This chapter felt different from all the prior ones, so this entry will probably do too. Mr. Tompkins was staying in Rome for a few days and Johnny Jay, his old boss, wanted him to meet with somebody that also happened to be in Europe at the time. Since Mr. T respected Mr. Jay so much, he agreed to meet the guy.
Dr. Abdul Jamid had been working on management dynamics. Specifically, on a program that could translate the gut feeling that has been mentioned several times throughout the book into a controlled model that would be able to simulate different scenarios in a similar way to an algorithm. Mr. T was unsure about the viability of such a program, but after a couple of demonstrations and about two days of exchanging thoughts with Dr. Jamid he was convinced and acquired the software.
The book states that the software used by Dr. Jamis is a visual programming language that was marketed as ‘iThink’. Its actual name is a little longer: Systems Thinking, Experimental Learning Laboratory with Animation (or STELLA for short). It was introduced way back in 1985 as a program for system dynamics modelling.
I had never heard the term ‘system dynamics’, so after doing a little research I found that that’s the name given to certain approach that allows to model and simulate the behavior of complex systems by using several components, including loops, time delays and table functions. This small website explains that one of the main applications of system dynamics in business is, precisely, one of the topics from my last entry: improving performance over time. It also explains that when system dynamics is translated into the business/management context it is sometimes called ‘strategy dynamics’.
iThink includes four main building blocks that create the basic structure of the models. These are:
- Flows: Measures change over time.
- Stocks: Model simple inventories or process delays.
- Connectors: Provide information flows.
- Converters: Apply gain factor or other formulae to variables.
This kind of programs seem to be very useful and really complicated, but I bet it’s just a matter of getting used to it to understand what’s actually happening and learn something from the simulations.
This time, Mr. Tompkins’ journal got very few new entries. They all basically say that you should model your hunches, run simulations with it and tune it to get more accurate results. Nonetheless, it was a very interesting chapter.