A scheduling methodology for dealing with uncertainty in construction projects

Purpose - The purpose of this paper is to address the problem of scheduling under uncertainty in construction projects. The existing methods for determining a project schedule are based on assumption of complete knowledge of project parameters; but in reality there is uncertainty in construction projects, deriving from a multitude of context-dependent sources and often provided as outcome of a risk analysis process. Thus, classical deterministic analysis might provide a schedule which is not sufficiently protected against possible disruptions. Design/methodology/approach - A quantitative methodology is developed for planning construction projects under uncertainty aimed at determining a reliable resource feasible project schedule by taking into account the available probabilistic information to produce solutions that are less sensitive to perturbations that occur on line. The methodology relies on a computer-supported system that allows to identify, analyze and quantify the schedule reliability and the impact of possible disruptions on the duration of the project. Findings - It is found that the proposed methodology can exploit more information about the uncertain parameters than the commonly-used deterministic method, and it provides an improved understanding of the schedule reliability in presence of uncertainty. The schedule generated with a classical deterministic method sets a completely unrealistic planned project delivery date of about 1,250 days, with a probability around 50 per cent to be exceeded. This behavior can be very unsatisfactory for construction projects for which high penalties are usually associated to heavy due date violations. Originality/value - This paper presents an approach for robust scheduling of construction project problem under uncertainty. We provide a tool able to support managers in developing a workable and realistic project schedule to be used as a guideline for project control and monitoring.