Renyi’s parking problem (or 1D sequential interval packing problem) dates back to 1958, when Renyi studied the following random process:
Consider an interval $I$ of length $x$, and sequentially and randomly pack disjoint unit intervals in $I$ until the remaining space prevents placing any new segment. The expected value of the measure of the covered part of $I$ is $M(x)$, so that the ratio $M(x)/x$ is the expected filling density of the random process. [1]
Renyi himself proves the following continuous recursion for $M(x)$:
and from this he deduces the asymptotic mean filling density
This number is known as Renyi’s Parking Constant.
The first “discretized” version of the problem, namely the expected density derived from sequential packings of non-overlapping neighboring pairs of integer points, i.e., edges or bonds, selected at random on a long segment of a 1D lattice was first given by Page. And [1] gives a better discrete approximation of the above problem.
Similarly but not trivially, we can also obtain an discrete approximation for the case when we have blocks of different types, a.k.a. $A_1$ of length $k_1$ and $A_1$ of length $k_2$. Plus, an elementary C++ code for simulation is given as the following:
1 |
|