TY - JOUR
T1 - Quantifying the risk of project delays with a genetic algorithm
AU - Pfeifer, Jeremy
AU - Barker, Kash
AU - Ramirez-Marquez, Jose E.
AU - Morshedlou, Nazanin
N1 - Publisher Copyright:
© 2015 Elsevier B.V. All rights reserved.
PY - 2015
Y1 - 2015
N2 - Of interest in project management is the (i) quantification of the risk associated with project performance and the (ii) identification of the project tasks that contribute most to that risk. Risk in this work addresses delays in project completion. The tasks and precendences are represented with nodes and links, respectively, in a project network whose tasks (i) have stochastic completion times that (ii) are subject to disruptions. An optimization problem is developed to maximize project delay subject to particular stochastic task disruptions, and a genetic algorithm is introduce to identify the critical tasks which lead to the maximum risk of project delay. A small project of 40 tasks and large project of 800 tasks are analyzed. Primary conclusions are (i) that critical tasks need not necessarily be on the critical path if they are subject to considerable uncertainty, and (ii) that project complexity (network topology) matters more in the performance of the algorithm than the number of tasks (network size). In fact, the genetic algorithm solution works well for large-scale projects whose schedules cannot be resolved with conventional techniques. Focus is given to the performance of the algorithm for this project risk context.
AB - Of interest in project management is the (i) quantification of the risk associated with project performance and the (ii) identification of the project tasks that contribute most to that risk. Risk in this work addresses delays in project completion. The tasks and precendences are represented with nodes and links, respectively, in a project network whose tasks (i) have stochastic completion times that (ii) are subject to disruptions. An optimization problem is developed to maximize project delay subject to particular stochastic task disruptions, and a genetic algorithm is introduce to identify the critical tasks which lead to the maximum risk of project delay. A small project of 40 tasks and large project of 800 tasks are analyzed. Primary conclusions are (i) that critical tasks need not necessarily be on the critical path if they are subject to considerable uncertainty, and (ii) that project complexity (network topology) matters more in the performance of the algorithm than the number of tasks (network size). In fact, the genetic algorithm solution works well for large-scale projects whose schedules cannot be resolved with conventional techniques. Focus is given to the performance of the algorithm for this project risk context.
KW - Genetic algorithm
KW - Project network
KW - Risk analysis
UR - http://www.scopus.com/inward/record.url?scp=84956971293&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84956971293&partnerID=8YFLogxK
U2 - 10.1016/j.ijpe.2015.09.007
DO - 10.1016/j.ijpe.2015.09.007
M3 - Article
AN - SCOPUS:84956971293
SN - 0925-5273
VL - 170
SP - 34
EP - 44
JO - International Journal of Production Economics
JF - International Journal of Production Economics
ER -