[ Main page | About Your System | Knowledge Bases | Rule Traces | Demand Modelling | About Fuzzy Logic | The Importance of Explainability | Commercialising ]
This is the main page for the Dynamic Pricing expert system and the simulation environment that demonstrates it. The expert system "lives" in a simple simulated environment where a traveller wants to fly from Oxford Kidlington in England to Alpenstadt, a German city much like Munich. Alpenstadt hosts numerous business fairs, concerts, and other events, including Alpenfest, a large beer festival not unlike Munich's Oktoberfest. These, plus the annual skiing season and the Christmas and Easter holidays, have a dramatic effect on demand for airline seats.
The expert system has access to several knowledge bases: that is, sets of rules which enable it to react to input variables such as airline capacity utilisation, time to departure when booking, and market segment. It can do this directly, or indirectly by using demand as a proxy.
The system is not a commercial product, but a Minimum Viable Prototype — an MVP. However, the techniques I used for it, namely fuzzy logic and the software that implements and interprets it, can be used in a full product. For this reason, I have taken some considerable care in demonstrating them, by making the simulation environment clear and easy to understand, and by showing how the rules operate within it. For instructions on how to run it, please go to About Your System. The knowledge bases that you can call upon are explained in Knowledge Bases.
You may want to check that the rules are working as you think they should, or work through a session with someone else. To this end, the system generates a trace every time you invoke a knowledge base. For an explanation, please see Rule Traces.
As hinted above, a knowledge base can react directly to data-specific variables such as seasonal weather, or indirectly by being passed the demand for seats as a proxy. The simulation has one knowledge base that does the latter, modulating it by travellers' health concerns and the current popularity of the events being held in Alpenstadt. Where it gets the demand curves from is explained in Demand Modelling.
Fuzzy logic rules can be understood at an intuitive level by reading them as though they were English. This indeed is a key selling point, meaning that while coding, you can easily share knowledge with people who are domain experts but not expert-system experts. However, there is a rigorous mathematics behind fuzzy logic, dating back to its invention in the 1960s. For more on this, please go to About Fuzzy Logic. This links to some non-technical background info for non-experts, a tutorial on fuzzy logic from two computer scientists who use mapping people's heights to shoe sizes as an example, an online demonstration of this written by me, and another tutorial also written by me. There is also a link to a LinkedIn tutorial on fuzzy logic for inventory optimisation, which I shall repeat here. By Bipin Reghunathan, it's an easy read, and includes a list of rules for optimising from demand variability and other attributes.
As mentioned above, ease of communication with colleagues is an important benefit of fuzzy logic. More generally, programs that use fuzzy logic — and other kinds of logic — are much better at justifying their advice than are other techniques such as neural networks and numerical optimisation. I regard this as very important, and as worth emphasising when you demonstrate to potential investors. Other AI experts back me up, most notably Donald Michie, "the Father of British AI". For his views, please go to The Importance of Explainability. You might want to keep this available to show to others.
Finally, you will want to know what steps to take in order to move from this MVP to a full product. I explain this at Commercialising.