Acta Scientiarum Polonorum

The simulation methods are one of the tools used to forecast the duration of a construction project. In order to
carry them out, a distribution of probability of the duration of construction works should be assigned to each
construction work. The disadvantage of such a procedure is that distributions are received in a very subjective
way. What is more, a shape of distribution of probability of the duration of construction works does not
show on the basis of which premises was established. It is quite problematic, because it is obvious that the
construction works are affected by risk factors whose impact after the shape of the distribution cannot be
seen. What’s more, each factor has parameters such as incidence, impact on the course of a given construction
work and impact on other risk factors. In order to take into account the impact of risk factors on the shape
of the probability distribution, it is possible to use fuzzy inference, which allows to obtain a distribution of
probability of the duration of construction works depending on the parameters of particular risk factors for
specific works. It will be in a discrete form. In order to obtain the best possible simulation result, a continuous
distribution should be obtained on the basis of a discrete distribution. This publication describes and shows
advantages and disadvantages of the point estimate, interval estimation and non-parametric estimation that
are used to solve the problem.