The paper outlines a method for finding a trace of events capable of completing
a flexible manufacturing system (FMS) task in minimal, although not necessarily
optimal time. The method proposed is an estimation of distribution algorithm (EDA).
In addition the paper gives experimental results for there algorithm which
suggests that there method can outperform other similar algorithms in many
situations, sometimes markedly so. I am not familiar enough with the area to
evaluate how generally the method developed can be applied, or whether the
method only has application to a very constrain instance of problems.

Issues:
A full page is dedicated to the longest common substring problem. The longest common substring problem
is both old and relatively well known. In addition the solution which is shown was
developed in the seventies, and for those who are interested in learning the
details of how to calculate the LCS there are numerous easily accessible
resources available which will explain how it is done. That is to say there is
no reason why 1/6 of the space available for this paper should be devoted to
explaining the LCS problem. That space should be used to describe the actual
contributions made in this paper. In particular certain aspects of the algorithm
are ill defined. For example the paper states that if an invalid trace is
generated the algorithm "Amend"s it so that it becomes a valid one, but it is not
described how this is done. I'm assuming that there is some standard way of
doing this for Flexible Manufacturing System petri net models. But no reference
is made to when "Amend" is first mentioned.
Similarly the local heuristic search aspect of the algorithm is only very broadly
described. Finally very little space is given to describing how E is calculated
this is in spite of the fact that E is what makes the algorithm an EDA rather
than a genetic algorithm which uses LCS in it breeding step.
Again the page spent describing LCS could have been far better spent giving more
details to the actual contributions of the paper, while giving a brief
description of LCS with references. 