We also went the simulation route:
One simulation was that one of the Toughbooks (laptops) used from programming our machines had a hard drive failure so the data was corrupted and we couldn’t program certain machines. We simulated that the setter discovered this as he was about to program said machine for a new order.
We recorded a timeline of events as we followed our procedures, and collect evidence of each task as we went along. Long story short, from start to finish it took less than two hours to get the laptop back up and running and ready to use.
The evidence included screen grabs of the back-ups, a record of who did what and when, the receipt from a local retailer for a hard drive, the old hard drive and even pictorial evidence of me cloning the new hard drive from the back up and installing the new hard drive, and the setter using the reinstalled Toughbook to program his machine.
I think as long as the simulation is as close to reflecting real life failures and events are recorded in real time, the evidence will prove if the contingency plan works or not. We made sure that Toughbook was not going to be needed on the day of the simulation. If it was, I would have stopped at the successful cloning of the hard drive from back up, which was surprisingly quick and easy.
And before anyone asks, No. No programming takes place at night, at the weekends or on bank holidays
Our CB auditor reviewed our simulations and said that the requirements of this clause had been met.