Traditional Critical Path Method (CPM) and its different generalizations have become
the prevailing technique in managing complex projects over recent decades. Among them,
Precedence Diagramming Method (PDM), with its four precedence relationships, has become
the most popular due to its increased flexibility in modeling complex project logic.
Despite the four well-known precedence relationships, CPM/PDM still suffers from modeling
shortcomings. Developments such as conditional logic, maximal relationships, point-to-point
relationships, continuous relationships, and non-linear activities attempt to remedy
these problems. However, these improvements require more complex and sometimes iterative
time analysis. Monte Carlo (MC) simulation, which calculates the effects of risks
on the project duration, also requires numerous repetitions of the deterministic time
analysis. It is especially true if the network contains the developments mentioned
above. In these cases, the time analysis speed is of great importance. The goal of
this study is a) to implement different algorithms for time analysis that perform
better than the traditional time analysis, b) to compare their speed on large-scale
artificially created projects, c) to make suggestions regarding which algorithms are
suitable in different cases d) to develop a database containing artificially created
networks to serve as a basis for benchmarking for researchers and developers working
in the field of project planning. Comparisons were based on networks with at least
1000 activities. Results show that, depending on the sample network's structure, different
algorithms perform well. Therefore, different time analysis algorithms are necessary
to implement into the scheduling tools, and they can decide which algorithms perform
better.
