Solving Complexity with Networks

I am fed up with the excuse that a problem is too complex, therefore it should not be bothered…

Ben Ferrum
2 min readJan 18, 2021

I hear this all the time from software engineering to the global economy…

What if this is not true, what if we use the right tools we could get insights and solve our “complex” problems?

For example, take the usual blanket problem of “communication”: there is a complex communication problem between 3 teams/departments.

What if we model the communication network and the information flow between the individuals in an example, keeping in mind the biggest symptoms: the process of buying a new widget is slow.

Map of current process/comm flow:

The old process of getting widgets

As we can see Bob has to go to Cecil, then Cecil has to go to David, then David asks Abigail to buy and bring the widget to Bob. This happens frequently and takes up a lot of time and follow up in a long chain.


New improved solution

After we draw up the communication and process network we realize there are too many people in this frequent chain, and there is no point of doing this every time, it is enough to check in monthly under a budget and allow the field operators to work with authority.

This is a silly simple example, however the more complex the situation, the better you could utilise your pattern recognition machine (your brain) to spot improvement opportunities.

What other problem could you model and solve with networks?