10 Engineering Tools for Thought
Becoming an engineer is tough! You have to study a lot of stuff ... classes on fluids, structures, thermodynamics, and so on. I want to save you all that effort, so here are the 10 key tricks that engineers use when faced with complex problems (which means pretty much any real-world problem).
1. Estimate. Engineers approximate, recognising that they have to rely upon incomplete or uncertain information. A precise answer is not required, usually the correct course of action depends on knowing the rough magnitude of certain key variables.
2. Use the 80:20 rule. The Pareto Principle that states that "for many events, roughly 80% of the effects come from 20% of the causes". Thus, for example in a machine, 20% of the components might typically give rise to 80% of the failures. Engineers learn that by concentrating on a relatively small number of the right issues they can secure large practical improvements.
3. Watch out for feedback loops. Systems that exhibit self-regulation have feedback, where the output of the system is fed back to control the performance of the system. Systems with feedback can be difficult to analyse and can give rise to unpredictable dynamic behaviours. Engineers understand that systems with feedback need careful modelling and tuning.
4. Use dimensional analysis. Most engineers recognise it is all too easy to get calculations wrong. Checking the fundamental dimensions (mass, length, time etc.) and the measurement units (which may combine different fundamental dimensions) provides a rough working understanding of the science that underpins the calculation and is an essential way of ensuring that an engineering calculation makes sense. It often does not.
5. Budget and trade-off. Engineers work in the context of scarce resources. You cannot get all you want, so you have to trade-off. This can be difficult but one powerful aid is to create a budget with a single commensurable basis. Thus, for instance, engineers might create a weight budget for a system and trade-off cost, performance and ease of manufacturing using the weight as a surrogate.
6. Reuse or repurpose existing solutions. Engineers try to avoid novel solutions wherever possible. They are always risky and require extensive analyses. It makes much more sense to find an existing solution and reuse, adapt or repurpose it if at all possible.
7. Think about the life of the system. When most people set up a new system they do so based on the immediate merits of that system. Engineers are trained to think about system life including maintenance, enhancement, decommissioning and disposal.
8. Identify boundaries and specify them. Engineers learn that it is very important to be clear on the boundaries of a system. It is essential to distinguish 'the machine' that is being designed and can be precisely specified, and what is assumed about 'the world' with which the machine interacts but cannot be directly controlled. This is an essential step in all engineering problem solving.
9. Test and test again. Engineers are taught rigorous formal analysis but they rarely fully trust it. They always test and they know that whenever a change is made they have to repeat all their tests. Ideally an engineer works out how to test a system before designing and building it.
10. Think about margins. Recognising the fact that the environment in which a system operates is liable to change, engineers build in margins. These margins are not arbitrary contingencies, that would be costly, but rather calculated appreciations of the uncertainties to which the system is exposed.