necessary
The short audit by the Defence Evaluation Research Agency was carried out after MPs on the transport sub-committee expressed concerns that they were not being told the full story about the IT problems at the New En Route Centre at Swanwick in Hampshire.
National Air Traffic Services had argued strongly against the need for an independent audit. Computer Weekly took the opposite view and gave oral and written evidence to the transport sub-committee aimed at showing that, because of the lack of hard information on the project's progress, such an assessment could prove helpful. MPs agreed with our view.
The audit report makes dozens of recommendations for strengthening management of the project. There have been no suggestions that it was a waste of time and money.
(Nb. This article was written by Tony Collins of CW)
A government audit of the £217m systems at the Swanwick air traffic control centre has concluded that the technical problems have been overcome. But does this mean the systems will work? Tony Collins reports
The main conclusion of an independent audit into the management of the world's most advanced air traffic control system at Swanwick in Hampshire, is that the project can be completed satisfactorily.
"In view of the good progress on the technical development of the system, we see no reason to abandon the project at this stage," says a report by the state-owned Defence Evaluation Research Agency.
This positive conclusion will be welcomed by airlines, the travelling public and the Government who collectively have paid about £450m so far for the New En Route Centre, including £217m so far for the IT systems.
The Swanwick centre was due to have all but replaced the ageing London Air Traffic Control Centre near Heathrow by 1996 but the date has been deferred four times, and is not expected to go live until 2001/2.
To compensate for the delay, the London air traffic control centre's IBM 4381-based systems have been upgraded. But there is only limited expansion potential.
Under pressure
Meanwhile, the increasing pressures on air traffic controllers and their systems has contributed to longer waits in departure lounges, flight delays, and fears about safety.
On the face of it, the Defence Evaluation Research Agency report assuages many of the concerns. It says there is no reason why the late introduction of the Swanwick centre should compromise safety, though there may be an increase in flight delays. The report's conclusions also suggest that the £217m which had been paid by April 1998 to Swanwick's main IT contractor Lockheed Martin has been money well spent.
"The Dera audit team is strongly of the opinion that the New En Route Centre at Swanwick will meet its design purpose," says the report.
But those who study the agency's report closely might wonder if some of the more positive conclusions were written by one set of people, and the report by a different group. These conclusions are politically sensitive because they are likely to be the only part of the 78-page report which is read by many MPs and national newspaper journalists.
For example the transport minister John Reid, in an interview on Radio Four's Today programme mainly referred to the report's more positive findings. He said that the report would lay to rest the "scare stories" that have been running in the media about Swanwick.
But within the body of the report are long lists of the specific risks that, if they materialise, could jeopardise delivery of the project in 2002 or even lead to its abandonment.
There was praise for Bill Semple, who became chief executive of National Air Traffic Services in 1997. But the agency also criticised his organisation's project management and expressed anxiety that testing of the system might be curtailed which could cause "serious problems" when the system is operational, and referred to the "high" number of bugs which posed a "significant" risk to the project's target date.
The agency also expressed concern that the safety case - which is aimed at convincing regulators of the Swanwick systems' safety in operational use - was "not as complete and mature as we would have expected at this stage". It added, "We judge there to be a significant timescale risk associated with its [the safety case] completion and approval."
Testing problems
The report goes on to disclose the difficulties involved in trying to test all of the sub-systems together. It says the final phase involves parallel running between the old London and new Swanwick air traffic control centres, but, "even here operators will not actually control aircraft in a way representative of the full system load . . . the first time this will really happen is when the centre goes operational".
Another of the agency's "main concerns" is the backlog of bugs - called programme trouble reports - and change requests. There are 1,260 "significant" bugs which are being cleared at a rate of 70 per week but new bugs "continue to be generated". This "high" number of bugs represents a "significant" risk to the target date.
There are also nearly 50 changes that need to be made to the design of the system which has already been subjected to major change. And further changes may be needed. This is because the end-users who originally helped to specify the systems were not "truly representative" of the whole population of air traffic controllers. "The result of this is that the original requirements fall significantly short of those that are now required for the New En Route Centre."
User testing
It adds that "in view of the considerable amount of user testing still to be carried out, we believe that the raising and clearance of change requests remains a considerable risk to the programme timescales".
Also, the new communications network may not be big enough to handle the loads and so might need to be upgraded urgently.
Page 46 of the report reveals that there are "significant risks still to be overcome," - a point which is not made in the report's summary, or in the announcements by the minister or the Department of Transport.
Despite the short time available for the audit, it is at least as penetrative as one would expect, which is an achievement. But it avoids any discussion on how National Air Traffic Services would cope with air traffic should the whole project prove unfeasible.
In addition the agency might be said to have hedged its bets. If the Swanwick systems are a success, the agency can say that it always said the systems would work. Should the project fail, the agency could argue that it had pointed out all the most significant risks in the body of the report.
And should the worst of all possible worlds occur, and there is a mid-air collision where a contributory cause is a failure of procedural and IT systems, the agency could argue that it never guaranteed that the existing systems were safe.
Indeed it could say it had recommended that a further study be carried out to assess the maximum safe capacity of the existing systems.
The transport minister John Reid finds the agency's report reassuring. Not everyone will agree.
Why the audit was necessary
The short audit by the Defence Evaluation Research Agency was carried out after MPs on the transport sub-committee expressed concerns that they were not being told the full story about the IT problems at the New En Route Centre at Swanwick in Hampshire.
National Air Traffic Services had argued strongly against the need for an independent audit. Computer Weekly took the opposite view and gave oral and written evidence to the transport sub-committee aimed at showing that, because of the lack of hard information on the project's progress, such an assessment could prove helpful. MPs agreed with our view.
The audit report makes dozens of recommendations for strengthening management of the project. There have been no suggestions that it was a waste of time and money.
(Nb. This article was written by Tony Collins of CW)
