Keywords: sensitivity, schedule risk analysis, Monte Carlo, construction, CPM.

Usually, when running a project, we aim to equip managers with the computer ability to discriminate between good and poor practice so they are able to run the best management processes to achieve the project goals; to achieve this we are concerned with duration sensitivity of a variety of different activities. As a result of the conservative approach most often taken in the construction industry, we usually think that the activities most likely to delay the project are those based on CPM or CPA, with a large amount of uncertainty. Based on the proposed construction feasibility management case study, and using computer abilities, this paper aims to establish a strong positive correlation between the duration of a given task and the project duration, and illustrates the importance of using such computer based analysis within the construction industry.

Schedule analysis is a generic term used for several project planning methods, of which two of the most popular ones are Critical Path Method (CPM) and Program Evaluation and Review Technique (PERT). Based on deterministic durations in CPM, the critical path is the longest duration path through the project. Any delay in this path will postpone the PPC (Project Practical Completion) accordingly [1]. With PERT's ability to deal with uncertainty in durations three time estimates are required for each task, these are: Optimistic, Likely and Pessimistic durations. Following this, with the assumption of a Beta probability distribution for activity durations and a Normal probability distribution for total schedule duration, the expected time for each activity can be estimated.

Although the construction industry is characterized by conservative thinking, CPM and PERT are being using widely for scheduling and project managing; however due to some limitations of these scheduling methods, simulation should be applied to the schedule program using a quantitative risk analysis method, which considers random activity durations for estimating PPC [2,3].

For a computer based analytical method, Monte Carlo simulation is an efficient technique for analyzing these types of problems. This simulation examines more than one critical path and creates an opportunity for sensitivity analysis. Also, schedule logic can include branches which are probabilistic and conditional cases.

After Schedule Risk Analysis, this paper tries to highlight the benefits of risk assessment and the pitfalls of relying on a CPM analysis alone in the proposed case study. Duration sensitivity risk analysis ranks the variables in turn, according to their contributions to the forecast. These results can lead to some dramatic changes and redirecting of the program. Occasionally this sensitivity analysis may show that the technical development is being directed in the wrong direction.

Sensitivity analysis provides a computer tool for analyzing a schedule network by combining CPM, PERT and DES (Discrete Event Simulation) scheduling methods into one system. CPM uses a deterministic approach, PERT uses a probabilistic approach and DES uses a discrete event simulation approach for scheduling construction projects [4].

In this case study, the highest duration sensitivity rank belongs to "Prepare Feasibility RFT" with 75%. 'Develop Feasibility PDP' and 'Site Inspection' have the next levels with 51% and 32%, respectively. By applying duration sensitivity analysis to this construction feasibility management project, we can concentrate on those inputs most likely to improve quality and accuracy . The process identifies the most promising opportunities, where additional work will help to correct the input range by using computer based risk analysis.

For this project the mean PPC date was 06/Dec/04. We reached 140 max hits when the Std Deviation was 11.56 and minimum and maximum PPC dates were 02/Nov/04 and 11/Jan/05 respectively. In our simulation, we used a Latin Hypercube sampling and the results converged in 200 iterations.

This paper shows that computer based risk analysis can support construction scheduling by providing a more reliable estimation for PPC and the preparation of duration sensitivity analysis. Following the Duration Sensitivity Analysis, the process should be repeated with more accurate estimates for activities with regard to those activities with have the highest duration sensitivity.

1

D. Lock, "Project Management in Construction", Ashgate Publication Co., Burlington, VT, USA, 2004.

2

N.T.H. Ahuja, V. Nandakumar, "Simulation Model to Forecast Project Completion Time", Journal of Construction Engineering and Management, ASCE, 111(4), 325-342, 1985. doi:10.1061/(ASCE)0733-9364(1985)111:4(325)

3

E. Douglas, E. David, "PERT and Simulation", Proceedings of the 1978 winter simulation conference, Harold J. Highlight, IEEE Press, Piscataway, N.J, 89-98, 1978.

4

D.E. Lee, J.J. Shi, "Statistical Analyses for Simulation Schedule Networks", Proceedings of the 2004 Winter Simulation Conference, Construction Engineering and Management Program, Department of Civil and Architectural Engineering, Illinois Institute of Technology, USA, 2004.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £135 +P&P)