Estimation of Construction Activity Duration Under Uncertainty Using Discrete Fuzzy Weighted Average Algorithm

Estimation of Construction Activity Duration Under Uncertainty Using Discrete Fuzzy Weighted Average Algorithm

Pejman Rezakhani, Kasim A. Korkmaz
Copyright: © 2022 |Pages: 19
DOI: 10.4018/IJPMPA.301598
OnDemand:
(Individual Articles)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

This paper presents a qualitative risk assessment tool based on fuzzy set theory to estimate the amount of activity duration overrun under the combinatory effect of multiple risk factors. Applying proposed methodology, a set of duration modifiers to calculate the optimistic, most-likely and pessimistic duration values under uncertainty are calculated. To elaborate the methodology, a simulated bridge project with ten risk factors affecting activities duration is presented. Proposed model contributes to knowledge that could help minimize the schedule overrun and improve risk mitigation strategies through providing risk-based duration estimates instead of discrete values. Advantages of the developed approach compared to existing models include (1) considering the combinatory effect of multiple risk factors on activity duration, (2) accounting for uncertainty in experts’ evaluations by employing the interval-valued fuzzy numbers, and (3) utilizing a discrete fuzzy weighted average algorithm which avoids creation of incorrect fuzzy membership functions.
Article Preview
Top

Literature Review

Existing approaches in applying fuzzy set theory to estimate the activity duration may be categorized as:

  • 1)

    Assessing the effects of risks on activity durations;

  • 2)

    Modeling activity durations under uncertainty combining fuzzy set theory with probabilistic and possibilistic means;

  • 3)

    Combining the crisp and fuzzy numbers to generate duration distributions under uncertainty.

Assessing the Effects of Risks on Activity Durations

(Rezakhani & Maghiar, 2019) developed a risk assessment tool based on Karnik-Mendel fuzzy analytical solution to estimate the activity duration overrun under combinatory effect of multiple risk factors. (Pawan & Lorterapong, 2015) proposed a fuzzy set theory-based framework to assess the required time contingencies for reducing the impacts of risks on scheduling. In this approach first, scheduler estimates the extension time needed to complete an activity in case a specific risk occurs using imprecise linguistic expressions, such as “Approximately 5 to 8 days”. Next, these linguistic evaluations are adjusted based on possibility of occurrence of each risk. Finally, the maximum of adjusted fuzzy time extensions for each activity is selected and added to normal duration to update the schedule. The major drawback of this approach is assumption of independency of risk factors. (Salah & Moselhi, 2015) also used fuzzy set theory to model and allocate contingency in project. In their proposed methodology, experts express their evaluations of total duration and needed contingency for each activity in numeric fuzzy numbers. Project contingency is then calculated by de-fuzzifying of these values. Not assigning weights to experts’ estimations and ambiguity in applied averaging method which may cause incorrect membership functions are two major shortcomings of this approach. (Ock & Han, 2010) presented a quantification method based on fuzzy set theory to estimate activity duration under uncertainty. In their approach, each activity duration is calculated by summing the normal duration and risk-associated duration. Fuzzy membership functions to assess risk-associated duration of each task are generated based on experts’ subjective duration estimates. A simple de-fuzzification method based on area of membership function is used to convert fuzzy risk-associated duration into a crisp value.

Complete Article List

Search this Journal:
Reset
Volume 11: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 10: 2 Issues (2022): 1 Released, 1 Forthcoming
Volume 9: 2 Issues (2021)
Volume 8: 2 Issues (2020)
View Complete Journal Contents Listing