
Ant-Inspired Allocation
■
263
represent the number of the agents at each site. The larger the circle, the higher the
agent population.
For the first set of simulations, the transition rate matrix K was set to K
u
, K
∗
,and
K
∗
(x
0
). The agents switch between sites with controllers derived from (10.5), i.e.,
no quorum activation is used. In all of these simulations, the equilibrium initiation
rate for a transition from site i to site j was set to 1. This is equivalent to bounding
the total number of idle trips at equilibrium. Also, k
max
= 12 in all simulations.
Figure 10.5 shows the fraction of misplaced agents over time for the three
different choices of K.