Interlocks provide a means of coordinating the function of different components so that task steps have to be performed in a specified sequence or certain conditions have to be met before a task can proceed. Valves used to create process isolations can be interlocked so that it is physically impossible to manoeuvre them in an incorrect sequence. This is often seen as a method of eliminating the potential for human error.


Advances in technology have allowed more extensive and complex interlocks to be used, which on the face of it, appears to provide the opportunity to make isolations safer than ever before. However, interlocks do not actually eliminate errors; and complexity can be a source of risk. In fact, when all factors are considered there may be an argument to say ‘less is more.’


Whilst there is some guidance available about when interlocks can or should be used; there is very little to say which or how many components should be interlocked. This leaves designers with a dilemma. Do they attempt to apply a ‘sensible’ approach, which may leave them open to criticism because their design is not totally ‘error proof?’ Or do they go to an interlock vendor and ask them to interlock everything,?


One of the problems is that the reason for using interlocks is not always clearly understood or defined. Are they provided to:
• Ensure a ‘spared’ item (e.g. relief valve, filter) remains available at all times and is not interrupted when changing over duty/standby?
• Ensure isolation valves are in the correct position before carrying out a task?
• Ensure the item has been fully isolated and prepared for the task by ensuring valves are manoeuvred in a defined sequence and secured in the correct position;
• All of the above?

Extensive and complex interlock systems are expensive to purchase, install and maintain. They are often only effective for performing one task, and so cause significant problems when other activities have to be performed or if a problem occurs (e.g. valve passes or pipework is blocked). Also, they can create a false sense of security that introduces human factors risks. This paper will discuss these issues using real life examples and suggest that less really can be more.

 

Presented at Hazards 2017