Change Management - Failures

[gautambangalore]

Hi,

I have a question on Change Management. If an organization say a bank has come up with an idea of having ATMs set up across a location and say 10 in number. Once this is implemented and 4 out of 10 fail, what would be the next steps involved in this change?

[thechosenone69]

the next step is called Rollback.

But that shouldn't be the scenario in banking. their should be a good build plan, testing/ piloting stage before the implementation.

If this is the scenario as you are describing, then the bank needs to get a new IT department. Might as well replace the board.

[Timo]

I agree... there has to be a provision for rolling back such implementation. As a user, you'd be well advised to change banks.

[swansong]

Hi

The fact that four of the ten ATM's failed suggests that the your live environment is not consistent across all locations.

If it was i would expect all the ATM's to work as expected, or all fail.

I would investigate why your live environment differs between locations and then determine whether this change would be best managed as ten seperate changes, one single change, or something in between.

[UKVIKING]

Swansong

Good answer. And almost close to being consulting..

gautambangalore,

This reads almost like an exam / self study question.

Is this one ?
_________________
John Hardesty
ITSM Manager's Certificate (Red Badge)

Change Management is POWER & CONTROL. /....evil laughter

[Timo]

Heh... come to think of it, if the change success criterion says that unless 5 ATMs fail this change is considered successful, then you have nothing to do there

[paulfixter]

You could also test the change prior to the "live" rollout.

Is there a test environment. Did any ATMs fail during the test?

How urgent it is to get the faulty ATMs up and running. You could deal with the faulty ones under incident management and get them up and running one by one.

If the root cause is not known and the impact is major, you could deal with under problem management.

Or you could just implement your back out plan, and do the change again later after finding out what went wrong.

The possibilities are endless.

I suppose "It Depends"

[asrilrm]

Hi,

Hope it's not too late but I wonder did you deploy 10 ATMs at the same time, assuming that you raised one RFC for all 10?
For me, it would be easier to raise an RFC for deploying 1-2 ATM at a time, monitor stability then proceed with the next 1-2, and so on.
Of course unless it is an urgent deployment.

Anyhow, there are many factors contributing to failure(s): could be the network, could be the comm line, could also be the ATM equipment itself.
Just a discourse.

Cheers,
Asril