If a customer asks me how best to provision spare parts and on-going support for their control system, the first question I ask is: “What kind of mean time to repair are you able to live with?” The second question is: “Does your control system hardware and software have a single integrator or multiple integrators able to support it?” Why is it no-one else thinks about these things?
Technology will always die in the end. Capacitors and resistors drift, dry and die over time and these are failure mechanisms that can not be prevented. Whilst every effort is made to protect equipment from damage, lightening strikes do occur, as do power surges and brown-outs and all of these are outside the control of the customer and can cripple their control system assets. Hence it is not a question of: “Well it’s really reliable and we’ve never had a problem with it” or “So and so have always supported us so that’s all we need to worry about”. Natural events, external factors, businesses going out of business are always going to happen. The real question is have you invested wisely in your control system software and hardware: beyond price and features? Most customers don’t.
I recently visted a process plant with a somewhat rare control system. When asked, “Do you keep spares on site?” their answer: “Why would be do that? The supplier/integrator (same people in this case) keeps them all for us”. Problem was, this supplier/integrator are the only people that supply, service and maintain their equipment and software and are thus a sole source of supply. If they decide to close up shop tomorrow they will have a completely unsupportable control system. This didn’t seem to bother them.
In the long run these ideas of an expensive single source of supply and support simplify the management side of the site, but leave the business depending upon that control system with no options if something ever goes wrong. It’s short-sighted. It’s dangerous. In the end, it will catch up to the customer and they will lose.