Resource-constrained Project Evaluation and Review Technique (PERT) : stochastic simulation and optimization

Abstract

In the construction industry, the classic Critical Path Method (CPM) and Project Evaluation and Review Technique (PERT) have been widely applied in modern project scheduling management systems since 1950s. The major limitation in conventional critical path analysis in handling practical construction project planning is identified as the potential confusions and misleading information associated with the constraints of resource availability and activity interruptions (such as multiple activity calendars, individual resource calendars, random resource breakdown or routine resource maintenance). In consequence, the determination of correct float values by using the industry norm software such as Primavera Project Planner (P3) and Microsoft Project is not guaranteed. The purpose of the research is to (1) develop a simulation tool for construction scheduling, named the Simplified Simulation-based Scheduling (short as S3), (2) apply an evolutionary optimization for resource-constrained construction scheduling analysis, (3) enable project cost analysis based on resource-constrained schedules and expose the implicit relationships between resources, project duration and cost, and (4) compare the functionality and special features of S3 andP3 in connection with the resource-constrained project scheduling. S3 allows resource, time and cost-integrated project optimization analysis under both deterministic and uncertain scenarios, lending itself well to planning construction projects featuring complicated resource and technology constraints. In this research, a classic textbook example serves to illustrate the limitations of the popular P3, together with the application of S3 developed from in-house research. In an effort to address ,P3's shortfalls, a new total float (TF) determination algorithm is introduced to check the validity of TF and derive the accurate TF for those activities whose TF are overstated by P3. This thesis also presents the Particle Swarm Optimization (PSO) algorithm for the optimization of resource-constrained scheduling problem and discusses the experimental results on case studies. S3 is capable of rapidly formulating the best project plan with respect of achieving the shortest duration or the least cost. Finally, a real-world construction project is utilized to demonstrate the application of S3 in practical settings. In conclusion, the methodology and prototype software of S3 provide a complete and valid alternative solution to CPM-based project scheduling, subject to the constraints of resource limits and activity interruptions.