Estimation of railway projects duration at initiating phase: a statistic approach

In Large Transportation Projects (LTPs) estimation of duration provided in tender and initiating phase are often contradicted in executing phase, due to a mix of technical and contractual problems arising during the project execution. This paper present a research that apply a statistic approach to this problem and, determining the key features of a LTP and a set of hi storic data coming from the railway field, present a quantitative solution. The first part of the paper makes an analysis of existing literature on this topic, then the methodology is discussed. In the second part, a statistical model is presented, to show how the set of identified relevant variables are used to estimate the duration according to project features avoiding other goals such as political goals of cutting the project duration. In fact, the common sense of technical advisers is that the duration of a project is often given by non -technical goals, but the drivers are more referred to a specific deadline such as elections or large public events (World Cups, Universal Exposition etc.), asking to cut and cut time and, as a result, projects become more expensive, hard to manage, and face technical problems in the operating phase. In the third part, this method is calibrated on a set of input coming from real railway LTPs of the last ten years. The aim of this paper is to let the decision makers know that, given a new project, the past experience says that it will have a given duration according to its features, and to ask for a shorter project duration would be risky and cause project failure, or let the project cost rise. It also allow to define a guideline to determine which contractual form is the best according to t he project.