Forecasting the Project Duration Average and Standard Deviation from Deterministic Schedule Information

Abstract

Most construction managers use deterministic scheduling techniques to plan construction projects and estimate their duration. However, deterministic techniques are known to underestimate the project duration. Alternative methods, such as Stochastic Network Analysis, have rarely been adopted in practical contexts as they are commonly computer-intensive, require extensive historical information, have limited contextual/local validity and/or require skills most practitioners have not been trained for. In this paper, we propose some mathematical expressions to approximate the average and the standard deviation of a project duration from basic deterministic schedule information. The expressions’ performance is successfully tested in a 4100-network dataset with varied activity durations and activity durations variability. Calculations are quite straightforward and can be implemented manually. Furthermore, unlike the Project Evaluation and Review Technique (PERT), they allow drawing inferences about the probability of project duration in the presence of several critical and subcritical paths with minimal additional calculation.

Integration of Cost and Work Breakdown Structures in the Management of Construction Projects

“…Delivering projects on time, within budget and with required quality is one of the most difficult and challenging goals for project managers [16,17]. In the construction industry, an incomplete scope definition in the early stages of the project life cycle may be a source for future problems [18].…”

“…Tracking is necessary at all project execution stages [17]. However, the approval and implementation of changes at the planning stage of the project usually have a lower cost impact [38].…”

Stability and accuracy of deterministic project duration forecasting methods in earned value management

“…More rounded values (e.g., SP = 10%, 20%, 30%, …) were not possible due to the fixed amount of activities within each network (30). Also, we did not consider perfectly serial (SP = 100%) or parallel (SP = 0) projects, as those configurations are not representative of real construction projects [see some average SP values of building, civil engineering, industrial and services projects in Ballesteros-Pérez et al (2020a, b), Table 9].…”

“…Apart from EVM project duration forecasting methods, other more advanced techniques have also been proposed (e.g., fuzzy logic, neural network analysis, Bayesian inference, Monte Carlo simulation, statistical learning and artificial intelligence methods, Kalman filter algorithms, and endless variants of PERT) (Bai et al , 2020; Ballesteros-Pérez et al , 2020a). However, most of these techniques are also more computer demanding and/or data-intensive than the EVM deterministic expressions analyzed here.…”

“…The Normal distribution of d i in equation (8) becomes Lognormal when exponentiated. The Uniform distribution in equation (10) acts as the Coefficient of Variation ( CV i = σ i / μ i ) whose values are set to resemble those of real construction projects as measured by Ballesteros-Pérez et al (2020a, b). Activity costs ( c i ) are also randomly generated for each project and activity i , but now introducing a duration-cost correlation coefficient ( δ i ):where most variables are analogous to previous equations (8–10), but now some of them with an apostrophe (') indicating that they refer to costs instead of durations.…”