Let’s start defining what refers Disaster Recovery to.
There are many definitions for “Disaster Recovery” and all of them are closely related to the computing world (a world that is well ahead in terms of recovery strategies).
Among these definitions, we find the following:
“It’s part of a major Business Continuity plan including process and solutions in order to restore critical applications, information, hardware, communications, networks and other infrastructures associated to information systems and technologies”
“The Disaster Recovery Institute International (DRII) states that disaster recovery is the area within business continuity dealing with technology recovery instead of recovery of sales operations”
“Disaster Recovery is the security planning area commissioned to protect the company from the effects of different negative events…”
We may think that Disaster Recovery may only be applied to computing systems at the business and office level (i.e., to office automation and computing as we know them), but that’s not true. From the latter definition, which technologies can we mention in our industrial “world”? We find, among them: PLCs, HMIS, SCADAs, shifters, robots, etc.
Data acquisition and transfer technologies have improved so much during the last 20 years that now it is very unlikely a computer not being part of a manufacturing process. Besides, all control devices (PLCs, DCS, shifters) are interconnected and this is because, among other reasons, to the need to gather more data to take better decisions.
As states the sentence “you can’t control what you can’t measure”, it’s no longer a secret that corporate decisions are increasingly based on data.
How to apply these recovery strategies in the industrial environment?
Are there technologies available for this purpose? The answer is yes. Several technologies and solutions focus on guarantee the shortest unscheduled downtimes and increase our productivity (and all KPIs that will analyse later those who take decisions from board panels or dashboards).
Among solutions allowing us to recover easily in case of disaster (i.e., to adopt Disaster Recovery strategies) we can mention the following:
- Redundant communication networks.
- SCADA software with application redundancy options at the application level, historical and display.
- Software to create backups and version histories of the industrial devices programs (PLCs, SCADAs, HMIs), so in case i.e. a PLC fails, we can recover the latest well known copy since it is stored on a server. Thanks to this, we only have to connect the new hardware and download the program.
- Hardware and software solutions for our application servers to be fault This will allow your systems to work even after a hardware failure.
- Display terminals based on“zero clients” that will allow you, i.e., to replace an operational terminal in less than 5 minutes.
This solutions are focused on simplifying operations at plant level and do not require advanced computing skills.
If any of the following situations are familiar to you, I think it’s a good moment to evaluate a strategy of what I would define as Industrial Disaster Recovery:
- I don’t know where is the latest copy of the main PLCs program. In any of them fails, I don’t know how long it would take the system to recover.
- A networked PC has been infected with a virusspread across the plant and we had to stop production. Recovery was not easy at all.
- The server hosting the traceability database had a hard disk failure. We lost very relevant information during all the time needed for recovery.
I could mention more scenarios, but I’m sure that at this very moment you are thinking about many more…