In sharp contrast, the same analysis performed with = 10 shows a great deal of organisation (see Figure 3).
Figure 3: Charts for the organized regime (30 agents visiting 3 shops, when the learning parameter and ). All charts and notation are the same as for figure 2, except for the scale of the order parameter plot ( y). a Starting from all equal to 1, all representative circles move to the triangle corners representing the preferred shops. b y, the order parameter, varies from 0.33 (equal interest for all shops) to nearly 1 (strong preference for only one shop). c Due to organization, fluctuations of performance attenuate in time.
The order parameter, y, steadily increases to 1 in 200 time steps. As seen on the simplex plot at time 50, each customer has built-up fidelity to one shop. Performance of shop number one also stabilizes in time, and variations from stationarity are not observed after 20 time steps.
The daily profit of buyers averaged over all buyers and over 100 days after a transition period of 100 days, is 0.0572, exactly the average profit per transaction for the buyer. Because buyers have not changed shops during the last 100 days, sellers learned to purchase the exact exact quantity needed to satisfy all their buyers, and they had no losses themselves: their daily average profit of sellers is 0.877.
By avoiding daily fluctuations in the number of customers visiting a shop, the ordered regime is beneficial to both customers and sellers, that is both obtain higher profits than in the disorganised situation.