X Close

UCLse Blog


Thoughts from the staff of the UCL Centre for Systems Engineering


Systems thinking

By Michael Emes, on 19 February 2014

The term ‘system’ crops up all over the place. But what does it mean?

I like to think of it as a way of looking at something – a lens you can choose to look through when you are thinking about an object or a process in a particular way.

A bicycle is the same collection of metal parts (usually with a few bits of plastic, and if you are really lucky, carbon fibre) whether you call it a bike or a personal transportation system. But when we choose to use the term system, it implies that it’s no longer just a homogeneous object that is the same throughout; it’s an object made up of multiple parts that are connected in a way that means that the whole thing has emergent properties or behaviour that are not exhibited by the parts in isolation. For the bicycle, the parts of wheels, frame, handlebars, brakes, pedals, etc. have little value in isolation. But when put together according to a particular design, my bicycle ‘system’ has the emergent property of enabling me to propel myself in an incredibly efficient way.

In making the choice of how we see the system, we also need to specify a system boundary, which defines the scope of the system. This may seem trivial, but it’s particularly important to make sure we are clear about which parts we will develop or optimise, and what we are assuming about external interfaces. An interface challenge for the Airbus A380 ‘superjumbo’ aircraft is that it is simply too big to be accommodated in most airports.

There is no rule to say whether something is a system or not – it depends on your point of view at the time. When I’m talking to my kids about how they are going to get to school – scooter or bike – I don’t think in terms of systems. But when I am thinking about how to repair the bike or make it work better, the system lens becomes useful.

At UCL Centre for Systems Engineering, we apply systems thinking to understand how to improve the design of spacecraft instrumentation, asking questions like ‘what are the difficult interfaces’, ‘how much can we compromise on detector resolution to save a bit of mass’ and ‘how might the system fail’? But it’s not just technology that benefits from systems thinking – we also apply the systems view to soft systems in which people and processes are the major focus of attention. For example, we have recently been undertaking a study with a local hospital to understand the tensions between keeping frail elderly patients in the hospital and seeking an early discharge with support in the community.

Next time you use the term ‘system’, challenge yourself: what is happening inside the system? What are the component parts (or subsystems)? Where are the tricky interfaces? How does it change over time? What higher level ‘systems’ does it work within? When you start to think in these terms, you really are ‘systems thinking’.

Leave a Reply